NI TECHNICAL REPORT ON THE PRE-FEASIBILITY STUDY OF THE KÉMAG IRON ORE PROJECT, QUEBEC FOR NEW MILLENNIUM CAPITAL CORP

Transcription

1 NI TECHNICAL REPORT ON THE PRE-FEASIBILITY STUDY OF THE KÉMAG IRON ORE PROJECT, QUEBEC FOR NEW MILLENNIUM CAPITAL CORP Prepared by Andre Allaire, Eng, M.Eng., PhD. John Dinsdale, Eng. Langis Charron, Eng. March 2, 2009 Montreal, Canada

2 Technical Report on the KéMag Pre-Feasibility Study Page ii TABLE OF CONTENTS 1. SUMMARY GENERAL HISTORY GEOLOGY DEPOSIT TYPE AND MINERALIZATION NML EXPLORATION AND DRILLING DATA CORROBORATION ADJACENT PROPERTIES METALLURGICAL TESTING MINERAL PROCESSING CONCENTRATOR PIPE LINE POINTE-NOIRE FACILITIES MINERAL RESOURCE ESTIMATE MINE DESIGN AND MINERAL RESERVE SCHEDULE INFRASTRUCTURE AND SUPPORT SYSTEMS MINE AND CONCENTRATOR SITE PIPELINE POINTE-NOIRE AND SEPT-ILES ENVIRONMENTAL AND PERMITTING FINANCIAL ANALYSIS CONCLUSIONS RECOMMENDATIONS INTRODUCTION AND TERMS OF REFERENCE INTRODUCTION TERMS OF REFERENCE SOURCES OF INFORMATION UNITS AND CURRENCY DISCLAIMER RELIANCE ON OTHER EXPERTS PROPERTY DESCRIPTION AND LOCATION PROPERTY LOCATION DESCRIPTION...30

3 Technical Report on the KéMag Pre-Feasibility Study Page iii 4.3 LAND OWNERSHIP AND RIGHTS OF WAY PERMITTING INFRASTRUCTURE AND PHYSIOGRAPH ACCESSIBILITY CLIMATE PHYSIOGRAPHY LOCAL RESOURCES AND INFRASTRUCTURE HISTORY GENERAL HISTORICAL DRILLING GEOLOGICAL SETTING REGIONAL GEOLOGY PROPERTY GEOLOGY GENERAL STRUCTURE DEPOSIT TYPES MINERALIZATION EXPLORATION DRILLING EXPLORATION PROGRAM DRILLING PROGRAM DRILLING RESULTS DRILLING PROGRAM DRILLING RESULTS DRILLING PROGRAM DRILLING RESULTS SAMPLING METHOD AND APPROACH GENERAL CORE HANDLING PROCEDURES LOGGING AND SAMPLING PROCEDURES SAMPLE PREPARATION, ASSAYING AND SECURITY ASSAYING AND TESTWORK QUALITY ASSURANCE/QUALITY CONTROL PROGRAM NML QA/QC PROGRAM MRC QA/QC PROGRAM CHECK SAMPLING...61

4 Technical Report on the KéMag Pre-Feasibility Study Page iv 14. DATA CORROBORATION ADJACENT PROPERTIES MINERAL PROCESSING AND METALLURGICAL TESTING METALLURGICAL TESTING AND PROCESS DEVELOPMENT WORK DRILL CORE SAMPLES TESTING LIBERATION TESTWORK SMALL SCALE PILOT PLANT TESTING FLOTATION OPTIMIZATION AT PILOT PLANT SCALE PELLETIZING TESTWORK MINERAL PROCESSING, CONCENTRATOR AND PELLET PLANT CONCENTRATOR TAILINGS MANAGEMENT PIPELINE PELLET PLANT FLOTATION PLANT CONCENTRATE FILTER PLANT PRODUCT STOCKYARD AND SHIP LOADING FACILITIES PROTECTION OF THE ENVIRONMENT MINERAL RESOURCE AND MINERAL RESERVE ESTIMATES INTRODUCTION DATA RECEIVED DRILL HOLE DATA BASIC STATISTICS OF SAMPLE DATA, PER SEAM, PER CUT-OFFS GEOLOGICAL INTERPRETATION COMPOSITING BLOCK MODEL CONSTRUCTION BLOCK MODEL GEOMETRY INTERPOLATION PARAMETERS RESOURCE ESTIMATION SPECIFIC GRAVITY USED RESOURCE CLASSIFICATION CLASSIFIED RESOURCES MINE DESIGN GENERAL PIT OPTIMIZATION AND MINE DESIGN PIT OPTIMIZATION CRITERIA AND PARAMETERS...127

5 Technical Report on the KéMag Pre-Feasibility Study Page v CUT-OFF GRADE CALCULATION MINE DESIGN AND MINERAL RESERVE LIFE OF MINE RESERVES PIT DESIGN MINE PLANNING MINING METHOD ANNUAL PRODUCTION REQUIREMENT BLENDING MINING PLAN DUMP DESIGN MINE OPERATIONS AND EQUIPMENT REQUIREMENT WORK SCHEDULE DRILLING & BLASTING LOADING & HAULING OTHER RELEVANT DATA AND INFORMATION INFRASTRUCTURE AND SUPPORT SYSTEMS MINE AND CONCENTRATOR PIPELINE POINTE-NOIRE SITE ELECTRICAL POWER SUPPLY MINE AND CONCENTRATOR PIPELINE POINTE-NOIRE SITE ENVIRONMENTAL CONSIDERATIONS PROJECT ENVIRONMENTAL APPROVAL REQUIREMENTS GENERAL TIMETABLE HARMONIZATION OF ENVIRONMENTAL IMPACT REGIMES BASELINE DATA STUDIES KÉMAG BASELINE STUDIES ENVIRONMENTAL MANAGEMENT PLAN MARKETING STUDY CRUDE STEEL PRODUCTION IRON ORE DEMAND FROM INTEGRATED STEEL MILLS IRON ORE DEMAND FROM THE DIRECT REDUCTION SECTOR IRON ORE SUPPLY PELLET SUPPLY...159

6 Technical Report on the KéMag Pre-Feasibility Study Page vi PELLET DEMAND LONGER RANGE PELLET DEMAND SUPPLY-DEMAND BALANCE FREIGHT ISSUES IRON ORE PRICE DEVELOPMENT PRICING PRICE ASSUMPTION MARKETING CAPITAL COST ESTIMATE SCOPE OF ESTIMATE BASIS OF ESTIMATE OPTIMIZATION AND UPSIDE POTENTIAL: EXCLUSIONS SUMMARY OF THE ESTIMATE OPERATING COST ESTIMATE BASIS OF ESTIMATE SUMMARY OF ESTIMATED OPERATING COST OPERATING SCHEDULES MANPOWER FINANCIAL ANALYSIS GENERAL REVENUES EXPENSES CAPITAL EXPENDITURES FISCAL CONSIDERATIONS AND DEPRECIATION RESIDUAL VALUE PROJECT FINANCING FINANCIAL RESULTS SENSITIVITY ANALYSIS PROJECT MASTER SCHEDULE CONSTRUCTION INTERPRETATION AND CONCLUSIONS RECOMMENDATIONS CERTIFICATES REFERENCES...204

7 Technical Report on the KéMag Pre-Feasibility Study Page vii LIST OF TABLES Table 1: Pellet Quality...14 Table 2: Mineral Resources at 18% DTWR...16 Table 3: Proven and Probable Reserves...17 Table 4: Results of Financial Analysis...21 Table 5: Summary of Terms and Abbreviations for Units...27 Table 6: Claims Covering the KéMag Property...30 Table 7: Historical Exploration on the Property...35 Table 8: Summary of Historic Diamond Drilling...36 Table 9: Regional Stratigraphic Column...40 Table 10: Stratigraphy of the Howells River Property...42 Table 11: Deposit Model for Lake Superior Type Iron Formation...46 Table 12: Comparison of Current and Past Producing Taconite Mines...47 Table 13: Results of Outcrop Sampling in 1950 and Table 14: 2006 Drilling Summary...51 Table 15: 2007 Drilling Summary...53 Table 16: 2008 Drilling Summary...55 Table 17: Summary of Core Samples Submitted to MRC in 2006, 2007 and Table 18: Howells River Iron Deposit 2007 Mineral Resource Estimate...64 Table 19: Drill Core Sample Results...66 Table 20: Liberation and Grindability Test Results Summary for the KéMag Deposit...68 Table 21: Liberation Grinding Comparison of the Deposits...69 Table 22: Summary of Results for High Pressure Grinding, Wet Screening and...69 Table 23: Results for Ball Mill Regrinding and Cleaner Magnetic Separation Part Table 24: Results for Ball Mill Regrinding and Cleaner Magnetic Separation Part Table 25: Estimated blast furnace pellet feed...71 Table 26: Estimated Direct Reduction Pellet Feed, based on Test Table 27: Estimated Direct Reduction Pellet Feed, based on Test Table 28: Estimated Direct Reduction Pellet Feed, based on Pilot Plant Testing...73 Table 29: Expected Blast Furnace Concentrate if Produced in an Industrial Plant...74 Table 30: Expected Direct Reduction Concentrate if Produced in an Industrial Plant...74 Table 31: Summary of Concentrator Material Balance...76 Table 32: Concentrator Design Criteria...76 Table 33: Pipeline System Design Criteria...96 Table 34: Typical Concentrate Analysis...97

8 Technical Report on the KéMag Pre-Feasibility Study Page viii Table 35: Pellet Quality...97 Table 36: Pellet Plant Design Criteria...98 Table 37: Typical Chemical Composition (%)...99 Table 38: Bentonite Chemical Analysis and Physical Properties Table 39: Fuel Oil Specification Table 40: Flotation Plant Design Criteria Table 41: Concentrate Filter Plant Design Criteria Table 42: Product Stockyard and Ship Loading Design Criteria Table 43: Lithological Codes used in KéMag deposit Table 44: Basic Statistics of Sample Data per Seam per Cut-Off Table 45: Basic Statistics of 3m Composites per Seam Table 46: Block Grid Geometry Parameters Table 47: Specific Gravity used at KéMag, per Seam Table 48: Classified Mineral Resources by DTWR cut-off grade Intervals Table 49: Classified Mineral resources per seam at 18% DTWR cut-off Table 50: Pit Optimization Parameters for LG 3D Table 51: LOM Reserve for the KéMag Project Table 52: Summary of Planned 25-year Production Schedule Table 53: Estimated Operating Hours Table 54: Major Mine Equipment Requirement Table 55: Environmental Impact Assessment Regimes Applicable Table 56: Planned Baseline Studies for the KéMag Project Table 57: Projection of Pellet Production Table 58: Pellet Supply-Demand Forecast Table 59: Freight Rate Differentials Table 60: Examples of Hourly Construction Rates Table 61: Summary of Capital Cost Estimate (Can$) Table 62: Summary of Estimated Total Operating Costs for Year 3 (Can$) Table 63: Manning by Category Table 64: Estimated Sales Tonnages and Revenues by Year Table 65: Estimated Capital Expenditure (Can $) Table 66: Results of Financial Analyses Table 67: Sensitivity of Project IRR to Variations in Key Parameters...186

9 Technical Report on the KéMag Pre-Feasibility Study Page ix LIST OF FIGURES Figure 1: Sensitivity Analysis (before tax) IRR%...21 Figure 2: Project Location...29 Figure 3: Claims Map...31 Figure 4: Regional Geology...38 Figure 5: Regional Geology Map Legend...39 Figure 7: Blaine Value versus the SiO2 Content...72 Figure 8: Primary and Secondary Crushing Circuit...81 Figure 9: High Pressure Grinding...83 Figure 10: Cobber Magnetic Sepearation...85 Figure 11: Rougher Magnetic Circuit...87 Figure 12: Finisher Magnetic Seperation Circuit...89 Figure 13: Tailings Flowsheet...91 Figure 14: Proposed Route of the Slurry Pipeline from Lake Harris to Pointe-Noire...95 Figure 15: Pellet Plant Flow Diagram: Acid Pellets (All units are in dry tonnes per hour) Figure 16: Flotation Circuit Flowsheet Figure 17: Typical Stratigraphic Column of the KéMag Deposit Figure 18: Location Map of KeMag Drill Holes per year Figure 19: Geological Interpretation on Section Figure 20: Geological Interpretation on Section Figure 21: Geological Interpretation on Section Figure 22: Geological Interpretation on Section Figure 23: Lateral Extent of the KeMag Resource Model Figure 24: Block Grid Geometry and Orientation Figure 25: Outline of Resource Classification used at KeMag Figure 26: 3D View of the of the Detail Pit Design Figure 28: Global Projection of Pellet production (Million Tonnes) Figure 29: Projected Pellet Demand to 2016 (million tonnes) Figure 30: Brazilian Sinter Feed Price Forecast (cents/mtu FOB) Figure 31: Tubarão BF Pellet Premium Forecast (cents/mtu) Figure 32: Tubarão BF Pellet Price Forecast (cents/mtu) Figure 33: Tubarão DR Pellet Price Forecast (cents/mtu FOB) Figure 34: Sensitivity of IRR to Variations in Key Parameters (Pre-tax) Figure 35: Preliminary Master Schedule...192

10 Technical Report on the KéMag Pre-Feasibility Study Page SUMMARY 1.1 GENERAL New Millennium Capital Corp. ("NML") holds a 100% interest in certain mineral claims ("the Property") that contain the KéMag magnetite-rich taconite iron ore deposit at Lake Harris, Nunavik, approximately 50 km north of Schefferville in the Province of Québec, Canada. The KéMag Deposit is located 18 km north of the LabMag Deposit at Howells River owned by the LabMag Limited Partnership; which is held 80% by NML. The KéMag Property covers a total area of approximately 81 km 2 and comprises 171 map-staked claims. As far as BBA is aware, no royalties for the Property are pending and NML has completed no property agreements concerning the Property. The KéMag Deposit was explored by the Iron Ore Company of Canada ("IOCC") from 1949 until 1971 using aeromagnetic surveys, geological mapping and test drilling. NML staked claims covering the taconite deposit in 2004 and carried out reconnaissance mapping and sampling in From June through October 2006, NML carried out a limited diamond drilling campaign on the deposit, to which it then gave the name KéMag Deposit. On the basis of this drilling, Geostat Systems International Inc. ("Geostat") completed a Technical Report compliant with the requirements of Canadian National Instrument , which was filed with the Ontario Security Commission in March, At this time the estimated Resources were used to prepare a Preliminary Assessment Study ( PAS ). Subsequently, following further drilling in 2007 and 2008 and updated mineral resources estimates and technical report by Geostat, NML mandated ( BBA ) to complete a Prefeasibility Study ("PFS") for the KéMag Iron Ore Project. This report was issued in January This Technical Report covers the technical, commercial and financial aspects of the Project to develop the KéMag Deposit and to construct and operate associated processing facilities to support annual production of 21.2 million tonnes of concentrate, with downstream processing to result in 15 million tonnes of pellets and 7 million tonnes of concentrate for sale. This scenario requires the establishment of the following facilities: A mine at Lake Harris with an average output of some 76 Mtpy of iron ore; A 21.2 Mtpy concentrator at the Lake Harris site; A 750 km-long slurry pipeline to transport 21.2 Mtpy of concentrate from the concentrator to the pellet plant located in Pointe-Noire; Facilities at Pointe-Noire, Sept-Îles, Quebec, on the north shore of the Gulf of St. Lawrence, including: o A 15 Mtpy pellet plant; o A flotation plant for the production of 5 Mtpy of low-silica DR pellets; o A filtration plant for dewatering 7 Mtpy of concentrate for export; o A stockyard to accommodate 1.5 million tonnes of concentrate and 3.1 million tonnes of pellets; o A jetty and ship loading facility, with annual capacity of 15 Mt pellets and 7 Mt concentrate, designed to permit the loading of Laker-size ships and vessels up to 360,000 DWT at the same time.

11 Technical Report on the KéMag Pre-Feasibility Study Page HISTORY Exploration in the area was conducted by IOCC in The initial work included an airborne magnetic survey and geological mapping. IOCC completed further surveys in to assess potential for direct shipping iron ores. This program included the drilling of a series of short holes mostly collared over frozen lakes. Several of these drill holes intersected taconite-type iron formation, but no residually enriched direct shipping iron deposits were located and holes intersecting taconite were not assayed. IOCC renewed exploration in the area in the early 1970s with more airborne magnetic surveying. This work was designed to identify the best taconite potential. IOCC acquired an exploration permit covering the area in 1972 but conducted no further work. In 2004, NML staked its first claims in the area and in 2005 conducted its initial reconnaissance mapping and sampling program. 1.3 GEOLOGY The Property is located on the extreme western margin of the Labrador Trough ("Trough") adjacent to Archean basement gneisses. The Property, for the most part, is overlain by deep overburden which is boggy and strewn with basement gneissic boulders. Recent drilling by NML has shown that units of the Knob Lake Group, including the Sokoman Formation, which is the major iron formation host in the Trough, underlie a major part, if not all, of the Property and comprise a north-northwest striking sequence of rocks that dips shallowly to the northeast at about 5 to 7 degrees. All of the 29 NML 2006 drill holes intersected sections of the Sokoman Formation. The Sokoman Formation in the area has undergone only slight, very low grade metamorphism and shows very few effects of structural deformation. Furthermore, it has been subject to minimal post-depositional leaching or weathering. No intrusives into the Knob Lake Group have been recognized on the Property. 1.4 DEPOSIT TYPE AND MINERALIZATION The KéMag Deposit is iron formation of the Lake Superior-type. Lake Superior-type iron formation consists of banded sedimentary rocks composed principally of bands of iron oxides, magnetite and hematite within quartz (chert)-rich rock with variable amounts of silicate, carbonate and sulphide lithofacies. Such iron formations have been the principal sources of iron throughout the world. The KéMag iron formation consists mostly of recrystalized chert and jasper with bands (beds) and disseminations of magnetite. Some martite, a type of hematite pseudomorphic after magnetite, may also occur. Hematite is also present in the KéMag Deposit, but it is not economic because it will not be recovered by the proposed magnetic beneficiation process. Other gangue minerals are present and these are mostly iron silicates and carbonates.

12 Technical Report on the KéMag Pre-Feasibility Study Page 12 The portion of the Property that has been explored by diamond drilling has a strike length of 9.5 km, oriented northwest-southeast, and iron formation is present along this entire length and continues beyond the Property boundary. 1.5 NML EXPLORATION AND DRILLING In the summer of 2005, NML undertook a preliminary mapping and outcrop sampling program was undertaken in the Lake Harris area, using a fly-in, fly-out camp. The mapping revealed the boggy nature of the area with few outcrops. A few scattered outcrops of iron formation were mapped and sampled near the south end of the Property. NML s 2006 drilling program was initiated to check airborne anomalies outlined by IOCC during the 1950s and in The 2006 program consisted of 29 holes aggregating 3,633.6 m. The majority of this meterage (2,224.7 m) was in the iron formation. All of the drill holes were drilled vertically and ranged in length from 59 m to 186 m. Originally the drilling was to be on a grid pattern 1,000 m by 500 m. However, due to boggy conditions, it was not possible to adhere to the planned pattern. A number of drill holes could also not be completed to their target depth. In the 2007 drilling program, 45 holes were drilled to depths of between 30 m and m for a total of m. 720 drill core samples were sent to the Midland Research Center ( MRC ) for analysis and testing. No major changes in the stratigraphy or the mineralogical characteristics of the economic strata were identified. Drilling continued in March and April 2008 to confirm the eastern extension of the deposit under Lake Harris and Lac de la Frontière and the swampy grounds to the south. Fifteen (15) holes were drilled for a total of 2216 m and 291 samples were collected for testing and analysis. The results of this last campaign confirmed that the ore body extends under Lac de la Frontière and that the stratigraphy, mineralogy and structure were similar to other parts of the deposit. 1.6 DATA CORROBORATION BBA Senior Metallurgist John Dinsdale visited the site and inspected the drill cores in storage at Schefferville in November, IOCC s abandoned mines in the area, the electrical substation, the northern terminal of the Tshiuetin Railway and other infrastructure and facilities in the town of Schefferville were also visited. BBA was not required to complete any validation drill core sampling because independent validation was already completed by Geostat during their Mineral Resource estimate. 1.7 ADJACENT PROPERTIES NML holds an 80% interest in the LabMag iron ore deposit located in Newfoundland and Labrador, approximately 18 km south of the KéMag Deposit. A positive PFS was completed for this deposit in 2006.

13 Technical Report on the KéMag Pre-Feasibility Study Page 13 Both NML and Labrador Iron Mines Ltd are carrying out development work for processing and shipping of Direct Shipping Ore from deposits in the Schefferville area that were previously mined by IOCC. The current iron ore mining areas at Wabush, Labrador City and Mount-Wright are within 250 km of the KéMag Property. Bedford Resource Partners (Bedford) staked 99 claims in north central Québec, 160 km north of Schefferville in the spring of The claims cover the Lac Otelnuk iron ore deposit, comprised of meta-taconite. Other iron ore projects are under investigation by Adriana Resources and Champion Minerals. Metco Resources Inc. announced in 2004 a planned exploration program for gold and polymetallic massive sulphides on Lac La Touche and Lac Gauthier properties some 50 km east-northeast of Schefferville. Virginia Gold Mines Inc. is exploring for gold, uranium, nickel and platinum group metals on properties 275 km northwest of Schefferville. 1.8 METALLURGICAL TESTING Liberation tests, including Davis tube and grindability testing, were performed by MRC on fourteen diamond drill core samples from the NML 2007 and 2008 drilling campaigns. Samples were then tested according to standard MRC procedures to provide indicators of the liberation and grindability characteristics of the samples tested. Test results showed that the average deposit properties are amenable for processing and that a grind of 86.2% -325 mesh (45 µm) would produce a concentrate with the targeted 3% SiO 2 content. A 3.9 tonne sample of the KéMag ore was sent to SGA Germany for small scale pilot plant testing. The test work showed that a final concentrate for a blast furnace feedstock with about 3.0% SiO 2 and a total weight recovery of roughly 26% is achievable by regrinding, screening and wet low intensity magnetic separation at Blaine values of about Further testwork showed that a final concentrate analyzing 1.6% SiO 2 could be obtained following a reverse flotation stage. Results from this test program are only indicative since the small size of the head sample meant that equilibrium operating conditions were not always achieved. Based on the results of the metallurgical testing, a projected weight recovery of 28% was used in the financial evaluation. Testwork for the similar LabMag Deposit demonstrated that good quality pellets could be produced to meet the required market specifications. Concentrate produced during the SGA test run with KéMag ore will be used for confirmatory pelletizing tests. 1.9 MINERAL PROCESSING The flowsheet for the KéMag concentrator is based on the bench and pilot scale metallurgical tests for the adjacent LabMag Deposit and confirmatory testwork with material from the KéMag deposit. The concentrator is designed to process an average of 76 million tonnes per year ("Mtpy") of ore to produce 21.2 Mtpy of

14 Technical Report on the KéMag Pre-Feasibility Study Page 14 concentrates at a grade averaging 69.1% Fe over the 25 year operation. Iron concentration is based on wet magnetic separation, a process used successfully in concentration of similar deposits in the Mesabi Iron Range in Minnesota, USA. The concentrator flowsheet is designed to recover only the magnetite, as hematite is considered too fine to be economically recovered. Overall iron recovery is therefore 60.5% and tailings grade will average 16.7% Fe. The concentrator flowsheet and mass balance has been detailed to a level to support equipment sizing and selection. Most of the unit operations and conceptual designs that have been incorporated are standard practices for the industry and operations in the region. Work will be necessary to confirm some of the processing requirements, especially with the screening and de-sliming operations and solid-liquid separations on the fine particle size of the process streams, in preparation for the final feasibility study. Based on this work, the final product quality is shown as follows in Table 1: Table 1: Pellet Quality Acid Pellets with 1% Limestone Fluxed Pellets with Basicity of 0.7 * Fe (%) SiO 2 (%) Compression (kg) Tumble Dynamic LTD ** * defined as CaO/SiO 2 ** low temperature disintegration CONCENTRATOR Ore from the mine will be crushed using primary gyratory crushers and secondary cone crushers, stockpiled, then fed to nine parallel processing lines. The ore will first be ground finer using roller presses operated in closed circuit with wet vibrating screens. The liberated magnetic mineral will be recovered from the screen undersize product using three stages of magnetic separators with the second stage of separators operated in closed circuit with screens and regrind ball mills. Concentrate from the final separator stage will be screened and dewatered in thickeners. The underflow from the concentrate thickeners will be pumped to storage tanks at the head of the pipeline. The non-magnetic product from each of the separator stages will be fed to tailings thickeners or in the case of coarse material from the first separator stage directly to the tailings pumpbox. The underflow from the tailings thickeners will be pumped to the tailings pumpbox and the combined tailings stream pumped to the disposal area. The overflow from the concentrate and tailings thickeners will be collected in the process water reservoir and recirculated in the process PIPE LINE From the concentrate storage tanks, slurry will be pumped at 65% solids density through a 750 km long pipeline to storage tanks at the Pointe-Noire installations. A booster pumping station will be provided 475 km down the

15 Technical Report on the KéMag Pre-Feasibility Study Page 15 line. The pipeline will be buried to prevent freezing and repair kits with the necessary excavation and other equipment will be pre-positioned at strategically-located points along the pipeline POINTE-NOIRE FACILITIES The concentrate received at Pointe-Noire will be sufficiently fine to be used as balling feed material without the need for further grinding. Two independent dewatering, balling, indurating and load-out processing lines will permit production of two different types of pellet if required. The concentrate slurry will be dewatered, firstly in thickeners and secondly in pressure filters, to provide a filter cake with a moisture content suitable for balling. Bentonite will be added to the filter cake as a binder. Balling discs operating in closed circuit with roller screens will provide a closely-sized greenball as induration feed. Two straight-grate furnaces will have the capacity to produce 15 million tonnes per year of acid pellets. The furnaces will be oil-fired and will have variable speed fans and will be equipped with electrostatic precipitators to meet environmental standards. Limestone and dolomite will be used as flux agents, either alone or in combination to meet a customer s required pellet quality. For the production of DR grade pellets, concentrate will first be upgraded in a flotation plant. The required output of the circuit will be 5.0 million tonnes in a year but during the circuit s operating campaigns, it will produce upgraded concentrate at the rate of 7.5 million tonnes per year to match the throughput of one pelletizing line. Flotation reagents will be added to the circuit to obtain the required reduction of silica. Flotation concentrate will be thickened and pumped to the filter feed tank prior to balling. The flotation tailings will be pumped to a tailings pond where the solids will settle. Surface water at the pond will be returned to the main pellet plant process water reservoir. Excess water will be sent to a treatment plant for clarification and polishing. The 7 million tonnes per year of concentrate produced exceeding the capacity of the pellet plant will be sold on the international market. To prepare the concentrate for storage and ship loading, the slurry will be dewatered in a thickener and will then be pumped to pressure filters. The resulting filter cake with a moisture content of 8% or less will be transferred by belt conveyor to the product stockyard. Pellets will be stored in three rows of uncovered stockpiles with a total capacity of 3 million tonnes of pellets of four types. In addition there is storage space for 1.5 million tonnes of concentrate. Ship loading facilities will consist of two separate berths and two shiploaders to permit the simultaneous loading of Laker-size and large ships. Ship loading will be at the rate of up to 16,000 tph on large ocean-going vessels and 4,500 tph on Laker-sized vessels MINERAL RESOURCE ESTIMATE This 2008 update of the resource model of the KéMag iron ore deposit follows an initial mineral resource estimation done by Geostat Systems International Inc. (Geostat) in March This mineral resource update takes into account the diamond drilling carried out during the winter of 2007 and 2008.

16 Technical Report on the KéMag Pre-Feasibility Study Page 16 In 2006, NML drilled a total of 28 holes for a cumulated length of 3,574m. The 2007 drilling program consisted of 45 diamond drill holes for a cumulated length of 4,964m. In 2008, the drilling program consisted in 15 holes drilled in winter conditions over Lake Harris and in swampy areas for a cumulated length of 2,216m. A total of 10,774m of drilling in 88 holes were used for this mineral resource estimate. The LabMag deposit is composed of a series of strata or seams slightly dipping (6 ) to the north-east. The seams lie flat, without significant deformation. The following layers are considered mineralized: LC, JUIF, GC, URC, PGC, LRC and LRGC. Two other seams, i.e. the MS and LIF layers, are considered barren and un-mineralized. The thickness of the deposit is limited to the contacts between LRGC and LIF. In order to carry out statistical analyses, it is important to regularize the sample lengths so that each sample has an equivalent representativity. This process is called compositing. The assays are composited into composites 3 meters in length. Regular down-the-hole compositing was used. No blending between seams occurred. Samples were composited seam per seam. The deposit s resources are estimated using a block modelling method. For the purpose of this study, KéMag has been interpolated using Inverse Distance interpolation. Mineral resources were classified using NI compliant nomenclature. Drilling density allowed Geostat to classify a significant portion of the deposit as measured. The classified Mineral Resources at an 18% DTWR cut-off as estimated by Geostat as of February 2008 are provided in Table 2. Table 2: Mineral Resources at 18% DTWR Category DTWR Cut-off (%) Total Tonnage (Mt) DTWR (%) Fe Head (%) Fe Conc. (%) SiO2 Conc. (%) Measured 18 1, Indicated Measured+Indicated 18 2, Inferred 18 1, MINE DESIGN AND MINERAL RESERVE The open pit design is based on a cut-off grade calculation determined from operating costs and sales of products. It is believed that these costs are reasonable for use in the study and are comparable with other similar operations in the region. Based on a material definition exercise using geological data available to-date, the life-of-mine ( LOM ) reserves for the ultimate pit design for KéMag were calculated and classified in the Proven and Probable categories in accordance with the criteria of the Canadian National Instrument ( NI ) for Standards of Disclosure of Mineral Project, of February 2001 and the classifications adopted by the CIM Council of August Table 3 below provides a summary of the LOM reserve for the KéMag project using a cut-off grade of 18% DTWR

17 Technical Report on the KéMag Pre-Feasibility Study Page 17 Reserve Category Table 3: Proven and Probable Reserves (Cut-off Grade 18% DTWR) DTWR Crude Fe Concentrate Tonnes (million) (%) (%) Fe(%) SiO 2 (%) Proven Probable Proven + Probable Inferred 73 Waste Inferred+Waste Waste : Ore Ratio 0.51 The designed pit is approximately 9.5 km in length with widths of one km typical and a maximum width of 1.5 km at its widest point. Total pit depth is 182 m with the lowest bench at m elevation. Incorporated in the pit design is a pit slope of 10% (6 o ) on the footwall contact side of the deposit to the west, and a conservative slope of 50 o on the hanging wall to the east. Single benching will be adopted in ore to maximize the recovery of the mineral resources and double-benching will be adopted in waste to minimize the amount of waste. Further analysis on pit slope stability in the hanging-wall will be undertaken to determine whether this is the optimal configuration with respect to rock structures and properties. The proposed KéMag iron ore mine will constitute a large-scale open pit operation using conventional drill, blast, load and haul process to mine an average of 76 million tonnes of ore and 30 million tonnes of waste per year. The combined ore and waste mining rate will range from 290,000 to 295,000 tonnes per day. The primary mine production equipment includes four 33 m 3 electric shovels and a large front end loader loading a fleet of 290 tonne trucks. In full production, truck requirements increase from 19 units in year 3 to a maximum of 34 trucks in years 9 through 17 of the mine plan, due to longer hauls and increased stripping requirements. Mine dilution was provided for in the 13 m bench height planned for mining of the designated blocks. Mining recovery is assumed to be 100%. Waste will be trucked to four principal waste dumps located to the west of the pit limit at distances varying from 1 to 4 km. The three designed waste dumps have a capacity of 31 Mt, 171 Mt, and 370 Mt. There is a fourth dump reserved for overburden, which has a capacity of 75 Mt. Some of these materials will be used for reclamation purposes at the end of mine life. Opportunities for in-pit waste backfilling may become available in mined out areas after nine years of mining; details have not yet been developed. The mine will operate year-round on a continuous basis, two 12-hour shifts per day. The workforce will be housed in on-site accommodations and will work a schedule based primarily on a two week fly-in/fly-out rotation.

18 Technical Report on the KéMag Pre-Feasibility Study Page 18 Run of mine ore will be crushed in two stages with two gyratory crushers and eight cone crushers in closed circuit taking the ore down to a top size of 63 mm. Nine high pressure grinding roll units will run in closed circuit with a wet screening operation to further reduce the ore to less than 6.3 mm. Wet magnetic separation, followed by a screening stage and regrinding in ball mills to 95% minus 325 mesh, will achieve the required liberation of the magnetic iron. Concentrate in slurry form from the circuit will be dewatered in three thickeners to 70% solids, prior to handling through agitated slurry surge tanks and subsequent pumping in the 750 km slurry line to the pellet plant SCHEDULE A period of 5.5 years is estimated for the time from when the decision is made to advance to the next stage of the Project and the start of production INFRASTRUCTURE AND SUPPORT SYSTEMS MINE AND CONCENTRATOR SITE An insulated, heated and air-conditioned building adjacent to the concentrator will contain a number of different facilities including an 8-bay garage for the maintenance of mining and other mobile equipment, mechanical and electrical workshops, conveyor belt repair shop, warehouse, a garage for the fire truck and ambulance and change rooms and lunch room. Administration offices housing the mine, concentrator and site superintendents and their staff will be located on the upper floor together with conference rooms, open office areas for clerical staff, space for computer facilities, map files and printing, and a vault for archives. A fully equipped laboratory will be built for analysis of exploration samples and production samples from the mine and concentrator. Offices will be provided for the laboratory supervisor and assistants. Dormitories, kitchens, eating areas, laundry and recreational facilities will be provided for some 1700 contractors employees during construction, after which the buildings will be used by the 776 mine, crusher and concentrator operations and maintenance personnel and administrative staff all of whom will be working on a two weeks in and two weeks out, fly-in/fly-out basis. Pump houses will be constructed on Lac Gillespie and Lac du Canoë to provide gland seal, boiler make-up and process make-up water in case of problems with the water recycling system. A branch off the pipeline carrying lake water to the plant will feed a treatment plant that will provide potable water to the plant and the camp. Reclaim water from the tailings disposal area will be pumped back to the concentrator for reuse. Surplus water will be treated and disposed of to the environment. Electrical power to the site will be provided by a 270 km long, 315 kv overhead power line from Hydro- Quebec s Brisay generating station. Diesel generators will provide emergency power. Road access will be provided by the construction of a 7 km section of unpaved highway to meet existing roads. An airport is located in Schefferville and a railway connects the town with Sept-Iles on the St Lawrence River.

