Technical Paper HIGH PURITY CALCIUM ALUMINATE BINDERS FOR DEMANDING HIGH TEMPERATURE APPLICATIONS

Transcription

1 Page : 1/14 HIGH PURITY CALCIUM ALUMINATE BINDERS FOR DEMANDING HIGH TEMPERATURE APPLICATIONS C. PARR, D. VEYRAT and Ch. WÖHRMEYER Kerneos SA, 8 rue des Graviers, Neuilly sur Seine, 92521, France J. P. LETOURNEUX Lafarge Centre de Recherche, St Quentin Fallavier, 38291, France Presented at Calcium Aluminate Centenary Conference, Avignon, July 2008

2 Page : 2/14 ABSTRACT In order to ensure the continued substitution of shaped refractories by monolithic products and at the same time render the downstream refractory monolithic products more robust there is an ongoing need to deliver more stable and regular calcium aluminates. This paper presents a study of the behaviour of calcium aluminates in the two types of refractory concretes from the initial placing and setting properties through to the installed thermomechanical properties. The study shows that as the quantity of calcium aluminate in model formulations decreases then the system becomes intrinsically less robust and more sensitive to variation in internal formulation parameters. In particular, the impact of variation within calcium aluminates on the functional behaviour is studied. A specific focus is given to new and analytical techniques that can be used to more comprehensively characterise interactions within the binder phase and provide a new insight into the underlying mechanisms. Conclusions show that refractory concretes can be rendered more robust through a careful management of the characteristics of the high alumina content calcium aluminates. Keywords: aluminate, cement, castable, high temperature applications, LCC, monolithic, refractory,

3 Page : 3/14 1 Introduction Calcium aluminate cements as binders for monolithic refractory castables, have a rich history of over 80 years. They have been the backbone for development of increasingly sophisticated refractory products offering enhanced durability. Monolithic refractory systems have evolved from high cement conventional formulations to the present era of reduced cement castable systems that can be placed using a variety of techniques such as casting, gunning, self flow, pumping and shotcreting [1]. The usage chain of monolithics containing calcium aluminate cement contains several steps such as mixing, placing and consolidation, curing and dry out and finally use in service. Each of these steps is intimately linked to the hydration process of calcium aluminate cement (CAC). The role of CAC depends to a great extent upon the system type. In the case of high cement conventional systems, the CAC acts as the fundamental binder, dictating to a significant extent, the placing and the placed properties. In the case of reduced cement or low cement castable (LCC) systems the situation is quite different. The (CAC) behaves in these systems, as more of a chemical reactive than a fundamental binder system. Therefore, it is to be expected that the placing properties can be perturbed by changes in the underlying chemical and physical reactions linked to the CAC hydration. In order to ensure reliable and repeatable placing properties these interactions need to be manipulated and controlled. The paper details the effect of two different 70% alumina content calcium aluminate binders and the effect of their basic properties upon two different castable systems. The behaviours of the two CAC in a conventional regular high cement content refractory castable are compared in terms of their impact upon placing properties and installed properties. This is contrasted with a similar analysis in a reduced cement deflocculated castable system. The study shows that as the quantity of calcium aluminate in model formulations decreases then the system becomes intrinsically less robust to internal formulation parameters. In particular, the impact of variation within calcium aluminates on the functional behaviour is studied. The underpinning mechanisms that explain the visible placing properties is investigated by an analysis of the chemical reactions that take place during the placing phase of deflocculated and reduced cement castables. An experimental procedure has been developed to study reactions within the matrix binder phase (calcium aluminate, alumina, fume silica, and additives) in a concentrated suspension similar to that used in complete deflocculated castables. This has provided new evidence of the interactions that occur during the placing phase of deflocculated castables. Measurement techniques of concentrated suspension conductimetry analysis, in parallel with existing techniques of ultrasonic velocity and thermal analysis are performed. During the placing and stiffening phase periodic extractions of the pore solution are made using a new approach to determine the various ionic concentrations. Conclusions show that refractory concretes can be rendered more robust through a careful management of the characteristics of the high alumina content calcium aluminates.

