This article, based on a paper presented at the 13th International Congress on the Chemistry of Cement at Madrid in July 2011, describes a radical, step-change approach to reducing CO2 emissions from cement manufacture.
A highly novel method of cement mineral synthesis using a molten salt process yields a fine-grained product that requires little or no grinding. The specific sustainability impacts that may arise from a molten salt route have been assessed and quantified against the conventional manufacturing route in terms of resource efficiency (energy consumption and waste output), supply chain influences and management of the production process.
World-wide, mankind produces over 3Bnt/yr of cement,1,2 most of which is used to make an estimated 10km3 of concrete – more than an order of magnitude greater than the combined volume of all other man-made materials (Figure 2). Although the carbon dioxide (CO2) embodied in concrete is low (0.4t/m3) compared to other common construction materials such as glass (2t/m3), steel (10t/m3) and polymers (40t/m3),3 the sheer volume of cement produced world-wide presents a serious climate change challenge simply because of the amount of CO2 emitted during its manufacture - almost 1t of CO2 per tonne of product.
Proof of principle
Despite recent advances in the design of cement kilns and alternative, low-energy clinker compositions, advances in the reduction of embodied CO2 to date have been only incremental and are now yielding diminishing returns.
With this in mind, on reading a journal article some 15 years ago about the synthesis of ceramic oxide compounds at relatively low temperatures in molten salts,4 one of the authors (AM) was inspired into speculating as to whether such an approach might be applicable to the low-energy production of hydraulic cements. This idea was believed to be highly novel and to represent a step-change in the reduction of process energy and of CO2 emissions.
A short experimental exercise was undertaken to demonstrate proof of principle. Financial support for a full-scale project was not obtained until 10 years later and results from the Ultra-Low Energy Cement Synthesis (ULECeS) project, funded by the UK EPSRC and carried out in the Department of Chemical
Engineering at University College, London, have been recently reported in considerable detail.5
Molten salts
Molten salts, also known as melts or fused salts, are a class of inorganic compounds that are made into liquids by heating. They possess a number of beneficial properties that make them attractive to industrial processes in which complex reactions take place and non-hazardous solvents are required. Molten salts are:
- Good solvents,
- Generally environmentally benign,
- Excellent heat-transfer media,
- Highly stable at high temperatures,
- Cheap and easy to prepare,
- Able to attain high temperatures for rapid reaction,
- Chemically stable at the reaction temperature,
- Non-volatile, with generally low vapour pressures.
Such salt systems, for example alkali metal nitrates at 200–600ºC, have been investigated by Douglas Inman and colleagues at Imperial college, London, as solvents for the synthesis of ceramic materials with considerable success.4 This work has shown substantial reductions in the temperatures required to synthesise a wide range of compounds in powder form, without the need for grinding, yielding a considerable saving in the total energy required.
Although advances in molten salt processing have been made by many other industries, notably the nuclear, metal-refining and ceramics sectors, this technology has not yet been applied to the commercial manufacture of other oxide powders such as cements.
Experimental approach
The reagents used comprised calcite (CaCO3) and α-quartz (SiO2) as reactants and sodium chloride (NaCl), which has a melting point of 801°C, as the molten salt solvent. Stoichiometric quantities of the reactants and the halide salt were homogeneously mixed, placed in alumina crucibles and introduced into a muffle furnace where the temperature was slowly raised over a period of between 2.5hr to 3.5hr up to 908°C. The samples were then allowed to react at this temperature for a further two to three hours. The heating was then switched off and the samples were allowed to cool slowly in the furnace.
The synthesis of β-Ca2SiO4 in molten NaCl was largely successful. Scanning electron microphotographs (Figure 1) show that the β-C2S has the shape of milky droplet globules. The C2S crystallites are clearly visible in the frozen salt, which is constrained by its own surface tension during cooling.
Even when the intended product was C3S, the major product was still β-C2S. In these samples, a certain quantity of unreacted CaO was detected in the solidified reaction mixture. This points to the need for temperatures higher than 1140°C in order to stabilise C3S, through a Lux-Flood type acid-base reaction where Ca2SiO4 behaves as the acid reacting with CaO as the conjugate base.
Results
An important result of the present work is the successful preparation of the cement compound β-C2S at a temperature of only 908°C over a period of just two to three hours. This temperature and reaction time are significantly lower and shorter than the 1000-1200°C (>12 hrs) used in a typical synthesis of belite by solid state reaction. Production of β-C2S was achieved without adding any doping elements, which is a common practice to stabilise this polymorph under ambient conditions, where the γ-C2S is normally dominant.
