Using Slag in Manufacturing Masonry Bricks and Paving Units

Transcription

1 Using Slag in Manufacturing Masonry Bricks and Paving Units Yasser Korany 1 and Salah El-Haggar 2 Steel production involves reducing the iron ore, after being sintered or belletized, in a blast furnace to obtain hot metal containing some 4% carbon and smaller quantities of other alloying elements. The slag generated from this process is refereed to as Blast Furnace Slag (BFS). The hot metal is then converted to steel in a basic oxygen furnace where the carbon percentage is controlled to obtain desired properties such as high strength. Another type of slag results from this process and is called Basic Oxygen Furnace Slag (BOFS). This production process is known as the integrated production method. An alternative method of production is the secondary processing, often called minimills, in which steel scrap is melted in an electrical arc furnace. The slag generated from the minimills is called the Electrical Arc Furnace Slag (EAFS). In Egypt, 136,000 imperial tons (600,000 metric tons) of BFS, 68,000 imperial tons (300,000 metric tons) of EAFS, and 45,000 imperial tons (200,000 metric tons) of BOFS are generated annually. Such slag is not only hindering the use of land for more useful purposes, but also contaminates it. This paper introduces green construction materials, whereby different slag types are proposed to replace coarse aggregates in producing cement masonry bricks and paving interlock units. Three different slag replacement levels were investigated, namely: 33%, 67%, and 100%. Masonry bricks were tested for bulk density, water absorption, compressive strength, and flexural strength. Paving units were examined for bulk density, water absorption, compressive strength, and abrasion resistance. Heavy metals content and water leaching tests were also conducted for all slag types under investigation to assess the health impact. ENVIRONMENTAL PROBLEMS The problem of such slag generated from the iron and steel industry is not only hindering the use of millions of square meters of land for more useful purposes, but also contaminating it. Many slag types contain some heavy metals such as barium, titanium and lead. Also, it is well known that toxic substances tend to concentrate in slag. Furthermore, due to the likely leaching, ground water is susceptible to serious pollution problems. 1 Ph.D. Canadidate, Department of Civil Engineering, McMaster University, 1280 Main St. W., Hamilton, ON, Canada L8S 4L7. 2 Professor of Energy and Environment, Engineering Department, The American University in Cairo, Cairo, Egypt. Besides the economic and technical importance of utilizing slag, this activity is of great importance from the environmental protection point of view. The first environmental impact of such utilization is the useful consumption of the huge stockpiles of these waste materials. When slag is used as a replacement for other products such as cement, the natural resources that serve as raw materials are preserved. Another important impact that has been almost completely overlooked is the considerable loss of energy contained in slag. The temperature of the molten slag at the furnace outlet is 1,500 C (2,732 F), then it drops to 1,300 C (2,372 F) when slag reaches the disposal yard. According to the information published by the Iron and Steelmaker Journal, iron-making produces a BFS that amounts to 20-40% of the hot metal production. In Western Europe and in Japan, virtually all slag produced is utilized either in cement production or as road filling. In Egypt, almost two thirds of the BFS generated is utilized in cement production. Some kg (220-9,700 lb) of BOFS is produced for every imperial/metric ton of steel made in the BOF, with an average value of 120 kg/metric ton (530 lb/imperial ton). At present, about 50% of BOFS is being utilized worldwide, particularly for road construction and as an addition to