4 Page : 4/14 2 Experimental details Model systems A series of model castable systems have been used to study the behaviour of two different 70% Alumina CAC. The details of the castable types are shown in Table 1. The LCC-FS system is a generic low cement castable based upon reduced cement content with a deflocculated matrix comprised of reactive alumina, fumed silica and calcium aluminate cement. The additive, sodium tripolyphosphate (TPP) is added in variable quantities from 0.02% to 0.14% based upon 100% castable dry weight. The RCC is a regular conventional castable based upon a graded aggregate and 15% calcium aluminate cement. In addition the study of fundamental reactions has necessitated the use of simplified mortar systems based upon the same binder phase and with a re calculated particle size distribution. A binder phase mortar (LCCFS2 mortar) comprising of the fine part of a model LCC (LCCFS2) has been derived from the full sized LCC. The particle size distribution has been recalculated from a maximum size of 10 mm to 3 mm in the case of the mortar (Table 2). The aggregate was changed to a very dense brown fused alumina (BFA) compared to the LCCFS model system as this would minimise effects due to aggregate porosity. Different water additions were studied with a w/c of the mortar ranging from 0.72 to The water content was determined on the basis of achieving at least 100% initial vibration flow (vibration 30 seconds at 0.5 mm amplitude with initial cone diameter of a lower diameter of 100 mm and an upper diameter of 70 mm. The height of the cone measures 50 mm.). The working time was measured for the whole castable system (WT=120 min). This is sometimes referred to as the setting time which is somewhat too ambiguous a term as it can lead to confusion with a setting time associated with hydrate formation in classical or conventional refractory concrete. Table 1. Composition of Model Castable systems and corresponding mortar Material (wt. %) LCC FS RCC LCC FS2 Mortar: LCCFS2 Sintered alumina, -7 mm to 0 80 Chamotte, 45% alumina 80 BFA: 10-1 mm BFA: 3-1 mm BFA: -1 mm Fume silica Reactive alumina, (3,3 m 2 /g) % alumina cement Sodium tripolyphosphate (TPP) variable Water (%) w/c CAC basic properties Phase quantification: is determined by a quantitative Rietveld phase analysis using a protocol previously developed [2]. Chemical composition is determined by x-ray fluorescence. Measurement of placing properties Placing properties for each system were characterized through the flow profile and the working time. For each sample two measurements of the placing properties were made and the average reported.

5 Page : 5/14 Flow value: determined using a cone with 100 mm base diameter, 50 mm high and 70 mm top diameter. The cone is placed on a vibrating table (according to the ASTM 230C) filled with the castable, then taken away and subjected to 20 seconds of vibration. Flow value resulting of the cake is calculated as a percentage as follows: FV (%) = (cake diameter initial diameter) / initial diameter * 100 Working time: is taken as the time after mixing at which the castable will not flow under vibration. Measurement of castable hardening kinetics: Hardening kinetics were followed by exothermic profiles. These were determined at 20 C with castable samples placed in an insulated chamber. A thermocouple (type K), imbedded in the cast sample, is linked to a data capture system and the temperature recorded as a function of time [3]. Conductimetry analysis The basic apparatus is described in numerous publications [4]. The circular electrodes are around the circumference of the conductimetry cell. A modified cell is used here which allows the safe removal of stiffened/hardened mortars without damaging the cell at low w/c ratios. A w/c of 1.44 is used which is somewhat higher than the real systems (0.72), but several factors lower than previous studies [4] and much closer to the highly concentrated matrix suspensions of real castables. The data acquisition and processing is automated and as in previous studies and publications. Exothermic profiles Hardening kinetics was followed by exothermic profile, determined at 20 C with castable samples placed in an insulated chamber. A thermocouple, imbedded in the cast sample, is linked to a data capture system and the temperature recorded as a function of time [5]. Ultrasonic profiles A special ultrasonic cell has been used to follow the wave propagation during the first hours after mixing of the concrete with water. A detailed description of the test method is given by Simonin, Wöhrmeyer and Parr (2005) [6]. The results are expressed as speed vs. time. These graphs allow the observation of structuring during the early phase after mixing and of the subsequent hydration process of the concrete. Pore solution extraction and analysis The mortar is wet mixed for a total of 8 minutes and the resulting wet mixed mortar is used to fill test tubes which are placed in a centrifuge. Samples are subsequently centrifuged at 4500 rpm for 3 minutes after a predetermined time interval. The liquid is recovered, filtered, diluted and acidified to precipitate the fume silica residue. These extractions are analysed by ICP-AES. It is possible that the extracted pore solution contains some colloidal particles that are not visible during the analysis. Similarly other residues (such as AH 3 ) may also be dissolved. The technique requires further refinement to eliminate these potential effects. Therefore a certain caution is needed when analysing the results. Hydrate characterisation and analysis Hydration is stopped by acetone-ether treatment [7]. At pre-determined intervals by washing/filtration of the mortar with acetoneether, drying and screening out the +80 micron residue. The resulting samples are analysed by XRD (if crystalline species can be identified), carbon and water content and via DTA