19 Technical Report on the KéMag Pre-Feasibility Study Page PIPELINE The main pumping station at the concentrator and the receiving station at the pellet plant will be supported by the infrastructure at the Lake Harris and the Pointe-Noire site respectively. At the booster pumping station, equipment and facilities will be housed in a metal frame building with insulated walls and roof. Facilities built for construction personnel will be retained for use by operating and maintenance personnel. A gravel road will be constructed that will connect the booster pump house to the existing highway 389. Fresh water will be pumped from a local stream or lake. Power from the Hydro-Québec Hart-Jaune hydro-electric generating station will feed the booster station substation at 34.5kV. The power will be stepped down to 4.16kV and 600V to feed the loads of the pumps and ancillary equipment. Diesel generators will provide emergency power to the booster pumps POINTE-NOIRE AND SEPT-ILES Construction personnel and operators of the Pointe-Noire facilities will be housed locally in Sept-Iles, a town with a population of and a well-developed infrastructure. An administration building will be leased in Sept-Iles to provide centralized services covering sales, finance and accounting, labour relations, purchasing, information technology, health and safety, human resources and the environment for the Project. Power from the Hydro-Québec Arnaud substation will be transmitted to the Pointe-Noire site by an existing 161 kv overhead line and diesel generators will provide emergency power. The site has ready access to the municipal road network and to Highway 138 which connects Sept-Iles with Quebec City. A rail-ferry service between Sept-Iles and Matane on the south shore of the Saint Lawrence River provides rail connection between Schefferville and the North American railway network. The airport provides a scheduled connection between Sept-Iles and Montreal. Potable water will be taken from the existing pipeline that delivers potable water from Lac des Rapides to existing Wabush and Alouette installations in the Pointe-Noire area. Wherever possible, recycling systems will be installed to reduce fresh water requirements ENVIRONMENTAL AND PERMITTING Since the Project is situated wholly within the Province of Quebec, it is subject to an Environmental Impact Assessment (EIA) in this Province only. No separate federal EIA will be required. An ongoing program will be implemented to monitor the effects of the Project on water quality, groundwater, effluents, fish population, benthic invertebrate communities, geotechnical matters and sediment quality during both the construction and operational periods. A rehabilitation plan will be submitted to the Quebec Ministry of Natural Resources and Wildlife prior to the start of production. No monetary provision will have to be provided in the first eight years of operations.

20 Technical Report on the KéMag Pre-Feasibility Study Page FINANCIAL ANALYSIS The financial evaluation for the KéMag Project is carried out by the preparation of a discounted cash flow model to which the capital and operating cost estimates as well as the production schedule developed in the mining section are input data. The Internal Rate of Return (IRR) on total investment and the Net Present Value ( NPV ) resulting from the net cash flows generated by the project have been calculated. The payback period is also indicated as a financial indicator. The main inputs to the financial analysis are summarized as follows: All costs are expressed in fourth quarter 2008 CDN$. Exchange rate: US$ 0.85/CDN$. Project timing: Construction phase of 3 years and production phase of 25. Financing plan: a mix of equity, suppliers credit and funds borrowed from commercial banks. The projected equity to debt ratio is 30:70. Income tax: both pre-tax and after-tax basis. The capital cost of the Project is approximately $4,451.1 million, including direct costs of $3,442.7 million and indirect costs of $1,008.4 million. The indirect cost includes an estimated amount of approximately $139.4 million for initial start-up requirement. For working capital purpose, accounts payable and receivable have been established at 30 days in average. Initial working capital is estimated at $ 31.0 million. Operating cost: owner-operated leased mining equipment, processing, port handling, environment and G/A have been estimated to average $20.4 per tonne of concentrate after reaching full production capacity in Year 3 after start of commercial production.. Selling prices: o o o Blast furnace: $ 66.5% Fe. DR pellet: $ 67.5% Fe (i.e. 11.7% premium over BF pellets) Concentrate: $ 69.1% Fe Net residual value of project is zero. The results of financial analyses with internal rate of return (IRR) and the Net Present Value (NPV) discounted at 8% before and after tax cases are presented in Table 4.

21 Technical Report on the KéMag Pre-Feasibility Study Page 21 Table 4: Results of Financial Analysis Before taxes After taxes Project IRR (%) Equity ROE (%) Payback (Years from production start-up) 5 5 Net Present Value ($ 8% 8,588 5,552 The results show that the Project generates sufficient funds to service debt and has an attractive return on investment. The results of the sensitivity analysis of the IRR for revenue (pellet and concentrate prices, ore grade and weight recovery), capital cost and operating cost are presented in Figure 1. This analysis shows that the project is mostly sensitive to revenue and less sensitive to operating cost. Figure 1: Sensitivity Analysis (before tax) IRR% 34% KéMag Iron Ore Project Sensivity Analysis (pre-tax) Internal Rate of Return (IRR) 32% 30% 28% IRR(%) 26% 24% 22% 20% 18% 16% 14% -20% -16% -12% -8% -4% 0% 4% 8% 12% 16% 20% Variation (%) OPEX CAPEX SALES 1.16 CONCLUSIONS Based on the findings and results presented of the Pre-Feasibility Study, BBA believes that the KéMag Iron Ore Project is a world-class deposit technically feasible and economically viable. Based on the assumptions presented in the Study, the KéMag Project, with an estimated initial capital cost of $4,451 million, can achieve

22 Technical Report on the KéMag Pre-Feasibility Study Page 22 an internal rate of return (IRR) of 25.2% and a net present value (NPV) of $8,588 million using a discount rate of 8%. The payback period after the start of the commercial production is 4 years before taxes. The level of accuracy of the capital and operating costs is +/- 25%. The mining equipment needed to start the production is provided as an operating lease and is added in the operating cost. Using the latest and modern technology available today and having no legacy burden, costs of mining, processing and pelletizing for the KéMag Project will be among the lowest-cost iron producer in North America. The KéMag Project also benefits from the low power cost in the Province of Quebec and the quality of the ore, which is hard but not difficult to concentrate. Furthermore, transportation of iron concentrate by pipeline is established to be technically feasible as demonstrated by successful existing mineral slurry pipelines in cold weather on similar projects and represents a substantial saving in operating cost in comparison with rail transportation. Based on an estimate of mineral resources produced by GEOSTAT in compliance with the National Instrument , the KéMag Project hosts a very large resource of iron ore. The mineral reserves are estimated by BBA at 2,141 million tonnes in the Proven and Probable categories and will be sufficient for over 25 years of operation at the presently planned rate to produce 10 Mtpy of blast furnace pellets, 5 Mtpy of DR pellets and 7 Mtpy of iron concentrate. To date, the difficulty in obtaining a bulk sample representative of the KéMag deposit has precluded any largescale metallurgical test work. This situation should be remedied for the next phase of test work during the Feasibility stage. The duration of the Project schedule is 5.5 years from the time a decision is made to continue engineering until the start of commercial production. Such period accounts for the identification of a strategic partner, the Feasibility Study, the financing, detailed engineering and construction. The project is located in the Province of Québec and it is expected that there will be no political or regulatory risk associated to the development of the KéMag Iron Ore Project RECOMMENDATIONS Prior to advancing the project through the Feasibility Stage, there are a number of areas where BBA is of the opinion that some opportunities exist where development and operation can be enhanced or the risk reduced. BBA believes that NML may benefit from studying all the possible ways to achieve the lowest capital and operating costs structure, as well as minimising risks. Recommendations for the Project include: Review and optimization of the process and layout including preliminary engineering activities for plant layouts;

23 Technical Report on the KéMag Pre-Feasibility Study Page 23 Review of tailings disposal solution and construction; A large bulk sample representative of the ore body should be collected for pilot plant test work during the feasibility study phase, for a final flowsheet optimization; Environmental permitting works to be initiated (i.e. Project Notification); Hydrogeology and geotechnical studies to be initiated; Hydro-Québec should be requested to begin detailed studies to provide power to the mine site; Detailed topographic mapping for the slurry pipeline routing and optimization should be undertaken prior the Feasibility Study; The waste rock is not believed to be acid generating and confirmatory test work should be carried out at the Feasibility Study Stage.

24 Technical Report on the KéMag Pre-Feasibility Study Page 24 2 INTRODUCTION AND TERMS OF REFERENCE 2.1 INTRODUCTION New Millennium Capital Corp. ( NML ) holds a 100% interest in the claims that constitute the KéMag taconite iron ore deposit at Lake Harris in the Province of Québec, Canada (the KéMag Deposit ). The property is located approximately 50 km north of Schefferville, Québec, as shown in Figures 2 The deposit was explored by Iron Ore Company of Canada ( IOCC ) from 1949 until 1971 using aeromagnetic surveys, geological mapping and test drilling. Based on the survey work, IOCC obtained an Exploration Permit to conduct a detailed investigation of the Lake Harris area in 1972, but made no such investigation. NML staked claims covering the taconite deposit in 2004 and carried out reconnaissance mapping and sampling in From June through October 2006, NML carried out a limited diamond drilling campaign on the deposit. In June, 2006, Met-Chem Canada Inc. completed a Pre-feasibility Study of the LabMag Iron Ore Project ( LIOP ), involving the development of an iron ore deposit on a property at Howells River in Labrador (the LabMag Deposit ) which is adjacent to the KéMag property in Québec. That study was prepared for LabMag GP Inc., a subsidiary company of NML, which granted NML the right to use it. Geostat Systems International Inc. ( Geostat ) prepared and updated the mineral resource estimates of the LabMag Deposit on which the LabMag Pre-feasibility Study and several NI Technical Reports were based. Early in 2007, given the firm s familiarity with the geology of the area, Geostat was again retained by NML, to prepare a mineral resource estimate for the KéMag Deposit The KéMag Deposit was not as advanced as the LabMag Deposit and did not have as much exploration data but, because of its similarity to the LabMag Deposit, Geostat relied on the knowledge gained from the LIOP and felt confident to apply it in the modelling of the KéMag resources even though at that time the KéMag Deposit only had 29 drill holes with 75% of the crosssections containing only one drill hole. In classifying the KéMag resources, Geostat used the same classification criteria as for the LabMag Deposit and as a result, Geostat was able to delineate Indicated mineral resources on the KéMag Property. A Technical Report compliant with the requirements of National Instrument (NI) was filed with the Ontario Security Commission in March 2007 and can be viewed on In that report, Geostat observed that The KéMag Deposit is located some 18 kilometres to the north of the LabMag Deposit. Although this distance may seem important with respect to the concept of adjacent properties, given the scale of both deposits they can be treated as neighbours. The KéMag Deposit and the LabMag Deposit are both iron formations of the Lake Superior type and are considered geologically and stratigraphically similar; The KéMag Deposit is considered to be the northern extension of the LabMag Deposit. Drilling done in 2007 intersected the same iron formations and indicates that the seven stratigraphic horizons of the KéMag Deposit are similar to those occurring at the LabMag Deposit.

25 Technical Report on the KéMag Pre-Feasibility Study Page 25 The overall thickness of the iron formation remains the same, although there are some minor changes in the individual thicknesses and in the magnetite content; Structurally, the taconite beds are dipping at the same average 6 towards the east. From the drill core, no discernable fault or shear zones could be observed. In July, 2007, employees of, and consultants to, New Millennium Capital Corp. ( NML ) prepared a study of the KéMag Deposit. The study was intended to be a Pre-Feasibility Study ( PFS ) and given the similarity of the two deposits and the extent of the work already done for the LabMag PFS, the accuracy of the resource estimates for the KéMag study was probably better than that of the LabMag PFS. Furthermore, the accuracy of the capital and operating cost estimates forming part of the study was considered by NML to be within the required limit of ± 25% for a PFS. However, Watts Griffis McOuat ( WGM), the firm that was mandated by NML in July/August 2007 to produce an updated NI Technical Report and Economic Analysis, considered that only if the results of the limited drilling carried out on the KéMag Deposit up to March 2007 were to be augmented by results from the drilling campaign planned to be carried out thereafter would the estimated Indicated Resources be sufficient to support a PFS. Hence NML was required by WGM to classify its study not as a PFS but as a Preliminary Assessment Study ( PAS ). The planned 2007 drilling campaign was completed and the results thereof were incorporated by Geostat into an updated Technical Report (the Geostat 2008 Report ). More drilling was carried out in 2008 and the results were used by Geostat in the preparation of a report entitled Update of the KéMag mineral resources September 2008 (the Geostat 2008 Update. Based on that report, NML has been able to update its PAS and improve its status to that of a PFS, of which this document is the Study Report. 2.2 TERMS OF REFERENCE Breton Banville and Associates ( BBA ) was retained by NML to complete it s PFS for the KéMag Project (the Project ) and document the study in an independent technical report prepared in compliance with the standards of the Canadian Securities Administrators National Instrument ( NI ), and definitions of the Council of the Canadian Institute of Mining, Metallurgy and Petroleum ( CIM Standards ). The preparation of this report was authorized by Mr. Dean Journeaux, Project Manager of KéMag, on June 25, This PFS covers the technical, commercial and financial aspects of the Project to develop the KéMag Deposit and to construct and operate associated processing facilities to support annual production of 21.2 million tonnes of concentrate with downstream processing to result in 15 million tonnes of pellets and 7 million tonnes of concentrate for sale. This scenario requires the establishment of the following facilities: A mine at Lake Harris with an average output of some 76 Mtpy of iron ore; A 21.2 Mtpy concentrator at the Lake Harris site;

26 Technical Report on the KéMag Pre-Feasibility Study Page 26 A 750 km-long slurry pipeline to transport 21.2 Mtpy of concentrate from the concentrator to the pellet plant; Facilities at Pointe-Noire, Sept-Îles, Quebec, on the north shore of the Gulf of St. Lawrence, including: o A 15 Mtpy pellet plant; o A flotation plant for the production of 5 Mtpy of low-silica DR pellets; o A filtration plant for dewatering 7 Mtpy of concentrate for export; o A stockyard to accommodate 1.5 million tonnes of concentrate and 3.1 million tonnes of pellets; o A jetty and ship loading facility, with annual capacity of 15 Mt pellets and 7 Mt concentrate, designed to permit the loading of Laker-size ships and vessels up to 360,000 DWT at the same time. 2.3 SOURCES OF INFORMATION Much of the material used to prepare this report has been excerpted from The Pre-Feasibility Study Report for the KéMag Iron Ore Project prepared jointly by BBA and NML in January A complete list of the material reviewed is found in the References section of this report. 2.4 UNITS AND CURRENCY Metric units are used throughout this report, unless specified otherwise, and all dollar amounts are quoted in Canadian currency ( C$ ) or United States currency ( US$ or US cents ). The exchange rate used in the PFS was US$0.85:C$1.00. Historical data and some government map data are generally in Imperial units All metal values are expressed in weight percent ( wt% or % ), unless otherwise stated. Total iron is expressed throughout the report as total iron ( Tfe or Fe ). The majority of the historical samples had assays done on both crude mineralization and Davis Tube ( DT ) concentrates with crude or Head assays being those assays done on non-concentrated material. For example Hfe% refers to percent iron in Head or crude samples and Fe%DTC refers to iron in the Davis Tube concentrate. Davis Tube refers to instrumentation and a procedure that produces a mineral concentrate high in magnetic iron by separating that portion of the sample that is magnetic from the portion that is non-magnetic, following sample comminution. Percent Davis Tube Weight Recovery ( %DTWR ) refers to the weight percent of the sample concentrated in the magnetic fraction using the Davis Tube procedure. This is roughly the same as percent magnetite in the crude sample. Davis Tube concentrates are also assayed for iron and other oxides expressed in weight percent and the % magnetic iron in the crude sample can be obtained by multiplying the %DTWR figure by the % iron in the Davis Tube concentrate. Total Iron Recovery ( Tfe Recovery ) is the %Tfe units recovered in the magnetic concentrate in the Davis Tube compared to the Tfe in the crude sample. Table 5 documents several of the commonly used abbreviations and acronyms in the text.

27 Technical Report on the KéMag Pre-Feasibility Study Page 27 Table 5: Summary of Terms and Abbreviations for Units Abbreviation Term % or Wt % Weight Percent Head or Crude or H Non-concentrated material Tfe or Fe Total Iron in the Head or Crude Sample DT, DTC or C Davis Tube, Davis Tube Concentrate, Concentrate %DTWR % Davis Tube Weight Recovery %Wt Recovery General term for magnetic weight recovery Tfe Recovery % Tfe units recovered in the Davis Tube 2.5 DISCLAIMER This report or portions of this report are not to be reproduced or used for any purpose other than to fulfil NML s obligations pursuant to Canadian provincial securities legislation, including disclosure on SEDAR, and if NML chooses to do so, to support financing, without BBA s prior written permission in each specific instance, all as discussed in Section 2.2 above. BBA does not assume any responsibility or liability for losses occasioned by any party as a result of the circulation, publication or reproduction or use of this report contrary to the provisions of this paragraph.

28 Technical Report on the KéMag Pre-Feasibility Study Page RELIANCE ON OTHER EXPERTS The authors have compiled this Technical report using information contained in the Pre-Feasibility Study Report, in consultants contributions retained by NML and other documents supporting the Pre-Feasibility Study. Consultants contributions are identified in the text. The authors have not carried out a thorough review of each consultant s work. The information provided to BBA was supplied by reputable consultants and BBA has no reason to doubt the validity of the information.

29 Technical Report on the KéMag Pre-Feasibility Study Page PROPERTY DESCRIPTION AND LOCATION 4.1 PROPERTY LOCATION The KéMag Property is situated in the municipality of Rivière Koksoak in Northern Québec, centred about 50 km to the northwest of the town of Schefferville, Québec. The Property is approximately 245 km north of Labrador City, Province of Newfoundland and Labrador, and 550 km due north of Sept-Iles, Québec. The location of the Property and other Project elements are shown on Figure 2. Figure 2: Project Location The area is centred at N Latitude and W Longitude on National Topographic Map reference 23O/03.

30 Technical Report on the KéMag Pre-Feasibility Study Page DESCRIPTION The Property covers a total area of approximately 81 km 2 and comprises 171 map-staked claims held 100% by NML. Map-staked claim means a claim giving the holder the exclusive right to explore for minerals in an area covered by the claim. A claim does not bestow any surface rights. The currently used definition of the Property is all those claims that are within 4.5 km of a claim on which NML has drilled. The claim group extends for a distance of about 15.5 km aligned on a north-northwest south-southeast axis as shown in Figure 4.2. The Property has not been legally surveyed but map-staked licences are defined on the basis of Universal Transverse Mercator (UTM) coordinates and consequently the Property location is accurate. The claim data are summarized in Table 6. Table 6: Claims Covering the KéMag Property Claim Number Number of claims Issuance Date Expiry Date Ownership CDC to CDC , incl January January % NML CDC to CDC , incl February February % NML CDC to CDC , incl March March % NML CDC to CDC incl May May % NML CDC to CDC incl June June % NML CDC to CDC incl June June % NML CDC to CDC incl August August % NML CDC to CDC incl May May % NML Total 171 In its current state, the Property has only been explored by surface drill holes. A camp was set up to accommodate drilling personnel, but there is no infrastructure. A trail usable by 4-wheel drive vehicles comes to within 7 km of the Property. Figure 3 shows the location of the KéMag claims and the drill holes.

31 Technical Report on the KéMag Pre-Feasibility Study Page 31 Figure 3: Claims Map 4.3 LAND OWNERSHIP AND RIGHTS OF WAY The mine, concentrator, tailings containment area, waste dump, camp and associated infrastructure at the KéMag Deposit will be located on Crown Land that NML will acquire from the Government of Québec ( GQ ). The slurry pipeline between the concentrator and the pellet plant will, for most of its length, be located on Crown Land and NML will acquire that land from GQ. At certain points, the pipeline will cross below the railway owned by Chemin de Fer Cartier. Such crossings will conform to the Standards Respecting Pipeline Crossings under Railways published by The Railways Association of Canada, and relevant Sections of the Canada Transportation Act will apply.

32 Technical Report on the KéMag Pre-Feasibility Study Page 32 In their currently planned locations, the flotation plant, filter plant, pellet plant, pellet stockyard, conveyor between the pellet stockyard and the jetty and wharf and associated infrastructure at Pointe-Noire will be partly on land that NML will acquire or lease from Wabush Mines and partly on Crown Land that NML will acquire or lease from GQ and from the Sept-Îles Port Authority ( SIPA ) that manages the land. Alternative locations in the Pointe-Noire area are under consideration and will be addressed in the Feasibility Study. The jetty and wharf and associated infrastructure at Pointe-Noire will be located on Crown Land that NML will acquire or lease from SIPA, which manages the land and seabed at the coast. For the AC electric power transmission lines on Crown Land between Brisay hydro-electric power station and the Lake Harris site and between Hart-Jaune hydro-electric power station and the pipeline booster pumping station, NML will obtain the necessary permission from GQ and Hydro-Québec for the construction and installation of the lines or cables, after which the lines or cables will be handed over to Hydro-Québec for operation and maintenance. 4.4 PERMITTING Permit No giving NML permission to drill at Lake Harris was issued by the Québec Ministry of Natural Resources in April 2006 and was valid for one year, until the end of March A new permit, No , was issued to cover the 2007 drilling campaign and subsequent exploration work was covered by Permit No , which is valid until 31 March 2009.

33 Technical Report on the KéMag Pre-Feasibility Study Page INFRASTRUCTURE AND PHYSIOGRAPH 5.1 ACCESSIBILITY The Property is accessible to its nearest point by a good gravel road for 25 km northwest of Schefferville past some former open pit mines and for a further 30 km by 4x4 pick-up truck or all-terrain vehicle over a trail that reaches Lac de la Frontière. From that point, a new 7 km access road will be built to reach the Lake Harris property. 5.2 CLIMATE The Schefferville area and vicinity, including the Harris and Gillespie lakes lowlands, have a sub-arctic, continental taiga climate with very severe winters. Daily average temperatures exceed 0 C for only five months a year. Daily mean temperatures for Schefferville average C and C in January and February, respectively. Mean daily average temperatures in July and August are 12.4 C and 11.2 C, respectively. Snowfall in November, December, and January generally exceeds 50 cm per month and the wettest summer month is July with an average rainfall of mm. Vegetation is boreal forest. 5.3 PHYSIOGRAPHY The Property has an average elevation of 535 m above sea level. It slopes gently from west to northeast, away from the height of land representing the Québec-Labrador border and towards Lake Harris and Lac Gillespie, more or less parallel to the dip of the rocks. Terrain on the Property is generally flat, with total relief of some 100 m. Streams to the east and west of the height of land in Québec flow into the Kaniapiskau watershed and then northward into Ungava Bay. 5.4 LOCAL RESOURCES AND INFRASTRUCTURE Schefferville, an incorporated municipality in the Province of Québec, continues to exist despite the closing of the iron ore mines of IOCC in 1982 and the subsequent demolition of a number of houses and original public buildings, including a recreation centre, hospital and churches. The present population is now about 250 nonnative residents, most of who work directly or indirectly for the First Nations. Some 700 members of the Nation Innu Matimekosh-Lac John live in the nearby Matimekosh community. The economy of Schefferville is based on hunting and fishing, tourism and public service administration. More than a dozen fishing and hunting camp operators are based in Schefferville and yearly several thousand people fly to various camps distributed about the region, chiefly for trout fishing and caribou hunting. In addition to the hunting and fishing outfitters, the population of the town consists mainly of motel, store and flying service operators, teachers, retired families and support staff for the town services.

34 Technical Report on the KéMag Pre-Feasibility Study Page 34 Kawawachikamach (Kawawa), a community located some 20 km north of the town of Schefferville, is the home of the Naskapi First Nation of Canada. The community was established in this location following the signing in 1978 of the North-eastern Québec Agreement between the Government of Quebec and the Naskapi Band of Québec. Since 1982, some 130 housing units have been built for the Naskapi people and there are now about 750 Naskapis living in the modern community that has its own school, medical clinic, recreational complex and swimming pool. Until the end of November 2005, the QNS&L railway, owned by IOCC, ran between Sept-Îles and Schefferville and offered weekly passenger and freight services. On December 1, 2005, that part of the rail line that runs from Emeril to the northern terminus at Schefferville was acquired from QNS&L by Tshiuetin Rail Transportation Inc. ( TRT ), which is owned in equal parts by the Naskapi Nation of Kawawachikamach, the Nation Innu Matimekosh Lac John and Innu Takuaikan Uashat mak Mani-Utenam. Today, TRT operates two trains per week between Schefferville and Sept-Îles for passengers and community freight. The region is also served by an airport, classified a Remote Airport under the National Airports Policy, which has a paved 1,500 metre runway. Service to Sept-Îles is offered six days per week and service to Québec City and Montreal is offered twice weekly Kawawachikamach receives its electricity by a 25 kv power line from Schefferville, which in turn is supplied by a 69 kv power line from the 15 MW hydro-electric generating station at Menihek Lake, Labrador, about 40 km south of Schefferville.

35 Technical Report on the KéMag Pre-Feasibility Study Page HISTORY 6.1 GENERAL Prior to staking of the Property by NML in 2004, all recorded exploration work had been carried out by IOCC. A brief summary of exploration work is presented in Table 7. Table 7: Historical Exploration on the Property Radar Geophysics Ltd. Conducted regional aeromagnetic surveys in the Lake Harris/Howells River area for IOCC. Most of the data from these surveys were not interpreted until 1966 by IOCC Geological mapping by IOCC on a scale of 1 =1000 and sampling were carried out by G. Perrault to the west of Howells River, extending to Gillespie Lake. During this period, exploration was mainly aimed at outlining enriched iron ore deposits During the winter of 1958, additional work was carried out by M. Belland (IOCC) in an area located near Boundary Lake (Lac de la Frontière), in the valley west of the Goodwood deposit. This area was investigated by a combined program of dip-needle survey and test drilling, to locate iron formation with possible enriched sections. The area surveyed and test drilled covers 28.5km 2, of which 19.5km 2 is in Québec. A total of 23 holes were drilled in Québec, mostly in lakes, to check the subsurface geology. Most of the holes intersected only slates (MS). Only three holes near the western shores of Harris and Gillespie Lakes encountered in iron formation (Lean Chert or LC) below MS and were not analyzed IOCC conducted a remnant magnetism study of the iron formations occurring within a 64km radius of Schefferville, which included the Lake Harris and Howells River areas. The main aim of this study was the evaluation of the magnetic taconite deposits in the area surrounding Schefferville An airborne electromagnetic and magnetic survey was flown by IOCC over a 518km 2 area of Howells River magnetic iron formation. The purpose of the survey was to outline the best economic taconite zones in terms of tonnage and grade between Astray and Gillespie Lakes Based on the results of the above cited surveys, IOCC obtained an Exploration Permit to conduct a detailed investigation in the Lake Harris area. However, no such investigation was carried out by IOCC NML staked claims covering the Lake Harris taconite deposit Reconnaissance mapping and sampling was carried out by geologists for NML.

36 Technical Report on the KéMag Pre-Feasibility Study Page HISTORICAL DRILLING Historical drilling on the Property consisted of test drilling of favourable dip-needle survey targets, mostly over the lakes. In 1958, IOCC drilled 23 holes during the winter time to locate enriched iron ore deposits. These were shallow holes designed to probe the iron formation. Sixteen holes were drilled on Harris, Gillespie and Jacques Lakes for a total of 246m (807 ). Only three holes intersected unleached UIF. Those samples were not analyzed and none of these historical holes was used in the current mineral resource estimation. Data of the 16 drill holes are summarized in Table 8. Year Company Number of Holes Table 8: Summary of Historic Diamond Drilling Drill hole Numbers Core Size Cumulative length (m) Cumulative length (ft) 1958 IOCC 16 Z1201c to Z1216c unknown

37 Technical Report on the KéMag Pre-Feasibility Study Page GEOLOGICAL SETTING 7.1 REGIONAL GEOLOGY The Property is located on the extreme western margin of the Labrador Trough (Trough) adjacent to Archean basement gneisses as shown in Figure 4. The Trough, otherwise known as the Labrador-Québec Fold Belt, extends for more than 1,000km along the eastern margin of the Superior craton from Ungava Bay to Lac Pletipi, Québec. The belt is about 100km wide in its central part and narrows considerably to the north and south. The Trough is comprised of a sequence of Proterozoic sedimentary rocks, including iron formation, volcanic rocks and mafic intrusions known as the Kaniapiskau Supergroup. The Kaniapiskau Supergroup consists of the Knob Lake group in the western part of the Trough and the Doublet Group, which is primarily volcanic, in the eastern part. The Knob Lake group is of interest on the Property and the stratigraphy is outlined in more detail in Figure 5. The principal iron formation, the Sokoman Formation, part of the Knob Lake group, forms a continuous stratigraphic unit that thickens and thins from sub-basin to sub-basin throughout this fold belt (details on Table 9). The southern part of the Trough is crossed by the Grenville Front. Trough rocks in the Grenville Province to the south are highly metamorphosed and complexly folded. Iron deposits in the Grenville part of the Labrador Trough include Lac Jeannine, Fire Lake, Mont-Wright and Mont-Reed and the Luce, Humphrey and Scully deposits in the Wabush area. The high-grade metamorphism of the Grenville Province is responsible for recrystalization of both iron oxides and silica in primary iron formation producing coarse-grained sugary quartz, magnetite, and specular hematite schists (meta-taconites) that are amenable to concentration and processing. The main part of the Trough north of the Grenville Front is in the Churchill Province and has been subjected to low-grade (greenschist facies) metamorphism. In areas west of Ungava Bay, metamorphism increases to lower amphibolite grade. The mines developed in the Schefferville area by IOCC exploited residually enriched earthy iron deposits derived from taconite-type protores.

38 Technical Report on the KéMag Pre-Feasibility Study Page 38 Figure 4: Regional Geology

39 Technical Report on the KéMag Pre-Feasibility Study Page 39 Figure 5: Regional Geology Map Legend

40 Technical Report on the KéMag Pre-Feasibility Study Page 40 Description PROTEROZOIC Helikian Shabogamo Group Gabbro, Diabase Intrusive Contact PROTEROZOIC Aphebian Kaniapiskau Supergroup Knob Lake Group Menihek Formation Purdy Formation Sokoman iron formation Wishart Formation Fleming Formation Denault Formation Attikamagen Formation Unconformity Table 9: Regional Stratigraphic Column Carbonaceous slate, shale, quartzite, greywacke, mafic volcanic rocks, minor dolomite and chert. Dolomite, developed locally. Oxide, silicate and carbonate lithofacies; minor sulphide lithofacies; interbedded mafic volcanic rocks (Nimish Formation); ferruginous slate and slaty iron formation, slate and carbonaceous shale. Feldspathic quartz arenite, arkose, minor chert, greywacke, slate and mafic volcanic rocks. Chert breccia, thin-bedded chert, limestone, minor lenses of shale and slate. Dolomite and minor chert. Green, red, grey and black shale, and argillite intrerbedded with mafic volcanic rocks. ARCHEAN Ashuanipi Complex Granitic and Granodioritic gneiss and mafic intrusives Note: Zajac (1974) redefined the Ruth Formation, located between the Wishart and Sokoman formations, as part of the Sokoman formation.

41 Technical Report on the KéMag Pre-Feasibility Study Page 41 The main part of the Trough north of the Grenville Front is in the Churchill Province and has been subjected to low-grade (greenschist facies) metamorphism. In areas west of Ungava Bay, metamorphism increases to lower amphibolite grade. The mines developed in the Schefferville area by IOCC exploited residually enriched earthy iron deposits derived from taconite-type protores. 7.2 PROPERTY GEOLOGY GENERAL The Property for the most part is overlain by deep overburden which is boggy and strewn with basement gneissic boulders; however, the extension of the Howells River stratigraphic sequence comprising the Knob Lake Group to the north-northwest onto the Property is well established based on: The sporadic exposures of the lower Sokoman Formation overlying the continuous outcrops of the lowermost unit of the Knob Lake Group along the south western margin of the deposit; The aeromagnetic response; The 2006 drilling results. Recent drilling by NML has shown that units of the Knob Lake Group, including the Sokoman Formation, which is the major iron formation host in the Labrador Trough, underlie a major part, if not all of the Property, and comprise a north-northwest striking sequence of rocks that dips shallowly to the northeast. All of the 29 NML 2006 drill holes intersected sections of the Sokoman Formation. Table 10, adapted after Fink (1972) and Klein and Fink (1976), presents the stratigraphic sequence and type descriptions developed for the Howells River area. Recent drilling by NML on the KéMag Property indicates that this stratigraphic sequence and descriptions are applicable to the KéMag Property. A few of the unit thicknesses reported in Table 10 vary slightly from those found for the KéMag Property in the 2006 drill holes, but differences are not significant.