cement kilns. EAF produces about 116 kg (450 lb) of slag for every imperial/metric ton of molten steel. Worldwide, about 77% of the slag produced in EAF is reused or recycled. The remainder is used as land-fill or dumped. Due to the relatively high iron content in EAFS, screens and electromagnetic conveyors are used to separate iron to be reused as a raw material. The remaining EAFS is normally aged for at least 6 months before being reused or recycled. EXPERIMENTAL PROGRAM The cement used in the study was ASTM type I normal Portland cement having a specific surface area of 3,133.5 cm 2 /gm (220,285 in 2 /lb) and initial and final setting times of 2.45 and 4.00 hours. Three different slag types were investigated namely; air-cooled BFS, BOFS, and EAFS. The chemical composition of slag types used in the study is given in Table 1 and the sieve analysis is shown in Table 2. The fine aggregate used was clean desert sand having a maximum particle size of 2.50 mm (0.1 in.) and fineness modulus of 2.1. The coarse aggregate used was dry dolomite having a bulk density of 1.61 t/m 3 (100 pcf), crushing strength of 18.2 MPa (2,639 psi), and maximum particle size of mm (0.5 in.). TMS Journal September

2 Table 1. Chemical Composition of BFS, BOFS, and EAFS Chemical Analysis (%) BFS BOFS EAFS Silicon Dioxide (SiO 2 ) Aluminum Oxide (Al 2 O 3 ) Iron (Fe) Ferric Oxide (Fe 2 O 3 ) Calcium Oxide (CaO) Magnesium Oxide (MgO) Sulfur (S) Sulfur Trioxide (SO 3 ) Manganese Oxide (MnO) Phosphorous Oxide (P 2 O) Phosphorous Pentoxide (P 2 O 5 ) Potassium Oxide (K 2 O) Chromium Oxide (Cr 2 O 3 ) Titanium Oxide (TiO 2 ) Sodium Oxide (NaO 2 ) Barium Oxide (BaO) Potassium Oxide (K 2 O) Basicity Index The mix proportions of masonry bricks were those used practically by most local cement masonry manufacturers. However, slight changes were made to achieve higher consistency and a workable batch suitable for manual casting. The cement content was increased by 20% and the water/ cement ratio was increased by 25%. These modifications were based on laboratory trial mixes. The mix proportions of the control batch used were 1: 3: 2.5: 0.5 for cement: sand: dolomite: water. BFS, BOFS, and EAFS having maximum and minimum particle sizes similar to those of dolomite were used to replace dolomite at three different levels: 33%, 67%, and 100%. Dolomite was replaced by slag on an equal volume basis. Table 3 shows the mix proportions of the different masonry brick groups under investigation. Table 2. Sieve Analysis of BFS, BOFS, and EAFS Sieve Cumulative Retained (%) No. BFS BOFS EAFS # 1/ # 3/ # # # # Eighteen specimens having the dimensions of 25 x 12.5 x 6 cm (10 x 5 x 2 3 / 8 in.) were cast for each group as shown in Figure 1. Twelve specimens were tested for compressive strength and 6 specimens were tested at 28-day age for flexural strength and absorption. All masonry brick specimens were air-cured until testing rather than water cured in order to simulate the curing practice followed by the industry. The specimens prepared for the water absorption test were checked for dimensions and weighed to determine the bulk density. The mix proportions of the paving interlock units were determined based on the information available from local producers in Egypt. Adjustments to the mix proportions were necessary for higher workability and ease of mold removal. The mix proportions of the control group were 1: 1.73: 1.33: 0.4 for cement: sand: dolomite: water. Table 4 shows the mix proportions of the different mixes and the details of the paving interlock groups. Eighteen paving specimens having the dimensions of 20 x 16 x 8 cm (8 x 6 3 / 8 x 3.25 in.), shown in Figure 1, were cast for each group. At 28-day age, 3 specimens were tested for water absorption according to ASTM C936 and another 3 were tested for abrasion resistance according to ASTM C Twelve specimens were tested for compressive strength according to ASTM C140 at different