6 Page : 6/14 3 Results Basic properties of two 70% Alumina CAC CAC mineralogy has a large impact on placing properties, the castable workability as well as the hardening. As shown in Table 2 the higher the C/A ratio the greater the reactivity of the CAC. However, all commercial products are comprised of a multi phase assemblages and when considered together the effect of each phase is not linear. This is illustrated using two different 70% alumina cements. Table 2: Primary mineralogical and chemical composition of two 70% alumina cements Main Phase Unit CAC A CAC B C 12A 7 /CA CA/CA Al 2O 3 CaO SiO 2 % % % 70,9 28,4 < 0,2 72,1 27,0 <0,2 Surface area/ Blaine cm 2 /g Conventional dense castables The placing and installed properties for the two CAC types in the model regular conventional castables are shown in Table 3. As can be seen the flow properties are somewhat similar in terms of absolute flow values and the flow decay. The installed properties as measured by compressive strengths are also similar for both CAC types. The differences in basic CAC characteristics did not manifest as significant differences in placing properties of regular conventional castables. Table 3: Placing and installed properties for RCC model castable system Property Unit CAC A CAC B Water demand % Vibration flow After mixing After 30 mins. After 60 mins Working time % % % mins Time to peak temperature mins Strength : CCS After 24 hours After drying 110 C After firing at 815 C MPa MPa MPa Reduced Cement Deflocculated castables CAC A and B were tested using the LCC-FS as the base castable with TPP as the sole additive at 5% water addition in both cases. The TPP addition was varied from 0 up to 0.14%. The results can be seen in Fig. 1.1 and 1.2. At low TPP additions large differences are seen in the placing properties between the two cements with CAC A always showing shorter working times. A similar trend (not shown) was observed for the flow values. As the TPP addition increases the working time values for the two cements tend to converge together. The same trend is seen for the exothermic profiles, with CAC B, which takes longer to reach the exothermic maximum. At lower TPP additions, CAC B gave extremely good placing characteristics with long working times in excess of two hours. The use of CAC A on the other hand, gave extremely short working times that would severely limit the ability to optimise or modify placing properties of this castable. Clearly, CAC A in the LCC-FS system is far from an optimum in terms of the trade off between placing properties and hardening. CAC A has more rapid flow decay than CAC B and presents more difficulties in finding optimum placing characteristics. CAC A presents less adapted placing properties which are difficult to optimise based upon a single additive. CAC B shows different characteristics in that the placing properties are excellent and the strength acquisition is only marginally slower than CAC A. These manifest differences show that the importance of basic CAC properties in these reduced cement castable systems.

7 Page : 7/14 In order to investigate further a study of a reduced cement binder phase system was studied with a view of understanding more clearly the different interactions in the LCC type system that could explain the observed differences. Working time (min) CAC/A CAC/B 0 0,02 0,04 0,06 0,08 0,1 0,12 0,14 TPP (%) Time to exothermic peak (h) ,02 0,04 0,06 0,08 0,1 0,12 0,14 TPP (%) Fig. 1.1 The effect of TPP on working time Fig. 1.2 The effect of TPP on exothermic peak time ( LCC FS formulation ) (LCC FS formulation) Analysis of binder phase One of the problems faced with the analysis of binder phase interactions in reduced cement castable systems is the highly concentred matrix phases with only 4 to 5% water added to the entire castable. A potential solution is to use a mortar based system which is derived from the full castable. However, it is important to validate that the mortar characteristics in terms of placing properties correspond to those seen with the complete castable. A single addition of TPP was chosen on which to base the analysis using the LCC FS2 model formulation. Three methods, conductimetry, ultrasonic profile and the exothermic profile have been compared simultaneously (Fig. 2) using the same mortar formulation as detailed in Table 1. During the first 2 hours a pseudo plateau is observed which reveals a slightly decreasing conductivity. The temperature increases slightly during this period and the ultrasonic propagation speed remains stable. At around 2 hours the measured values of the three techniques increased rapidly indicating [4] an increased ionic activity, chemical activity and the start of stiffening. At 5 hours, the conductivity can be seen to increase to a maximum prior to a rapid decrease. There is also a rise in the propagation speed and at the same time the temperature rises to a maximum (28-29 C) during the period 5 to 8 hours. Thus, it can be seen that the different events as, measured by the three techniques, are simultaneously linked to the developing structure. The event at 2 hours corresponds with the working time of the complete LCC. It should be noted that this is longer than in the previous example shown in Figure 1 due to differences in the choice of aggregate type and a different ratio of reactive alumina. The initial flow deceases during the period up to 120 minutes.