42 Technical Report on the KéMag Pre-Feasibility Study Page 42 Table 10: Stratigraphy of the Howells River Property Unit Est d Avg True Description Thickness & Range (m) Youngest Diabase Menihek Formation >79.2 Dark grey to black shale with minor interbedded greywacke and carbonate lithofacies, carbonaceous pyritic shale. THRUST FAULT Sokoman Formation UIF Member Lean Chert Sub-member (LC) 25.0 Greenish, green to grey-green and pink-grey magnetite-chert iron Silicate Facies ( ) formation with local zones of laminated to shaley bedded (siderite-magnetite) chert iron formation. This unit contains a stromatolite-bearing purple-red and green chert band with magnetite less than 3 m thick. Stilpnomelane-bearing magnetiterich shales occur both above and below the stromatolitic band. Jasper Upper Iron Formation (JUIF) Magnetite-Carbonate Facies Green Chert (GC) Magnetite-Carbonate Facies MIF Member Upper Red Cherty (URC) Hematite-Carbonate Facies Pink-Grey Cherty (PGC) Magnetite-Carbonate Facies Lower Red Cherty (LRC) Hematite-Carbonate Facies LIF Member Lower Red Green Cherty (LRGC) Magnetite-Carbonate Facies Lower Iron Formation (LIF) Silicate Facies Ruth Formation (RF) Sulphide Facies 26.2 ( ) 3.8 ( ) 8.1 ( ) 12.6 ( ) 8.6 (0-18.6) 21.2 (0-46.0) 8.2 ( ) 5.2 ( ) Layered to laminated, magnetite-chert iron formation. Red-greypink in colour, red chert and oolites. Silicate-rich, green chert unit, laterally continuous and an excellent marker horizon. Predominantly arenitic oxide facies. Oolitic and granular texture with cross bedding, abundant iron oxides throughout with more jasper near the top (URC) and bottom (LRC) of unit. Massive to layered, jasper-magnetite-chert iron formation. Redgrey to reddish purple. Disseminated magnetite-chert iron formation. Grey to pink-grey to green-grey. Layered magnetite-chert iron formation. Red-grey to reddish purple. Lower contact transitional. Layered silicate-magnetite-carbonate, magnetite-chert iron formation. Pink to reddish-grey to green-grey. More silicate in lower part, more oxide in upper part. Lower contact transitional with LIF. Massive to layered green to grey-green silicate-carbonatemagnetite-chert iron formation. Thin bedded to laminated chert-siderite, with thin interbeds of shale. Note Zajac (1974) argues the term Ruth Formation should be abandoned because it is for most part equivalent to LIF. Wishart Formation (Qte) 17.7 ( ) Black Chert 1.4 m ( m) Quartzites and/or re-crystallized cherts.

43 Technical Report on the KéMag Pre-Feasibility Study Page 43 UNCONFORMITY Ashuanipi Complex Archean Granitic and Granodioritic gneiss and mafic intrusives. Paleosol on contact between Proterozoic Assemblage and Archean basement. Adapted after Fink (1972) and Klein and Fink (1976). The Sokoman sequence lies unconformably on Archean granitic gneisses (Ashuanipi Complex). These basement rocks were intersected in the bottom of five of the 2006 drill holes: 1001, 1006, 1008, 1023 and A sharp angular unconformity marks the contact between the gently dipping Knob Lake Group and the steeply foliated Archean basement rocks. The lowermost unit of the Knob Lake Group found on the Property is composed of the feldspathic quartzites and conglomerates of the Wishart Formation. In the drill logs, (drill holes: 1001, 1006, 1008, 1022, 1023 and 1040) and on the Property geology maps, these Wishart quartz-rich sedimentary rocks are designated Qte. This Wishart Formation is overlain conformably by the Ruth and Sokoman Formations. Zajac (1974) redefined the Ruth Formation as part of the Sokoman, however, historical IOCC drill logs and other descriptions for the area use the term Ruth Formation to describe shales located between the Sokoman and Wishart and WGM believes it best to maintain use of this term for describing rock types and stratigraphy for the Property. The contact between the Wishart and the Ruth Formation is commonly marked by a Black Chert ( BC ) horizon 0.6 m to 3 m thick containing zones of disseminated pyrite and carbonate. All three Sokoman members: Lower Iron Formation ( LIF ), Middle Iron Formation ( MIF ) and Upper Iron Formation ( UIF ) defined by IOCC and Zajac (1974) are present on the Property in areas drilled by NML. Each of these three members is in turn broken down into individual stratigraphic units called sub-members. Drill hole logs and all geological work conducted on the Property use these sub-member names to classify samples and describe geology. The Green Chert unit that is the basal unit of the UIF is very characteristic and readily identified. James (1954) proposed, on the basis of his work on iron formations in the Lake Superior region, a division of iron formation into four facies: sulphide, silicate, carbonate and oxide. Klein and Fink (1976) have classified the various sub-members of the Sokoman Iron Formation in the Howells River area into sulphide, silicate, magnetite-carbonate and hematite-carbonate facies. Although Zajac (1977) disagreed, Klein and Fink considered the Ruth shale to represent sulphide facies. The silicate facies in the Howells River area according to Klein and Fink is represented by the LIF and LC sub-members, while the LRGC, PGC, GC and the JUIF submembers are magnetite-carbonate facies and the LRC and URC are hematite-carbonate facies, where magnetite and hematite are present in nearly equal amounts, or hematite is more prevalent than magnetite. Bulk chemical data for each of the sub-members is provided by Klein and Fink (1976). The Sokoman Formation in the area has undergone only slight, very low grade metamorphism and shows very few effects of structural deformation. Furthermore, it has been subject to minimal post-depositional leaching or weathering. According to Klein and Fink (1976), the Howells River area may well represent one of the least altered and best preserved sections of the Sokoman Iron Formation. Exploration work conducted in 2006 by NML shows distinct similarities between the Howells River/LabMag Property and the KéMag Property and

44 Technical Report on the KéMag Pre-Feasibility Study Page 44 WGM is in agreement with NML in that Klein and Fink s descriptions also apply equally well to the Harris Lake area. Upper units intersected in drill holes often show pock marked features due to the weathering out of carbonates. Minor fractures in the rock throughout the sequence are often lined with a superficial layer of limonite, but no substantial zones of supergene enriched iron oxides/hydroxides have been identified. The Sokoman Iron Formation is in turn overlain by the Menihek Formation, comprised of dark grey to black shales. Menihek Formation shales form the uppermost rock unit along the northeast margin of the Property and were intersected in 11 drill holes of the 2006 campaign. No intrusives into the Knob Lake Group have been recognized on the Property STRUCTURE Drilling results indicate that the Wishart and Sokoman Formations on the KéMag Property dip at about 5 to 7 degrees to the northeast, however, only three cross sections in the central part of the drilled area contain multiple drill holes. Core angles for bedding structures in the remaining drill holes appear to be in accordance with this interpretation of shallow dip. This dip is similar to that delineated on the much more densely drilled LabMag Property to the south. Therefore, it can be concluded that the KéMag iron formation most probably dips at a shallow angle, uniformly, to the north east. The contacts between various sub-members are gradational. The GC unit occurring immediately under JUIF is distinctive and a good marker.

45 Technical Report on the KéMag Pre-Feasibility Study Page DEPOSIT TYPES The KéMag Deposit is iron formation of the Lake Superior-type. Lake Superior-type iron formation consists of banded sedimentary rocks composed principally of bands of iron oxides, magnetite and hematite within quartz (chert)-rich rock with variable amounts of silicate, carbonate and sulphide lithofacies. Such iron formations have been the principal sources of iron throughout the world (Gross, 1996). Table 11, after Eckstrand, editor (1984), presents the salient characteristics of the Lake Superior-type iron deposit model. Lithofacies that are not highly metamorphosed or altered by weathering are referred to as taconite. The KéMag Deposit is magnetite-rich taconite containing a lesser component of hematite. Strongly metamorphosed taconites are known as meta-taconite or itabirite. The iron deposits in the Grenville part of the Labrador Trough in the vicinity of Wabush are meta-taconite. The deposits at Schefferville, which were mined by IOCC prior to mine shutdown in 1982, are supergene residual deposits formed by the leaching of silica and the concentration of iron oxides from what was originally taconite (also called protore ). Lake Superior type taconite deposits have not been mined in Canada, but are a major part of the iron mined in the Great Lakes region of the United States. Salient characteristics of ores from the Mesabi Range mines are listed in Table 12.

46 Technical Report on the KéMag Pre-Feasibility Study Page 46 Commodities Examples: Canadian Foreign Importance Table 11: Deposit Model for Lake Superior Type Iron Formation (after Eckstrand, 1984) Fe (Mn) Knob Lake, Wabush Lake and Mont-Wright areas, Que. And Lab. Mesabi Range, Minnesota; Marquette Range, Michigan; Minas Gerais area, Brazil. Canada: the major source of iron. World: the major source of iron. Typical Grade, Tonnage Geological Setting Host Rocks or Mineralized Rocks Associated Rocks Form of Deposit, Distribution of Ore Minerals Minerals: Principal Ore Minerals - Associated Minerals Age, Host Rocks Age, Ore Genetic Model Ore Controls, Guides to Exploration Up to billion s of tonnes, at grades ranging from 15 to 45% Fe, averaging 30% Fe. Continental shelves and slopes possibly contemporaneous with offshore volcanic ridges. Principal development in middle Precambrian shelf sequences marginal to Archean cratons. Iron formations consist mainly of iron- and silica-rich beds; common varieties are taconite, itabirite, banded hematite quartzite, and jaspilite; composed of oxide, silicate and carbonate facies and may also include sulphide facies. Commonly intercalated with other shelf sediments: black Bedded chert and chert breccia, dolomite, stromatolitic dolomite and chert, black shale, argillite, siltstone, quartzite, conglomerate, redbeds, tuff, lava, volcaniclastic rocks; metamorphic equivalents. Mineable deposits are sedimentary beds with cumulative thickness typically from 30 to 150 m and strike length of several kilometres. In many deposits, repetition of beds caused by isoclinal folding or thrust faulting has produced widths that are economically mineable. Ore mineral distribution is largely determined by primary sedimentary deposition. Granular and oolitic textures common. Magnetite, hematite, goethite, pyrolusite, manganite, hollandite Finely laminated chert, quartz, Fe-silicates, Fe-carbonates and Fesulphides; primary or. Metamorphic derivatives Precambrian, predominantly early Proterozoic (2.4 to 1.9 Ga). Syngenetic, same age as host rocks. In Canada, major deformation during Hudsonian and, in places, Grenvillian orogenies produced mineable thicknesses of iron formation. A preferred model invokes chemical, colloidal and possibly biochemical precipitates of iron and silica in euxinic to oxidizing environments, derived from hydrothermal effusive sources related to fracture systems and offshore volcanic activity. Deposition may be distal from effusive centres and hot spring activity. Other models derive silica and iron from deeply weathered land masses, or by leaching from euxinic sediments. Sedimentary reworking of beds is common. The greater development of Lake Superior-type iron formation in early Proterozoic time has been considered by some to be related to increased atmospheric oxygen content, resulting from biological evolution. 1. Distribution of iron formation is reasonably well known from aeromagnetic surveys. 2. Oxide facies is the most important, economically, of the iron formation facies.

47 Technical Report on the KéMag Pre-Feasibility Study Page Thick primary sections of iron formation are desirable. 4. Repetition of favourable beds by folding or faulting may be an essential factor in generating widths that are mineable (30 to 150 m).. 5. Metamorphism increases grain size, improves metallurgical recovery. 6. Metamorphic mineral assemblages reflect the mineralogy of primary sedimentary facies. 7. Basin analysis and sedimentation modelling indicate controls for facies development, and help define location and distribution of different iron formation facies. Author G.A. Gross Table 12: Comparison of Current and Past Producing Taconite Mines Mine-Deposit Location Crude SR Concentrate % Tfe % MagFe Waste:Ore %Wt Recovery Butler Minnesota Minntac USS Minnesota Hibbing Taconite Minnesota Northshore Mining Minnesota Ispat (formerly Inland) Minnesota Keewatin Taconite (formerly Minnesota National Steel) Empire Michigan Source: Skilling Review 2000, 2002, 2003, WGM files; SR: Stripping Ratio waste:ore, tons to tons For iron formation to be mined economically, iron content must generally be near 30%, but also the iron oxides must be amenable to concentration (beneficiation) and the concentrates produced must be low in manganese and deleterious elements such as silica, aluminium, phosphorus, sulphur and alkalis. For bulk mining, the silicate and carbonate lithofacies and other rock types interbedded within the iron formation must be sufficiently segregated from the magnetite.

48 Technical Report on the KéMag Pre-Feasibility Study Page MINERALIZATION The KéMag iron formation consists mostly of recrystalized chert and jasper with bands (beds) and disseminations of magnetite. Some martite, a type of hematite pseudomorphic after magnetite, may also occur. Fink (1972) reports that martite is the second most common iron oxide present at Howells River, but no mineralogical study has been completed for the KéMag deposit so identifications are tenuous. Hematite is also present in the KéMag Deposit, but it is not economic because it will not be recovered by the magnetic beneficiation process proposed. Other gangue minerals are present and these are mostly iron silicates and carbonates. The iron silicates may mainly be minnesotaite and stilpnomelane but mineralogical studies are required to be definitive. Chamosite is also mentioned in the drill logs. The NML drill logs report occasional narrow bands (beds) and blebs of cream coloured carbonate which is likely siderite. Klein and Fink (1976) provide a detailed description of the mineralogy of the Sokoman Iron Formation at Howells River. Magnetite and gangue concentration is dominantly controlled by sub-member geology and hence stratigraphy, but other controls are also evidently in-force because a simple stratigraphic interpretation, defined by submember stratigraphy, fails to completely explain variations in iron and magnetite concentration. The PGC submember consistently contains the highest concentrations of magnetite. LC and JUIF also contain higher concentrations of magnetite, while hematite is most common in LRC, URC and JUIF sub-members. Silicate iron minerals are most prevalent in LC, just beneath the Menihek Formation, and in LIF. Silicate iron minerals also give GC, the lowermost sub-member of UIF, its defining colour. Siderite is common in LC and LIF submembers and manganese carbonates are also likely present. Calcite fills some fractures. Goethite and limonite are also common as fracture facings and are likely due to percolating groundwater, but no significant concentrations of iron hydroxides have been identified. The portion of the Property that has been explored by diamond drilling has a strike length of 9.5 km, oriented northwest-southeast, and iron formation is present along this entire length and continues beyond the Property boundary. The formations dip to the northeast, and to a large extent, the magnetite concentration as described above is controlled mostly by the formational sub-member stratigraphy and large and small-scale sedimentary processes. To the northeast, the Sokoman dips under the Menihek shale.

49 Technical Report on the KéMag Pre-Feasibility Study Page EXPLORATION The exploration programs consisted of a diamond drilling program described under Section 11.

50 Technical Report on the KéMag Pre-Feasibility Study Page DRILLING EXPLORATION PROGRAM In the summer of 2005, a preliminary mapping and outcrop sampling program was undertaken by NML in the Lake Harris area, using a fly-in, fly-out camp. The mapping revealed the boggy nature of the area with few outcrops. A few scattered outcrops of MIF and LIF were mapped and sampled near the south end of the property. Table 13 gives the outcrop location and the analyses. It also includes data for samples collected in 1950 and the results of analyses. Sample Rock Type Table 13: Results of Outcrop Sampling in 1950 and 2005 UTM 1 UTM Sample Concentrate Easting Northing %Fe DTWR %Fe %SiO 2 By IOCC In PGC PGC PGC PGC LRC By NML In LIF LIF LRC PGC LIF LRC DRILLING PROGRAM The 2006 drilling program was initiated by NML to check airborne anomalies outlined by others during the 1950s and again in Since there are no exposures of iron formation on the Property, drilling is the only means of obtaining subsurface information. A total of 3,633.6 m in 29 holes was drilled on KéMag in 2006, out of which 2,224.7m intersected the iron formation. The drilling contractor was Heath & Sherwood (1986) Inc. of Kirkland Lake, Ontario, which provided two JKS 300 diamond drills. Due to the ground conditions, a helicopter had to be used to move the drills to the sites, to move between holes and to provide crew transportation. Canadian Helicopters Ltd of Goose Bay, Newfoundland and Labrador (NL) provided a B2 helicopter for the duration of the program. 1 Based on NAD83

51 Technical Report on the KéMag Pre-Feasibility Study Page 51 The drilling started on June 9, 2006 and concluded on October 14, All of the drill holes were drilled vertically and ranged in length from 59m to 186m. Core size for most drilling was BTW (42mm diameter) and BQ (36.4mm diameter). Holes were spotted using a Global Positioning System ( GPS ) receiver. No downhole directional or geophysical surveys were carried out. At the end of the program all the drill holes were surveyed by N. E. Parrott Surveys Ltd. Of Happy Valley, Goose Bay, NL. Table 14 provides a list of the 2006 drill-holes and their coordinates. Figure 6 shows the locations of the holes drilled in the 2006 program, together with those of holes drilled in the subsequent 2007 and 2008 programs. Originally the drilling was to be at the intersection of lines spaced 1,000m apart and section lines 500m apart. However, due to boggy conditions, it was not possible to adhere to the planned grid. Moreover, due to mechanical problems and inefficient crews, the drilling was suspended from August 14 to 7 September 7, The subsequent revised drilling program fell short of the initial objective DRILLING RESULTS The drilling done in 2006 indicated that the seven economic stratigraphic horizons are similar to those occurring at the LabMag deposit in the Howells River area. The overall thickness of the iron formation remains the same as in the Howells river area, but there are some minor changes in the individual thicknesses and in the magnetite content. The LC unit has very lean sections at the upper levels and the JUIF unit is thin but shows higher magnetite content in certain sections. The LRC unit for the most part grades into LRGC and as expected, the PGC unit shows higher DTWR% and higher concentrate silica. Structurally, the taconite beds are dipping at the same average 6º towards the east. From the drill core no discernable fault or shear zones could be observed. Table 14: 2006 Drilling Summary Drill-Hole No. Depth (m) Northing Easting Elevation (m) 06HL1001D HL1002D HL1004D HL1006D HL1008D HL1009D HL1010D HL1011D HL1013D HL1015D HL1016D HL1018D HL1019D

52 Technical Report on the KéMag Pre-Feasibility Study Page 52 06HL1021D HL1022D HL1023D HL1026D HL1027D HL1028D HL1029D HL1030D HL1031D HL1037D HL1038D HL1039D HL1040D HL1042D HL1047D HL1048D Total: 29 holes for 3,633.6m DRILLING PROGRAM The 2007 program, during which 45 holes were drilled for a total of m, was a follow-up of the 2006 program with the objective of completing the drilling in the regular adopted pattern of 250 m x 300 m grid. This was accomplished in the northern part of the deposit but in the southern part, the drilling continued in the same pattern as used in The drilling contractor was CABO Drilling (Ontario) Corp., 34 Duncan Ave. North, Kirkland Lake, Ontario, P2N 3L3. Canadian Helicopters Ltd. Of Goose Bay, NL, again provided a B2 helicopter for the duration of the program, to move drills and their crews to different sites. Drilling started on July 18 and concluded on October As in 2006, N. E. Parrott Surveys Ltd. Of Happy Valley, Goose Bay, NL surveyed all the drill holes at the end of the program. Table 15 provides a list of the 2007 drill-holes and their coordinates. Figure 6 shows the locations of the holes drilled in the 2007 program, together with those of holes drilled in the 2006 and 2008 programs. 720 drill core samples were collected and sent to MRC for analysis and testing. As an internal check on the analysis, 71 duplicate samples were also sent to MRC. One duplicate sample was sent to Lerch Bros Laboratory in USA for conducting DTWR simulation (MRC) tests.

53 Technical Report on the KéMag Pre-Feasibility Study Page DRILLING RESULTS Logging and subsequent analytical results of the drill core samples indicated no major changes in stratigraphy or the mineralogical characteristics of the seven economic units. Structurally, the units are dipping at 8º towards east at the northern part and around 6º in the southern part of the deposit. The deposit is narrow and shallower in the central part, while deeper and wider in the southern and northern parts, indicating the presence of two shallow basins separated by a plateau region in the central part of the deposit. Table 15: 2007 Drilling Summary Hole No. Depth (m) Northing Easting Elevation (m) HL1003D HL1005D HL1007D HL1012D HL1014D HL1015D HL1017D HL1020D HL1024D HL1025D HL1032D HL1033D HL1034D HL1035D HL1036D HL1041D HL1043D HL1044D HL1045D HL1046D HL1049D HL1050D HL1051D HL1052D

54 Technical Report on the KéMag Pre-Feasibility Study Page 54 Hole No. Depth (m) Northing Easting Elevation (m) HL1053D HL1054D HL1055D HL1056D HL1057D HL1059D HL1060D HL1061D HL1062D HL1063D HL1064D HL1065D HL1067D HL1068D HL1069D HL1071D HL1072D HO1074D HL1075D HL1076D HL1077D Total: 45 holes for 4,883.6m DRILLING PROGRAM From March 5 to April 30, 2008, CABO Drilling (Ontario) Corp continued drilling on the southern part of the KéMag deposit, to confirm that the eastern extension of the deposit lies under Lake Harris and Lac de la Frontière and the swampy grounds to the south. 15 holes were drilled on lines spaced 250 m apart for a total of 2216m. A total of 291 samples were collected for testing and analysis. The samples were analyzed at MRC and at SGS Lakefield Research Limited, Lakefield, Ontario, Canada ( SGS ). Holes were spotted using a Global Positioning System ( GPS ) receiver and are to be surveyed by N. E. Parrott Surveys Ltd. Of Happy Valley-Goose Bay, NL.

55 Technical Report on the KéMag Pre-Feasibility Study Page 55 Table 16 provides a list of the 2008 drill-holes and their coordinates. Figure 6 shows the locations of the holes drilled in the 2008 program, together with those of holes drilled in the preceding 2006 and 2007 programs DRILLING RESULTS The results of the drilling confirm that the ore body continues beyond the western shores of Lac de la Frontière, dipping at angles of 6 to 8 towards the east, and that the stratigraphy, mineralogy and structure are similar to the other parts of the KéMag deposit. The results of the analyses confirm the overall geological grade of the deposit at 31.2% Total iron ( Tfe ), although the DTWR for units JUIF, PGC and LRC are lower than those for the western half of the deposit even though they show high total Fe values. These units occurring between LC and LRGC have an average thickness of 23.3m. The reason for the lower DTWR is martitization, i.e. oxidation of the magnetite, but the extent of the area of oxidation remains to be defined. Table 16: 2008 Drilling Summary Hole No. Depth (m) Northing Easting Elevation (m) HL1078A HL1078D HL1079D HL1080D HL1081D HL1082D HL1083D HL1084D HL1085D HL1086D HL1087D HL1088D HL1089D HL1090D HL1091D Total: 15 holes for 2,216m

56 Technical Report on the KéMag Pre-Feasibility Study Page 56 Figure 6: Drill-hole Locations (2006, 2007 and 2008)

57 Technical Report on the KéMag Pre-Feasibility Study Page SAMPLING METHOD AND APPROACH 12.1 GENERAL The same method of sampling was used in the sampling of cores produced during the 2006, 2007 and 2008 drilling programs CORE HANDLING PROCEDURES At the drill site, the drill core was extracted from the core barrel and was laid out by the drill contractor in threecompartment, 4.6 m capacity core boxes. Each core box was clearly marked by the drill assistant/helper with the hole number, the box number and the starting and ending meterages for each box. Blocks recording hole depth in metres were inserted at the termination of each 3 m drill core run. The core box remained capped all the time when not receiving the core. When the box was full, it was capped and secured at both ends. All full core boxes were delivered to the core storage facility in Schefferville and stacked outside at the end of each shift. The core boxes received at the core storage building were sorted by hole and stored in the building until required by the geologists. The boxes were transferred as complete holes to the core logging facility, as requested by the logging geologists. Following logging and splitting, the core trays containing the split half core saves were returned to the core storage facility for permanent storage on the racks. A metal tag identifying the hole number and the box number was affixed to one end of each core tray LOGGING AND SAMPLING PROCEDURES At the request of the core logging geologist, the core boxes for a complete drill hole were opened from the top of the drill hole and laid out on the logging tables, five boxes at a time. The core was then checked to confirm that the entire core for the drill hole was present. If required, the core was cleaned with a brush and water prior to logging. The descriptive core logging procedure began with the recording of the overburden depth and identification of the stratigraphic units based on the mineralogical assemblage. The overall thickness, magnetism, texture, colour of the chert bands and structural characteristics such as bedding thickness, banded and or massive nature of the units, fault zones were all determined and described. Rock Quality Index (RQD) logging was also done at the same time as the descriptive logging. RQD measurements were made for the entire length of the core. The core recovery percentage was measured and the core loss intervals were recorded. Once the contacts between the stratigraphic units were established, they were clearly marked and a tag was inserted delineating the unit contacts. The core logging geologist then selected and marked the sampling intervals on the core and also placed a tag at the end of the sample interval in the core tray, showing the drill hole number, sample number, sample interval, and the starting and ending depths. Each stratigraphic unit was sampled separately, with sample lengths varying

58 Technical Report on the KéMag Pre-Feasibility Study Page 58 from 1.6 m to a maximum of 9.05 m. All units (LC, JUIF, GC, URC, PRG, LRC, LRGC and LIF) were sampled except MS and RF. Even though LIF is considered as waste, it was also sampled. The sampling interval was based on the extent of magnetite/hematite mineralization and the width of lean cherty zones. If the lean low iron oxide zones exceeded 3 m, they were sampled separately, although individual sample lengths seldom exceeded 6 m in both mineralized and waste zones. Once the sampling intervals were clearly marked, magnetic susceptibility measurements were made at 0.3 m intervals along the core for each sample length. This procedure of selecting sample intervals was repeated for the entire length of the core. Each set of core trays was also digitally photographed by the logging geologist. The core was then sent for splitting and sampling. All logging and sample descriptions were recorded on paper forms for later transfer to digital records based on Microsoft Excel spreadsheets. The cores were split using a hydraulic core splitter and the half core for assaying was placed in a canvas sample bag with a tag showing the drill hole number, sample number, sample interval, sample width and the analysis required. The sample bags were properly tied, with a tag showing the drill hole number, sample number and the sample interval. All the collected samples were sent in wooden boxes to the processing laboratory every two weeks. The split half save of the core was placed on the original core tray that was then returned to the core storage building.

59 Technical Report on the KéMag Pre-Feasibility Study Page SAMPLE PREPARATION, ASSAYING AND SECURITY 13.1 ASSAYING AND TESTWORK From the 2006, 2007 and 2008 drilling programs, a total of 1,570 samples, including 88 check samples, were sent to MRC for chemical and Davis Tube ( DT ) analysis. The following test work and sample analyses were completed: Head assay for Tfe; Determination of %DTWR on -325 mesh DT concentrates; Determination of iron and silica in all DT concentrates. The entire LIF drill core was sampled. Table 17 summarizes the cores submitted in the three sampling programs. No. of check samples included Table 17: Summary of Core Samples Submitted to MRC in 2006, 2007 and 2008 Number of samples No. of DT Tails Total. Fe% DTWR%, CrudeF Crude Crude DT Conc. DT Conc. DT Conc. or For For Assay For Fe ICP-12 For Fe++ Assays for Fe Fe++ LOI SG and SiO , In addition, 21 samples on three fractions (Crude, DT Concentrate and DT Tails) were analyzed for trace elements and sulphur. MRC s sample preparation and analysis procedure consisted of the following steps: Individual core samples crushed to ⅜ with a 4 x6 jaw crusher; Split 1,500 g for test work; Save the balance; Roll crush 1,500 g to 100% -10 mesh; Split 50 g for Davis Tube test and Head sample analysis; Save the balance; Stage grind 50 g to -325 mesh as per MRC procedure (Hanna Procedure); DT % Weight Recovery test on g sample as per the procedure provided by MRC (Hanna Procedure);

60 Technical Report on the KéMag Pre-Feasibility Study Page 60 Analyze DT concentrate sample for Tfe and SiO 2 ; (non-mercury titrimetric method for total iron; SiO 2 determination using hydrofluoric acid); and Analyze Head sample for Tfe; save the balance. The security measures to protect the samples integrity were adequate and consisted in identifying sample bags with drill hole name, From, To and sample number, referencing sample locations in core boxes and direct shipment of sample bags containing half core pieces to the MRC laboratory. No sample preparation was done on site. MRC is holding the sample rejects until further notice from NML QUALITY ASSURANCE/QUALITY CONTROL PROGRAM NML QA/QC PROGRAM In 2006, in order to control the quality of the laboratory results, NML selected a total of 13 samples from selected drill hole intersections to be re-assayed blindly at MRC. The samples were taken from the remaining half core and were assigned a new drill hole name and a new sample number and were sent in the stream of sample bags. For the same reason and in a similar fashion, 71 check samples were submitted in 2007 and a further 4 were submitted in MRC QA/QC PROGRAM MRC had its own internal QA/QC blind assaying program in which samples were randomly selected and reassayed. The details of the MRC QA/QC protocol are as follows: Run standards at the start of procedure to calibrate the test /equipment. 4% of samples are submitted by management as blind samples by following the procedure outlined below to check the analytical accuracy of the work: 1. Randomly pick pulp to be assayed and place in an envelope. 2. Assign a new number. 3. Record old and new numbers in a folder that is not in the lab. 4. Submit for analysis. 5. Record old and new assays for comparison purposes. In 2006, Geostat received check results of 10 head check assays and 10 concentrate assays. Although the total of 20 checks represents 4% of the samples, comparisons were actually made of groups of 10 samples, or 2% of the sample set. MRC also sent 60 selected samples to be checked by an external laboratory, Lerch Brothers Inc. of Minnesota.

61 Technical Report on the KéMag Pre-Feasibility Study Page CHECK SAMPLING RESULTS OF NML CHECK SAMPLING In 2006, although there was only a very limited set of 13 check samples, the check results reported consistent values with the exception of sample 4953 where a 14.41% Tot. Fe sample returned a check value of 33.02% Tot. Fe. NML requested re-assays of sample 4953 and its blind counterpart and the new assays showed more realistic results. It was assumed that the original assay of sample 4953 was erroneous and passed through MRC internal quality control procedures. RESULTS OF MRC INTERNAL CHECK SAMPLING As mentioned in previously, MRC has an internal QA/QC blind assaying program in which samples are randomly picked, renumbered and re-inserted in the stream of samples. This check sampling was done on the prepared original pulps and no check assaying was done on the rejects. In 2006, a total of 20 check assays were selected, 10 for head assays and 10 for Davis Tube concentrates. MRC concluded that all check assays returned values well within normal ranges and that no significant bias was observed. RESULTS OF LERCH BROTHERS INC. CHECK SAMPLING In 2006, a selection of 60 sample pulps was sent from MRC to the Lerch laboratory for an external assaying check. 29 pulps were assayed for Fe in head and 30 pulps were assayed for Fe in concentrate. In this case again, the number of pairs was small and more control samples would be required to be more conclusive. In reviewing the results of the Lerch assays, Geostat s findings were that: %Fe in head does not present a statistical bias. %Fe in concentrate and %SiO 2 in concentrate do present a bias. However, although a bias seems to exist, it is a small one, -0.6% on an average of 69% for Fe in concentrate and +10.9% on an average of 3% for SiO 2 % in concentrate. Geostat recommended increasing the number of pairs in each group and further investigating the potential bias between the two laboratories in future sampling campaigns. The biases had no impact on the current mineral resource estimates because they were based solely on DTWR% but Geostat considered that, for future campaigns, the subject of bias should be addressed by the laboratories concerned. OVERALL RESULT OF CHECK SAMPLING The detailed results of check sampling are given in the Geostat 2007 Report. Except for the group of 30 samples sent to Lerch, which was a minimum in its opinion, Geostat could not derive significant conclusions as to whether or not the assays were biased. As noted above, the Lerch sample groups did present a bias for Fe and

62 Technical Report on the KéMag Pre-Feasibility Study Page 62 SiO 2 in concentrate which, although not large, was considered statistically significant. Geostat recommended increasing the number of check assays and investigating the potential bias source in future sampling campaigns. Geostat also noted that DTWR was not internally checked at MRC nor was it at Lerch. Only the NML blind half-core check samples were checked. Since DTWR is a critical component of the mineral resource, being the element on which the cut-off grade is applied, Geostat recommended adding DTWR to its QA/QC program. Geostat s opinion was that although there did not appear to be a bias, the small number of pairs compared could not lead to a significant conclusion as to bias. In conclusion, Geostat considered that the quality of the samples used was sufficient to support mineral resource estimation and a classification of the resources at the Indicated level but that a more extensive QA/QC program will be required to support the estimation of Measured Resources.

63 Technical Report on the KéMag Pre-Feasibility Study Page DATA CORROBORATION BBA Senior Metallurgist John Dinsdale visited the site in November, It was not possible to access the mine site because of the bad weather conditions but an inspection was made of the drill core still in storage in Schefferville. IOCC s abandoned mines in the area, the electrical substation, the northern terminal of the Tshiuetin Railway and other infrastructure and facilities in the town of Schefferville were also visited. BBA was not required to complete any validation drill core sampling because independent validation was already completed by Geostat for its Mineral Resource estimate.

64 Technical Report on the KéMag Pre-Feasibility Study Page ADJACENT PROPERTIES The LabMag Deposit located approximately 18 km south of KéMag in Newfoundland and Labrador owned by the LMLP, in which NML holds an 80% interest, is an extension of the KéMag Deposit. A positive PFS was completed for the LabMag Deposit in Mineral Resources for the LabMag or Howells River Deposit are summarized in Table 18. Table 18: Howells River Iron Deposit 2007 Mineral Resource Estimate Block Resource Classification Tonnes (Millions) DTWR% %Fe Head %Fe DTC %SiO 2 DTC A Measured 2, Indicated Measured + Indicated 2, Inferred B Measured 1, Indicated Measured + Indicated 1, Inferred C Measured Indicated Measured + Indicated Inferred Total Measured 3, Indicated Measured + Indicated 4, Inferred 1, Note: Data is adapted after NML Press release dated 11 July NML is aware that exploration for similar types of iron deposits in the immediate area has been, or is being, carried out as follows: o o NML itself is carrying out exploration work for its DSO Project, whereby it will process and ship Direct Shipping Ore from deposits in the Schefferville area that were previously mined by IOCC. Labrador Iron Mines Ltd. is also developing a similar project to sell Direct Shipping Ore from deposits in the Schefferville area that were previously mined by IOCC. o The current iron ore mining areas at Wabush, Labrador City and Mont-Wright are within 250 km of the KéMag Property.

65 Technical Report on the KéMag Pre-Feasibility Study Page 65 o Bedford Resource Partners (Bedford) staked 99 claims in north central Québec, 160 km north of Schefferville in the spring of The claims cover the Lac Otelnuk iron ore deposit, comprised of meta-taconite. The area is also being actively prospected for poly-metallic deposits. Active in this sphere are: o o Metco Resources Inc. who announced in 2004 a planned exploration program for gold and polymetallic massive sulphides on Lac La Touche and Lac Gauthier properties some 50 km eastnortheast of Schefferville. Virginia Gold Mines Inc. who are exploring for gold, uranium, nickel and platinum group metals on properties 275 km northwest of Schefferville.