ages. All specimens were air-cured until testing applying the curing practice followed by the manufacturers. The specimens Table 3. Mix Proportions of Masonry Brick Groups Group Cement Sand Dolomite Water Slag Slag Repl. Desig. lb (Kg) lb (Kg) lb (Kg) lb (Kg) lb (Kg) Type (%) R c 485 (220) 1,455 (660) 1,213 (550) 242 (110) R (220) 1,455 (660) 811 (368) 242 (110) 417 (189) BFS 33 R (220) 1,455 (660) 401 (182) 242 (110) 844 (383) BFS 67 R (220) 1,455 (660) 242 (110) 1,261 (572) BFS 100 R (220) 1,455 (660) 811 (368) 242 (110) 485 (220) BOFS 33 R (220) 1,455 (660) 401 (182) 242 (110) 983 (446) BOFS 67 R (220) 1,455 (660) 242 (110) 1,468 (666) BOFS 100 R (220) 1,455 (660) 811 (368) 242 (110) 485 (220) EAFS 33 R (220) 1,455 (660) 401 (182) 242 (110) 983 (446) EAFS 67 R (220) 1,455 (660) 242 (110) 1,468 (666) EAFS TMS Journal September 2001

3 1.50" 8.00" 3.00" 6.50" 2.50" 1.50" 4.50" 3.25" 10.00" 5.00" Masonry Brick Unit Paving Interlock Unit Figure 1 Typical Test Specimens (1 in. = 25.4 mm) prepared for the water absorption test were also checked for dimensions and weighed to determine the bulk density. Twenty-four cubes 50 x 50 x 5 0 mm (2 x 2 x 2 in.) were cast for heavy metals and water leaching tests and divided into 4 groups as shown in Table 5. Two different media were prepared: fresh deionized water having a ph value of 6.25, and 10% concentration sulphuric acid solution having a ph value of In addition, another 3 groups of ground BFS, BOFS, and EAFS were tested for comparison purposes. Half the test specimens were immersed in 300 ml (10 oz) of the 10% sulphuric acid for 2 weeks for the determination of heavy metals content according to DIN S 7, while the other half was immersed in 300 ml (10 oz) of the deionized water for 4 weeks for the water leaching test according to DIN S 4. A pure sample of the fresh deionized water was also tested to determine the background levels of heavy metals in water. After the conditioning period, the media were filtered using No. 50 filter paper and analyzed for heavy metals concentrations by the atomic absorption spectrometry. EFFECTIVENESS OF SLAG IN PRODUCING MASONRY BRICKS Table 6 shows the average bulk density and water absorption results of masonry brick groups. Figure 2 shows a comparison between the absorption of the different slag types. It was found that all slag types resulted in higher bulk density values than the control group. However, a slight decrease in bulk density was noticed for BFS groups. At all replacement levels, BOFS and EAFS groups showed similar bulk density values higher than that of the control group. The relatively similar bulk density values of BFS groups and the control group may be attributed to the similar bulk densities of BFS and dolomite. The same reason may be applied to explain the close bulk density values of BOFS and EAFS groups and to explain the higher density values of both BOFS and EAFS groups over those of the control and BFS groups. The highest increase in bulk density was 14.6% for group R 6 where BOFS was used. Comparing the results of Table 4. Mix Proportions of Paving Interlock Groups Group Cement Sand Dolomite Water Slag Slag Repl. Desig. lb (Kg) lb (Kg) lb (Kg) lb (Kg) lb (Kg) Type (%) P c 992 (450) 1720 (780) 1323 (600) 397 (180) P (450) 1720 (780) 886 (402) 397 (180) 452 (205) BFS 33 P (450) 1720 (780) 437 (198) 397 (180) 9217 (418) BFS 67 P (450) 1720 (780) 397 (180) 1376 (624) BFS 100 P (450) 1720 (780) 886 (402) 397 (180) 529 (240) BOFS 33 P (450) 1720 (780) 437 (198) 397 (180) 1072 (486) BOFS 67 P (450) 1720 (780) 397 (180) 1600 (726) BOFS 100 P (450) 1720 (780) 886 (402) 397 (180) 529 (240) EAFS 33 P (450) 1720 (780) 437 (198) 397 (180) 1072 (486) EAFS 67 P (450) 1720 (780) 397 (180) 1600 (726) EAFS 100 TMS Journal September