8 Page : 8/14 A further validation was performed by comparing the ultrasonic curves of the complete LCC and the corresponding mortar. The results showed that the different events occurred at the same time but the final propagation speeds were higher in the case of the complete LCC. Fig.2. Comparison of ultrasonic, exothermic and conductivity curves for the LCC mortar at 1.44 w/c Analysis of pore solution In order to analysis the pore solution a mortar (table 2 mortar LCC FS2), concentrating the binder phase was constructed from the LCC FS2 model formulation. Periodic extractions were made of the pore solution of the LC mortar during the workable period and prior to the rapid loss of flow and stiffening associated with the arrival of the end of the working time. The first extractions were made after 10/15 minutes from water addition. Due to mixing effects it was not possible run extractions earlier.

9 Page : 9/14 Fig. 3. Ionic concentration as a function of time after mixing for the LCC mortar Progressive extractions were made between 15 and 120 minutes from mixing and the change in ionic concentration of each species (Figure 3) measured via ICP. During this period significant change in ionic concentrations is observed which is somewhat contrary to the picture from the conductimetry analysis where an initial pseudo plateau is seen. A number of observations can be made: 1. Alumina passes progressively into solution with an accelerated dissolution between 60 and 80 minutes 2. Calcium is present in an initially higher concentration than alumina increases up to 110 minutes then decreases 3. Phosphate (from the TPP) has a high initial value which decreases progressively then more rapidly at the same time (from 80 minutes) as the calcium.. Magnesia and silica in relatively low concentrations follow the same trend as the calcium and phosphate 5. The concentration of the alkalis remains relatively stable albeit at a high level in the case of sodium (from TPP) After the period of rapid stiffening it is no longer possible to extract pore solution by centrifugal forces and another technique would have to be developed to study this period

10 Page : 10/14 4 Discussion Diagnostic studies [8] dating back to 1989, using calorimetry and conductivity, proposed a mechanism whereby the TPP formed an insoluble precipitate with the calcium and the fume silica adsorbed calcium on to its surface. [1] The initial pseudo plateau was interpreted as a blocking of the calcium aluminate dissolution. The mechanism was postulated that the insoluble phosphate was precipitated on the surface of the cement as an impermeable or semi-impermeable layer of calcium phosphate and or C-S-H gel. The TPP is slowly consumed via the formation of this layer. The current experiments and analysis conducted in a more concentrated solution than previous work clearly raises questions about the validity of this hypothesis. More recent studies [9-11] have also cast new light on the role of fume silica, and the nature of the interactions between CA-TPP FS. Role of sodium tripolyphosphate Reference is made to figure 3 and the change in ionic concentration with time. The concentration of TPP added to the mortar equates to 26.4 mm/l of Na 5 P 3 O 10. This corresponds to a theoretical maximum value of 66mM/l of Na 2 O and 71mM/l of P. An initial concentration of 63mM/l of Na 2 O is measured. The initial concentration of P is measured at 60 mm/l this decreases to 40mM/l at 90 minutes and only 1mM/l at 117 minutes which, is end of the workability period. Thus, the phosphate is removed from solution at the same time as the workability decreases to zero. Both the initially measured concentration of sodium and phosphate would suggest that effectively all of the TPP has passed into solution. The difference between the initially measured value 60mM/l and the theoretical maximum of 71mM/l is presumed to be principally linked to the quantity of phosphate adsorbed on to the surfaces. The calcium ion concentration is initially 15mM/l and rises to a maximum of 21.6mM/l before dropping to a minimal value of 2.25mM/l. It appears as if the precipitations of calcium and phosphate are intimately linked. It should be noted that during the time the TPP is in solution the calcium is complexed by the NaTPP and does not necessarily exist in the form of a free ionic species. The concentration of alumina (which is a major constituent of the calcium aluminate cement) is only 4.8mM/l at 13 minutes and increases slowly to a value of 40.6 at 117 minutes. Silica increases to 7.4mM/l, decreases slowly then drops rapidly at the same time as the calcium and the phosphate The concentrations of calcium and alumina that are found are clearly very different from the situation of a calcium aluminate binder by itself where the lime and alumina concentrations increase simultaneously. If the measured alumina is assumed to come from the mono calcium aluminate phase then the following observations can be made: The initial concentration of alumina should be equivalent to that of the calcium and more than 10mM/l. It is possible that some of the alumina has already passed into solution and been precipitated as AH 3 in a gel form. This being promoted by the drop in ph that is observed which reduces the alumina solubility. It cannot be said, as previously imagined that there is a blockage of dissolution of the calcium aluminate phases. At the end of the working time the alumina concentration reaches 40 mm/l whereas the calcium drops to effectively zero. Logically, the calcium concentration should be of a similar value. If this 40 mm/l is considered plus the additional 10 mm/l (corresponding to the part of alumina precipitated as AH 3 ) then a total of 50 mm/l of calcium has been precipitated.