66 Technical Report on the KéMag Pre-Feasibility Study Page MINERAL PROCESSING AND METALLURGICAL TESTING 16.1 METALLURGICAL TESTING AND PROCESS DEVELOPMENT WORK This Section contains information from the Geostat 2007 Report on the resource estimates; the MRC 2007 and 2008 Reports on liberation and grindability testing; the SGA Report on testing of the KéMag ore in a mini pilot plant; and the SGS 2008 Report on bench scale and pilot plant testing using flotation to optimize the production of direct reduction concentrate DRILL CORE SAMPLES TESTING The drill core from drilling done to-date was split and halves were sent to MRC for the determination of various parameters such as head (crude) Fe content, magnetite content, DTWR and concentrate Fe and SiO 2 content. Samples of each drill hole and of the seven stratigraphic horizons, which are similar to those occurring at the LabMag Deposit, were sent to MRC and Geostat used the results to build a block model and evaluate the resources of the KéMag Deposit. Results of the resource estimate from Geostat show an average DTWR of 27.8% with 2.96% silica in the concentrate for resources above 18% magnetite cut off grade. The results from the Geostat 2007 Report are summarized in Table 19 below. Resource Classification Table 19: Drill Core Sample Results Million DTWR Crude Ore Concentrate Tonnes % %Fe %Fe %SiO 2 Indicated Mineral Resources per Seam at 18% DTWR Cut-off LC JUIF GC URC PGC LRC LRGC Inferred Mineral Resources per Seam at 18% DTWR Cut-off

67 Technical Report on the KéMag Pre-Feasibility Study Page LIBERATION TESTWORK Liberation tests, including Davis tube and grindability testing, were performed by MRC on diamond drill core samples from the NML drilling campaigns of 2007 & 2008 on the KéMag deposit. Each of the drill core samples designated by NML was processed by MRC and, in some cases, samples from multiple core intervals were composited by MRC to provide feed material for the liberation and grindability testing. The selected samples were then tested according to standard MRC procedures that are intended to provide indicators of the liberation and grindability characteristics of the samples tested. Summary test results for each of the drill core samples or composites tested in the MRC programs are given in Table 20. Details of the liberation grinding studies are given in the MRC 2007 and 2008 Reports The liberation properties vary between the various layers and within layers, but the average deposit properties are considered acceptable for processing. The geological control of the plant feed is based only on weight recovery. The resources proportions shown in Table 20 have been used to calculate the average liberation grind required to meet 3% silica grade which is estimated at 86.2% -325 mesh (45 microns). A comparative grindability of the LabMag block A & B and KéMag deposits is presented in Table 21. As can be seen, the LabMag Block B and KéMag deposit liberation is similar. However, this is for the tested samples. The representativity of those samples compared to the average of the deposit is unknown. It is difficult to collect a bulk sample from the KéMag deposit as there is a thick overburden layer on top compared to the LabMag deposit which is exposed. Large scale pilot plant testwork has not been done for KéMag. However, a mini pilot plant has been used by SGA in Germany on core samples rejects from MRC to confirm the selected processing flowsheet and compare liberation with the LabMag previous pilot plants. The SGA work is addressed in Section

68 Technical Report on the KéMag Pre-Feasibility Study Page 68 Table 20: Liberation and Grindability Test Results Summary for the KéMag Deposit Hole Number Rock Unit Samples Composited Length Feet % -325 Mesh % Wt. Rec. HL-1018D LC HL-1029D LC HL-1082D LC HL-1083D LC % of Total Average LC HL-1018D JUIF HL-1029D JUIF HL-1082D JUIF HL-1083D JUIF Average JUIF HL-1018D GC HL-1029D GC HL-1082D GC HL-1083D GC Average GC HL-1018D URC HL-1029D URC HL-1082D URC HL-1083D URC Average URC HL-1018D PGC HL-1029D PGC HL-1082D PGC HL-1083D PGC Average PGC HL-1082D LRC HL-1083D LRC Average LRC HL-1018D LRGC HL-1029D LRGC HL-1082D LRGC HL-1083D LRGC Average LRGC Remarks 3% SiO 2 was not attained, used 100% 3% SiO 2 was not attained, used 100% 3% SiO 2 was not attained, used 100% Average Liberation Grind 86.2 For Samples Tested Average Weight Recovery 25.9 For Samples Tested

69 Technical Report on the KéMag Pre-Feasibility Study Page 69 Table 21: Liberation Grinding Comparison of the Deposits Deposit Liberation grind (% -45 microns) LabMag block A 92.7 LabMag block B 84.5 KéMag SMALL SCALE PILOT PLANT TESTING Mineral processing and metallurgical test work on KéMag ore was done by SGA in Germany at a "mini" pilot plant scale. Results indicate that the grind and liberation characteristics of KéMag material and the quality of concentrate produced are very similar to those of LabMag ores. This testing also provided some material for use in preliminary pelletizing test work. The test work was carried out with a 3.9 t sample of the KéMag iron ore deposit provided by NML. The sample was delivered pre-crushed to <12.0 mm and showed a d 80 value of 6.8 mm which is rather low for a roller press crude iron ore feed sample. The objectives of the testwork were to produce iron ore concentrates with about 3.0% SiO2 suitable for blast furnace use and with less than 2.0 % SiO 2 +Al 2 O 3. for direct reduction use. In the first series of tests, roller press grinding in closed circuit with either 1.0 or 2.8 mm wet screens followed by cobber wet low-intensity magnetic separation was investigated and a certain amount of pre-concentrate for regrinding was produced. The results are summarized in Table 22. Cut size (mm) Table 22: Summary of Results for High Pressure Grinding, Wet Screening and Roller Press Grinding in Closed Circuit Magnetic Separation on Screen Underflow (last cycle) Max Weight Magnetite Concentrate Spec. Energy Specific Recirc H press demand 2 O throughput load Content Recovery Fe Fe content force crude ore Conc. Tails (kn/mm 2 (Ts/hm) (%) tails tails content recovery of feed ) (kwh/t net ) (%) (%) (%) (%) (%) (%) (%) The observations from this first test series were that: The 78.9% recirculating load with the 2.8 mm screen increased to 183.3% with the 1 mm screen; A moisture content of up to 2.9% in the roller press feed had no significant influence on the performance of the roller press; The recirculation with a normal roller press feed up to 50 mm in size will be higher than observed in these tests; With the 2.8 mm screen 42.1% by weight of tailings containing of 7.7% of the magnetite was rejected in the cobbers. By decreasing the particle size to <1.0 mm, the amount of rejected

70 Technical Report on the KéMag Pre-Feasibility Study Page 70 tailings increased to 47.7% at slightly decreased magnetite losses of 6.5%, as a result of the better liberation of the mineral phases. The power requirement for grinding to <2.8 mm was 2.0 kwh/t net. 3.0 kwh/t net was required to grind to <1.0 mm cut size. Industrially, with a feed in the range of mm, the energy input needed will be somewhat higher. In a continuous operation it is expected that the magnetite losses can be reduced to about 50 % or even less of the above results. In the second phase of the pilot plant testing, with the ball mill regrinding in closed circuit with wet screening, rougher wet low intensity magnetic separation, hydroseparation and cleaner magnetic separation for the recovery of final magnetic concentrates was undertaken. The production of the pre-concentrate for regrinding was performed at a cut size of 2.8 mm using a magnetic separator with a somewhat higher field strength compared to that used for the batch tests and therefore magnetite rejection to tailings was reduced to 3.1% (magnetite content 2.6%), but the weight rejection was also reduced, to 34.8%. In total, three regrinding tests were performed at different feed rates or energy inputs. The results are summarized in Tables 23 and 24. Pilot Plant Test (PPT) Table 23: Results for Ball Mill Regrinding and Cleaner Magnetic Separation Part 1 Rougher magnetic separation tailings Hydroseparation overflow Grinding energy Magnetite Magnetite Magnetite Magnetite demand Weight Weight content recovery content recovery (kwh/t net ) (%) (%) (%) (%) (%) (%) Table 24: Results for Ball Mill Regrinding and Cleaner Magnetic Separation Part 2 Cleaner magnetic separation tailings Final concentrate PPT Weight (%) Magnetite content (%) Magnetite recovery (%) Weight (%) Fe content (%) SiO 2 content (%) Magnetite recovery (%) Blaine (cm 2 /g) The observations from the second series of tests were that: The amount of material available for the tests was insufficient to permit stable operating conditions to be achieved. The three product formula was used to calculate a material balance for Tests 1 and 2. No balance was possible for Test 3.

71 Technical Report on the KéMag Pre-Feasibility Study Page 71 Pilot plant Test 1 was performed at an energy input of 20.0 kwh/tnet on ball mill feed basis % by weight was recovered to the final concentrate assaying 68.95%Fe and 3.29% SiO2., slightly above the 3.0% for blast furnace feedstock. The Blaine value was measured at 2118 cm2/g. In Pilot Plant Test 2, the feed rate was increased slightly, resulting in an energy input of 15.4 kwh/tnet and a coarser concentrate with 1976 Blaine. The Fe content of the concentrate decreased to 68.6% and the SiO2-content increased to 3.62%. In Pilot Plant Test 3, the energy input was increased to 26.7 kwh/tnet, achieving a concentrate with an Fe content of 69.65% and a SiO2 content of 2.62%. The Blaine value was measured at a rather high value of A final concentrate with about 3.0% SiO2 is achievable by regrinding, screening and wet low intensity magnetic separation at Blaine values of about 2300 with a total weight recovery of roughly 26%. It is expected that in an industrial-sized operation, processing weight recovery can be increased to about 28% or even higher. From the pilot plant test work results, the overall metallurgical criteria for the production of a blast furnace pellet feed material were estimated as show in Table 25. Weight recovery ref. to crude ore (%) Table 25: Estimated blast furnace pellet feed Fe (%) Fe recovery ref. to crude ore (%) Magnetite recovery ref. to crude ore feed (%) SiO 2 (%) Blaine (cm 2 /g) An overview of the relationship of Blaine values to the SiO 2 -content of the final magnetic concentrates is given in Figure 7.

72 Technical Report on the KéMag Pre-Feasibility Study Page 72 Figure 7: Blaine Value versus the SiO2 Content The Pilot Achieving a concentrate with <2.0% SiO 2 +Al 2 O 3 by wet low intensity magnetic separation only is difficult because even in the <0.025mm fractions of the final concentrate the SiO 2 contents were measured in the range of %. If a feedstock for direct reduction is envisaged, values of about 1.5% SiO 2 (<2.0% SiO 2 +Al 2 O 3 ) can only be reached by using reverse flotation. In phase 3 testwork, concentrate from Test 3 was used for laboratory scale and pilot plant testing. LABORATORY SCALE TESTS The best results were achieved in laboratory Test 9 by the addition of 250 g/t of starch, conditioning time 5 min and the use of 75 g/t of the collector Lilaflot 811 M from Akzo Nobel. Total flotation time was 5 minutes at a ph-value of 9.6. Under these conditions, 88.8% by weight was recovered as final flotation concentrate with a SiO 2 content of 1.35 % and a high Fe-recovery of 90.7%. The inclusion of a scavenger stage to improve recovery was tried but resulted in the production of a concentrate with 5.98 % SiO 2 The overall balance for grinding, multistage magnetic separation, as well as flotation, considering the laboratory scale flotation Test 9 is shown in Table 26. Weight recovery ref. to crude ore (%) Table 26: Estimated Direct Reduction Pellet Feed, based on Test 9 Fe (%) Fe recovery ref. to crude ore (%) Magnetite. recovery ref. to crude ore feed (%) SiO 2 (%)

73 Technical Report on the KéMag Pre-Feasibility Study Page 73 The SiO 2 +Al 2 O 3 content is calculated to be somewhat less than 1.7 % which is in the desired range for direct reduction feedstock. PILOT PLANT TESTS Test 10 pilot plant flotation test work was performed at roughly the same conditions as for laboratory Test 9. For the pilot plant test, only 70.3% by weight and 71.5% by Fe was recovered to the rougher concentrate and SiO 2 was measured at 1.58%. Due to the rather low quantity of available feed material and therefore short testing time, the flotation process could not be optimised. By scavenger flotation, another 11.3% by weight of concentrate could be recovered at a higher SiO 2 content of 3.42%. Performance parameters for the production of a combined rougher and scavenger concentrate are shown on Table 27. Table 27: Estimated Direct Reduction Pellet Feed, based on Test 10 Weight recovery ref. to crude ore (%) Fe (%) Fe recovery ref. to crude ore (%) Magnetite. Recovery ref. to crude ore feed (%) SiO 2 (%) The 1.87% calculated value for %SiO 2 +Al 2 O 3 is still in the acceptable range. For pelletizing, a certain amount ( %) of bentonite normally is added to improve the green balling and strength. Bentonite contains about 50 to 60% SiO 2 +Al 2 O 3, therefore the calculated value of 1.87% SiO 2 +Al 2 O 3 is just at the limit. To improve the SiO 2 content of the scavenger concentrate, this product from the pilot plant test was reground to about <0.025 mm and processed by 2-stage cleaner magnetic separation. After mixing the reground and cleaned scavenger concentrate containing 1.8% SiO 2 with the rougher concentrate, the metallurgical performance given in Table 28 was calculated. Weight recovery ref. to crude ore (%) Table 28: Estimated Direct Reduction Pellet Feed, based on Pilot Plant Testing Fe (%) Fe recovery ref. to crude ore (%) Magnetite. Recovery ref. to crude ore feed (%) SiO 2 (%) The SiO 2 +Al 2 O 3 content was calculated to be 1.615%. This value meets the DR feed stock specifications with respect to the SiO 2 +Al 2 O 3 content.

74 Technical Report on the KéMag Pre-Feasibility Study Page 74 CONCLUSIONS In general, the test work showed that it is possible to produce low-silica content concentrates for blast furnace as well as direct reduction use. The recovery figures for iron and magnetite referred to the crude ore estimated from the pilot plant test results are low but no optimisation could be performed because of the small amount of material available for testing. Furthermore, the small laboratory-scale test units are less efficient than the larger pilot plant and industrial scale units. For the recovery of a blast furnace concentrate applying only multistage grinding and magnetic separation, the balance expected for an industrial application is shown in Table 29. Weight recovery ref. to crude ore (%) Table 29: Expected Blast Furnace Concentrate if Produced in an Industrial Plant Fe (%) Fe recovery ref. to crude ore (%) Magnetite. Recovery ref. to crude ore feed (%) SiO 2 (%) If a direct reduction feedstock is wanted, the balance is expected to be as shown in Table 30. Table 30: Expected Direct Reduction Concentrate if Produced in an Industrial Plant Weight recovery ref. to crude ore (%) Fe (%) Fe recovery ref. to crude ore (%) Magnetite. Recovery ref. to crude ore feed (%) SiO 2 (%) To more accurately estimate recovery figures for an industrial scale plant, large scale pilot plant tests will have to be performed with about 50 to 60 tonnes of crude ore. For industrial processing of magnetite ores using multistage grinding and magnetic separation, the magnetite recovery typically is in the range of 95% or even 96%. Applying an additional flotation stage to recover a direct reduction feedstock, 4 to 5% units of magnetite recovery will be lost and therefore magnetite recovery is expected to be in the range of 91% FLOTATION OPTIMIZATION AT PILOT PLANT SCALE During the spring of 2008, a new testing program to optimize the direct reduction concentrate production via flotation was carried out by SGS Lakefield. The program used LabMag block B concentrate which was still available from previous pilot plant testwork done in The program was aimed at defining the best potential

75 Technical Report on the KéMag Pre-Feasibility Study Page 75 flowsheet for flotation at bench scale and then validate that it could be applied at pilot plant scale in continuous operation. Under optimized pilot plant conditions, the final direct reduction concentrate produced with a combination of flotation, magnetic separation and regrinding, reached 1.45% SiO 2 at a weight recovery of 95.1% and Fe recovery of 96.5%. This investigation, although carried out on LabMag blast furnace concentrate, is considered valid for the production of direct reduction concentrate from KéMag ore. Previous flotation investigations have shown that LabMag ore is slightly more difficult to process in flotation than is KéMag ore. It is also to be noted that pilot plant flotation performances are always difficult to optimize and a full-scale industrial plant normally achieves better performance because it can be fine-tuned over long periods of operation PELLETIZING TESTWORK There has been no pelletizing testwork carried out on concentrate from the KéMag Deposit. Testwork has been performed on the adjacent and similar LabMag Deposits that demonstrated that good quality pellets could be produced to meet the required market specifications. Concentrate from the SGA mini pilot testwork has been saved for pelletizing testwork at the Feasibility Study stage MINERAL PROCESSING, CONCENTRATOR AND PELLET PLANT CONCENTRATOR The mineral processing facilities will be located near the mine. The material balance for the concentrator, summarized in Table 31, was developed from the metallurgical test work and was the basis for the design criteria presented in Table 32. The concentrator has been designed to produce 21.2 million tonnes per year of high-grade magnetite concentrates at 69.2% Fe. The estimated plant weight recovery will be 28.0%, and this will require a feed rate of 75.7 million tonnes of crude ore per year. The plant will operate 365 days per year, 24 hours per day at an overall utilization of 95%. The average feed rate to the mill will be 9,098 tph.

76 Technical Report on the KéMag Pre-Feasibility Study Page 76 Item Table 31: Summary of Concentrator Material Balance % Weight % Magnetite Magnetite Units % Magnetite Distribution Feed (Crude) , Cobber Magnet Concentrate , Cobber Magnet Tailings Cobber Magnet Concentrate , Rougher Screen Oversize , Final Screen Oversize , Total Ball Mill Feed , Rougher Magnet Feed , Rougher Magnet Concentrate , Rougher Magnet Tailings Rougher Screen Feed , Rougher Screen Undersize , Rougher Screen Oversize , Deslimer Feed , Deslimer Overflow Deslimer Underflow , Finisher Magnet Feed , Finisher Magnet Tailings Finisher Magnet Concentrate , Final Screen Feed , Final Screen Oversize , Final Screen Undersize , Combined Tailings SG Table 32: Concentrator Design Criteria Item Design Value Unit General Ore crushed (dry) 75.7 mtpy Concentrate production (dry) 21.2 mtpy Weight recovery 28.0 % Mill feed head grade 32.5 % Fe 27.9 % Fe 3 O 4 Concentrate grade 69.2 % Fe 95.8 % Fe 3 O 4 Magnetite recovery 96.2 % Ore specific gravity 3.41

77 Technical Report on the KéMag Pre-Feasibility Study Page 77 Item Design Value Unit Ore moisture content 3 % Annual operating time 365 days/year Operating hours per day 24 Operating shifts per day 2 Concentrator utilization 95 % Concentrator annual operating time 8,322 hours Concentrator feed rate (dry) 9,098 tph Concentrate production (dry) 2,547 tph Primary Crushing Circuit Surge Pile Crusher type Gyratory 60 x 110 (1525 x 2795) Number of primary crushers 2 inches (mm) Primary crusher feed size < 55 (1400) inches (mm) Crushed product size < 12 (305) inches (mm) Crusher operating shifts per week 13 Availability 11 hours per shift Utilization 85 % Primary crushing annual operating time 6321 hours Capacity 25,000 tonnes Secondary crushing circuit autonomy 2 hours Secondary Crushing & Screening Circuit Stockpile Crusher type & size standard cone MP1000 Number of secondary crushing lines 8 Crusher operating shifts per week 13 Availability 11 hours per shift Utilization 85 % Secondary crushing circuit annual operating time 6321 hours Screen size 10 x 20 (3050 x 6100) feet (mm) Screened product < 2.5 (63) inches (mm) Screening medium urethane Capacity 300,000 tonnes High Pressure Grinding Roll (HPGR) Circuit HPGR model/size 33 hours Polycom 20/15-7 Number of HPGR lines 9

78 Technical Report on the KéMag Pre-Feasibility Study Page 78 Item Design Value Unit HPGR operating time per day 24 hours Utilization 95 % Total HPGR throughput (new feed) 9,098 tph Number of sizing screens per line 2 Screen size 8 x 16 (2440 x 4880) feet (mm) Screening medium urethane Screened product < 0.12 (3) inches (mm) Cobber Magnetic Separator Circuit Separator type LIMS, counter rotation Drum size 48x125 (1219x3175) inches (mm) Number of cobber units per line 6 Number of lines 9 Separator feed percent solids 45 % Cobber weight rejection to tails 42 % of mill feed Regrind Mill Circuit Number of lines 9 Number of cyclone clusters per line 1 Number of cyclones per cluster 9 Cyclone diameter 20 (508) inches (mm) Cyclone feed percent solids 50 % Regrind ball mill size 24 x 35 (7.3 x 10.7) feet (m) Number of mills 9 Ball mill discharge percent solids 75.5 % Rougher Magnetic Separator Circuit Separator type LIMS, counter rotation Drum size 48x125 (1219x3175) inches (mm) Number of rougher units per line 10 Number of lines 9 Separator feed percent solids 45 % Rougher separator weight rejection to tails 25 % of mill feed Sizing screen type Stacksizer Number of screens per line 10 Screening medium urethane Screen opening 150 (100) mesh (µm) Deslimers Number of units per line 1 Number of lines 9 Separation size to overflow < 200 (74) mesh (µm) Deslimer weight rejection to tails 2 % of mill feed Underflow percent solids 50 % Finisher Magnetic Separator Circuit

79 Technical Report on the KéMag Pre-Feasibility Study Page 79 Item Design Value Unit Separator type LIMS, triple drum, semi-c.c. Steffensen type Drum size 48x125 (1219x3175) inches (mm) Number of finisher units per line 4 Number of lines 9 Separator feed percent solids 35 % Finisher separator weight rejection to tails 3 % of mill feed Finishing screen type Multiscreen Number of screens per line 16 Screening medium Concentrate Thickeners wire mesh Screen opening 325 (45) mesh (µm) Thickener type Tailings Hydrosizers high rate Number of units 3 Thickener diameter 116 (35.4) feet (m) Underflow percent solids 70 % Flocculant addition Type Tailings Thickeners Tailings Disposal none high rate Total number of units 9 Hydrosizer diameter 25 (7.6) feet (m) Separation size to underflow > ¼ (6.5) inches (mm) Underflow percent solids 65 % Thickener type high rate Number of units 3 Thickener diameter 135 (41.2) feet (m) Underflow percent solids 60 % Flocculant addition Amount of tailings produced 6551 tph Tailings slurry percent solids 61 % yes ORE CRUSHING, STORAGE AND RECLAIM Run-of-mine ore will be hauled by trucks to two identical primary gyratory crushers. Both crushers will be housed in an enclosed building with an overhead crane capable of lifting out the heaviest components. The building will have a maintenance area, a control room, the required electrical, sanitary and other services, a dust collection system and two hydraulic-arm rock breakers for occasional use. Ore from the crushed ore pocket of each crusher will discharge onto two apron feeders, the crushed ore (about -12 top size or -305 mm), will be fed onto two wide belt conveyors that will carry the ore from underground to a conical surge pile.

80 Technical Report on the KéMag Pre-Feasibility Study Page 80 From the surge pile, the ore will be reclaimed with apron feeders located in a concrete tunnel, transferred on a conveyor and distributed by a tripper conveyor to eight 500-tonne feed bins, each ahead of one of eight MP secondary cone crushers. The secondary crushers will operate in closed circuit with vibrating screens to produce a 2½ (-63 mm) product. Screen oversize material will be recycled to the crusher feed bins and the screen -2½ (-63 mm) undersize product will be conveyed to the covered ore storage building. The ore storage building will have a capacity of 300,000 tonnes, equivalent to 33 hours consumption by the concentrator. The flowsheet for this particular area is shown on Figure 8.

81 Technical Report on the KéMag Pre-Feasibility Study Page 81 Figure 8: Primary and Secondary Crushing Circuit

82 Technical Report on the KéMag Pre-Feasibility Study Page 82 HIGH PRESSURE GRINDING Crushed ore from the belt feeders underneath the ore storage building will be distributed to nine separate processing lines. HPGR machines operated in closed circuit with wet vibrating screens will be used to reduce the size of the ore from -2½ (-63 mm) to 3 mm. Screen oversize and edge materials will be recycled to the feed bins. The screen undersize from each HPGR will be pumped to a group of six cobber magnetic separators. The flowsheet for this particular area is shown on Figure 9

83 Technical Report on the KéMag Pre-Feasibility Study Page 83 Figure 9: High Pressure Grinding.

84 Technical Report on the KéMag Pre-Feasibility Study Page 84 COBBER MAGNETIC SEPARATION About 42% of the feed material will be rejected as tail in this first stage of separation. The tailings will flow to a 25' (7.6 m) hydrosizer to separate the fine fraction from the coarse. The fine fraction will overflow to a transfer pump-box and be pumped to the tailings thickener. The material collected at the hydrosizer underflow is too coarse to be fed to the thickener and will be pumped directly to the tailings pump-box for disposal. The magnetic concentrate from each group of six single drum separators will flow by gravity to a cyclone feed pump-box and will be joined by the rougher screens oversize. The flowsheet for this particular area is shown on Figure 10.

85 Technical Report on the KéMag Pre-Feasibility Study Page 85 Figure 10: Cobber Magnetic Separation

86 Technical Report on the KéMag Pre-Feasibility Study Page 86 BALL MILL CIRCUITS There will be one wet overflow type ball mill on each of the nine processing lines. Magnetic concentrate from the cobbers and oversize from the rougher screens oversize will be fed to a cluster of densifying cyclones to control the slurry density inside the mill for good grinding efficiency. The finisher screens oversize will also be returned to the ball mill. The ball mill will grind in closed circuit with the rougher magnetic separators and classifying stacksizer screens. Each mill will discharge to a pump-box and the slurry will be pumped to 10 rougher magnetic separators. The cyclone overflow will flow to the ball mill discharge pump-box to dilute the feed to the rougher magnetic separators. The flowsheet for this particular area is shown on Figure 11.

87 Technical Report on the KéMag Pre-Feasibility Study Page 87 Figure 11: Rougher Magnetic Circuit

88 Technical Report on the KéMag Pre-Feasibility Study Page 88 ROUGHER MAGNETIC SEPARATION The placement of the magnetic separation step in the ball mill closed circuit is designed to remove gangue as soon as it is liberated and thus save grinding energy. An additional 25% of the plant feed will be removed as tailings at this stage of the process and will flow to the tailings thickener via the tailings launder. The concentrate from each single drum magnetic separator will be fed by gravity to a stacksizer screen. The screen oversize will be collected in the cyclone feed pump-box. The screen undersize will flow by gravity to a deslimer to remove as much slimes as possible ahead of the finisher magnetic separators. Deslimer overflow will be collected with the hydrosizer overflow in a transfer pump-box from where it will be pumped to the tailings thickener via the tailings launder. Deslimer underflow will be pumped to the finisher magnetic separator pulp distributor. The flowsheet for this particular area is shown on the previous Figure. FINAL UPGRADING On each of the nine processing lines, the deslimer underflow will be distributed between four triple drum finisher magnetic separators. Finisher tailings will flow by gravity to one of the two tailings thickeners that will service all six processing lines. The concentrate from each finisher magnetic separator will be fed by gravity to a distributor ahead of four multifeed screens. The screen oversize will be pumped back to the ball mill while the screen undersize will be the final concentrate and will flow by gravity to one of three concentrate thickeners that will service all nine processing lines. The flowsheet for this particular area is shown on Figure 12.

89 Technical Report on the KéMag Pre-Feasibility Study Page 89 Figure 12: Finisher Magnetic Separation Circuit

90 Technical Report on the KéMag Pre-Feasibility Study Page 90 CONCENTRATE THICKENING AND SLURRY STORAGE There will be one concentrate thickener for each three processing lines. Thickener overflow will flow to the process water reservoir and thickener underflow, at 70% solids by weight, will be pumped to agitated slurry storage tanks ahead of the slurry pipeline pumping station. The surface of the thickeners will be covered with Styrofoam or similar blocks to minimize evaporation in the summer and heat loss in the winter. The flowsheet for this particular area is shown on the previous Figure. TAILINGS THICKENING There will be one tailings thickener for each three processing lines. Flocculant will be added to the thickener feed to obtain the desired overflow water quality. The overflow will be collected in the process water reservoir. The surface of the thickeners will be covered with Styrofoam or similar material to minimize evaporation in the summer and heat loss in the winter. The flowsheet for this particular area is shown on Figure 13.

91 Technical Report on the KéMag Pre-Feasibility Study Page 91 Figure 13: Tailings Flowsheet

92 Technical Report on the KéMag Pre-Feasibility Study Page 92 TAILINGS PUMPING The combined tailings thickener and hydrosizer underflow will be pumped to a tailings containment basin, the eastern edge of which will be some 1.25 miles (2 km) to the northwest of the concentrator. As more fully described in Section , the basin will be served by three pipelines, two transporting tailings to the basin and one handling the return of water to the concentrator process water reservoir. At the beginning of concentrator operations, only one centrifugal pump with a variable speed drive will be required on each tailings line, but later on as the heights of the dams increase, a second pump will be added. REAGENTS The concentrator will not be using any chemical reagents other than a small quantity of flocculants to produce a clear overflow in the tailings thickeners TAILINGS MANAGEMENT This Section is based on information set out in the Barr Report. The tailings basin will be located relatively close to the concentrator, to minimize capital and operating costs, and its capacity will be sufficient for a minimum service life of 25 years. The tailings containment system will operate in closed circuit with the concentrator and water in the tailings basin will be recycled to the concentrator to meet process and slurry pipeline needs. The design will meet all existing safety and environmental standards. SITE SELECTION Barr Engineering ( Barr ) considered a number of options for tailings disposal. Option 5 in which a single tailings basin is located north of the concentrator plant was selected by NML as it met the requirement that the tailings containment basin be constructed on land that does not have resources with the potential for economic development and that is on claims held by NML. The possibility of constructing two basins in the northern area, to give operating flexibility, will be addressed in the next stage of the Project. As operations progress and a sufficiently large area of the pit is mined out, NML will consider using part of the pit bottom for tailings storage. SYSTEM DESCRIPTION Tailings Characteristics: The size distribution of tailings is expected to be similar to that used for the LIOP and similar to that of tailings in Minnesota, USA, where upstream construction techniques using tailings are applied.

93 Technical Report on the KéMag Pre-Feasibility Study Page 93 Based on an analysis of drill cores, there are only trace amounts of sulphur in the mineralized rock and therefore it is unlikely that the tailings will generate acid. This will be confirmed during operations by an ongoing monitoring program on representative tailings samples. Tailings Basin: The initial construction of a basin for tailings containment will consist of a rock starter dam, built using a maximum of 1,400,000 m 3 of mine waste rockfill with 160,000 m 2 of impervious membrane liner. This configuration will be spread over three years (Years -3, -2 and -1) and will provide sufficient containment for the initial tailings recirculation water storage at Year 0. Placement of the membrane and adjacent filter zones will be restricted to the summer months of June to October. During the first three years of operations, the tailings rock dam will be raised to a height of about 200' (60 m) using downstream construction methods. Tailings will be spigotted off the top of the dam, and the length of the dam will be extended periodically as the elevation increases to provide containment up the existing valley slope and to provide a working area to start spigotting tailings in new areas. At elevations higher than the crests of the rock starter dams, dams will be constructed of tailings by spigotting tailings at points along the dams and pushing coarse tailings with a dozer. A filter consisting of a geotextile bedded on a 24" (610 mm) layer of material that is finer than blast rock and covered with an additional 24" (610 mm) of similar material, will reduce seepage and the potential for piping of tailings through the rock dam. At its upper elevations, after some 20 years, the basin may begin to discharge seepage to drainage area on the west side of the basin footprint as that side rises above natural ground elevations. Any such seepage will be captured in ditches and returned to the tailings basin. If required, channels will also be constructed to divert natural drainage routes that may be affected by the northern boundaries of the basin Tailings Lines: A graded access road suitable for heavy equipment circulation will be built to the tailings basin and will serve as access for the construction of the tailings dams in the year prior to, and the year of, the start-up of the concentrator. Space will be provided at the side of the road for two 34 (860 mm) rubber-lined tailings pipes resting on sleepers. To protect the lines from freezing during shut-downs, they will be self draining and without low points along their route. In the case of an emergency, the lines will drain into a dammed area adequate for four shut-downs. During the winter, the emergency dumps will most likely freeze before the material can be reclaimed, and in the spring the water will be pumped back to the basins and the solids will be reclaimed by a front-end loader and trucked to the basins. The maximum slope of the roads and lines has been set at 10%. Reclaim Water System: Starting from a vertical turbine pump installed in a pump house floating in the tailings basin, a reclaim water line will run to the concentrator process water reservoir. Final pump capacity will be defined at the Feasibility Study stage of the Project when more is known about the hydrology of the tailings basin area.

94 Technical Report on the KéMag Pre-Feasibility Study Page 94 CLOSURE PLAN At the end of the mine life, the material in the tailings basin will be re-sloped and covered, firstly with a layer of waste rock over the last tailings to eliminate dust generation and then a layer of organic material retained during initial stripping at the mine will be used to promote plant growth PIPELINE This Section is based on information set out in the PSI 2007 Report. A tonnage of 21.2 mtpy of concentrate will be pumped by pipeline from the concentrator at Lake Harris to the pellet plant at Pointe-Noire, Sept-Îles, on the Gulf of St. Lawrence. The total length of the pipeline will be approximately 465 miles (750 km). The main pumping station at the head of the pipeline will be located in a building adjacent to the concentrator and a booster station will located some 295 miles (475 km) along the pipeline towards Pointe-Noire and just south of the site of the old town of Gagnon, where highway 389 crosses the Hart Jaune River. A map showing the proposed route of the slurry pipeline in Québec is shown in Figure 14. For approximately the first 185 miles (300 km) from Lake Harris, the pipeline will pass through an area with no existing roads until it reaches highway 389 near Mont-Wright. The pipe will run southward along the side of that highway for some 125 miles (200 km) to Relais Gabriel and then for some 95 miles (150 km) along logging roads east of highway 389, crossing the QNS&L railway, until it reaches the high-quality SM3 Dam road that it will run alongside until it intersects with highway 138. Finally, the pipe will run in an easterly direction along the side of highway 138 and then turn off at an appropriate point to join up with storage tanks ahead of the pellet plant at Pointe-Noire.