4 Table 5. Mix Proportions of Heavy Metals Content and Water Leaching Test Groups Group Cement Sand Dolomite Water Slag Slag Repl. Desig. lb (Kg) lb (Kg) lb (Kg) lb (Kg) lb (Kg) Type (%) C c 485 (220) 1455 (660) 1213 (550) 243 (110) C (220) 1455 (660) 243 (110) 1261 (572) BF 100 C (220) 1455 (660) 243 (110) 1261 (572) BOF 100 C (220) 1455 (660) 243 (110) 1261 (572) EAF 100 masonry brick groups with those of the commercial brick group (R com ), it was found that the bulk densities of all test groups are below those of the commercial brick except for groups R 6 and R 9, where BOFS and EAFS fully replaced dolomite. From Table 6 and Figure 2, the water absorption percentages were found to be either comparable to or slightly higher than those of the control and commercial bricks. However, it should be noted that all specimens exhibited much lower absorption percentages than the ASTM limit of 13%. The lowest increase in water absorption was 1.14% for group R 7 where EAFS was used, while the highest increase was 26% for group R 3 where BFS was used. The higher water absorption percentages of slag groups may be directly attributed to the higher porosity of all slag types over the dolomite and aggregates used in producing the commercial bricks. Since the coarse aggregates represent more than 35% of the weight of the constituent materials of masonry bricks, they have substantial influence on the overall porosity. Table 7 summarizes the average compressive and flexural strength results for all masonry brick groups. The development of compressive strength at 100% replacement level is shown in Figure 3. The test results showed markedly higher compressive strength for all groups over both the control group and the commercial bricks. All masonry brick groups have reached higher compressive strength than the ASTM C requirement of 4.14 MPa (600 psi) for nonload-bearing masonry units at only 3-day age. All groups showed higher compressive strength values than the ASTM C90-97 requirement of 1.31 MPa (1,900 psi) for load-bearing masonry units at 28-day age. The compressive strength at 100% replacement level was close to or higher than the ASTM limit for load-bearing units at 7-day age. At all ages, increasing the slag replacement level resulted in increasing the compressive strength. The increase in compressive strength for BFS at 100% replacement level was 52% over the control, while for BOFS the increase in compressive strength was found to be 60%. The highest increase was 82% for group R 9, where EAFS was used. The higher compressive strength of masonry brick made of slag over that of the control and commercial bricks may be attributed to the higher crushing strength of slag over dolomite and aggregates used in producing the commercial bricks. The higher compressive strength of EAFS groups over that of BOFS groups and the higher strength of BOFS groups over that of BFS groups may be also attributed to the differences in crushing strength. Table 6. Average Bulk Density and Water Absorption of Masonry Brick Groups Group Bulk Density Dry Weight Wet Weight Absorption Desig. pcf, (t/m 3 ) lb (gm) lb (gm) (%) R c 141 (2.26) 8.52 (3,862) 8.89 (4,032) 4.40 R (2.24) 8.60 (3,896) 9.00 (4,083) 4.79 R (2.21) 8.66 (3,931) 9.12 (4,135) 5.18 R (2.20) 8.75 (3,966) 9.23 (4,187) 5.57 R (2.37) 8.87 (4,023) 9.27 (4,204) 4.49 R (2.48) 9.23 (4,184) 9.65 (4,375) 4.58 R (2.58) 9.58 (4,345) (4,548) 4.67 R (2.36) 8.91 (4,043) 9.31 (4,223) 4.45 R (2.45) 9.31 (4,223) 9.73 (4,413) 4.51 R (2.54) 9.71 (4,404) (4,605) 4.56 R com * 146 (2.34) 9.28 (4,210) 9.48 (4,300) 4.28 * Commercial samples tested for comparison. 100 TMS Journal September 2001