11 Page : 11/14 Thus, at the end of this working time it is logical to suggest that the calcium has precipitated as a form of phosphate with a concentration of 50 mm/l of calcium and 70 mm/l of phosphate. Considering different forms of calcium phosphates, the most likely is calcium tripolyphoshate ( Ca 5 (P 3 O 10 ) 2 ). This can exist in several forms of a gel and as a crystallised species. Based on the change of concentration it is presumed that the formation of calcium tripolyphoshate (TPP-Ca) is progressive, and there is an ionic exchange between the sodium and the calcium. In summary, the TPP-Na is believed to adsorb on the particle surfaces to ensure a deflocculation. The TPP which isn t absorbed forms complexes with calcium and magnesium, which helps to stabilise the solution rich in alumina and silica. The phosphate forms calcium tripolyphosphate by ionic exchange which results in the removal of phosphate from solution and this provokes the rapid stiffening of the castable at the end of the working time. Therefore the dissolution of the calcium aluminate is not blocked but regulated by the formation of the tripolyphosphate complexes which are then precipitated. Unfortunately there is no experimental protocol available today to confirm the formation of TPP-Ca and this hypothesis. Furthermore further analysis at different TPP concentrations is merited. Role of the fume silica The presence of fume silica gives rise to a relatively high concentration of silica in solution with values rising to 8 mm/l which equates to 300 mg/l. This is only possible as the calcium is complexed with the phosphate otherwise the silica would normally precipitate as a C-S-H species when calcium ion concentrations are present up to 20 mm/l. In the same manner a high concentration of silica would precipitate very rapidly any calcium ions coming from a dissociation of the TPP complexes or further dissolution of calcium aluminates. It can be imagined that the soluble silica plays a role of regulator and retarder with respect to calcium that is free in solution. It also delays the formation off TPP- Ca which as a consequence lengthens the working period. From the concentrations measured (Figure 3) it appears as if the alumina does not precipitate. It is not unreasonable to imagine the formation of C-S- H species as a result of soluble silica reacting with calcium. Implications for 70% alumina cement From this analysis it can be deduced that the reactivity, in terms of the chemical activity, of calcium aluminate cements must be carefully controlled if stable application placing properties are to be achieved. Compared to regular conventional systems which are more robust to small changes in CAC properties, reduced cement castables require very stable CAC characteristics as a pre-requisite to predictable placing properties. Small changes in CAC reactivity can manifest themselves as large changes in placing properties of the castable. Logically, improvements targeting a reduction in variability of the 70% alumina cement as a primary objective would yield significant benefits for the user and installer of low cement castables. During the last decade a significant reduction in standard deviations of the basic cement properties, of a commercial 70% alumina cement, with respect to flow and set have been achieved as shown in Figure 4. This reduction in variability of cement reactivity is shown in figure 4 with RT 1 representing initial variability and RT2 representing reduced variability at the end of the decade in The corresponding variation induced in castable placing properties as a result of initial cement variability (RT 1 ) is schematically represented by Cv. The reduction in variability shown as RT 2 yields a much lower variation in castable placing properties represented by Cv 2. Through continuous improvement programmes, advances in process, manufacturing technology and analytical