95 Technical Report on the KéMag Pre-Feasibility Study Page 95 Figure 14: Proposed Route of the Slurry Pipeline from Lake Harris to Pointe-Noire Wherever practicable, the pipeline will be buried with its top some 3' (1.0 m) to 5' (1.5 m) below ground. PSI is of the opinion that, given the heat that will be created by friction in the pipeline, the insulation provided by this depth of cover is sufficient to prevent freezing during normal operating conditions. The pipeline will only be insulated where it is above ground and a slurry heating system will only be used if the slurry temperature, which will be continuously monitored along the pipeline, drops below 35.6 F (2 C). The subject of maintaining the slurry at an above-freezing temperature will be addressed in greater detail in the Feasibility Study stage of the Project. In May 2006, the LIOP Project Manager, now the KéMag Project Manager, visited the pipeline installations of Norilsk in Siberia, Russia, at 69ºN latitude, where eight slurry pipelines, each some 20 miles (30 km) long, have been operated successfully for 25 years under climatic conditions that are similar to, or more severe than, those to be experienced in Québec. Because of permafrost throughout the area, the Norilsk lines are above ground and insulated with 2.75" (70 mm) of rigid foam covered by a galvanized jacket. A 163 mile (262 km) long, 20" (508 mm) diameter buried coal/water slurry pipeline with 6' (2 m) of cover operated in Siberia from the Inskaya Coal Mine in the Belovo District to Novosibirsk at 55ºN latitude over three winters from 1989 to 1991 without freezing problems before being shut down due to slurry fluid flow problems.

96 Technical Report on the KéMag Pre-Feasibility Study Page 96 The Simplot phosphate slurry pipeline in the mountains of Idaho and Wyoming at about 7,000' (2,130 m) elevation has been operating for many years with winter temperatures as low as -45ºF (-43ºC). Except for one water crossing, the pipeline is buried along its entire length, mostly at a depth of 6' (2 m) or more but some times at as little as 2' (610 mm). The company states it has never had a problem with the pipeline freezing. The basic design criteria for the pipeline system are set out in Table 33 Table 33: Pipeline System Design Criteria General Item Design Value Unit Pipeline design capacity 22 MTPY Utilization 95 % Storage capacity at concentrator 12.0 hours production Slurry percentage solids for pumping 65 % Material fineness, P (45) mesh (µm) Pipeline length 465 (750) miles (km) Pipeline nominal diameter 28 (700) inches (mm) PELLET PLANT The design and scope of supply was based on a turnkey quotation received from a reputable supplier. The pellet plant, to be located at Pointe-Noire, near Sept-Îles, Québec, will produce 15 million tonnes per year of pellets containing an average of approximately 67% Fe from some 14.2 million tonnes of concentrate received in the form of slurry in storage tanks at the end of the pipeline. Two balling and induration lines will have the capability of producing acid, fluxed, low silica and DR quality pellets for sale in the global market. While other technologies are available for indurating, it was decided to use the oil-fired straight grate system with closed circuit balling discs at the PFS stage of the Project. The concentrate will be composed mainly of magnetite with approximate iron and silica contents of 69.2% and 3.75% respectively, and will be an ideal feedstock for the production of high-grade pellets at low fuel and electrical power consumption rates. The typical chemical composition of the concentrate is shown in Table 34.

97 Technical Report on the KéMag Pre-Feasibility Study Page 97 Table 34: Typical Concentrate Analysis % Fe SiO Al 2 O CaO MgO TiO 2 <0.001 Mn Na 2 O K 2 O P S - Based on pelletizing tests on LIOP concentrate at SGA in Germany, NML anticipates that it will produce pellets with the qualities given in Table 35. Table 35: Pellet Quality Acid Pellets with 1% Limestone Fluxed Pellets with Basicity of 0.7 Low Silica Pellets Fe (%) SiO 2 (%) Compression (kg) Tumble (%) Dynamic LTD (%) The latest technology and controls will be employed to minimize manpower requirements and consumable usage while achieving the highest possible product quality. The plant will operate on a 24 hour basis and have an overall operating utilization of 92%. Each line will be composed of a thickener for slurry dewatering, two slurry storage tanks, a filter feed tank, five pressure filters, a mixer, ten balling discs each with roller screens, a feed end roller screen, an induration machine with multiclone and electronic precipitator, a segregation bin, a hearth layer and chips screening system and a product conveyor and surge pile. Basic design criteria for the pellet plant are given in Table 36.

98 Technical Report on the KéMag Pre-Feasibility Study Page 98 Table 36: Pellet Plant Design Criteria Item Design Value Unit General Total pellet production 15.0 MTPY Total concentrate requirement 14.2 MTPY Number of pelletizing lines 2 Machine type straight grate Equipment utilization 92 % Nominal total production rate fired pellets tph Concentrate Thickeners Thickener type high rate Number of units per line 1 Thickener diameter 95 (29) feet (m) Underflow percent solids 72 % Concentrate Filters Filter type plate & frame Chamber size 42 (3.9) ft 2 (m 2 ) Number of units per line 5 Filter cake moisture required < 8 % Filter cake storage capacity per line 16 hours Balling Circuit Agglomerator type disc Number of parallel balling circuits per line 10 Required for normal production 9 Greenball sizing roller screen On-size greenballs (8-16) inches (mm) Induration Number of machines 2 Machine width 13 (4) feet (m) Heating zones 4 After-firing zone 1 Cooling zones 2 Grate area 8000 (744) ft 2 (m 2 ) Grate factor 2.9 (30) t/ft 2 /day (t/m 2 /day) Pellet surge pile capacity - each 8,000 tonnes Induration Clean-up Grinding ball mill Classification screw classifier Pellet Additives Fluxes limestone & dolomite

99 Technical Report on the KéMag Pre-Feasibility Study Page 99 Item Design Value Unit Flux grinding wet ball mill in closed circuit with cyclones Limestone addition rate 0-90 (0-41) lb(kg)/tonne pellets Dolomite addition rate (superfluxed pellets) 0-95 (0-43) lb(kg)/tonne pellets Binder FEED PREPARATION bentonite Bentonite grinding vertical roller mill Bentonite addition rate approx 9 (4) lb(kg)/tonne pellets Concentrate that has already been ground to 90% 325 mesh will be received from the concentrator through a pipeline, in the form of a slurry at 65% solids and will be stored in agitated storage tanks. Concentrate will be dewatered in two stages using thickeners and pressure filters prior to balling. Concentrate slurry destined to be upgraded to become the feed for low silica pellet production will be pumped from one of the slurry storage tanks to the flotation circuit. Limestone and dolomite will be used as flux agents, either alone or in combination to meet a customer's required pellet quality. Fluxes will be wet ground in a ball mill in closed circuit with hydro-cyclones to pelletizing fineness and stored in agitated tanks. Flux will be transferred from the storage tanks by a variable speed pump and added in the required proportions to the concentrate in the agitated filter feed tank ahead of each pelletizing line. Typical analyses of dolomite and limestone are shown in Table 37. FILTERING AND MIXING Table 37: Typical Chemical Composition (%) Dolomite Limestone CaO MgO SiO Al 2 O Moisture 8 8 The fluxed concentrate slurry will be pumped from the filter feed tank to pressure filters on each pelletizing line. Pressure filter test work on LIOP slurry indicated that a cake moisture content of 8% or less was easily achievable with this type of filter. An emergency dump for filter cake will be provided to ensure that an extended shutdown at the pellet plant does not impact the operation of the concentrator or the slurry pipeline. Concentrate will be reclaimed from the dump with mobile equipment and loaded into a hopper over an extension of one of the stockyard belts. The filtrate from the pressure filters is expected to contain only a minor amount of solids and will be re-circulated to the thickeners to be further clarified.

100 Technical Report on the KéMag Pre-Feasibility Study Page 100 Pelletizing-grade imported bentonite will be used as the green pellet binder. Crude bentonite will be dry ground in a vertical roller mill plant and pneumatically distributed to day-bins in the mixer area. A typical analysis of Indian bentonite is given in Table 38 as are the typical physical properties of the material. Table 38: Bentonite Chemical Analysis and Physical Properties Chemical Analysis Physical Properties Content % Type % Silica SiO Moisture 15 Alumina Al 2 O Structure 1" 100 Ferrous Oxide Fe 2 O 3 4 Structure 100 m 5 Slaked Lime CaO 1.7 Grit Test 325 m 5 Magnesia MgO 2 ph 9.4 Sodium Oxide Na 2 O 3 Colloids Content 70 Loss on Ignition H 2 O 5 Viscosity (Marsh Yield) 96 Combination Na 2 O/CaO 1.7 Swelling (Enslin Value) at 8 hrs. 650 Alkalis Na 2 O/K 2 O 4 Swelling (Enslin Value) at 18 hrs. 850 The filter cake will be transferred by belt conveyors to storage bins ahead of each pelletizing line. Filter cake and bentonite will be intensively mixed in horizontal mixers and the mixed material conveyed to a feed bin before each balling module, BALLING Each pelletizing line will have 10 balling discs with nine discs normally in operation. Each disc will discharge onto a roller screen, which will remove undersized 0.32" (-8 mm) and oversized +0.63" (+16 mm) green balls which will be recycled to the balling disc feed bins. The on-size screened green balls will be fed onto a collecting conveyor and transferred to the induration furnace, where the balls will be placed on a reciprocating conveyor that will spread them evenly onto a 13' (4 m) wide belt. Just before the greenballs are fed onto the furnace grate another roller screen will remove any green balls broken in transit. INDURATION AND HEARTH LAYER The two 13' (4 m) wide straight-grate furnaces will have the capacity to produce 15 million tonnes per year of acid pellets. Concentrate samples have been sent to a specialized laboratory for balling and pot grate test to define the grate factor and product quality. The results will be available for the Feasibility Study stage of the Project. The furnaces will be oil-fired. Typical Bunker C fuel specifications are given in Table 39.

101 Technical Report on the KéMag Pre-Feasibility Study Page 101 Table 39: Fuel Oil Specification Characteristics Bunker C Relative Density at 15 C Flash Point PMCC C Min. 66 Kinematic Viscosity CST 30 at 100 o C Pour Point C 60/72 Sulphur % Weight Max. 1.0 Sediment % Weight Max Ash % Weight Max 0.1 Water Content in Vol. Max. 1.0 Gross Calorific Value MJ/litre (average) 45 Green balls will be dried in two stages, updraft followed by downdraft drying. In the following zones of pre-heat and firing, the dried pellets will be progressively heated to between 2,282 F (1,250 C) and 2,372 F (1,300 C) to calcine the flux, initiate magnetite oxidation and achieve the correct pellet strength. An after-firing section will provide time for the heat front to completely penetrate to the bottom of the bed and help in ensuring a more consistent product quality. Thereafter, two stages of cooling will reduce the pellet temperature to approximately 212 F (100 C). Five variable speed process fans will provide the ability to control process gas flows as required. Sensible heat from the second cooling will be recovered through a fan system and will be used in the first drying zone while the hot gas from the first cooling zone will be recovered directly through the diluent air header above the machine hood, heated with Bunker C in the down-comers and used to fire the pellets. Waste process gas will be cleaned in electrostatic precipitators and discharged to atmosphere through a stack. The dust from the hearth layer system and the machine discharge will be recovered in wet scrubbers and returned as slurry to the thickeners. Pellets discharged from the induration machine will be sized and the larger pellets will be used as hearth layer to help minimize the plugging of the pallet car grate and to protect the grate bars and the sidewalls of the pallet car from excessive heat. When DR grade pellets are being made, they will be sprayed with a fine dolomite coating mixture prior to being stocked. INDURATION FLOW DIAGRAM Figure 15 on the following page shows the mass balance for the production of acid pellets. Producing the rated 7.5 mtpy of acid pellets on one line will require 7.2 mtpy of fresh feed. The mass balances for fluxed basic, low silica and DR grade pellets are not significantly different from the example shown for acid pellets and, for a ⅓:⅓ ⅓ mixture of acid, fluxed and DR grade pellets, it is estimated that the required total concentrate will be of the order of 14.2 millions tonnes per year with chips regrind, but this figure is subject to adjustment after the results of pot-grate tests on KéMag material are known.

102 Technical Report on the KéMag Pre-Feasibility Study Page 102 Figure 15: Pellet Plant Flow Diagram: Acid Pellets (All units are in dry tonnes per hour)

103 Technical Report on the KéMag Pre-Feasibility Study Page FLOTATION PLANT This Section is based on information set out in the SGS 2008 Report and on the SGA Report. The first of those report covers locked cycle tests and pilot plant work carried out by SGS on LabMag material having characteristics that are similar to those of KéMag material, whilst the second report covers confirmatory test work using a sample from the KéMag deposit. Concentrate as received via the slurry pipeline from the concentrator will contain between 3.4% and 5.0% SiO 2. After processing, the resultant pellets will have a SiO 2 level below 3.9% and will therefore be excellent blast furnace feed material. However, the iron ore market also calls for pellets with more particular specifications such as low silica fluxed pellets with 2% SiO 2, or DR quality pellets with 1.5% SiO 2. Therefore, a flotation circuit will be installed next to one of the two pellet lines at Pointe-Noire and will be operated as and when required, with the length of campaigns being dictated by the tonnage required to produce pellets with reduced silica content. Basic design criteria for the flotation plant are given in Table 40. Table 40: Flotation Plant Design Criteria Item Design Value Unit General Nominal total annual plant capacity 5.0 MTPY concentrate Plant design capacity 7.5 MTPY concentrate Number of flotation lines 2 Equipment utilization 92 % Design throughput tph Weight recovery 95 % Silica content in flotation feed % Silica content of flotation concentrate 1.5 % Rougher Flotation Cell type mechanical Cell size 3500 (100) ft 3 (m 3 ) Number of cells per line 10 Flotation feed percent solids 36 % First Cleaner Flotation Cell type mechanical Cell size 1050 (30) ft 3 (m 3 ) Number of cells per line 5 Second Cleaner Flotation Cell type column Cell size 13 dia.x 33 (4 x 10) feet (m) Number of cells per line 2 Regrind Circuit Dewatering unit magnetic separator

104 Technical Report on the KéMag Pre-Feasibility Study Page 104 Item Design Value Unit Number of dewatering units per line 1 Separator type Feed to separator counter rotation rougher float + 2nd cleaner non-float products Number of regrind mills per line 1 Mill feed material Mill type Operation separator magnetic product IsaMill Open circuit Mill discharge percent solids 50 % Flotation Concentrate Thickening Thickener type High rate Number of units per line 1 Thickener diameter 115 (35) feet (m) Underflow percent solids 72 % FLOTATION CIRCUIT When the flotation circuit is in use, slurry will be diverted from one of the two slurry storage tanks at the end of the pipeline to the flotation circuit conditioners where the slurry will be diluted from 65% solids to 36% solids. The flotation circuit will use mechanical cells and flotation columns. It will consist of two parallel banks of rougher cells. Froth from the rougher cells will be dewatered and reground in an IsaMill operated in open circuit. The ground material will be floated in a bank of first cleaner cells. The froth from the first cleaner cells will be re-cleaned in two second-cleaner flotation columns. The froth from the second cleaner will be final tailings and the sink will be recycled with the rougher froth to the first cleaner feed. The non-float product from the rougher flotation, flotation concentrate, will be pumped to a thickener. The thickener overflow will be a clear liquid and will be partly recycled to the flotation process for dilution of the rougher feed. The surplus will retain some traces of flotation reagents and will be sent to the tailings with the second cleaner froth. The thickener underflow will be pumped to a concentrate slurry storage tank from where it will be fed into the agitated filter feed tank ahead of the pressure filters of one pelletizing line. The flowsheet for this area is shown on Figure 16.

105 Technical Report on the KéMag Pre-Feasibility Study Page 105 Figure 16: Flotation Circuit Flowsheet

106 Technical Report on the KéMag Pre-Feasibility Study Page 106 FLOTATION TAILINGS The flotation tailings will be pumped to a tailings pond where the solids will settle. Surface water at the pond will be returned to the main pellet plant process water reservoir. Surplus water will be sent to a treatment plant for clarification and polishing CONCENTRATE FILTER PLANT Of the 21.2 mtpy of concentrate slurry that will be transported through the pipeline, 14.2 mtpy will be processed in two filtration units, one ahead of each pelletizing line, as described in Section The remaining 7 mtpy will be destined not for the pellet plant but for sale on the international market. To prepare the concentrate for stacking and ship loading, the concentrate solids will be thickened and filtered in a manner similar to that for concentrate that is to be pelletized. The resulting filter cake with a moisture content of 8% or less will be transferred by belt conveyor to the product stockyard. The filtrate from the pressure filters is expected to contain only a minor amount of solids and will be recirculated to the thickener to be further clarified. Design criteria for the concentrate filter plant are given in Table 41. Table 41: Concentrate Filter Plant Design Criteria Item Design value Unit General Concentrate for direct shipping 7.0 MTPY Equipment utilization 92.0 % Filter plant throughput rate tph Number of processing lines 1 Concentrate Thickener Thickener type high rate Number of units 1 Thickener diameter 95 (29) feet (m) Underflow percent solids 72 % Filters Agitated storage tanks 2 Storage capacity 8.0 Hours of filter plant production Filter type plate & frame Chamber size 42 (3.9) ft 2 (m 2 ) Number of units 5 Filter cake moisture < 8 % PRODUCT STOCKYARD AND SHIP LOADING FACILITIES The text of this Section is based in part upon Case A of the Howe 2007 Addendum II Report.

107 Technical Report on the KéMag Pre-Feasibility Study Page 107 A pellet and concentrate stockyard and ship loading facilities capable of operating year-round with sufficient capacity to support the annual sale of 15 million tonnes of pellets and 7 million tonnes of concentrate will be constructed in the Pointe-Noire area on the Baie des Sept Îles, close to the pellet plant. Three rows of uncovered stockpiles will be capable of containing 3.0 million tonnes of four different types of pellet plus off-specification material together with 1.5 million tonnes of concentrate. Both pellets and concentrate will be stockpiled simultaneously. Ship loading facilities will consist of two separate berths and two shiploaders to permit the loading of Laker-size and large ships at the same time. The ship loading berths will be located clear of existing navigation channels and east of the existing Wabush Dock. With a water depth at low tide of approximately 24 m, the berthing facility will be able to accommodate ships from 25,000 DWT to 360,000 DWT. Ship loading will be at the rate of up to 16,000 tph on large ocean-going vessels and 4,500 tph on Laker-sized" vessels. Repairs and major maintenance of equipment will be carried out by pellet plant personnel using pellet plant facilities as appropriate. Design criteria for the product stockyard and ship loading facilities are given in Table 42. Table 42: Product Stockyard and Ship Loading Design Criteria General Item Design Values Units Pellet storage capacity 3.0 Million tonnes Concentrate storage 1.5 Million tonnes Loading season - all vessels 330 days per year Loading season - Lakers 246 days per year Effective operating loading time per day 22 hours Pellet stacker/reclaimers 4 Stacking rate (per machine) 2,400 tph Reclaim rate (per machine) 8,000 tph Maximum ship loading rate 16,000 tph Maximum loading rate for Lakers 4,500 tph Effective loading rate 85 % of rated capacity PRODUCT STORAGE Pellets will be withdrawn from 8,000 tonne capacity stockpiles at the discharge end of each of the two pellet lines and discharged onto a single conveyor which in turn will discharge onto one of the four belt conveyors running the length of the stockyard between the stockpiles and each serving a travelling and slewing stacker/reclaimer. The stacker/reclaimers will be mounted on rails running the length of the stockyard between the three rows of stockpiles. Each stacker will discharge at a rate of up to 2,400 tonnes per hour in a specified location to one side of the conveyor, using the slewing feature to form a flat-topped storage pile suitable for future reclaim by the machine in its bucket-wheel reclaim mode. A separate conveyor will transport concentrate from the filter plant to the feed conveyor system in the stockyard.

108 Technical Report on the KéMag Pre-Feasibility Study Page 108 Each stacker/reclaimer will be capable of operating at the rate of 16,000 tph in reclaiming mode to match the capacity of the shiploader. Reclaimed material will be conveyed some 2½ miles (4 km) to a transfer tower at the landward end of the jetty. There, pellets destined for Laker-size vessels will be fed by a tripper conveyor into three 10,000-tonne silos and pellets destined for other, much larger ships will be transferred onto another covered conveyor belt running along the jetty. SHIP LOADING From the three silos, a covered conveyor belt will transport pellets at the rate of 4,500 tph along the jetty to a transfer tower from where the pellets will be conveyed to a shuttle type traversing and slewing shiploader for loading into Lakers. The 16,000 tph capacity belt conveyor will transport pellets along the jetty from the transfer tower to a quadrant shiploader for large vessels PROTECTION OF THE ENVIRONMENT Materials handling system operations will comply with relevant Canadian, provincial and local environmental guidelines, regulations and laws. All operations will comply with relevant Canadian, provincial and local environmental guidelines, regulations and laws. Health and safety rules and regulations will be strictly enforced. Lighting for the crusher, concentrator and camp and for the material handling and ship loading equipment and around the stockyard, dock office, jetty and berth will be designed using the latest available technology to avoid diffusion and using photovoltaic cells to regulate the degree of lighting applied at any time. To minimize dust, transfer towers and surge bins at the port will be fitted with dust extraction units, the stackers and the reclaimer will be fitted with water sprays, and the shiploader will have a flexible spout that the operator will maintain at a minimum height above the surface of the pellets in ships holds. Collected dust will be returned to the pellet plant. Surface water management systems will be developed on all sites to channel contaminated run-off water to sedimentation basins. Mineral particles will be retained in the basin and the clear overflow directed towards natural drainage basins. Wherever possible, equipment will be purchased so that noise emission will not be greater than 85 dba at a horizontal distance of 39" (1 m). Personal ear protection will be provided as required.

109 Technical Report on the KéMag Pre-Feasibility Study Page 109 Slurry pipeline operations will comply with relevant Canadian, provincial and local environmental guidelines, regulations and laws. Exhaust gas from the generators at the booster station will comply with regulations. Based on data supplied by the SCADA system, a leak detection system will use real-time flow, pressure, density and temperature measurements at various points along the pipeline to continually check the pipeline s integrity.

110 Technical Report on the KéMag Pre-Feasibility Study Page MINERAL RESOURCE AND MINERAL RESERVE ESTIMATES 17.1 INTRODUCTION This update of the resource model of the Kémag iron ore deposit follows an initial mineral resource estimation done by Geostat Systems International Inc. (now SGS Geostat Ltd) in March 2007 and an update in May A technical report entitled Estimation of the Mineral Resources of the KéMag Iron Ore deposit, New Millennium Capital Corp., Technical Report was produced on March 20, A second report entitled Update of the resource model of the Kemag iron ore deposit, Quebec, New Millennium Capital Corp. was produced on May 15, This mineral resource update takes into account the diamond drilling carried out during the winter All of the data and information was provided to Geostat by New Millennium Capital Corp. (NML). Geostat did not verify the assay data. It is assumed that the data used is correct and free of errors DATA RECEIVED DRILL HOLE DATA In 2006, NML drilled a total of 28 holes for a cumulated length of 3,574m. The 2007 drilling program consisted of 45 diamond drill holes for a cumulated length of 4,964m. In 2008, the drilling program consisted in 15 holes drilled in winter conditions over Lake Harris and in swampy areas for a cumulated length of 2,216m. In total, for this mineral resource update there are 88 holes totalling 10,774m. NML provided Geostat with the drill hole database in the form of a Microsoft Access file directly compatible with the Geostat software library (GeoBase format). Geostat did not have to convert or process the data. NML carried out its own quality control over the drill hole data. Geostat did not independently verify it. Figure 18 presents the layout of the drill holes covering the deposit. The database content is summarized below. COLLAR INFORMATION: Northing, Easting and Elevation coordinates, Hole length All holes drilled vertically, no deviation tests available. ASSAY INFORMATION: Assays limits (From-To) Sample number

111 Technical Report on the KéMag Pre-Feasibility Study Page 111 Fe in head Davis Tube Weight Recovery (DTWR) Fe in concentrate SiO2 in concentrate LITHOLOGY INFORMATION: Limits (From-To) Lithological summary code Lithological description The lithological code represents the seam or strata intersected by the hole. The stratigraphic column is presented in Figure 17. A total of 10 strata are intersected (Table 43): Table 43: Lithological Codes used in KéMag deposit Strat Lithology Code a 1 Overburden, Rubble OBR 2 Menihek Slate MS 3 Lean Cherty LC 4 Jasper Upper Iron JUIF Formation 5 Green Cherty GC 6 Upper Red Cherty URC 7 Pink, Grey Cherty PGC 8 Lower Red Cherty LRC 9 Lower Red Green Cherty LRGC 10 Lower Iron Formation LIF

112 Technical Report on the KéMag Pre-Feasibility Study Page 112 Figure 17: Typical Stratigraphic Column of the KéMag Deposit

113 Technical Report on the KéMag Pre-Feasibility Study Page 113 Figure 18: Location Map of KeMag Drill Holes per year

114 Technical Report on the KéMag Pre-Feasibility Study Page BASIC STATISTICS OF SAMPLE DATA, PER SEAM, PER CUT-OFFS Following is a table (Table 44) presenting the basic statistics per seam, per DTWR cut-off. Cut-off DTWR Fe (head ) %DTW R Table 44: Basic Statistics of Sample Data per Seam per Cut-Off Fe (conc. ) SiO2 (conc. ) Min Max Avg Min Max Avg Min Max Avg Min Max Avg Coun Seam t LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC

115 Technical Report on the KéMag Pre-Feasibility Study Page LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC LC JUIF GC URC PGC LRC LRGC 17.3 GEOLOGICAL INTERPRETATION The LabMag deposit is composed of a series of strata or seams slightly dipping (6 ) to the north-east. The seams lie flat, without significant deformation. For the purpose of resource modelling, each seam is modelled separately. The surface contacts between the seams are constructed from the lithological information available in the drill holes. In each hole, the elevation of each lithological contact is derived from the geological interpretation. In each cross-section and for each seam, the contact points are linked together to form contact lines. These lines are further extrapolated at both ends of the section at an angle of 6 to cover the lateral extent of the resource model. The contact lines from all the interpreted cross-sections are then combined together into a triangulated surface, one for each seam contact. The following 4 figures (Figure 19, 20, 21, 22) present the geological interpretation on cross-sections. The following layers are considered mineralized: LC, JUIF, GC, URC, PGC, LRC, LRGC. Hence, the MS and LIF layers are considered barren and no resources come from them. We have hence limited the depth of the deposit to the contact between LRGC and LIF.

116 Technical Report on the KéMag Pre-Feasibility Study Page 116 Figure 19: Geological Interpretation on Section 2 Figure 20: Geological Interpretation on Section 6

117 Technical Report on the KéMag Pre-Feasibility Study Page 117 Figure 21: Geological Interpretation on Section 26 Figure 22: Geological Interpretation on Section 27 In order to cover the totality of the lateral extent of the deposit, we have extrapolated the geological interpretation of the first and last cross-sections. In fact, we have duplicated these cross-sections at a distance of 250 meters at both ends. In addition to the vertical limits of the deposit imposed by the seam layout, a lateral outline has been laid out to limit the extent of the deposit. The lateral outline extends around the drill holes to a maximum of 500m as shown in Figure 23. The colored areas (red, green and blue) form the total lateral extent of the resource model.

118 Technical Report on the KéMag Pre-Feasibility Study Page 118 Figure 23: Lateral Extent of the KeMag Resource Model 17.4 COMPOSITING The original samples vary in length. In order to carry out statistical analyses, it is important to regularize the sample lengths so that each sample has an equivalent representativity. This process is called compositing. We have composited the assays into composites 3 meters in length. Regular down-the-hole compositing was used. Moreover, during this process, no blending between seams occurred. In fact, we have composited the samples, seam per seam. Obviously, the last seam composite never reaches 3 meters in length. As a rule, we discarded all composites that did not contain at least 1.5m of assays to preserve a relative constant representativity. Basic statistics have been calculated for each element and for each seam. Following is a table (Table 45) presenting the statistics of the 3m composites.

119 Technical Report on the KéMag Pre-Feasibility Study Page 119 Table 45: Basic Statistics of 3m Composites per Seam DTWR Number Minimum Maximum Average Seam LC JUIF GC URC PGC LRC LRGC Fe (head) Number Minimum Maximum Average Seam LC JUIF GC URC PGC LRC LRGC Fe (conc.) Number Minimum Maximum Average Seam LC JUIF GC URC PGC LRC LRGC SiO2 (conc.) Number Minimum Maximum Average Seam LC JUIF GC URC PGC LRC LRGC 17.5 BLOCK MODEL CONSTRUCTION The deposit s resources are estimated using a block modelling method. Basically, this method consists in filling the space within the seams with rectangular blocks laid out on a regular grid oriented along the principal axis of the deposit. Each block is assigned grades by interpolation from the surrounding composites. For the purpose of this study, KéMag has been interpolated using Inverse Distance interpolation. The study of the LabMag deposit by Geostat where geostatistical methods were compared to Inverse Distance interpolation showed that the latter gave results very similar to kriging and that Inverse Distance interpolation was

120 Technical Report on the KéMag Pre-Feasibility Study Page 120 appropriate. Considering the similarities between KéMag and LabMag, Geostat elected to use Inverse Distance interpolation BLOCK MODEL GEOMETRY The block grid has been established to cover the entire deposit. The deposit lies oblique to the UTM coordinate system used. We have defined a local grid system so that the grid north is oriented along the seams strike direction as shown in Figure 24. Origin of the local grid system: 6,109, N, 597, E Local grid orientation: Azimuth 330. The local grid north is oriented along UTM azimuth 330. This origin has been chosen in order to coincide with the drill hole 06HL1040D. Block size: 25m across strike, 50m along strike, 13m vertical In the local coordinate system, the block grid parameters are as shown in Table 46: Table 46: Block Grid Geometry Parameters Local Local Z East North Origin Min. extent KéMag 6 Max extent 0 Block size Note: the coordinates above refer to block centroids Figure 24: Block Grid Geometry and Orientation

121 Technical Report on the KéMag Pre-Feasibility Study Page 121 Note that in the vertical direction, the origin is located at the top of the grid and level numbers increase downward. The block grid origin coincides with the center of block (1,1,1) INTERPOLATION PARAMETERS Each block of the block grid is interpolated from the surrounding composites. A search ellipse method is used to define the neighbourhood within which the composites are selected to interpolate a block. A single search ellipse has been used for all the variables in all the seams. The search ellipse used is as follows: Ellipse: 700m by 700m dipping 6 toward the local east or to the UTM northeast. SEARCH RESTRICTIONS In addition to the search ellipse, we impose a series of restriction to the composite selection to optimize the interpolation. We limit the number of composites inside the ellipse to 12. Only the 12 composites closest to the block center are used. A block is interpolated if a minimum of 1 composite is found in the ellipse. Moreover, a maximum of 4 composites per drill hole is used. Also, an octant search method is applied to the ellipse. We limit the number of composites in any one octant to 4. No restriction is applied to the minimum number of octants with at least 1 composite. The block model has been interpolated by Inverse Distance. All elements have been interpolated with a power of 1 applied to distance weighting. Moreover, the distance used is distorted by the search ellipse axis ratios. Each seam is interpolated independently from one another. To interpolate a seam, only the composites belonging to that seam are used RESOURCE ESTIMATION The resources are simply the accumulation of those block volumes and tonnes with their corresponding average grades SPECIFIC GRAVITY USED A series of specific gravity (SG) measurements have been done at the MRC laboratory on selected KéMag drill core samples. NML has elected to average the SG data per seam and use these average values to derive tonnes from volumes. It is interesting to compare the specific gravity values used at KéMag and those used at LabMag. Since no new measurements are available from the 2008 drilling campaign, the specific gravity data used in the May 2008 study are used as shown in Table 47: Table 47: Specific Gravity used at KéMag, per Seam Seam KéMag KéMag Difference 2006 LC % JUIF % GC %

122 Technical Report on the KéMag Pre-Feasibility Study Page 122 URC % PGC % LRC % LRGC % LabMag supplied Geostat with the specific gravity data to use. Geostat has not verified them RESOURCE CLASSIFICATION Resource classification is an exercise by which the resources are assigned a relative quality. One can intuitively assume that in an orebody, the resources are not equally estimated. There are areas where the uncertainty is greater than in others and most of the time uncertainty is intimately related to drilling density. Areas densely drilled are usually better known than areas with sparse drilling. The experience gained by Geostat from the LabMag resource estimation is used on KéMag due to the fact that it is considered that both deposits are similar and an extension of each other. Both deposits feature the same stratigraphy, structure, orientation and dip. As can be seen from the assay values, the iron content is similar in both deposits as is the specific gravity. Hence Geostat considers that it is reasonable to apply to KéMag the classification scheme used on LabMag. Based on LabMag knowledge, Geostat considers that where a drill hole intersects the iron formation, a classification on indicated material is applied provided that a spacing of 500m separated the drill holes. As can be seen from the drill layout, an area of the drilling pattern has drill holes on a tighter grid. Geostat has classified this area as measured. As was done on LabMag, a fringe of inferred material has been added all around the drill hole layout since it is reasonable to expect that the iron formation extends beyond the limits of the drilling. A 250m fringe was added, as shown in Figure 25. Geostat uses the terminology defined in National Instrument as prescribed by the Canadian Institute of Mining and Metallurgy. The definitions of Measured, Indicated and Inferred resources are as follows: MEASURED MINERAL RESOURCE A Measured Mineral Resource is that part of a Mineral Resource for which quantity, grade or quality, densities, shape, physical characteristics are so well established that they can be estimated with confidence sufficient to allow the appropriate application of technical and economic parameters, to support production planning and evaluation of the economic viability of the deposit. The estimate is based on detailed and reliable exploration, sampling and testing information gathered through appropriate techniques from locations such as outcrops, trenches, pits, workings and drill holes that are spaced closely enough to confirm both geological and grade continuity. INDICATED MINERAL RESOURCE

123 Technical Report on the KéMag Pre-Feasibility Study Page 123 An Indicated Mineral Resource is that part of a Mineral Resource for which quantity, grade or quality, densities, shape and physical characteristics can be estimated with a level of confidence sufficient to allow the appropriate application of technical and economic parameters, to support mine planning and evaluation of the economic viability of the deposit. The estimate is based on detailed and reliable exploration and testing information gathered through appropriate techniques from locations such as outcrops, trenches, pits, workings and drill holes that are spaced closely enough for geological and grade continuity to be reasonably assumed. INFERRED MINERAL RESOURCE An Inferred Mineral Resource is that part of a Mineral Resource for which quantity and grade or quality can be estimated on the basis of geological evidence and limited sampling and reasonably assumed, but not verified, geological and grade continuity. The estimate is based on limited information and sampling gathered through appropriate techniques from locations such as outcrops, trenches, pits, workings and drill holes.