5 Water Absorption, (%) BF Slag BOF Slag EAF Slag Commercial Brick Replacement Level, (%) Figure 2 Comparison Between Absorption Ratios of the Different Slag Types Used in Producing Masonry Bricks All masonry brick groups showed higher flexural strength than the control and the commercial bricks. Flexural strength was found to increase along with the increase in the replacement level. EAFS groups showed the highest increase in flexural strength, followed by BOFS groups, then BFS groups. EFFECTIVENESS OF SLAG IN PRODUCING PAVING INTERLOCKS The average bulk density and water absorption results for all paving interlock groups were calculated and summarized in Table 8. All paving interlock groups resulted in bulk density values either comparable to or slightly higher than those of the control and commercial groups. It was found that BFS groups have bulk density values similar to that of the control group while BOFS and EAFS groups showed relatively higher bulk density values. Group P 6 prepared with BOFS at a 100% replacement level showed the highest bulk density value. The higher bulk density values of BOFS groups and EAFS groups over BFS groups and the control group may be attributed to the higher bulk density of BOFS and EAFS over that of BFS and dolomite. All paving interlock groups resulted in water absorption percentages either comparable to or slightly higher than that of the control group. A significant increase in the water absorption values was observed for all groups compared to the commercial samples. It should be noted that these commercial samples are high quality, factory-made with small size gravel. All groups showed water absorption ratios far below the ASTM limit of 13%. Increasing the replacement level resulted in increasing the water absorption percentage. The highest increase in water absorption over the control group was found for group P 3, at 100% replacement of BFS. The lowest increase was observed for group P 7, at 33% replacement of EAFS. The higher water absorption values of slag groups over those of the control and commercial groups may be attributed to the higher porosity of slag aggregates over that of dolomite and gravel. The average compressive strength results for all paving interlock groups are given in Table 9. Figure 4 shows a comparison between the development of compressive strength of paving interlock groups at 100% replacement level. The test results revealed higher compressive strength values for all paving interlock groups than the reference group. However, all groups resulted in lower compressive strength than the commercial samples. This may be attributed to the high quality hard-gravel used in producing this particular product, which is not the common practice. The adjustments made to the mix design could be another factor. Table 7. Average Compressive and Flexural Strengths of Masonry Brick Groups Group Compressive Strength, psi (MPa) Flexural Strength Desig. 3-day 7-day 14-day 28-day psi (MPa) R c 713 (4.92) 1,192 (8.22) 1,430 (9.86) 1,786 (12.32) 75 (0.52) R (5.18) 1,251 (8.63) 1,520 (10.48) 1,876 (12.94) 83 (0.57) R (5.82) 1,405 (9.69) 1,662 (11.46) 2,108 (14.54) 91 (0.63) R 3 1,086 (7.49) 1,813 (12.50) 2,227 (15.36) 2,716 (18.73) 100 (0.69) R (5.51) 1331 (9.18) 1,659 (11.44) 1,998 (13.78) 93 (0.64) R (6.15) 1,500 (10.35) 1,715 (11.83) 2227 (15.36) 109 (0.75) R 6 1,140 (7.86) 1,686 (13.61) 2,337 (16.12) 2,850 (19.66) 126 (0.87) R (6.21) 1,500 (10.35) 1,755 (12.10) 2,252 (15.53) 99 (0.68) R 8 1,012 (6.98) 1,686 (11.63) 2,075 (14.31) 2,530 (17.45) 115 (0.79) R 9 1,302 (8.98) 2,175 (15.00) 2,572 (17.74) 3,257 (22.46) 130 (0.90) R com * * * 1,740 (12.00) 73 (0.50) * Not available TMS Journal September

6 25 3,625 R c R3 R6 R9 20 2,900 Compressive Strength, (MPa) Commercial Brick ASTM Loadbearing 2,175 1,450 Compressive Strength, (psi) 5 ASTM Non-Loadbearing Age, (days) Figure 3 Development of Compressive Strength for Masonry Brick Groups at 100% Replacement Level All groups made using either BOFS or EAFS showed higher compressive strength values than the ASTM value of 55.2 MPa (8,000 psi). Groups made using BFS and the control group showed lower compressive strength, up to 18% less than the 55.2 MPa (8,000 psi) limit. The higher compressive strength of BOFS and EAFS groups may be directly related to the high crushing strength of BOFS and EAFS compared to BFS and dolomite. At all ages, increasing the