12 Page : 12/14 capabilities a significant advance in the management of the intrinsic reactivity of 70% alumina cements has been achieved. Fig. 4. Continuous Improvement to reduce variation in reduced cement castable placing properties which translates into a significant progress in For example, extensive studies have been conducted in order to develop a detailed knowledge of the crystallographic structures, the effect of impurities on solid solutions and minor phases of calcium aluminate binders. This has been coupled with the use of new X- Ray detectors. These provide high signal to noise acquisition spectra leading to improved accuracy and reproducibility for quantitative determination of major and minor critical phases such as CA, CA 2 and C 12 A 7. This can be illustrated by the analysis of commercial cements where small amounts of terms of measurement capabilities. Therefore all calcium aluminate phases, even those present in minor amounts, can be determined quantitatively with a higher degree of precision than was previously possible. This developed capability of X-Ray analysis and quantitative calcium aluminate phase analysis in terms of hard and software has produced a constant evolution in terms of enhanced regularity and product stability. Consequently, more stable placing properties of reduced cement castable systems are a result. These improvements have been made in response to the evolution synthetic C 12 A 7 were added to confirm that this of castable technology. The historical phase can be accurately measured down to advances in CAC technology have often been +/-0,1% in the case of high purity 70% in parallel with the development of new alumina cements. The reproducibility at 9 castable types different concentrations of C 12 A 7, comparing measured values to added values, gives a correlation coefficient of more than 0.995

13 Page : 13/14 5 Conclusion High performance low cement castables are more sensitive to small differences in reactivity of calcium aluminate cements. This is due to the nature of chemical reactions that occur in the binder phase during the placing. Whilst regular conventional concretes can be considered as more forgiving, in that small differences in basic CAC properties do not have a dramatic effect upon placing properties. New techniques have allowed the study of chemical interactions during the initial stages of castable placing This provides a new light on the different matrix interactions and in particular provides a much clearer explanation of the mechanism of action between the phosphates; fume silica and calcium aluminate phases which control the period of flow decay and initial stiffening; The sodium tripolyphosphate absorbs on the surface of the particles in order to deflocculate them. Residual phosphate in solution forms calcium and magnesium complexes and is able to stabilise the solution in the presence of high ionic concentrations of alumina and silica. The progressive formation of calcium polyphosphate by ionic exchange with the sodium results in a reduction of phosphate concentration and eventually the end of working is reached when no more phosphate is in solution. A portion of the fume silica is transformed into soluble silica in solution and this limits the quantity of non complexed calcium and slows down the formation of calcium tripolyphosphate hence it can be considered to play a part in the regulation of working time. Furthermore, the analysis has shown that the pseudo plateau seen during conductimetry analysis, which had previously been attributed to the blockage of the Calcium aluminate phases, is in fact a coincidence. Unfortunately no techniques are available to characterise and confirm the existence of the calcium polyphosphate and the C-S- H hydrates. Therefore the hypothesis established to explain the mechanisms remains unconfirmed. In addition, it is possible that the extracted pore solutions contain some colloidal particles that are not visible during the analysis. Similarly other residues may also be dissolved. Therefore a certain caution is needed when analysing the results. One of the limits of this approach is that it is not possible to extract pore solutions by centrifugation after stiffening has occurred. The technique also requires further refinement to eliminate the potential effects of colloid particles This analysis explains why small differences calcium aluminate cement reactivity with water can yield significant differences in placing properties due to chemical interactions in the binder phase. Stable installation properties of LCC castables can only be achieved through the use of 70% alumina cements produced to exacting standards with tightly controlled and restricted permissible variation.

14 Page : 14/14 6 References 1 W.E. Lee, W. Vieira, S. Zhang, K. Ghanbari-Ahari, H. Sarpoolaky, C. Parr: Intl. Mat. Reviews, 46 [3] pp (2001) 2 T. Füllman, G. Walenta, T. Bier, B. Espinosa, K.L. Scrivener: World Cement Research, June pp (1999) 3 C. Alt, L. Wong, C. Parr, Measuring castable rheology by exothermic profile. UNITECR, Cancun, Mexico, pp (2001). 4 Th. A. Bier, A. Mathieu, B. Espinosa and J. P. Bayoux, The Use of Conductimetry to Characterise the Reactivity of Calcium Aluminate Cements, UNITECR, Sao Paulo, Brazil, pp (1993) 5 C. Alt, L. Wong, C. Parr, Measuring castable rheology by exothermic profile, Refractories applications, Vol. 8, No. 2, pp (2002) 6 F. Simonin, Ch. Wöhrmeyer, C. Parr, A new method for assessing calcium aluminate cements, UNITECR, Orlando, USA, pp (2005) 7 Möhmel, W. Gessner, B Ködderitzsch, A comparison of two different methods for stopping the hydration of monocalcium aluminate at 20 C, International Cement Conference, Goteborg, Sweden, vol.3,3iii007 (1997)