124 Technical Report on the KéMag Pre-Feasibility Study Page 124 Figure 25: Outline of Resource Classification used at KeMag In red: Measured resources In green: Indicated resources In blue: Inferred resources CLASSIFIED RESOURCES Table 48 presents the resources for a variety of DTWR cut-off grades. The per-seam resources presented in Table 49 are the accumulation of block tonnages above given DTWR cut-offs in each seam, as if each seam could be considered independently. No dilution is taken into account. Table 48: Classified Mineral Resources by DTWR cut-off grade Intervals DTWR Cut-off Fe Conc. (%) Total Tonnage(Mt) DTWR (%) Fe Head (%) (%) Global Measured Mineral Resources SiO2 (%) Conc.

125 Technical Report on the KéMag Pre-Feasibility Study Page , , , , , Global Indicated Mineral Resources 0 1, , Global Measured + Indicated Resources 0 2, , , , , , Global Inferred Resources 0 1, , , , Table 49: Classified Mineral resources per seam at 18% DTWR cut-off Total Tonnage Fe Conc. SiO2 Seam (Mt) DTWR (%) Fe Head (%) (%) (%) Measured Mineral Resources per Seam at 18% DTWR cut-off LC JUIF GC URC PGC LRC LRGC Indicated Mineral Resources per Seam at 18% DTWR cut-off Conc.

126 Technical Report on the KéMag Pre-Feasibility Study Page 126 LC JUIF GC URC PGC LRC LRGC Measured + Indicated Mineral Resources per Seam at 18% DTWR cut-off LC JUIF GC URC PGC LRC LRGC Inferred Mineral Resources per Seam at 18% DTWR cut-off LC JUIF GC URC PGC LRC LRGC

127 Technical Report on the KéMag Pre-Feasibility Study Page MINE DESIGN 18.1 GENERAL The PFS is based on a large-scale open pit operation utilizing shovels and trucks to move an average of 76 million tonnes of ore and 30 millions tonnes of waste per year over a 25 year mine life. The reserves were determined from the mineral resources in the ore body block model developed as described in the section on mineral resources. The development of the reserves was carried out using a computerized pit optimization method based on the Lerchs-Grossman 3D ( LG 3D ) algorithm. The LG 3D algorithm is a true pit optimizer used for determining the optimal ultimate limit for open-pit mines PIT OPTIMIZATION AND MINE DESIGN The Lerchs-Grossman 3D ( LG 3D ) pit optimization algorithm in the MineSight software was used to develop the configuration of the open pit at the end of its economic life, based on the total of estimated Measured and Indicated resources. The LG 3D algorithm is a true pit optimizer based on the graph theory in operations research. It operates on a net value calculation, i.e. concentrate sales revenue less operating costs, as follows: Sales Revenue = Weight Recovery (%) x Concentrate Grade (Fe%) x Sales Price ($/dmtu) Operating Cost = Mining + Stripping + Concentrator + Transport and Shipping + General & Administrative costs PIT OPTIMIZATION CRITERIA AND PARAMETERS It is a requirement of the Canadian NI Standards of Disclosure for Mineral Projects that only ore blocks classified in the Measured and Indicated categories can be used to drive the pit optimizer. Therefore, the fringe of Inferred resource blocks that had been added to the layout of Measured and Indicated blocks was not used in the pit optimization exercise, which involved parameters derived from an internal cost database and information from the LIOP. The main technical and economic input parameters used are presented in Table 50, as well as other parameters such as the exchange rate, concentrate price and pit slope angle. The costs were compared with those of similar iron ore operations and are considered acceptable.

128 Technical Report on the KéMag Pre-Feasibility Study Page 128 Table 50: Pit Optimization Parameters for LG 3D Economic Data Item Exchange Rate US$/Can$ 0.85 Price of Concentrate US$/dmtu 0.85 Price of 69.1% Fe Concentrate $49.92 per tonne Operating Cost Cost (Can$/t) Mine Run of Mine 1.75 Mine and Concentrator Site Services Run of Mine 0.12 Concentrator Run of Mine 2.80 Pipeline Run of Mine 0.36 Filter Plant Run of Mine 0.18 Stocking and Ship Loading Run of Mine 0.26 Pointe Noire Site Services Run of Mine 0.10 Administration Run of Mine 0.08 Other Run of Mine 0.13 Metallurgy Weight Recovery % In Block Model Pit Parameters Overall Pit Slope Degree CUT-OFF GRADE CALCULATION An economic cut-off grade based on several economic parameters such as iron ore price, costs of mining, processing, transport, stacking and reclaiming and ship loading was used to categorize the material inside the pit design as ore or waste. Using the operating costs and sales price presented in Table 51 and a profit margin of $1.00 per tonne of ROM, the DTWR cut-off grade for KéMag material was calculated to be 15%. For the purpose of estimating the KéMag mineable resources, a cut-off grade of 18% DTWR was used, consistent with previous work on LIOP MINE DESIGN AND MINERAL RESERVE LIFE OF MINE RESERVES Based on a material definition exercise using geological data available to-date, the life-of-mine ( LOM ) reserves for the ultimate pit design for KéMag were calculated and classified in the Proven and Probable categories in accordance with the criteria of the Canadian National Instrument ( NI ) for Standards of Disclosure of Mineral Project, of February 2001 and the classifications adopted by the CIM Council of August The Table 51 provides a summary of the LOM reserve for the KéMag project using a cut-off grade of 18% DTWR.

129 Technical Report on the KéMag Pre-Feasibility Study Page 129 Reserve Category Table 51: LOM Reserve for the KéMag Project (Cut-off grade of 18% DTWR) Tonnes DTWR Crude Fe Concentrate (million) (%) (%) Fe(%) SiO 2 (%) Proven Probable Proven + Probable Inferred 73 Waste Inferred+Waste Waste : Ore Ratio PIT DESIGN The detailed pit design work was carried out using the optimum pit outline obtained from the LG 3D algorithm as a guide. The detailed mine design took into account the following parameters: HAUL ROAD All in-pit ramps will be 35m wide to accommodate 290-tonne off-highway class rear-dump trucks. This ramp width is sufficient to support the two-way traffic system required to maintain an uninterrupted haulage cycle. Based on hauling experience from other mines operating under similar freezing conditions in winter, the maximum recommended ramp gradient will be 8% for all mine roads. Temporary ramps will be used in the early years of mine operations to shorten haulage distances to the primary crushers or to the waste dumps. As the pit depth increases with time, a final ramp will be developed to access the ore located at lower elevations. GEOTECHNICAL PIT SLOPE Based on the geometry of the deposit, the pit wall slope was divided into two sectors as follows: In the footwall, the west side, the designed pit slope will be approximately 6 with single benching, equivalent to the dip of the geological formations; In the hanging wall, the east side, an overall slope angle of 50 was used for the pit design based on experience gained in the past from operating iron ore mines in the Schefferville area. Single benching will be adopted in ore to maximize the recovery of the mineral resources and double-benching will be adopted in waste to minimize the amount of waste. Further analysis on pit slope stability in the hanging-wall will be undertaken to determine whether this is the optimal configuration with respect to rock structures and properties.

130 Technical Report on the KéMag Pre-Feasibility Study Page 130 DILUTION The block model incorporated some dilution in the 13m bench compositing process where sub-economic assays intersections were used in the block grade interpolation. Therefore, no additional dilution factor was added to the reserve estimate. MINING RECOVERY Mining recovery was assumed to be 100% for the mining schedule. A 3D view of the pit design is presented in Figure 26.

131 Technical Report on the KéMag Pre-Feasibility Study Page 131 Figure 26: 3D View of the of the Detail Pit Design

132 Technical Report on the KéMag Pre-Feasibility Study Page MINE PLANNING MINING METHOD The KéMag Deposit will be mined using conventional open pit mining methods based on truck/shovel operations. The rock will be drilled, blasted and loaded into haul trucks that will deliver run-of-mine ( ROM ) ore to the primary crushers located near the final pit edge half way along the strike length of the deposit. By condemning an insignificant amount of potential mineral resource in the outcrop area, the primary crushers were positioned so as to minimize haulage distances ANNUAL PRODUCTION REQUIREMENT The mine will be developed to support a nominal capacity of iron concentrate of 14.8 million tonnes per year ( mtpy ) in Year 1, tpy during Year 2 and 21.2 tpy when in full production in Year 3 and thereafter. Out of a total of 21.2 million tonnes of concentrate, 5 million tonnes DR grade pellets and 10 million tonnes of BF pellets will be produced. Based on the results of metallurgical testing and geological block modelling work, the average Davis Tube Weight Recovery ( DTWR ) was determined to be 27.0 %, equivalent to a concentration ratio of 3.71 tonnes of ROM per tonne of concentrate. This resulted in an estimated average feed requirement of 78.6 mtpy of crude ore to produce 21.2 mtpy of concentrate BLENDING As described in the preceding Section 17, the KéMag Deposit contains seven economic layers of iron ore, each layer having different physical and chemical characteristics. Blending of material from the different layers in the pit will be carried out in order to: Ensure that concentrate produced contains less than 3% SiO 2 ; Optimize the use of high weight recovery ore to maximize the concentration ratio and optimize the Project s net present value; Feed the mill with a constant ore hardness to facilitate stable process plant operation MINING PLAN The detailed pit is designed for mining the total of the Measured and Indicated Resources. As the top of the orebody is near to the surface, little pre-production stripping will be required in preparation for mining. However, sufficient waste rock will be mined during the pre-production period to build the tailings dam, ROM pad and other general construction work such as site access roads. Generally, the most easily accessible areas of the deposit will be scheduled for development in the initial years as this will maximize cash flow and minimize the stripping ratio. However, some less easily accessible areas may be mined in the initial phase to meet blending requirement. Pit staging using pushback technique will be

133 Technical Report on the KéMag Pre-Feasibility Study Page 133 used to accommodate the increased stripping work required with the deepening of the pit on the northeast side. As a general rule, the stripping ratio will gradually increase until years 9 to 11, after which time stripping will be decreased in order to minimize operating costs associated with waste mining. Some minor clearing of organic material from the surface will be required and the material will be stored for future reclamation and land restoration purposes. For blending purposes, it is likely that three to four production faces will be developed in ore at any one time on two or three benches. A 25-year mine production schedule was prepared, on an annual basis for the first five years and by 3-year increments in from Year 6 to Year 20 and a 5-year increment for Years 21 to 25. Table 52 summarizes the 25-year plan in terms of ore feed with Fe% of the crude ROM, and concentrate production, DTWR% and SiO 2 content. The total waste mined, total material moved and stripping ratio is also provided. Table 52: Summary of Planned 25-year Production Schedule ROM Concentrate Waste Total Material Moved Waste :Or Year Tonnes Fe Tonnes DTWR SiO 2 Tonnes Tonnes Tonnes/Year e Stripping Ratio ('000) (%) ('000) (%) (%) ('000) ('000) ('000) (t:t) P-P 2, ,111 16,165 16,165 - Year 1 48, , ,659 59,092 59, Year 2 66, , ,254 82,342 82, Year 3 77, , , , , Year 4 75, , , , , Year 5 75, , , , , Years , , , , , Years , , , , , Years , , , , , Years , , , , , Years , , , , , Years , , , ,184 99, Total 1,937, , ,987 2,683, , DUMP DESIGN Dumps that will serve as storage for material that is currently considered to be sub-economical will be located to the west of the pit, in non-mineralized areas to be defined based on the results of condemnation drilling yet to be carried out. The waste mined will be hauled to the closest accessible dump in order to minimize costs. As shown on the Layout Drawing presented on Figure 27, four dumps will be created: An overburden dump, containing surface material for later use in mine reclamation;

134 Technical Report on the KéMag Pre-Feasibility Study Page 134 Three waste dumps, containing in-pit materials which are non-economical (<18% DTWR). Using a compacted swell factor of 25%, rock dumps were designed to hold the waste for the 25-year mine plan, taking into consideration the material to be used for the tailings dam and site construction. The waste dumps will be progressively rehabilitated as their individual capacities are reached. The main design parameters for the overburden and waste dumps were as follows: Face angle of 35º; Overall slope of 27º; Bench height of 20m; Berm width of 10m. Potential in-pit back-fill will be reviewed at the Feasibility Study level.

135 Technical Report on the KéMag Pre-Feasibility Study Page 135 Figure 27: Ultimate Pit Design and Facilities

136 Technical Report on the KéMag Pre-Feasibility Study Page MINE OPERATIONS AND EQUIPMENT REQUIREMENT The sizing and the quantity of the mining equipment required for the development of the KéMag project were based on the following: The production schedule shown on Table 52; The measurement of the annual haul profiles; The physical and metallurgical characteristics of the various layers of the orebody for ore blending requirement and operational flexibility WORK SCHEDULE The mine will operate two twelve-hour shifts per day, seven days per week, and 365 days per year. Crews will rotate on a fly-in/fly-out basis, with two weeks in and two weeks out. The operating hours used was based on the schedule presented in Table 53. Table 53: Estimated Operating Hours Operating Time per Shift (Minutes) Schedule/shift (12 hrs) 720 Scheduled Non-productive Time Start-up (shift change) 5 Inspection 15 Coffee Break 15 Lunch 30 Net Scheduled Operating Time 655 Job Efficiency (83% or 50min/hr) Net Productive Operating Time (9.10 hours) DRILLING & BLASTING Both ore and waste will be drilled using 15 (381 mm) diameter holes with a drilling pattern of 9.5 m x 9.5 m. Drilling will be done on a single pass on 13 m bench height and 1.5 m sub-drilling. The drill pattern will be reviewed as more data on blast fragmentation and rock characteristics become available.

137 Technical Report on the KéMag Pre-Feasibility Study Page 137 Blasting will be executed under contract with an explosives supplier that will store and provide all the blasting materials and technology required by the mine. It has been assumed that the explosives supplier will also provide a down-the-hole service. Due to the large volume of explosives required and the remote nature of the operation, a bulk explosives plant will be constructed on site. Emulsion will be used for blasting and preliminary blasting simulation and fragmentation analysis indicates that an average powder factor of 0.40 kg/t for ore and waste will achieve a fragmentation of 95% passing 150 cm, which is adequate for the size of 60 x109 primary crusher and rope shovels selected for the KéMag mining operation LOADING & HAULING The shovel and truck fleet was selected to optimize the loading and hauling operations. Due to their proven reliability in similar applications, diesel-electric haul trucks and electric cable shovels were selected for the base case. The principal loading and hauling equipment will be as follows: Electric shovel (33m 3 ); Front end loader (27m 3 ); Haul truck (290 tonnes). In estimating the quantity of equipment required, the average haul distances to the primary crusher and dumps for ore and waste, respectively, were measured for each year using the mining plan. The major mine equipment requirements during the life of the mine, based on the factors presented above, are summarized in Table 54. Table 54: Major Mine Equipment Requirement Year Equipment Type Electric Drill (381mm) Cable Shovel (33m 3 ) Haulage Truck(290 t) Wheeled Loader(27 m 3 ) Wheeled Dozer(627HP) Tracked Dozer (850HP) Grader (16 ft) Water Truck (30,000 gallon)

138 Technical Report on the KéMag Pre-Feasibility Study Page OTHER RELEVANT DATA AND INFORMATION 19.1 INFRASTRUCTURE AND SUPPORT SYSTEMS MINE AND CONCENTRATOR ACCESS ROAD The 42 km upgraded road from Schefferville will serve not only the mine but also the concentrator, camp and associated infrastructure. SITE ROADS AND DRAINAGE Site roads will be 9 m wide, including a 1.5 m shoulder for drainage. The roads will be built using operations equipment during development of the mine and will consist of a compacted base covered with sized gravel. A surface water management system will be developed on the site to channel run-off in areas where it could become contaminated. A simple design involving protection trenches around the perimeter of waste evacuation areas, with low points directed towards natural drainage basins, and retention dikes for concentrator waste will be sufficient. Collected water will pass through a treatment plant prior to being returned to the process water reservoir. A storm drainage system will be created that will exploit the natural drainage on the site with open ditches and culverts that will connect with a clarification pond at the upstream end of Lac Gillespie, and buildings will be located on existing natural rises to avoid any risk of damage by flooding. In specific places in treatment facilities and storage areas for reagents and fuel, a hard coating and containment embankments will be constructed as required. HELICOPTER PAD AND HANGAR An illuminated landing pad, hangar and refuelling facilities will be provided for the helicopter that will service the plant, be used to inspect the pipeline and be available for emergency evacuation from the Lake Harris site. It is assumed that the helicopter will be leased and the owner will provide all servicing and repair at its own facilities. GATEHOUSE It is planned that a gatehouse will control pedestrian and vehicle access to the mine and concentrator site. The gatehouse will operate around the clock and access to facilities will generally be controlled with a manual barrier. The building will include a waiting room and washrooms, and visitor parking will be provided inside the barrier. A weighbridge will be located next to the roadway.

139 Technical Report on the KéMag Pre-Feasibility Study Page 139 GARAGE, WORKSHOPS, WAREHOUSE AND ADMINISTRATION OFFICES A single insulated, heated and air-conditioned building adjacent to the concentrator will contain a number of different but centralized facilities: A garage for the maintenance of mining and other mobile equipment. It will have four bays for mine trucks, two bays for loaders and bulldozers and other auxiliary mobile equipment, one for changing tires and one for servicing light vehicles. The garage will be 50' (15m) high to accommodate two 50-tonne overhead bridge cranes and a 15-tonne overhead crane underneath which a truck can fully raise its dump box. It will be fully equipped with air compressors, welding sets, tools and equipment for changing tires and an area will be provided for storage of used lubricants; Mechanical and electrical workshops and a conveyor belt repair shop, to serve both the mine and the concentrator; A warehouse fitted with shelving and storage bins for items that need to be in a sheltered, secure place. In addition to the closed warehouse, there will be an outdoor concrete platform for containers and a fenced area for storing tires and other supplies that can be left in the open. A cold warehouse will also be available to store material that needs to be protected from rain but does not require being in a heated building; Garages for a fire tender and an ambulance and offices for the plant security and fire protection departments; Separate change rooms, showers and toilets for men and women and a lunch room equipped with refrigerators, microwave ovens and coffee facilities; On the upper floor, served by an elevator as well as staircases, administration offices housing the Mine, Concentrator and Site Superintendents and their staff. The facilities will include conference rooms, a canteen that will also be used for training, open office areas for clerical staff, space for computer facilities, map files and printing, and a vault for archives. TEST LABORATORY A test laboratory will be built for analysis of exploration samples and production samples from the mine and concentrator. Offices will be provided for the laboratory supervisor and assistants, and the building will also include the following main rooms: A sample preparation room equipped with a laboratory-sized crusher, a pulverizing mill, a dust collector and an air compressor;

140 Technical Report on the KéMag Pre-Feasibility Study Page 140 An atomic absorption room equipped with an atomic-absorption spectrometer and XRF analyzer; A properly ventilated fire assay lab with a pyroanalysis furnace and an oven for carbon calcinations; A wet laboratory; A precision scale room; A metallurgical laboratory; A library. COMMUNICATIONS The following systems are planned for the mine and concentrator: A telephone system, connected by land line to existing telecommunication facilities in Schefferville, will provide voice, fax and data transmission facilities; Hand-held closed-circuit radios operating on three separate wavelengths will provide a local communication system for operations, security and maintenance staff; A local computer network, infrastructure and workstations will provide access to the Internet; The fibre-optic system of the slurry pipeline will connect the Lake Harris site communication system with that of the Pointe-Noire pellet plant and product storage and ship loading facilities. CAMP Living facilities will be provided for contractors employees during construction, after which the buildings will be used by mine, crusher and concentrator operations and maintenance personnel and administrative staff, all of whom will be working on a two weeks in and two weeks out, fly-in/fly-out basis. The insulated and heated facilities will include: Dormitories able to accommodate some 1,700 people during construction and some 400 people thereafter, slightly more than the maximum number of persons expected to be simultaneously on site; A mess hall block that will contain comfortable eating areas, including kitchen areas for food preparation and cooking, food storage, refrigeration and freezing facilities and a delivery dock

141 Technical Report on the KéMag Pre-Feasibility Study Page 141 and service areas. The cash desk will sell newspapers, toiletries, cold drinks, snacks and sundries, and the mess hall will also be available as a lounge between mealtimes; A laundry room with washers, dryers and ironing facilities; A training room, pool tables, other recreational facilities and other services. The camp will be operated by a contractor who will be responsible for providing operating and maintenance personnel and all supplies other than utilities. The contractor will also be responsible for the cleaning of the administration offices and the different buildings on site. TRANSPORTATION A local company will be contracted to provide bus transportation between the Schefferville Airport and the camp. WATER SUPPLY Process Water: For a production level of 21.2 mtpy, the concentrator will have an average annual requirement of some 46 million cubic metres ( Mm 3 ) of water, which it will consume at the rate of about 1.5 m 3 per second based on 8322 operating hours per year for the concentrator and for the pipeline. Of those 46 Mm 3 per year, about 34.3 Mm 3 will be pumped out to the tailings disposal area and can be re-used in part and some 11.4 Mm 3 per year will be pumped in slurry or as water batches down the pipeline to the pellet plant and will not be returned. Principal sources of water for the concentrator will be from water contained in the ore processed; from fresh water used for pump gland seals, equipment cooling, boiler make-up etc; from collected site run-off and the water which drains from the deposited tailings material. As thickeners will be covered, evaporation losses from the concentrator will not be significant. An average of 54.6 million tonnes of tailings will be stored each year, trapping some 6.1 Mm 3 of water, and evapo-transpiration from the basin will average some 4.9 Mm 3 per year. Including the water lost with the concentrate in the pipeline, total water losses will therefore total approximately 22.4 Mm 3 per year, equivalent to 0.75 m 3 per second. In a dry year, total precipitation on the combined area of the tailings basins and the pit will exceed the total system losses by some 700,000 m 3 and in a wet year, the excess will rise to 9.0 Mm 3. A controlled overflow from the clarification ponds will therefore be necessary and the water will be processed through a treatment plant before discharge to Lac Gillespie. Water will be received in a process water reservoir near the concentrator from three different sources:

142 Technical Report on the KéMag Pre-Feasibility Study Page 142 Mine dewatering: Rainfall, snow melt and groundwater in the open pit will accumulate in one or more mine sumps from where it will be removed by portable float-mounted electric submersible pumps (Flygt type) and piped to the process water reservoir. As the pit grows in area and depth, additional pumps will be added and water from the pit will gradually increase, reducing the need for makeup water from a nearby lake. In winter, most of the water in the pit will freeze and pumping will be suspended. Tailings deposition: Water from the tailings containment area will be returned through a pipe to the process water reservoir. Clarification basins at the tailings ponds will hold enough water to supply the needs of the concentrator until the spring snow melt. Pumping stations: The tailings basins and clarification ponds will be built prior to the start-up of the concentrator and water will need to be available to meet plant demand during start-up. Therefore, pump houses will be constructed on Lac Gillespie and Lac du Canoë to provide gland seal, boiler make-up and process make-up water in case of problems with the water recycling system. Each pump house will contain two operating pumps and one stand-by pump, each with a capacity of approximately 6000 usgpm (380 litres per second). Potable Water: A branch off the pipeline carrying lake water to the plant will feed a treatment plant that will provide potable water to the plant and the camp. WASTEWATER AND SEWAGE TREATMENT Domestic wastewater and sewage from all mine, crusher, concentrator and camp sanitary facilities will be piped to a central treatment plant that will be built to service 1,700 people during the construction phase of the Project. The treatment system will be based on the rotating biological contactor technology which produces an effluent in compliance with provincial regulations for discharge in a body of water. The plant effluent will be reused in the process or for toilet flushing. The treatment plant will produce no noticeable odour and can be located relatively near the camp. FIREFIGHTING An independent piping circuit will be connected to the process water reservoir to distribute water for fire fighting. A buried pipe will connect all the buildings, and secondary pipes will be connected to supply particular zones. A certain volume at the bottom of the reservoir will be a dedicated fire reserve. Process water pumps will be connected above this reserve, and fire pumps with electric and diesel motors will be connected to the bottom of the reservoir to provide water to fight fires. The reserve volume and the pumping rates will be defined and approved at the project design stage by the insurers of the complex.

143 Technical Report on the KéMag Pre-Feasibility Study Page 143 Fire hydrants, each equipped with an isolating valve, a riser and a length of pipe, will be located at strategic points. Portable extinguishers will be located as needed in such areas as the laboratory, workshops and offices. A fire tender with a double capacity for water and foam will also be provided. SERVICE VEHICLES Pick-up trucks, an ambulance and a fire tender, snowploughs, water trucks, mobile cranes, forklifts, bulldozers, graders, etc. are provided. DIESEL, LIGHT FUEL OIL AND GASOLINE Diesel fuel oil will be required for emergency power generators, portable pumps, other light equipment and service vehicles. Light fuel oil will be used to supply the boilers that will provide steam to heat the buildings during cold weather and gasoline will be required for light equipment. The use of electricity instead of oil for building heating will be evaluated at the Feasibility stage of the project. Fuel will be transported by rail from Sept-Îles to Schefferville where rail tank cars, each holding 45,000 litres, will empty directly into road tankers of the same capacity. A dedicated rail siding will be required for the purpose of handling petroleum products at Schefferville railhead but no storage facility will be necessary. Fuel will be transported to a storage and filling station located next to the mine dispatch building. On-site storage capacity will be four weeks for each type of petroleum product. The tank farm will meet environmental regulations and will be an island protected by an impervious geotextile and elevated berm to prevent run off in case of spillage. There will be two points on opposite sides of the diesel tank from where fuel will be distributed. One service point will be for the exclusive use of the mine production trucks and will be equipped with high volume 700 litres per minute pumps and quick-connect couplings to prevent spillage. The other service point will be used by all other equipment, including light vehicles, and by the tankers bringing fuel from the railhead. Mine equipment such as dozers and compressors will be re-fuelled in the pit by a fuel truck that will be refilled at the tank farm. It is estimated that diesel fuel consumption will be 750,000 litres per week and two road tankers will be capable of bringing that quantity of diesel to the tank farm in two days. A third road tanker will be dedicated to the transport of light fuel oil for building heating PIPELINE The main pumping station at the concentrator and the receiving station at the pellet plant will be supported by the infrastructure at the Lake Harris and the Pointe-Noire site respectively. Sub-sections through therefore apply only to the booster pumping station near the Gagnon site.

144 Technical Report on the KéMag Pre-Feasibility Study Page 144 COMMUNICATIONS A fibre-optic backbone will carry all pipeline communications, including SCADA data and video surveillance of the booster and valve stations, as well as office data and voice telephone channels between the concentrator and the pellet plant. The multi-fibre optical cable will be installed in a conduit in the pipeline trench. Should maintenance crews be required to work on the pipeline, hand-held closed-circuit radios will provide a local communication system. ACCESS ROAD A gravel road will be constructed that will connect to the existing highway 389. BUILDINGS At the booster pumping station, equipment and facilities will be housed in a 165' (50 m) by 70' (21 m) metal frame building with insulated walls and roof. CAMP Facilities built for construction personnel will be retained for use by operating personnel and by personnel arriving by vehicle or flown in by helicopter for planned maintenance or in the event of a breakdown. WATER SUPPLY A pump house with one operating and one stand-by pump will provide water from a local stream or lake. WASTEWATER AND SEWAGE TREATMENT A septic system will be provided to handle wastewater and sewage. FIREFIGHTING Hand-operated fire extinguishers will be provided. DIESEL AND GASOLINE Fuel for diesel-driven emergency generators, portable pumps and other light equipment at the booster station will be transported by truck via highway 389 from Sept-Îles or Baie-Comeau and stored in tanks that will be in a secure area, set on a concrete slab and surrounded by dikes lined with a geoliner, ensuring a retention capacity of 125% of one tank.

145 Technical Report on the KéMag Pre-Feasibility Study Page 145 CONTROL The entire pipeline system will be controlled and managed by a supervisory control and data acquisition (SCADA) system linking sensors at the main pumping station, the booster station and the slurry receiving station with each other and with computers in the concentrator and pellet plant control rooms POINTE-NOIRE SITE ACCESS ROADS The pellet plant site and stockyard area will be connected by an asphalted road to an existing municipal road. An asphalted road will be constructed to connect the dock office and jetty to a different existing municipal road. SITE ROADS AND DRAINAGE Site roads will be 30' (9 m) wide, including a 5' (1.5 m) shoulder for drainage. The roads will consist of a compacted base covered with sized gravel. A gravel road will be constructed to provide access along the length of the covered conveyor from the stockyard to the ship loading jetty. A storm drainage system will be created that will exploit the natural drainage on the site with open ditches and culverts, and buildings will be located on existing natural rises to avoid any risk of flooding in these areas. A surface water management system will be developed on the site to channel run-off in areas where it would otherwise become contaminated. A simple design involving protection trenches around the perimeter of the plant, with retention dikes to contain solid waste and overflow points directed towards natural drainage basins will be sufficient. In specific places in treatment facilities and storage areas for bentonite and fluxes and fuel, a hard pad will be provided and containment embankments will be erected as required. FENCES AND GATEHOUSES The pellet plant site, the stockyard, and the dock office and jetty, will be surrounded by security fences, with gatehouses with manual and card-operated barriers that will control pedestrian and vehicle access. Gatehouses will include a waiting room and washrooms, and visitor parking will be provided inside the barrier. Weighbridges will be located next to the access roadways to the pellet plant and stockyard.

146 Technical Report on the KéMag Pre-Feasibility Study Page 146 BUILDINGS AND OFFICES The following facilities will be within or in close proximity to the pellet plant: Employee change-rooms, washrooms and toilets, canteen, offices, fire brigade and first aid stations; General warehouse; Maintenance and pallet repair workshops; LABORATORY. At the landward end of the jetty, a 20 m by 40 m dock office building will provide space for a vessel agent, communications and security personnel, storage for emergency materials handling spares, spares for navigation aids and moorings, and toilet facilities. COMMUNICATIONS A telephone system serving not only the pellet plant but also the pellet handling, storage and ship loading equipment, and the dock office, will be connected to existing telecommunication facilities in the Sept-Îles area and will provide voice, fax and data transmission facilities. The PABX telephone equipment at the pellet plant will also be connected to similar equipment at the concentrator by the fibre-optic cable that will be installed in a conduit in the pipeline trench, as described in Section Hand-held closed-circuit radios operating on three separate wavelengths will provide a local communication system for operations, security and maintenance staff. WATER SUPPLY Process Water: Clarified water from the thickeners will be received in a process water reservoir near the pellet plant, from which it will be recycled through a separate pumping system for use in plant clean-up, dust collection and sump dilution. Excess water will be further clarified and polished and the cleaned water will be piped to the Bay of Sept-Îles or made available to local industries. Water for dust suppression sprays in the stockyard and loadout will be pumped from the process water reservoir at the pellet plant. Potable Water: Connection will be made to the 8 diameter pipe that delivers potable water from Lac des Rapides to existing Wabush and Alouette installations in the Pointe-Noire area. The water will be used by personnel, for gland and compressor seals and equipment cooling, as make-up to the process system during start-ups and as make-up for dust suppression sprays. Wherever possible, enclosed systems will be installed to reduce fresh water requirements. Bulk Water for Ships:

147 Technical Report on the KéMag Pre-Feasibility Study Page 147 Bulk water will be made available to ships at the loading berth and the local bunkering company serving the berth will also provide the required storage tanks, piping, pumps, meters and other equipment. WASTEWATER AND SEWAGE TREATMENT Wastewater and sewage from all the sanitary facilities in the various areas of the pellet plant, product stockyard and product handling and ship loading areas will be pumped to an on-site treatment plant. As at the Lake Harris site, the treatment system will be based on the rotating biological contactor technology which produces an effluent in compliance with provincial Environmental Control Water and Sewage Regulations for discharge in a body of water. The plant effluent will be reused in the process or for toilet flushing. The treatment plant will produce no noticeable odour and can be located relatively near a residential area. FIREFIGHTING An independent piping circuit will be connected to the process water reservoir to distribute water for fire fighting. A buried pipe loop will connect all buildings, and secondary pipes will be connected to supply particular zones. A certain volume at the bottom of the process water reservoir will be a dedicated fire reserve. Process water pumps will be connected above this reserve, and pumps with electric and diesel motors will be connected to the bottom of the reservoir to provide water to fight fires. The reserve volume and the pumping rates will be defined and approved at the project design stage by the insurers of the plant. Fire hydrants, each equipped with an isolating valve, a riser and a length of pipe, will be located at strategic points. Portable extinguishers will be located as needed in such areas as the laboratory, workshops and offices. A fire tender with a double capacity for water and foam will also be provided. SERVICE VEHICLES Pick-up trucks, an ambulance and a fire tender, snowploughs, mobile cranes, forklifts, bulldozers, graders, etc. will be provided. BUNKER C, DIESEL FUEL OIL AND PROPANE The induration process will operate with Bunker C fuel oil and the bentonite plant, emergency power generators, furnace pilot burners and service vehicles will operate on diesel. Therefore, two fuel handling systems will be installed that will include the reception, storage, filtering and distribution of fuels. The Bunker C system will feed the induration machine burner and a steam plant supplying steam for slurry heating before filtration and for general building heating. All fuel will be delivered to the pellet plant in rail tank cars by a local supplier who will also provide storage tanks in a secure area, on concrete slabs and surrounded by dikes lined with a geoliner,

148 Technical Report on the KéMag Pre-Feasibility Study Page 148 ensuring a retention capacity of 125% of the largest tank. Propane required for the burners during cold start up of the machine will be fed from propane tanks provided and refuelled by a local supplier. Bunker fuel will be made available to ships at the loading berth and the local bunkering company serving the berth will also provide the required storage tanks, piping, pumps, meters and other equipment. The tanks will be in a secure area, set on concrete slabs and surrounded by dikes lined with geoliner, ensuring a retention capacity of 125% of one tank ELECTRICAL POWER SUPPLY MINE AND CONCENTRATOR The estimated electrical power required at the site for the mine, crushers, concentrator, slurry pumping station and associated infrastructure totals some 290 MW. Due to the large proportion of Variable Frequency Drive ( VFD ) fed motors and synchronous motors in the Project load, the overall power factor will be very high at The project MVA requirement is estimated to be 305 MVA. No additional power factor compensation equipment will be required. The nearest source of power in Québec is the Brisay hydro-electric power station, some 270 km from the Lake Harris site, and Hydro-Québec has confirmed that the required 305 MVA can be made available for the Project. To provide power from the Brisay power station to the Lake Harris site, the traditional method of transmission based on the use of overhead lines was adopted for the purpose of this PFS. Based on a review of the topography, it was considered that the best route for an overhead power line between Brisay and the Lake Harris site is some 270 km long. To transmit 235 MW across such a distance will require a 315 kv single-circuit or 230 kv double-circuit, AC transmission line. Given that power is available in the Brisay substation at 315 kv, a single-circuit line with a bundle of two Besfort sub-conductors per phase will provide an acceptable voltage regulation for the site loads across this distance. The power meter will be installed at the load end. BRISAY SUBSTATION Two breaker bays with connections to the substation 315 kv tubular busbars and to both circuits of the transmission line will be added. LAKE HARRIS SUBSTATION Six 315 kv breakers will ensure good protection, personnel safety and ease of operation. Continuity of operation will be sustained in case of maintenance to, or failure of, one transformer. A prefabricated electrical room will to house protection relays and racks, batteries and rectifiers, as well as monitoring equipment.