slag replacement ratio resulted in increasing the compressive strength. The highest compressive strength was found to be 66 MPa (9,570 psi) for group P 9, made using EAFS at 100% replacement level. The lowest compressive strength was found to be MPa (6,665 psi) for group P 1 made using BFS at 33% replacement level. The average abrasion resistance results for all groups estimated based on ASTM C are given in Table 10. Figure 5 shows the effect of slag type on the abrasion resistance of paving interlocks. The test results revealed that all groups have higher abrasion resistance than the reference group. The abrasion resistance values of only groups P 6, P 8 and P 9 were comparable to or higher than that of the commercial group. All groups showed much lower abrasion coefficient than the ASTM limit of 15 cm 3 / 50cm 2 (1 in. 3 / 8 in. 2 ). The lowest abrasion coefficient was found to be 1.90 cm 3 /50cm 2 (0.12 in. 3 /8in. 2 ) for group P 9, while the highest was found to be 5.15 cm 3 / 50cm 2 (0.33 in. 3 / 8in. 2 ) for group P 1. Increasing the slag replacement level resulted in increasing the abrasion resistance. The high abrasion resis- Table 8. Average Bulk Density and Water Absorption of Paving Interlock Groups Group Bulk Density Dry Weight Wet Weight Absorption Desig. pcf, (t/m 3 ) lb (gm) lb (gm) (%) P c 128 (2.06) (4,611) (4,801) 4.13 P (2.06) (4,608) (4,821) 4.52 P (2.05) (4,605) (4,822) 4.71 P (2.05) (4,603) (4,824) 4.80 P (2.09) (4,675) (4,887) 4.45 P (2.12) (4,738) (4,955) 4.57 P (2.14) (4,802) (5,023) 4.60 P (2.10) (4,917) (5,129) 4.31 P (2.15) (4,815) (5,025) 4.38 P (2.20) (4,713) (4,923) 4.45 P com * 134 (2.15) (4,813) (4,932) 2.97 * Commercially available samples tested for comparison. 102 TMS Journal September 2001

7 Table 9. Average Compressive Strength of Paving Interlock Groups Group Compressive Strength, psi (MPa) Desig. 3-day 7-day 14-day 28-day P c 2,574 (17.75) 4,296 (29.63) 5,153 (35.54) 6,555 (45.21) P 1 2,665 (18.38) 4,440 (30.62) 5,330 (36.76) 6,665 (45.97) P 2 2,752 (18.98) 4,583 (31.61) 5,503 (37.95) 6,880 (47.45) P 3 2,842 (19.60) 4,733 (32.64) 5,680 (39.17) 7,105 (49.00) P 4 2,991 (20.63) 4,991 (34.42) 6,062 (41.81) 7,489 (51.65) P 5 3,062 (21.12) 5,101 (35.18) 6,125 (42.24) 7,656 (52.80) P 6 3,168 (21.85) 5,220 (36.00) 3,865 (26.66) 7,827 (53.98) P 7 3,225 (22.24) 5,380 (37.10) 6,460 (44.55) 8,068 (55.64) P 8 3,526 (24.32) 5,881 (40.56) 6967 (48.05) 8,819 (60.82) P 9 3,828 (26.40) 6,378 (43.99) 7,847 (54.12) 9,570 (66.00) P com ** * * * 8,092 (55.81) * Not available. ** Commercially available samples tested for comparison. tance of EAFS may be attributed to its higher wear resistance and higher bulk density over those of the other slag types. HEALTH IMPACT The test results of heavy metals content for all groups and the pure slag samples are listed in Table 11 while the results of water leaching tests are summarized in Table 12. A significant difference between the heavy metals concentration of pure slag samples and that of mortars containing slag was observed. It was noticed that BFS is nearly free 70 of lead (Pb) and cadmium (Cd). A very small amount of barium (Ba) was found for pure BFS samples while almost negligible concentrations were found for mortars containing BFS. Relatively high concentrations of chromium (Cr) and titanium (Ti) were obtained for pure BFS samples and groups containing BFS. It was found that the concentration of Cr and Ti has been increased by 93.6% and 21.5%, respectively, over the concentrations of the control group when BFS was used. The test results revealed minor concentrations of Ba and very small concentrations of Pb and Cd in pure BOFS samples and mortar groups containing BOFS. High con- 10,150 Compressive Strength, (MPa) Commercial Paving Stone ASTM Limit 8,700 7,250 5,800 4,350 2,900 Compressive Strength, (psi) 10 1,450 P c P 3 P 6 P Age, (days) Figure 4 Development of Compressive Strength of Paving Interlock Groups at 100% Replacement Level TMS Journal September

8 Table 10. Average Abrasion Resistance of Paving Interlock Groups Group Desig. Original Final Abraded Bulk Abrasion Abrasion Weight Weight Weight Specific Hardness Coefficient lb, (gm) lb, (gm) lb, (gm) Gravity Value in 3 /8in 2, (cm3/50cm 2 ) P c 9.92 (4497) 9.44 (4281) (216) (5.86) P (4499) 9.50 (4309) (190) (5.15) P (4500) 9.54 (4328) (172) (4.67) P (4502) 9.71 (4402) (150) (4.06) P (4622) 9.79 (4438) (184) (4.81) P (4748) (4596) (152) (3.84) P (4874) (4794) (120) (2.94) P (4626) 9.83 (4456) (170) (4.43) P (4753) (4628) (125) (3.10) P (4883) (4803) (80) (1.90) P com * (4845) (4785) (120) (2.94) * Commercially available samples tested for comparison. centrations of Cr and Ti were found for BOFS and mortars containing BOFS. It was found that embedding BOFS in a cement matrix slightly reduces the concentration of Cr while greatly reducing the concentration of Ti. Concentrations of Cr and Ti for the pure BOFS samples were found to be higher than those of the pure BFS samples. The heavy metals results showed minor Ba content for EAFS and small concentrations of both Cd and Pb. Higher concentrations of Cr and Ti were measured for the pure EAFS sample and mortars containing EAFS over that of BFS and BOFS. Utilizing EAFS as a coarse aggregate resulted in a decrease in Cr and Ti concentrations of 42% and 52.4%. Water leaching test results showed very small concentrations of heavy metals for all groups. It was clear that there is almost no trace of Ba in all test groups. This result was expected due to the minor Ba content in the different slag types. Pb and Cd leaching concentrations were very low and could hardly be detected. The Pb and Cd concentrations for all slag groups were almost identical to those of the control group. Only Cr and Ti showed some measurable concentrations. Comparing the results of water leaching tests with the World Health Organization guideline values for drinking Abrasion Hardness Value, (Ha ) EAF Slag BOF Slag BF Slag Commercial Paving Stone ASTM Limit Replacement Level, (%) Figure 5: Comparison Between The Abrasion Resistance of Slag Types Used in Producing Paving Interlock Units 104 TMS Journal September 2001

9 Table 11. Test Results of Heavy Metals Content Group Element Concentration, grain/gal (mg/l) Desig. Cd Pb Cr Ba Ti C c (0.000) (0.004) (4.044) (0.010) (2.568) C (0.000) (0.005) (7.829) (0.059) (3.119) C (0.520) (1.536) (30.669) (0.046) (31.650) C (0.721) (1.599) (49.850) (0.028) ( ) BFS* (0.000) (0.000) (12.683) (1.890) (19.653) BOFS* (1.525) (2.244) (39.664) (0.970) (81.631) EAFS* (1.666) (2.741) (86.135) (0.238) ( ) * Pure samples. water [WHO, (1992)], it was found that all groups showed Cd concentrations below the guideline value of mg/l ( grain/gal) and Pb concentrations well below the guideline values of 0.01 mg/l ( grain/gal). All groups except the pure BOFS and EAFS samples showed lower Ti concentrations than the guideline value of 0.1 mg/l ( grain/gal). On the other hand, all test groups resulted in higher Cr leaching concentrations than the value of 0.05 mg/l ( grain/gal). CONCLUSIONS This study was aimed at utilizing the slag generated from the iron and steel industry in producing environmentally friendly building materials. The chemical and physical properties of such slag, together with the previous research work, have led to the idea of utilizing these waste materials in producing masonry bricks and paving interlocks. Based on the results and observations of the experimental study presented, the following conclusions can be drawn: All masonry brick units made by utilizing BOFS and EAFS showed higher bulk density values than the control. The highest increase was found for the units made using BOFS. Water absorption values for all masonry brick groups were either comparable to or slightly higher than those of the control and commercial bricks. All groups exhibited absorption percentages well below the