149 Technical Report on the KéMag Pre-Feasibility Study Page 149 MAIN ELECTRICAL ROOM The substation transformer secondary windings will be connected by cables to the main electrical room which will also house the following equipment: 34.5 kv switchgear - Power at this level will be distributed over the whole facility from this location; 13.8 kv switchgear - This will supply the ball mill motors and the large VFD step-down transformers. Given the level of power, it was considered convenient to group the distribution by two sets of three lines; 4.16 kv switchgear - This will is used to feed motors which are uneconomical at 13.2 kv and yet are large enough to be fed at medium voltage; 34.5 kv Load Interrupter switchgear - This will be used to feed a number of 2 MVA transformers with 600 V secondary voltage. MINE PIT Two 34.5 kv overhead lines along the southwest edge of the deposit will bring power to the mining equipment and, on the way, will also serve various facilities such as the tank farm, mine dispatch and explosive warehouse. These lines will have sectionalizing switches in at least six locations to isolate faults when they occur, such as during dynamiting, and to ensure continuity of power. There will be two movable mining substations, each with a 34.5 kv circuit breaker feeding a 34.5/13.8 kv, 5 MVA transformer and various 13.8 kv feeder breakers. Switch houses, some with transformers to step-down to 600V, elements to quickly build a 3.8 kv overhead line, SHD-GC mining cable and plugs will also be provided. Substations and switch houses will be mounted on skids or blocks and, together with overhead lines, will be moved around to best serve mining operations. PRIMARY CRUSHERS Two 34.5 kv feeders from the main electrical room will provide power to the gyratory crushers. Transformers, switchgear, starters and miscellaneous equipment will be provided to energize the loads of this area and to provide the necessary building services. SURGE PILE The loads of the surge pile will be fed from the electrical room at the gyratory crushers. SECONDARY CRUSHERS

150 Technical Report on the KéMag Pre-Feasibility Study Page 150 Two 34.5 kv circuit breakers will feed the two 34.5/4.16 kv transformers of these crushers. An electrical room will house the electrical equipment required to operate the facility. CRUSHED ORE STOCKPILE The stockpile loads will be fed from the secondary crusher s substation. CONCENTRATOR PRODUCTION LINES The large loads of the components will be grouped for each of the two production lines and each group will be fed from a set of three 34.5 kv feeders originating from the main electrical room. Additional 34.5 kv feeders will be used for the concentrator 4.16 kv and 600 V loads. BUILDINGS 34.5kV feeders with 34.5/0.6kV distribution will feed 600V loads for the garages, workshops, warehouse, laboratory and administration building. TAILING RECLAIM WATER, CAMP SITE, WATER TREATMENT FACILITIES The camp site, tailings water reclaim and water treatment facilities will be fed by drops from a 34.5 kv overhead line. Breakers, fused interrupters and step-down transformers, sometimes pole-mounted, will be used in the various electrical rooms of these facilities. EMERGENCY POWER A set of six 3.15 MVA diesel generators will be provided to energize essential loads in case of power failure PIPELINE Power from the Hydro-Québec Hart-Jaune hydro-electric generating station will feed the booster station substation at 34.5kV. The power will be stepped down to 4.16kV and 600V to feed the loads of the pumps and ancillary equipment. A set of two 3.5MVA diesel generators will provide emergency power to the booster pumps POINTE-NOIRE SITE Power from the Hydro-Québec Arnaud substation will be transmitted to the Pointe-Noire site by an existing 161kV overhead line. A tap will be made by Hydro-Quebec to meet the needs of the pellet plant and of the stockyard, handling and ship loading facilities.

151 Technical Report on the KéMag Pre-Feasibility Study Page 151 MAIN SUBSTATION AND ELECTRICAL ROOM The substation will include two step-down transformers to 25 kv and a number of 161 kv breakers. It will be designed so that operation will be possible with only one transformer working. A prefabricated electrical room will house the substation protection and operation equipment as well as 25 kv metal-clad switchgear and two feeders will be provided for stockyard, handling and ship loading facilities. PELLET PLANT Four 25 kv feeders will energize all electrical equipment on both pellet plant lines. STOCKYARD AND SHIP LOADING FACILITIES Two feeders will run from the main site substation to the main conveyor motor controllers, to the stacker and reclaimer cable reel feeding points. Power will also be fed to the ship loader and to ancillary equipment such as hi-mast lighting for the stockyard and berth, sodium vapour roadway lighting, industrial lighting along conveyor galleries and range lights for navigation. EMERGENCY POWER Two 1,000 kw, 4160 V emergency diesel generators will be located near the main electrical room to give emergency power to critical loads in the pellet plant as well as elsewhere on the site ENVIRONMENTAL CONSIDERATIONS PROJECT ENVIRONMENTAL APPROVAL REQUIREMENTS The Project is expected to trigger the environmental assessment regimes of general application established by the Canadian Environmental Assessment Act ( CEAA ) 2, the Environment Quality Act ( EQA ) (Chapter I, Division IV.1), the provincial and federal regimes established by Sections 22 and 23 of the James Bay and 2 In Moses v. Canada, the Québec Superior Court ruled on 30 March, 2006, that the CEAA does not apply in the Territory defined in Section 22 of the JBNQA. Both sides have appealed that judgment. Pending the outcome of those appeals, the CEAA continues to apply 2 In Moses v. Canada; the Québec Superior Court ruled on 30 March, 2006, that the CEAA does not apply in the Territory of the JBNQA, because the assessment process established by the CEAA was incompatible with that established by section 22 of the JBNQA. The Court of Appeal upheld that ruling. Both sides appealed that judgment. Pending the outcome of those appeals, the CEAA continues to apply. The application for leave to appeal was granted by the Supreme Court of Canada on 16 October, 2008, and the appeal is scheduled to be heard on 9 June,

152 Technical Report on the KéMag Pre-Feasibility Study Page 152 Northern Québec Agreement ( JBNQA ) (EQA, Chapter II, Divisions II and III) and the provincial and federal regimes of Section 14 of the Northeastern Québec Agreement (Règlement sur l évaluation et l examen des impacts sur l environnement dans une partie du Nord-est québécois) ( NEQA ). The Golder Report, sets out a roadmap for obtaining the required environmental permits and authorizations for the LIOP and also serves as an important planning resource for the KéMag Project insofar as the permits/authorizations required from the Government of Canada ( GCan ) and the Government of Québec ( GQ ) are concerned GENERAL TIMETABLE The timetable for the Environmental Impact Assessment ( EIA ) cannot be established until the dates of the tabling of the Project Notice and of the start of collection of baseline data have been established. Table 55 below presents the environmental impact assessment regimes that apply. Most of the required baseline data can be collected within twelve months of the start of baseline data-collection. However, two seasons of baseline data collection may be required in certain cases such as micro-mammals at the Lake Harris site. The timing of the authorization of the start of baseline data-collection will influence the time required to prepare the Environmental Impact Statement ( EIS ). In the case of archaeology, for example, a desktop evaluation of the archaeological potential of affected areas is a precondition for field surveys. If the authorization to start work is received in fall or early winter, it will probably be possible to complete the study of archaeological potential in the following months and to initiate field surveys during the next summer. Thus, the archaeological component would be completed in approximately twelve months. If, however, authorization to start baseline data-collection is received only in the spring, the field surveys will probably not be able to take place until the following summer. In that event, the archaeological component might take 18 months or more to complete. Table 55: Environmental Impact Assessment Regimes Applicable (1) (2) Environmental Assessment Regime Provincial regime of JBNQA Section 23 and EQA Chapter II, Division III. Kativik Environmental Quality Commission (5 Québec members, including Chair; 4 Inuit members). Federal regime of JBNQA Section 23. Environmental and Social Impact Review Panel (3 federal members, including Chair; 2 Inuit Infrastructure Lake Harris site. Major sand and gravel pits north of 55 th parallel. Transmission line north of 55 th parallel (if > 75 kv). Access road north of 55 th parallel. Pipeline north of 55 th parallel. New significant sewage and wastewater collection and disposal system north of 55 th parallel. Solid waste collection and disposal (including land fill and incineration) north of 55 th parallel. Note: Air and ground reconnaissance, surveying, mapping and core sampling by drilling do not require impact statements. Lake Harris site. Transmission line north of 55 th parallel. Access road north of 55 th parallel.

153 Technical Report on the KéMag Pre-Feasibility Study Page 153 Environmental Assessment Regime Infrastructure members). Pipeline north of 55 th parallel. May apply to other infrastructure listed under (1) above. Provincial regime of JBNQA Section 22 and (3) EQA Chapter II, Division II. Evaluating Transmission line south of 55 th parallel and west of 69 th Committee (2 Québec members, 2 Canada meridian (if > 75kV). members, 2 Cree members; rotating Chair). (4) Federal regime of JBNQA Section 22. As for (3) above. (5) Provincial regime of NEQA Section 14. Regime of EQA Division IV.1 with special consultation with Naskapi Village of Kawawachikamach. Pipeline between 55 th and 53 rd parallels and east of 69 th meridian. (6) Federal regime of NEQA Section 14. As for (5) above. (7) EQA Chapter I, Division IV.1. Pipeline south of 53 rd parallel. MDDEP claims jurisdiction over all of the proposed infrastructure at Pointe-Noire. A legal opinion of 12 December, 2006, from M e Robert Daigneault argues that EQA does not apply to the deep-water dock. M e Daigneault s opinion on the application of the EQA to the other infrastructure at Pointe-Noire, particularly that outside the land of the Sept-Îles Port Authority, is not known. NML has not yet decided how to proceed in the light of M e Daigneault s opinion. (8) CEAA. Subject to the outcome of Moses v. Canada, all major infrastructure in the Territory as defined at Sub-section 1.16 of the JBNQA, the pipeline south of 53 rd parallel 3. Deep-water dock at Pointe-Noire. Possibly other infrastructure (including pellet plant) at Pointe-Noire. The analysis of the first season s data and the drafting of the EIS will be able to proceed during the second season of baseline data-collection. The data from the second season will, if possible, be integrated into the EIS. If that is not possible, they will either be submitted in the form of an addendum to the EIS, or the tabling of the EIS will have to be delayed. Drafting the EIS will take 6 to 9 months, starting some 9 months after the beginning of baseline data collection. The duration of the regulatory process starting from the tabling of the EIS to the granting of the Certificates of Authorization ( CoAs ) is estimated to take 12 to 18 months. 3 The 53 rd parallel is the approximate southern limit of the Territory as defined in the JBNQA in the area where the pipeline will pass.

154 Technical Report on the KéMag Pre-Feasibility Study Page 154 In summary, the EIA process is likely to take a minimum of 27 months from the date of authorization of the start of baseline data-collection, broken down as follows: Baseline data collection ("BDC"): months; Drafting EIS (starting 9 months after start of BDC): months; Regulatory process (after EIS tabled) months HARMONIZATION OF ENVIRONMENTAL IMPACT REGIMES Where several EIA regimes apply to a single project, it is increasingly the practice to harmonize their application, often by creating a multi-jurisdictional body for the purpose of public hearings. Harmonization between the governments of Canada and Québec was formalized by the execution of the Canada-Québec Cooperation Agreement in May Precedents for harmonizing regimes include the reviews of low-level military flying in the Québec-Labrador Peninsula, the Great Whale Hydroelectric Project, the Voisey s Bay Nickel Project and the Eastmain-1-A and Rupert Diversion Hydroelectric Project. There is no precedent for harmonizing up to eight EIA regimes. Official discussions on the form of the harmonized assessment for the Project are expected to start once the Project Notice has been filed BASELINE DATA STUDIES The Coordinator of Environmental & Social Affairs, assisted by advisors in specific disciplines and a quality assurance/control team, oversees the work of the lead consultants, who in turn coordinate the work of their respective associate consultants. The lead/associate consultants have been or will be identified on the basis of specific expertise or extensive knowledge of the Project area KÉMAG BASELINE STUDIES Baseline studies that have been completed for LIOP and are applicable to the KéMag Project are listed in Table 56. Experts in such areas as gender and sustainable development will review work plans and draft reports and will contribute directly to the EIS. Any further baseline studies required by the regulators will be conducted.

155 Technical Report on the KéMag Pre-Feasibility Study Page 155 Table 56: Planned Baseline Studies for the KéMag Project Baseline Study Archaeological Potential Terrestrial Archaeological Fieldwork Terrestrial Archaeological Fieldwork Subaquatic Palaeontology - Literature Review Palaeontology Fieldwork Land/Resource Use & TEK Ashuanipi Land/Resource Use & TEK Betsiamites Land/Resource Use & TEK Naskapi Land/Resource Use & TEK Inuit Land/Resource Use & TEK Crees Socio-Economic Impact Assessment Aboriginal Health Economic Modelling Visual Impact Assessment Terrestrial/Avifauna Team - Literature Reviews for Large Mammals, Furbearers, Small Mammals & Predators Winter Habitat Use Survey Marine Bird Survey * Passerine, Waterfowl & Raptor Surveys Nocturnal Owl Survey Micro-mammals, Herpetofauna, Chiroptera, Insects Inter-Basin Aquatic Microbe Transfer Terrestrial Ecosystem Mapping Climate Hydrology - Low Flows, Rain/Stream Gauging, Snow Depth/Density Surface Water & Sediment Quality Geology Hydrogeology Testing Rocks & Ore Ambient Air/Noise Sampling Project Component Pipeline & Transmission Line Corridors Pipeline & Transmission Line Corridors Pellet Plant & Ship-loading Facility Mine Site Mine Site improbable Mine Site; Part Pipeline & Transmission Line Corridors Part Pipeline & Transmission Line Corridors Mine Site; Part Pipeline & Transmission Line Corridors Mine Site; Part Pipeline & Transmission Line Corridors Part Transmission Line Corridor All All All(Robichaud 21 December 2007 to be updated) Mine Site; Transmission Line (if aerial); Pipeline Pumping Station; Pellet Plant & Ship-loading Facility All Mine Site; Pipeline & Transmission Line Corridors Pellet Plant & Ship-loading Facility Mine Site; Pipeline & Transmission Line Corridors Mine Site; possibly Pellet Plant & Shiploading Facility Mine Site; Pipeline & Transmission Line Corridors Mine Site; Pellet Plant & Shiploading Facility Mine Site; Pipeline & Transmission Line Corridors Mine Site Mine Site; Pipeline & Transmission Line Corridors Mine Site; Pipeline & Transmission Line Corridors Mine Site; Pipeline & Transmission Line Corridors Mine Site Mine Site Pipeline Pumping Station

156 Technical Report on the KéMag Pre-Feasibility Study Page 156 Baseline Study Physical Estuarine Environment * Permafrost ** Freshwater Environment Habitat Characterization of Stream Crossings Freshwater Environment Fieldwork Biological Estuarine Environment 1 Hydroacoustics Fieldwork Hydroacoustics - Marine-Mammal Analysis Risk Assessment Project Component Pellet Plant & Shiploading Facility Mine Site; Part Pipeline & Transmission Line Corridors Mine Site; Pipeline & Transmission Line Corridors Mine Site; Pipeline & Transmission Line Corridors Pellet Plant & Shiploading Facility Pellet Plant & Shiploading Facility Pellet Plant & Shiploading Facility All * Survey report for 2006 to be completed. ** If needed, will be addressed through terrestrial ecosystem mapping ENVIRONMENTAL MANAGEMENT PLAN An Environmental Management Plan will form part of the EIS and will provide a detailed description of environmental surveillance and monitoring practices and of mitigation measures during all phases and in all areas of the Project. The Plan will address the following: Waste management/minimization; Air emissions management; Water quality management; Tailings management; Fish habitat management to ensure compliance to the "No Net Loss" principle of the Department of Fisheries and Oceans ( DFO ), Canada; Pipeline management; Ship loading and marine transportation management; Dredging management; Wildlife management; Social issues;

157 Technical Report on the KéMag Pre-Feasibility Study Page 157 Emergency response and contingency, Site rehabilitation; Occupational health and safety management; Stakeholders communication; Commitment policies to such issues as Native employment and contracting, training, access control to site, public safety, local sourcing, etc MARKETING STUDY CRUDE STEEL PRODUCTION As is well known, the principal growth in crude steel production in the recent past has been in China whose growth has overshadowed virtually all other developments. Chinese crude steel production reached 487 million tonnes ( mt ) in 2007, representing compound annual growth over of almost 22% Chinese crude steel production is forecast to increase further by 10%. Looking forward to 2016, the following trends are expected: A gradual shift in crude steel production from North America and Europe to Brazil with an increase in shipments of steel slabs from Brazil to North America and Europe. Eastern Europe and the CIS shift unevenly and slowly from downsizing and modernization to growth with the likelihood that the EAF share of crude steel production will increase. Japanese crude steel production will remain in the mtpy range with a negative trend during the second half of the forecast period. Growth in South Korean crude steel production will moderate from past levels as increased demand is increasingly met from Korean-owned offshore plants in India, China, Malaysia, etc. India will follow in China s footsteps with increasingly rapid growth in crude steel production, reaching 75 mtpy by the end of the forecast period. This will be under-pinned by GDP growth as well as by foreign-owned steel mill investments. China s crude steel production will continue to grow at a significant rate in the short to medium term, although growth is likely to moderate significantly from the >20% per year rates of recent years as Chinese mills focus on consolidation, greater added value and cost reduction.

158 Technical Report on the KéMag Pre-Feasibility Study Page 158 There is a wide range of forecasts about the future development of Chinese crude steel production. There are those which assume that the government will be successful in implementation of its policy of limiting capacity and production and those that take the opposite view, as well as various shades of grey in between. Taking the various factors and issues into account, this report s forecast is based on Chinese crude steel production growth of 10% in 2008, falling to 7.5% in and falling thereafter to 5% per year in This means crude steel production of 650 million tonnes in 2011 and 830 million tonnes in IRON ORE DEMAND FROM INTEGRATED STEEL MILLS A forecast of blast furnace iron production has been derived from the crude steel forecast and, in turn, a forecast of incremental iron ore demand has been developed. This shows incremental iron ore demand from the integrated steel mill sector of 373 Mtpy by 2011 and 692 Mtpy by IRON ORE DEMAND FROM THE DIRECT REDUCTION SECTOR Another important - and growing - segment of the iron ore market is the direct reduction sector whose principal iron ore consumption is in the form of DR pellets. Global DRI/HBI production reached nearly 67 mt in Global DRI/HBI production is forecast to grow to 87 mtpy by 2011 and to 112 mtpy by Iron ore feedstock for DRI/HBI production is predominantly DR pellets and some lump ore, in the average ratio of 85-90% pellets and 10-15% lump. Thus, with incremental DRI/HBI production of 20 mtpy by 2011 and 45 mtpy by 2016, incremental iron ore demand from the DR sector will be approximately 30 mtpy and 67 mtpy respectively of which 31 mt and 61 mt will be pellets IRON ORE SUPPLY Global iron ore production, including China, in 2007 has been reported by UNCTAD at 1.63 billion tonnes, with Chinese production data corrected to 64% Fe content. The Compound annual growth rate ( CAGR ) between 1998 and 2007 was 6.8%. International trade was reported at 833 mt exports, of which 780 mt was seaborne. The market share of the three major producers CVRD, Rio Tinto and BHP Billiton was >70% of seaborne trade. The three major producers have large expansion programmes under way which in aggregate are expected to add iron ore capacity of at least 350 mt by In the case of Australia, there is some risk that Rio Tinto and BHP Billiton may not achieve their expansion plans in the timeframes envisaged. There are many other projects at various stages of development in Brazil, Australia and elsewhere. The average grade of Chinese domestic iron ore production is about 30% Fe which compares with the average 62-64% Fe of imported ore. As demand has increased, average grade has decreased with incremental production in the range of 10-20% Fe. Production at actual Fe content in 2007 was 707 million tonnes, equivalent to 332 mt at 64% Fe. A key issue for the longer term is the sustainability of this high level of production in China. There is no shortage of press reports of new Chinese iron ore projects - a natural response to increasing demand, tight

159 Technical Report on the KéMag Pre-Feasibility Study Page 159 market conditions and perhaps also a desire for less dependency on the international market. However, as lower cost imported ore becomes more available and as the international price finds its new trend level, growth in Chinese ore production can be expected to stagnate or even decline, meaning that increases in crude steel production will eventually be largely if not wholly dependent on imported iron ore. Indian iron ore production has grown enormously on the back of Chinese demand, reaching a reported 207 mt in Chinese imports from India in 2007 were 78 mt. Indian iron ore [as well as other non-traditional sources] has filled the gap created by the inability of traditional suppliers to meet the surge in Chinese demand, encouraged by the emergence of a spot market in China where prices for imported ore reached unprecedented levels. The current level of Indian ore exports to China seems unsustainable in the longer term for a number of reasons, not the least of which is the planned growth in Indian steel production. In summary, it would appear that, given the existing expansion plans of the major producers and the various other projects in the pipeline, iron ore supply through to 2016 is likely to be sufficient to meet demand, although it is unlikely that there will be a significant over-supply - the major suppliers can be expected to manage their production and the speed of their expansions to maintain a sensible balance in the market. Beyond 2010, CVRD, Rio Tinto and BHPB and others have scope to expand further PELLET SUPPLY Global pellet production in 2007 is estimated at 330 mt [including 41 mt in China], of which 138 mt was exported. There are many projects involving additional supply to the pellet market, both brownfield and greenfield. A forecast of potential production out to 2015 has been prepared, reaching 393 mtpy by 2010 and 408 mtpy by This forecast includes only existing plants, approved expansions and certain new plants. A summary of this forecast is shown in the following chart, Figure 28. Figure 28: Global Projection of Pellet production (Million Tonnes)

160 Technical Report on the KéMag Pre-Feasibility Study Page PELLET DEMAND 2007 Estimated pellet demand from integrated steel mills in 2007 was 272 million tonnes, and the breakdown of this tonnage by region is given in the following table (Table 57). Table 57: Projection of Pellet Production Region Country million tonnes NAFTA USA + Canada 67.0 Mexico 4.0 sub-total 71.0 S. America Brazil 4.1 other 1.0 sub-total 5.1 Europe Total 54.5 CIS Total 45.5 Asia / Oceania India 11.6 Japan 9.9 China 66.0 South Korea 1.8 Taiwan 3.3 Australia 3.4 sub-total 96.0 Total Source: Ferrum In the direct reduction sector, with reported DRI/HBI production of 67 mt in 2007, total iron ore consumption is estimated at 101 mt, of which pellets is estimated at 65 mt LONGER RANGE PELLET DEMAND Pellet demand out to 2015 is shown in the chart below. This forecast is based on the assumptions for crude steel/iron and DRI/HBI production. Specific pellet consumption is expected to increase in certain markets due to the expected tightening in lump ore supply, notably in Europe, an important consumer of traded lump ore. Pellet demand is forecast to increase from 337 mt in 2007 to 402 mt by 2011 and to 474 mt by 2016 (Figure 29).

161 Technical Report on the KéMag Pre-Feasibility Study Page 161 Figure 29: Projected Pellet Demand to 2016 (million tonnes) SUPPLY-DEMAND BALANCE The combination of the supply and demand forecasts is shown in Table 58. Table 58: Pellet Supply-Demand Forecast This forecast suggests a very tight market in the period The main drivers of this are China, Europe and the direct reduction market. The deficit of 6.7 mt in 2007 has not been examined in great detail and is not hugely significant in the context of this report and the longer term focus of the client. It could well be that there was an inventory draw down in The deficits in 2008 and 2009 suggest that some demand may go unsatisfied, although ramp up rates for newer plants may be faster than assumed and the short term assumption for China may be too conservative. After 2009 the gap between supply and demand starts to widen significantly, indicating the new for new capacity in and after By 2016 there will be the need for the equivalent of as many as eight or nine large scale plants.

162 Technical Report on the KéMag Pre-Feasibility Study Page FREIGHT ISSUES Freight rates softened in first half 2006 due to the increase in supply of dry bulk vessels being greater than the increase in dry bulk trade. The rates have climbed steadily during the second half of Post-2006, growth in dry bulk trade is expected to overtake dry bulk fleet growth, driven by such factors as increasing Chinese iron ore imports and changes in coal trade patterns, meaning that freight rates are likely to continue rising post In general, a direct reduction plant buyer will assess the competitiveness of the various grades of iron ore on the basis of the landed cost at the relevant port. In order to determine the competitiveness or otherwise of NML DR pellets at selected direct reduction plants, a matrix of freight rates for the principal loading port and discharge ports has been developed by Fearnley Consulting A/S, part of the Astrup Fearnley shipping services group. This matrix reflects current freight rates and is not a forecast of future rates. The resultant freight rate differentials are shown in Table 59, where a negative number represents a freight disadvantage for Sept Îles. Table 59: Freight Rate Differentials Bahrain Mangalore Tubarao PDM Narvik Nouadhibou Al Jubail Bin Qasim Umm Said Cigading Labuan El Dikheila Misurata Point Lisas Hamburg IRON ORE PRICE DEVELOPMENT The outcome of the 2008 price negotiation campaign was, to say the least, unexpected. The +65% outcome for Brazilian fines exceeded most forecasts, while the outcome for Brazilian pellets at +86% was dramatic indeed and reflects the extreme tightness of the current market. The resulting BF pellet premium of approximately 86 cents per metric tonne unit is unprecedented and represents a quantum increase. The DR pellet premium has been maintained at 10%. In order to develop a longer range European price forecast for Itabira SSF Fines: It is assumed that there will be increases of 10% in 2009 and 5% in 2010 and nominal price decreases of 5% in 2011, 12.5% in 2012, 10% in 2013 and 7.5% in both 2014 and 2015; After a rollover in 2016, from 2017 iron ore prices will grow at a trend rate of 2.5% per year in nominal terms. This forecast is shown in the first chart on the following page (Figure 30). The straight line development during the period is indicative only - in practice there will doubtless be cyclical variations. This is a single

163 Technical Report on the KéMag Pre-Feasibility Study Page 163 point forecast - most of the underlying assumptions are variables in themselves and it would make sense to consider higher and lower case scenarios. As far as the pellet premium is concerned, the issue that now has to be addressed is whether 2008 s surge in the pellet premium will turn out to have been a spike - like the 53 cents premium in or if it will be sustained. The increasing gap between pellet supply and demand indicates the need for additional pelletising capacity to come on stream in and after admittedly some of this will most likely happen in China. Against this background, the spike argument is more difficult to justify than in 2005/06 when there was some softness in the pellet market. The sustainability argument on the other hand is supported by the projected future supply deficit. As a matter of prudence, the authors feel that it would be appropriate to build in some element of correction in the pellet premium, but in the medium to longer term, they feel that there is a strong case for a quantum increase in the trend level compared to the previous forecast. Figure 30: Brazilian Sinter Feed Price Forecast (cents/mtu FOB) Source: Ferrum A progressive correction back to 65 cents/mtu by 2012 has therefore been factored into the forecast as the expected correction in the fines price will have a washover impact on the pellet premium. Thereafter, upward growth in the pellet premium is forecast at the trend rate of 2.5% per annum. The resultant pellet premium and price forecast for Tubarão BF pellets in the European market are shown in the following two charts (Figure 31 and Figure 32).

164 Technical Report on the KéMag Pre-Feasibility Study Page 164 Figure 31: Tubarão BF Pellet Premium Forecast (cents/mtu) Source: Ferrum Figure 32: Tubarão BF Pellet Price Forecast (cents/mtu) Source: Ferrum The traditional premium for Brazilian DR pellets in relation to BF pellets was 7.5% of the BF pellet price. In 2005 however, this DR premium was increased to 10% and has been maintained at this level ever since. The authors feel that it will be more likely that the current 10% premium will be sustained. The forecast for the Tubarão DR pellet price is shown in the following chart (Figure 33).

165 Technical Report on the KéMag Pre-Feasibility Study Page Figure 33: Tubarão DR Pellet Price Forecast (cents/mtu FOB) PRICING For the purpose of this PFS, it is assumed that the NML annual production of 15.0 million tonnes of pellets will be sold as follows: o 10 million tonnes of acid and fluxed basic pellets at the FOB price for acid pellets in the Western European market; o 5 million tonnes as DR grade pellets at a price 10% higher than that for acid pellets. 7 mtpy of concentrate will be sold as pellet feed fines. Prices for iron ore fines and pellets remained strong in 2008 due to tight market conditions, with fines gaining a further 65% in 2008 and pellets gaining 86.7%. Pellet premium increased to 85.8 /mtu over the price of fines. This large increase in premium was an indication of shortages in the pellet market. However, since the end of summer 2008, the severe financial crisis which started with the banking industry in the USA, started to affect the global economy. Steel companies in the USA, Europe and China are reducing production in the face of declining demand. As a result, major iron ore and pellet producers are also reducing their output to prevent inventory build-up. The USA and some major Western European countries have been in recession since mid Because of the severity of the economic crisis, global growth is expected to slow down in The USA and major western countries are expected to remain in recession during the first half of 2009 and it is also projected that growth in

166 Technical Report on the KéMag Pre-Feasibility Study Page 166 China will slow down from double digits to 8% or below. Analysts are projecting a 20% decline in demand, due mainly to the slowing growth and high ore inventory levels in China. Some are even projecting a decline of up to 50% in prices PRICE ASSUMPTION Because of a tight market condition and forecast of continued tightness until , Analysts had upwardly revised their long term price forecast. This was based on the fact that opportunities for brownfield expansion were becoming rarer and therefore higher capital will be required to develop Greenfield projects in order to balance supply with projected demand. Although the current economic realities will dampen the near-term prospect for iron ore, it is generally believed that the measures undertaken by major economies and China will stabilize the economy and restore the demand by Based on Credit Suisse s long term price forecast, the following prices are assumed for the study: Pellet feed fines Acid pellet price: DR grade pellet price US 85.0 /mtu US /mtu based on US 50 /mtu premium a 10% premium over acid pellet price. The above prices are about 60% of the 2008 contract prices for European markets MARKETING The sales and marketing plan implemented for the LIOP will be extended to cover KéMag products. However, under the terms of the binding agreement that NML entered into with Tata Steel ( Tata ) October 1, 2008, Tata has an option, valid until 30 June, 2009, to propose a transaction regarding its interest in participating in a pelletizing project For the purposes of this PFS, NML has assumed that Tata will be interested in taking 100% of the production of the KéMag pelletizing project. Any decision by NML to approach other potential customers is dependent upon Tata s decision regarding its option CAPITAL COST ESTIMATE SCOPE OF ESTIMATE The capital cost estimate of the Project includes the entire cost of engineering, construction, installation and development work required for starting up a mining operation and building, commissioning and starting up a concentrator, a slurry pipeline, a filtration plant, a pellet plant, a flotation plant, a, product storage and ship

167 Technical Report on the KéMag Pre-Feasibility Study Page 167 loading facilities, and all associated facilities, to produce and sell 15 million tonnes per year ( mtpy ) of pellets at an average 66.3% Fe, plus 7 mtpy of concentrate as pellet feed fines at an average 69.1% Fe The Project will consist of the following: A mine at Lake Harris to produce over 76 mtpy of iron ore; A concentrator at the Lake Harris site to produce 21.2 mtpy of concentrate; A 750km-long slurry pipeline to transport 21.2 mtpy of concentrate from the concentrator to a new 15 mtpy pellet plant at Pointe-Noire, Sept-Îles; on the Baie des Sept Îles, part of the Gulf of St. Lawrence; A filtration plant at Pointe-Noire to produce 7 mtpy of concentrate with less than 8% moisture for export; A flotation plant at Pointe-Noire with a design capacity of 7.5 mtpy to produce 5 mtpy of low silica concentrates for the production of DR pellets;. Also at Pointe-Noire, product storage space sufficient to permit 1.53 million tonnes of concentrate to be stocked in addition to 3.06 million tonnes of pellets and a handling system and ship berths that have the capacity to load 22 mtpy of product BASIS OF ESTIMATE The cost estimate covers: The direct cost of equipment, materials and labour for the construction of the facilities and for the electrical power supply and distribution systems. The cost estimates are based on the assumption that equipment will be purchased at competitive rates and that erection and construction work will be awarded in set packages and on a competitive basis. The indirect costs such as marine and inland freight, temporary power, water and communications, engineering, procurement and construction management, lodging of construction workers, owner s costs and commissioning. CURRENCY Estimates are shown in Canadian dollars ( Can$ ). Conversion from US dollar prices was made at US$0.85 per Can$1.00 and from Euro prices at Euro0.625 per Can$1.00 to remain conservative due to the inherent volatility of the rates.

168 Technical Report on the KéMag Pre-Feasibility Study Page 168 DATE Due to the inherent volatility of prices for commodities such as steel and copper and changes of demand on the suppliers market, the Capital cost estimates were based on prices for major equipment and materials as per the last quarter of 2006 and, when applicable, salary rates at the of end 2008 according to the Association de la Construction du Québec, heavy industrial sector in Baraquement. ACCURACY Based on the quality of information obtained from outside sources and from an in-house databank, the capital cost estimate is considered to have an overall accuracy of ± 25%. METHOD OF ESTIMATION APPLIED Mine and Concentrator Site: Contractors unit costs for construction were mainly based on rates obtained by the traditional bidding method for recent ongoing projects. Due to significant recent decreases in commodity prices and lower demand for process equipment, some 2008 unit prices were adjusted to reflect the anticipated changes which should bring back the unit costs to the 2006 level. The construction sites considered for reference were of medium size (±$500M) and located in the Québec-Labrador area with similar climatic conditions. Major Equipment: Due to significant decreases in commodity prices and lower demand for process equipment, the 2006 budget prices obtained previously by NML from major suppliers were maintained for the estimate. It is to be noted that the mine equipment required in the initial years of operation will be leased and the cost thereof is excluded from the capital cost estimate and included in the operating cost estimate. However, replacement equipment and additional equipment to increase the size of fleet will be purchased by NML and the associated costs are included in the capital cost Although the camp will be built and occupied by construction personnel in the three years construction period prior to the start of production, the inclusion of its capital cost in the estimate is deferred to the first year of production, when the camp will be occupied by operations personnel. Work Schedule: Labour costs were based on construction crews working ten hours per day, six days per week, for a total of 60 hours per week on a schedule of 40 consecutive days on site followed by 10 days off. Labour Rates:

169 Technical Report on the KéMag Pre-Feasibility Study Page 169 Rates for construction personnel include a heavy industrial premium allowance as follows: 20 hours per week of overtime, 16 hours of travelling time per 40 days, CSST (Quebec Workers Safety and Security Commission), safety material, small tools, consumables, airfares, site overhead, site administration fees, mobilization, demobilization, construction permanent equipment (except for heavy civil), contractor overhead and profit. The hourly rates have been verified with contractors rates used in the same area. All the hourly rates described above will be part of the contractor s contract and paid by the contractor. Examples of the rates are given in Table 60. Table 60: Examples of Hourly Construction Rates Discipline Average hourly rate ($) Earthwork (excluding equipment)/concrete 90 Structural steel 110 Architectural 120 Mechanical, Piping, Electrical, Automation 140 The cost for room and board is excluded from the hourly rates. Provision for services including workers camp, catering and janitorial services, and camp administration will be made in Indirect Costs. Electrical Power Supply to Mine and Concentrator Site: The direct cost for the 270 km long Brisay to Mine and Concentrator Site OHL at 315kV (single circuit, with bundles made of 2 Besfort conductors per phase) was estimated and compared with a similar 315 kv OHL project built by Hydro-Quebec in Slurry Pipeline: The cost estimate is as given in the PSI 2007 Report. adjusted by NML for the pipeline length being increased to 750 km. Pellet Plant: The cost estimate given in the Danieli turnkey Proposal, was discounted by 15% to bring it into line with the known actual costs of the ongoing, similar, 7 mtpy Samarco project. Shiploading and Stockyard: The cost estimate was maintained as given in the Howe 2007 Addendum Report Tailings Disposal: The cost estimate as given in the Barr Report was modified to reflect construction of an initial Tailings Starter Dam during Year -3 to Year 0.

170 Technical Report on the KéMag Pre-Feasibility Study Page 170 COST ELEMENTS In preparing the estimate, two categories of costs were developed. Direct Costs The estimated direct costs were based on the general layouts and process diagrams developed during the study, the list of major mechanical process equipment, electrical single-line diagrams, a description of major electrical equipment, and the dimensions and general layout plans of the larger buildings and structures. The prices of equipment and materials were based on quotations provided by potential suppliers as well as inhouse data which reflect experience from previous similar work. In general, quotations were obtained for all major mine and process equipment, FOB from a major port in the country of origin. The cost of civil work and local supplies was based on unit prices supplied by qualified contractors with prior experience in similar fields. The cost of civil work and local supplies was based on unit prices supplied for the LabMag PFS by qualified contractors with prior experience in similar fields. Equipment erection and installation costs were determined on the basis of a percentage of the value of the equipment in some cases, and on labour estimates in other cases. Process and service piping was estimated based on the cost of equipment to which it is related, using a percentage based on experience from other similar projects. For equipment and supplies originating abroad, shipping costs from the port of origin to the port of Sept-Îles were added, as were railroad transportation costs between Sept-Îles and Schefferville and Schefferville to the Lake Harris site. The overall freight cost was estimated to be 5.0% of the direct cost of supplies. Indirect Costs EPCM: Expected costs for project management, detailed engineering, procurement and construction management as well as commissioning and start-up were estimated based on experience in providing such services on similar projects A number of area-specific rates were applied to calculate estimated costs: Mine and Concentrator Site installations 10 % Slurry Pipeline 10 % Pellet Plant 3 % (Turnkey) Shiploading & Stockyard 7.5 % (according to Howe)

171 Technical Report on the KéMag Pre-Feasibility Study Page 171 Based on suppliers recommendations, provisions were made for spare parts and initial fills of oils, greases and grinding media, and the cost of replacement parts used for commissioning and start-up. Owner s costs: Estimates were included for the cost of supplemental work required prior to beginning basic engineering, including: Feasibility and environmental studies; Pilot plant campaign prior to final mill design; Equipment vendor and product consumer testing of KéMag concentrate and pellets; ABA and NAG tests and ICP scans on tailings; Hydrogeological study at the tailings basin sites; Topographic survey of the route of the pipeline; Soil investigations. Construction Indirects: A number of area-specific provisions were made for construction indirects for setting up sites. These items covered office trailers and equipment, temporary water supply and electricity service, communications, security guard services, etc. The provisions also cover the cost of obtaining the permits and authorizations required by government agencies and regulatory bodies. Mine and Concentrator Site 10 % Pellet Plant 0.5% of the Danieli turnkey price Shiploading & Stockyard 0.5% of the capital cost as given in the Howe 2007 Report No provision for construction indirects was made for the Slurry Pipeline, including the booster station, as they were already included in the PSI capital cost. The estimated cost for Room & Board in a Construction Workers Camp, based on a four-year construction period, was included. The operational cost for Room & Board in a Construction Workers Camp for the Slurry Pipeline was already included in the PSI cost estimate.

172 Technical Report on the KéMag Pre-Feasibility Study Page 172 A Construction Workers Camp is not required for the Pellet Plant and Shiploading Facilities at Pointe-Noire Contingencies: An overall contingency allowance was determined based on the estimated uncertainties for the various Project components and by attributing specific contingency rates to cover such uncertainties, based on the study team s best judgment. An average contingency allowance of 11.0% of direct capital costs (based on 15% for most components but only 3% for the Pellet Plant that is to be supplied on a turnkey basis, was included to cover undefined items or work that needs to be done or cost elements which will be incurred as part of the defined scope of the work covered by the estimate which cannot be anticipated or explicitly described when the cost estimate is made due to a lack of complete, precise and detailed information. The contingency reserve is thus an integral part of the cost estimate. It does not compensate for any lack of accuracy in the cost estimate, nor is it intended to cover items such as a potential change in Project scope, natural disasters, extended labour strikes beyond the Project Manager s control, currency fluctuations, or cost indexing in relation to the estimated rates. Other provisions: It was assumed that the Project is exempt from import taxes, sales tax and any other charges of a like nature. INFLATION No provision has been made for inflation. The estimate is as at the end of 2008 except for the cost of equipment, which was based on the 2006 market value. QUANTITIES Quantities used in developing the cost estimate were determined as follows: For the process plants, civil and structural quantities were estimated using general plans, process flow diagrams, building dimensions, and data from other similar projects; Fresh water and sewer system pipes as well as concentrator and pellet plant process water pipes were based on similar projects; The transmission lines, substations, other electrical equipment and components were based on singleline diagrams and specifications; Instrumentation and automation costs were factored from electrical costs, based on experience from other similar projects OPTIMIZATION AND UPSIDE POTENTIAL: An overall discount of 7.5% was applied on all direct costs. This discount is based on the following efforts to be undertaken during the Feasibility Study phase.

173 Technical Report on the KéMag Pre-Feasibility Study Page 173 VALUE ENGINEERING Process optimization and integration; Layout optimization; Tailings alternate solutions. OTHER OPPORTUNITIES Market Competitiveness (pellet plant, slurry pipeline); Slurry Pipeline routing; Use of impervious till instead of membrane; Political issues such as Plan Nord du Québec (2008); o To obtain additional rebates from the Government of Quebec for the 270 km, 315 kv, OHL from Brisay to the Mine and Concentrator Site; o To share costs with the Government of Quebec for the 250 km construction road (alongside the Slurry Pipeline from the Mine and Concentrator Site to the point where the pipeline will leave the existing road near Fermont and for the 60 km from that Site to the Town of Schefferville. FOREIGN SOURCED AND USED EQUIPMENT Discounts can be attained on foreign or used equipment and materials, but this was not addressed at the PFS stage of the Project at the request of NML EXCLUSIONS The following items are not included in the cost estimate: Cost increases due to the indexing of wage rates, the price of materials or the cost of equipment; Cost increases resulting from fluctuations in currency exchange rates; Transportation of material from borrow pits, quarries or gravel pits that are beyond five km from the point of delivery;

174 Technical Report on the KéMag Pre-Feasibility Study Page 174 Winter construction work at the Mine and Concentrator Site. Only the Concrete batch plant will be winterized. Training before and after hiring; Working capital; Project financing and interest during construction; Taxes and duties SUMMARY OF THE ESTIMATE Table 61 provides the summary of the capital cost estimate.

175 Technical Report on the KéMag Pre-Feasibility Study Page 175 Table 61: Summary of Capital Cost Estimate (Can$)

176 Technical Report on the KéMag Pre-Feasibility Study Page OPERATING COST ESTIMATE BASIS OF ESTIMATE The basis for the operating cost estimate is the requirement for: Energy, in the form of electricity, diesel oil, heavy fuel oil, propane, etc; Manpower; Maintenance and repair including wear parts such as drill rods and bits, liners, tires, grinding media and other replacement parts. These major cost components are well defined and: The price of energy was assumed to be $0.042/kWh for electricity, $0.85/litre for No. 2 diesel fuel and $0.40/litre for No. 6 Bunker C fuel oil delivered to the mine site, based on $0.38/litre at Sept-Iles. The cost of manpower was based on the latest collective bargaining agreement between similar operations in the region and its unionized labour; Maintenance and repair costs were estimated from manufacturer's data or based on the LIOP and other similar projects. Estimates are shown in Canadian dollars ( Can$ ). Conversion from US dollar prices was made at US$0.85 per Can$1.00 to remain conservative due to inherent volatility of the rate SUMMARY OF ESTIMATED OPERATING COST A summary of the operating cost estimate of $ million for year 3, the first year of full operation, is given in Table 62. The estimated operating costs for the mine, concentrator, pellet plant and product stockyard and ship loading facilities, expressed per tonne of pellets, are lower than published costs for similar operations in North America.

177 Technical Report on the KéMag Pre-Feasibility Study Page 177 Table 62: Summary of Estimated Total Operating Costs for Year 3 (Can$) (Million $ per year and $ per tonne of product) Cost Item Mine Mine & Conc. Site Services Concentrator Pipeline Flotation Pellet Plant Concentrate Filter Plant Pointe- Noire Site Services Stocking/ Loading Admin. Total Electricity Fuel (1) Manpower (2) Maintenance Supplies (1) Explosives Consumables (1) * Additives Other Total $ million $ per tonne of concentrate $ per tonne of BF pellets $ per tonne of DR grade pellets (1) Includes cost of transport to mine and concentrator site, where appropriate. (2) Includes cost of room and board and fly-in fly-out for personnel at camp, where appropriate. * Tires

178 Technical Report on the KéMag Pre-Feasibility Study Page OPERATING SCHEDULES The manpower requirements were based on the assumptions that the mine, concentrator, slurry pipeline, pellet plant and product stockyard will be in continuous operation 24 hours per day, seven days per week, but maintenance and repair will generally be done on a day shift basis. Ship loading operations will be governed by the presence of ships. At the mine and concentrator site, crews will work 12-hour shifts, seven shifts per week and will rotate on the basis of two weeks at work with two weeks off work, therefore requiring four complete crews for the mine, crushers, concentrator and pipeline pumping station. Some clerical employees, engineers and technicians will work only on the day shift for a total of 42 hours per week, and various tradesmen in the workshops and in the process plant will work 12-hour day shifts, seven days per week. The only overtime paid will be the hours worked above the normal 40 hours per week averaged on an annual basis, and this overtime will be paid at the rate of 1.5 times the base rate. The pellet plant will operate 24 hours per day, and crews will work 8-hour shifts, seven shifts per week therefore requiring four complete crews. Maintenance and repair will generally be done on day shifts MANPOWER MANNING LEVELS It is estimated that in the first full year of operation, the Project will employ a total of 1,230 persons. The numbers for the various categories of persons are given in Table 63. The Project will draw upon the resources of Schefferville, Kawawachikamach, Matimekush, Fairmont, Betsaimites, Uashat, Mani-Utenam, Sept-Îles and elsewhere in Québec to provide the required manpower. TRAINING At each of the Project locations in Québec, every effort will be made to train and employ the greatest number of Aboriginals and Quebecers. Well before commissioning begins, a master training program to provide suitable employment training will be set up and will include structured training in partnership with aboriginal and regional institutions. In addition, arrangements will be made for on-the-job internships to enable employees to acquire experience in companies having similar operations; and for on-the-job training. It is anticipated that equipment suppliers will actively participate in the latter process. Training expenditures were assumed to be necessary during the year ahead of commissioning and during the first year after personnel has been hired. However, it is understood that various sources of financing for training

179 Technical Report on the KéMag Pre-Feasibility Study Page 179 may be available through the federal government and, possibly, the provincial government, and training costs were therefore excluded from the estimate.

180 Technical Report on the KéMag Pre-Feasibility Study Page 180 Table 63: Manning by Category Superintendents, Managers & General Foremen Supervisors & Engineers Operators Maintenance Personnel Technicians Others Total Mine Mine & Concentrator Site Services Crushers Concentrator Pipeline Flotation Plant Pellet Plant Concentrate Filter Plant Pointe-Noire Site Services Stockyard & Shiploading Administration Total ,230 * Operator/Repairman

181 Technical Report on the KéMag Pre-Feasibility Study Page FINANCIAL ANALYSIS GENERAL This Section describes the method of analysis, the basic assumptions made, and the findings of the analyses to evaluate the viability of the KéMag Iron Ore Project to produce and sell 15 million tonnes per year of pellets, plus 7 million tonnes per year of concentrate as pellet feed fines at an average 69.1% Fe. The analyses were performed using estimates of capital and operating costs, an estimated construction schedule and an estimated production schedule, all as set out elsewhere in preceding Sections of this report. In addition, financial conditions of required funds were examined. All financial amounts were expressed in fourth quarter 2008 Canadian dollars. Prices obtained in American dollars were converted at the rate of US$ 0.85 = Can$ The financial analyses were made on the basis of a financing structure assuming a mix of equity, suppliers credit and funds borrowed from commercial banks. The estimates and assumptions were fed into a financial model constructed on COMFAR III-Expert software, developed by UNIDO. The COMFAR software produced an Income and Cash Flow Statement, a Balance Sheet, and other financial schedules for the chosen financial structure. The Internal Rate of Return ( IRR ) was calculated according to the discounted cash flow methodology, and sensitivity analyses were undertaken REVENUES Details related to tonnages and sales for iron ore pellets and concentrate are given in Table Following discussions between NML management and experts in the global iron ore market, it was decided that the FOB prices of pellets and concentrate used for this Pre-Feasibility Study would be based on projected long-term prices provided by Crédit Suisse. Therefore, for pellets at 66.5% Fe, the average price used was Can$ per tonne and for concentrate at 69.1% the average price was Can$69.10 per tonne. DR grade pellets sales assumed an 11.7% premium over the price for blast furnace pellets with an Fe content of 67.5% For the purposes of the financial analyses, no inflation was applied to those prices, which were assumed to be constant for the life of the Project. There will be no revenue during construction from Year 1 to the end of Year 3. From Year 4 and in subsequent years, estimated sales tonnages and revenues are as shown in Table 64.

182 Technical Report on the KéMag Pre-Feasibility Study Page 182 Table 64: Estimated Sales Tonnages and Revenues by Year Year Year 4 Year 5 Year 3-Year 28 (per yr) Blast Furnace Tonnes 7,000,000 9,000,000 10,000,000 Pellets $ , ,559 1,056,176 DR Grade Tonnes 3,500,000 4,500,000 5,000,000 Pellets $ , , ,632 Concentrate Tonnes 4,900,000 6,300,000 7,000,000 $ , , ,700 Total $ 000 1,490,656 1,916,558 2,129, EXPENSES Operating expenses were generated on a year-to-year basis in fourth quarter 2008 Canadian dollars. Expenses were developed on a year-by-year basis for the mine, to reflect the evolution of the pit, and as a yearly average for other sectors of the operation CAPITAL EXPENDITURES The capital cost of the Project was estimated to be approximately $4,451.1 million, including direct costs of $3,442.7 million and indirect costs of $ million. The indirect cost includes an estimated amount of approximately $139.4 million for initial start-up requirement. Initial working capital estimated at $ 31.0 million is not included in the capital cost estimate. In addition, for working capital purpose, accounts receivable and receivable have been established at 30 days. Table 65 provides a breakdown of the estimated capital expenditure.

183 Technical Report on the KéMag Pre-Feasibility Study Page 183 Table 65: Estimated Capital Expenditure (Can $) Component Million $ Mine (*) Crushers (including HPGR) Concentrator Pipeline 1,023.1 Pellet Plant Port Facilities Power Supply Railway 5.9 Sub-total Direct costs 3,442.7 EPCM Owner s costs Contingency Sub-total Indirect costs 1,008.4 Total 4,451.1 (*) Mining equipment is leased and not included in the initial capital cost Details of the capital expenditures, presented in Section 19.9, also appear in the financial model under Sources and Application of Funds, as well as capitalized interest charges during construction of $240.7 million. The levels of capital expenditures were not adjusted for future inflation and the initial expenditures stated above were scheduled for disbursement over the entire period of construction. In addition to the initial capital expenditures, a provision of $688.3 million was made for sustaining capital to cover equipment replacement and increasing mining fleet and tailings handling and storage requirements FISCAL CONSIDERATIONS AND DEPRECIATION Given the complexity of the taxation issue with regard to the various taxation levels and numerous allowances involved, special software supplied by a taxation expert was used for tax evaluation. The results obtained from that software were then input into COMFAR for the after-tax financial analysis. The following fiscal conditions were assumed to apply: For federal and Québec provincial corporate income tax:

184 Technical Report on the KéMag Pre-Feasibility Study Page 184 o Federal income tax rate of 15%; o o o o Provincial Income tax rate of 11.9% for subsequent years; Accelerated depreciation of 25% per year up to 100% on Class 41A mining concentrator, pipeline, pellet plant and power supply assets,; Depreciation of 25% on the declining balance for Class 41B mining and port installation assets; Canadian development expenditure depreciation on the basis of 30% per year; o Canadian exploration expenditure depreciation on the basis of 100%: o Mining duties. For Québec provincial mining tax: o Mining tax rate of 12%; o o Processing allowances of 15% per year of the cost of processing assets, up to a maximum of 65% of the profit for the year; Northern mine allowance of % of the cost of processing assets, deductible in the first ten years of production; o Exploration and development expenditures deductible at 100%; o Depreciation up to 100% of capital expenditures. The combined effects of depreciation and allowances result in tax holidays of about six years for corporate income tax and eight years for the provincial mining tax. The impact of taxes on project profitability is demonstrated by the After Tax IRR shown in Table RESIDUAL VALUE For the purpose of this financial evaluation, it was assumed that any revenue realized from the sale of fixed assets at the end of the Project would be offset by the cost of rehabilitating the sites. Therefore, a net residual value of zero was used PROJECT FINANCING The financial model was created to address the case where equity is assumed to be around 30% of the Project capital cost. To the extent possible, the disbursement of the borrowed funds was delayed in order to reduce the amount of interest payable.

185 Technical Report on the KéMag Pre-Feasibility Study Page 185 As financing was not yet in place, for the 30% equity analyses the annual interest rate was assumed to be 7.00% (Average U.S. Prime for 2005, 2006 and first quarter %) for the Project. The mix of long-term borrowed funds was assumed to be: Suppliers credits arranged with their equipment as collateral for a period of seven years and amounting to $1,781 million or 38% of the total long-term project financing; One or more loans from one or more commercial banks for a period of 10 years, totalling $1,516 million or 32% of the total long-term project financing. The assumed terms of repayment were single payment at the end of each civil year starting at the end of the first full year of production. It is important to note that no additional work was done to confirm these assumptions with potential lenders. The financing terms are used only to show the positive leverage of financing FINANCIAL RESULTS The results of financial analyses with the Net Present Value ( NPV ) discounted at 8% for each of the before and after tax cases are presented in Table 66. The results show that the Project generates sufficient funds to service debt and has an attractive return on investment. Table 66: Results of Financial Analyses Before taxes* After taxes Project IRR (%) Equity ROE (%) Payback (Years from production start-up) 5 5 Net Present Value ($ 8% 8,588 5, SENSITIVITY ANALYSIS A sensitivity analysis was prepared by measuring the effect of variations of up to ± 20% in key parameters on the Project IRR for the case Before corporate and provincial mining taxes. The selected parameters were: Revenue Capital Expenditure Annual Operating Costs As shown in Table 67 and Figure 34, for the pre-tax case the viability of the Project is most sensitive to variations in Revenue, and least sensitive to variations in Annual Operating Costs.

186 Technical Report on the KéMag Pre-Feasibility Study Page 186 Sensitivity analyses were also carried out for the case After corporate and provincial mining taxes and are included in the financial. Table 67: Sensitivity of Project IRR to Variations in Key Parameters Variation (%) Sales Revenue (%) Capital Cost (%) Annual Operating Costs (%)

187 Technical Report on the KéMag Pre-Feasibility Study Page 187 Figure 34: Sensitivity of IRR to Variations in Key Parameters (Pre-tax) 34% KéMag Iron Ore Project Sensivity Analysis (pre-tax) Internal Rate of Return (IRR) 32% 30% 28% IRR(%) 26% 24% 22% 20% 18% 16% 14% -20% -16% -12% -8% -4% 0% 4% 8% 12% 16% 20% Variation (%) OPEX CAPEX SALES

188 Technical Report on the KéMag Pre-Feasibility Study Page PROJECT MASTER SCHEDULE GENERAL Assuming that a Feasibility Study can be completed and Environmental Permitting obtained in the next two years, it is estimated that the Project can be engineered such that: Site preparatory work can start early spring year -4; Construction will start in the early spring of year -3; Start-up will begin in the last quarter of year -1; Sale of pellets will start in the first quarter of year 1; The full production rate will be achieved in mid-year 1. The major activities to be undertaken in that timeframe are described hereafter and shown on the bar-chart presented as Figure 35 at the end of this Section. The scheduled development of the mine and construction of the concentrator hinge on the timely transportation of all construction material and equipment from the port of Sept Îles to Ross Bay Junction on IOCC s QNS&L railway and from there on the TRT railway to Schefferville, where the Project includes a railway yard with sidings, offloading ramps, explosive siding and mobile handling equipment. It has been assumed that TRT will take the necessary measures to ensure that the railway is upgraded to meet KéMag requirements and schedule and that TRT will acquire all the required rolling stock. NML considers that the QNS&L railway system can handle all the required material movement. FEASIBILITY STUDY The Feasibility Study should begin in the second quarter of year -5 and carry on for five quarters based on the results of the Preparatory Studies listed below. In order to improve efficiency during the Feasibility Study, Preparatory Studies should be made of a number of key Project elements, including: Slurry Pipeline final routing and scope refinement; 315 kv Overhead line ( OHL ) final routing and scope refinement; Pellet Plant scope review, tender document content and confirmation of final bidders list; Tailings Containment final scope review and construction scope definition;

189 Technical Report on the KéMag Pre-Feasibility Study Page 189 Stockyard and Ship loading site selection; Concentrator plant final site selection; Flowsheets review and finalization; Layouts review and finalization; Downtime probability; Options Analysis followed by Value Engineering; Metallurgical test work; Geotechnical analysis; Aerial surveys; Feasibility Study planning; Issuing of some tender documents. ENVIRONMENTAL WORK AND PRE-ENGINEERING Pre-release Engineering: This engineering includes: Engineering required to obtain Pre-release permits, i.e. the permits delivered for site access and temporary work such as construction workers camp, generating station, etc. Engineering required to initiate process Project Release permits, i.e. the permits delivered for the construction of permanent installations such as the crushers, concentrator, slurry pipeline, 315 kv OHL, pellet plant, tailings containment dams, pumping stations, etc. Preparation and issue of procurement packages for some major long-lead equipment. Pre-release Environmental Work: This work comprises: Environmental studies required to obtain Pre-Release permits referred to above.

190 Technical Report on the KéMag Pre-Feasibility Study Page 190 Project Release Environmental Work: This work comprises: Environmental studies required to initiate the process to obtain Project Release permits referred to above Permit Application for construction (Pre-release): A period of time is required to request and obtain Pre-release Permits. Permit Application for construction (Project release): A period of time is required to request and obtain Project Release Permits CONSTRUCTION ENGINEERING, PROCUREMENT AND CONSTRUCTION MANAGEMENT ( EPCM ) This activity covers the detailed engineering, procurement and construction management of all elements of the project. It must start in the first quarter of year -4 and continue to the end of the construction phase. CONCENTRATOR SITE PREPARATORY WORK Construction is expected to take 16 months and must start early in the spring of year -4. POWER TRANSMISSION LINE Construction is expected to take 30 months and must start early in the spring of year-3. CRUSHERS AND CONCENTRATOR Construction is expected to take 30 months and must start early in the spring of Year -3. Construction will be controlled in part by the long delivery items such as crushers, HPGRs, mills, mill gears and transformers, all of which will need to be purchased early in the detailed engineering phase. SLURRY PIPELINE Three construction packages are contemplated, one for the main pumping station at the concentrator and the booster pumping station some halfway along the pipeline, one for the pipeline itself, which may be split between two contractors; and one for the slurry reception facility at the pellet plant site. Construction is expected to take 30 months, and advantage will be taken of the frozen state of muskeg areas in the winter. Pipeline construction will start simultaneously at the concentrator and at the pellet plant, with temporary construction camps located

191 Technical Report on the KéMag Pre-Feasibility Study Page 191 so that travel time to the construction sites is never more than one hour. Pipeline construction must be completed to permit start up during Year-1 so that the line is in operation before the winter sets in. Further details are given in the PSI 2007 Report. FLOTATION PLANT Construction will be such that the plant will be ready for start-up at the same time as the pellet plant. PELLET PLANT Based on the Danieli offer, construction is expected to take 30 months and must start early in the spring of Year -3. PRODUCT STOCKYARD AND SHIP LOADING FACILITIES Based on the Howe 2007 Report, construction is expected to take 30 months and must start early in the spring of Year -3. COMMISSIONING Commissioning will be performed in a timely and orderly manner but will have to be completed by the end of Year -1 START-UP A team of NML operations personnel will carry out the start up of all plants and facilities in a timely and orderly manner

192 Technical Report on the KéMag Pre-Feasibility Study Page 192 Figure 35: Preliminary Master Schedule

193 Technical Report on the KéMag Pre-Feasibility Study 20. INTERPRETATION AND CONCLUSIONS Based on the findings and results presented of the Pre-Feasibility Study, BBA believes that the KéMag Iron Ore Project is a world-class deposit technically feasible and economically viable. Based on the assumptions presented in the Study, the KéMag Project, with an estimated initial capital cost of $4,451 million, can achieve an internal rate of return (IRR) of 25.2% and a net present value (NPV) of $8,588 million using a discount rate of 8%. The payback period after the start of the commercial production is 4 years before taxes. The level of accuracy of the capital and operating costs is +/- 25%. The mining equipment needed to start the production is provided as an operating lease and is added in the operating cost. Using the latest and modern technology available today and having no legacy burden, costs of mining, processing and pelletizing for the KéMag Project will be among the lowest-cost iron producer in North America. The KéMag Project also benefits from the low power cost in the Province of Quebec and the quality of the ore, which is hard but not difficult to concentrate. Furthermore, transportation of iron concentrate by pipeline is established to be technically feasible as demonstrated by successful existing mineral slurry pipelines in cold weather on similar projects and represents a substantial saving in operating cost in comparison with rail transportation. Based on an estimate of mineral resources produced by GEOSTAT in compliance with the National Instrument , the KéMag Project hosts a very large resource of iron ore. The mineral reserves are estimated by BBA at 2,141 million tonnes in the Proven and Probable categories and will be sufficient for over 25 years of operation at the presently planned rate to produce 10 Mtpy of blast furnace pellets, 5 Mtpy of DR pellets and 7 Mtpy of iron concentrate. To date, the difficulty in obtaining a bulk sample representative of the KéMag deposit has precluded any large-scale metallurgical test work. This situation should be remedied for the next phase of test work during the Feasibility stage. The duration of the Project schedule is 5.5 years from the time a decision is made to continue engineering until the start of commercial production. Such period accounts for the identification of a strategic partner, the Feasibility Study, the financing, detailed engineering and construction. The project is located in the Province of Québec and it is expected that there will be no political or regulatory risk associated to the development of the KéMag Iron Ore Project Technical Report March 2009

194 Technical Report on the KéMag Pre-Feasibility Study 21. RECOMMENDATIONS Prior to advancing the project through the Feasibility Stage, there are a number of areas where BBA is of the opinion that some opportunities exist where development and operation can be enhanced or the risk reduced. BBA believes that NML may benefit from studying all the possible ways to achieve the lowest capital and operating costs structure, as well as minimising risks. Recommendations for the Project include: Review and optimization of the process and layout including preliminary engineering activities for plant layouts; Review of tailings disposal solution and construction; A large bulk sample representative of the ore body should be collected for pilot plant test work during the feasibility study phase, for a final flowsheet optimization; Environmental permitting works to be initiated (i.e. Project Notification); Hydrogeology and geotechnical studies to be initiated; Hydro-Québec should be requested to begin detailed studies to provide power to the mine site; Detailed topographic mapping for the slurry pipeline routing and optimization should be undertaken prior the Feasibility Study; The waste rock is not believed to be acid generating and confirmatory test work should be carried out at the Feasibility Study Stage Technical Report March 2009

195 Technical Report on the KéMag Pre-Feasibility Study CERTIFICATES Technical Report March 2009

196

197

198 CERTIFICATE OF QUALIFIED PERSON I, André Allaire, Eng., do hereby certify that: 1. I am currently employed as Director Mining and Metals and partner in the consulting firm: Breton Banville et Associés () 630 boul. René-Lévesque Blvd. W Suite 2500 Montréal, Québec Canada H3B1S6 2. I graduated from McGill University of Montréal with a B. Eng in Metallurgy in 1982, a M.Eng. in 1986 and a Ph.D. in I am in good standing as a member of the Order of Engineers of Québec (#38480) and a member of the Canadian Institute of Mining Metallurgy and Petroleum. 4. I have practiced my profession continuously since my graduation. 5. I have read the definition of qualified person set out in the National Instrument ( NI ) and certify that as a result of my education, affiliation with a professional association (as defined in NI ) and past relevant work experience, I fulfill the requirements to be a qualified person for the purposes of NI I have not visited the project site. I prepared Section 18.0, 19.8 and I prepared Section 1.0 and with co-author John Dinsdale. I prepared Section 19.9 with coauthor Langis Charron. I prepared Section 20.0 to 21.0 with co-authors John Dinsdale and Langis Charron. 7. I have not had prior involvement with the property that is subject of the Technical Report. 8. As of the date of this certificate I am not aware of any changes in fact or circumstances with respect to the subject matter of this report which materially affects the content of the report or the conclusion reached. 9. I am independent of the issuer applying all of the tests in Section 1.5 of National Instrument I have read National Instrument and Form F1, and the Technical Report has been prepared in compliance with that instrument and form.

199

200 Technical Report on the KéMag Pre-Feasibility Study Page 200 CERTIFICATE OF QUALIFIED PERSON I, Langis Charron, Eng., do hereby certify that: 1. I am currently employed as Project Manager and partner in the consulting firm: Breton Banville et Associés () 630 boul. René-Lévesque Blvd. W Suite 2500 Montréal, Québec Canada H3B1S6 2. I graduated from Cole de Technologie Supérieure of Québec with a B. Eng Technology in 1979 and a B. Eng in Electrical Engineering in I am in good standing as a member of the Order of Engineers of Québec (#103088). 4. I have practiced my profession continuously since my graduation. 5. I have read the definition of qualified person set out in the National Instrument ( NI ) and certify that as a result of my education, affiliation with a professional association (as defined in NI ) and past relevant work experience, I fulfill the requirements to be a qualified person for the purposes of NI I have not visited the project site. I prepared Section 19.6 with co-author André Allaire. I prepared Section 20.0 to 21.0 with co-authors John Dinsdale and André Allaire. 7. I have not had prior involvement with the property that is subject of the Technical Report. 8. As of the date of this certificate I am not aware of any changes in fact or circumstances with respect to the subject matter of this report which materially affects the content of the report or the conclusion reached. 9. I am independent of the issuer applying all of the tests in Section 1.5 of National Instrument I have read National Instrument and Form F1, and the Technical Report has been prepared in compliance with that instrument and form.

201

202

203