ASTM limit of 13%. Substantially higher compressive strength results were reached for all masonry groups at 28-day age compared to the control and commercial bricks. All groups showed higher compressive strength than the ASTM limit of 4.14 MPa (600 psi) for nonloadbearing masonry units at 3-day age. At 100% replacement level, all groups resulted in compressive strength higher than the ASTM requirement of 13.1 MPa (1,900 psi) for load-bearing units at 7-day age. Normal weight paving interlock units can be produced from slag with bulk density values comparable to the control and the commercial units. All slag types resulted in paving interlock units having water absorption values far below the ASTM limit of 13%. The absorption percentages of slag groups were comparable to or slightly higher than that of the control group. Table 12. Test Results of Water Leaching Group Element Concentration, grain/gal (mg/l) Desig. Cd Pb Cr Ba Ti C c (0.000) (0.000)** (0.093) 0.000** (0.000)** C (0.000) (0.000)** (0.124) 0.000** (0.000)** C (0.000)** (0.005) (0.124) 0.000** (0.000)** C (0.000)** (0.007) (0.193) 0.000** (0.000)** Water (0.000) (0.000) (0.000) (0.000) BFS* (0.002) (0.000) (0.230) 0.000** (0.041) BOFS* (0.002) (0.007) (0.310) 0.000** (1.123) EAFS* (0.025) (0.011) (0.541) 0.000** (1.524) * Pure samples. ** Measurement is under the detection limit of the instrument. TMS Journal September

10 All paving interlock groups showed higher compressive strength than the reference group. Both BOFS and EAFS groups resulted in compressive strength higher than the specified ASTM value of 55.2 MPa (8,000 psi) while BFS groups reached 82% of this limit. All slag types used in producing paving interlock units resulted in higher abrasion resistance values than dolomite used for the control group. The test results revealed much lower abrasion coefficients for all groups than the ASTM limit of 15 cm 3 /50cm 2 (0.33 in. 3 /8in. 2 ). Only at high replacement levels, BOFS and EAFS groups showed abrasion resistance values comparable to those of the commercial units. Based on the results of heavy metals content, considerable reduction in the heavy metals concentrations was observed for all slag types when used as aggregates for mortars. The highest heavy metals concentrations measured were chromium and titanium. Water leaching concentrations were found to be too small to be harmful according to the WHO guideline values for drinking water. ACKNOWLEDGEMENTS The authors would like to thank El-Ezz Steel Rebar Company and The Egyptian Iron and Steel Company for their generous contributions. The atomic absorption spectrometry analysis was performed at the Desert Research Institute, Faculty of Agriculture, Cairo University. Full thanks to all the technicians of the engineering department at the American University in Cairo where the experimental program took place. REFERENCES Akinmusura, J. O., Potential Beneficial Uses of Steel Slag Wastes for Civil Engineering Purposes, Resources, Conservation and Recycling, 1991, pp Featherstone, W. B. and Holliday, K. A., Slag Treatment Improvement by Dry Granulation, Iron and Steel Engineer, 1998, pp Malhotra, S. K. and Tehri, S. P., Development of Bricks from Granulated Blast Furnace Flag, Construction and Building Materials, V.10 No. 3, 1996, pp Naik, T., Singh, S., and Tharaniyil, M., Application of Foundry By-product Materials in Manufacture of Concrete and Masonry Products, ACI Materials Journal V. 93, 1996, pp Szekely, J., A Research Program for The Minimization and Effective Utilization of Steel Plant Wastes, Iron and Steelmaker, 1995, pp Taylor, H. F. W, Mohan, K. and Moir, G.K., Analytical Study of Pure and Extended Portland Cement Pastes: Pure Portland Cement Pastes and Fly Ash and Slag-Cement Pastes, Journal of The American Ceramic Society, V. 68, 1985, pp World Health Organization Report, Guidelines for Drinking Water Quality, Geneva, Switzerland, TMS Journal September 2001