It is also interesting to note that no evidence of calcium chlorosilicates is found in the X-ray diffractogram (Figure 3). This finding may reflect the relatively low activities of calcium and silicate ions in the melt such that the solubility limit of these phases was not reached. This is encouraging, even though chloride-containing cements based on the mineral alinite, whose clinker composition typically includes 1–3% chloride, have been used extensively in the USSR, suggesting that complete removal of chloride from the product may not be necessary for commercial use.
Sustainability implications
It is generally accepted that there is diminishing scope for improving conventional kiln calcination for manufacturing OPC, even though 10% or more of input heat must be lost through the refractories to avoid softening the mild steel casing. The process relies on nodularisation for efficient heat transfer and therefore the speed of reaction, which renders alternative technologies such as fluidised beds less effective. It also relies heavily on fossil fuel burning, mainly because alternative fuels are not available in sufficient quantity, despite some (eg: rubber tyres) possessing moderately high calorific values.
The molten salt route proposed is shown in outline in Figure 4. As to whether this approach offers a genuine advantage over conventional cement production is an interesting question to which we cannot yet offer a complete answer. Were it to be developed commercially, it would be dependent on the efficient removal of the reaction products from the solvent, which remains the subject of an on-going study. Nevertheless, we envisage a number of step-change improvements in our proposed process, which include:
- Low reaction temperature that greatly reduces thermal energy requirements, allowing combustion of non-fossil fuels (such as bio-fuels),
- Because molten salts are conducting, the process could be powered entirely by elecricity made from renewable sources,
- Grinding can be eliminated or at least considerably reduced,
- Novel cement types, compositions and particle-size distributions are all feasible,
- A smaller, more versatile plant can be envisaged,
- In principle any waste of a suitable chemical composition could be used as a raw material.
As far as process steps are concerned, the principal advantage with a molten salt process lies in the possibility of significantly reducing the heating energy, together with eliminating clinker grinding, which would save a further 10% of total energy requirements.6 Furthermore, the reaction temperature is low enough for the process to be powered entirely by electrical means, with a source that could feasibly be completely renewable and fossil-fuel-free.
Comparative sustainability evaluation
Despite the difficulty of comparing a novel process so far carried out batch-wise in the laboratory on the scale of a few grammes with that involving a conventional kiln producing continuously several thousand tonnes of cement a day, we believe that it is important to assess its potential sustainability at this early stage. Furthermore, we frankly do not yet have a clear vision of how our proposed process might be configured on a commercial scale.
Because of these obstacles and since it is difficult to assign a precise numerical value to some of the impacts, sustainability data has been presented in the form of a risk-based model developed from the principles of safety hazard analysis. This model is assembled by assigning a numerical rating (1 to 5 for 'negligible' to 'very high') for the likelihood of each sustainability impact actually occurring. A numerical rating (5 to 1 for 'negligible' to 'complete') for a possible mitigating strategy is then added to offset the likelihood of the occurrence.
Similarly, numerical impacts and offsets are assigned for the severity of the consequence of the impact occurring. The overall sustainability impact is calculated as a probability by multiplying likelihood by severity, resulting in impacts with numerical values in the range 4-100%.
A range of 16 process industry sustainability metrics were identified as significant, representing four themes of sustainability, namely 1. resources, 2. emissions, 3. economics and 4. social issues. For example, with conventional kiln calcination, the likely use of fossil fuel as the primary energy source is rated very high (5), with little scope for mitigation (4): the severity of the impact is high (5), again with little scope for mitigation (4), giving a sustainability impact score of (5+4) x (5+4) = 81. For the proposed molten salt process (which could be electrically powered) the likelihood of fossil fuel use is very low (rating 1) with a high degree of mitigation possible (2): the severity of the impact is moderate (3) with no need for mitigation (3), giving an impact score of (1+2) x (3+3) = 18. Data for the other 15 metrics are compiled in the same way to produce the summary shown as Figure 5.
Conclusion
At this early stage in the project, it is probably not of great value to read too much significance into Figure 5 except that it highlights the advantages and limitations of a molten salt process and suggests directions for further investigation. Nevertheless, the clear advantages of the new process in reducing fossil fuel consumption and CO2 emissions are very evident.
It must also be noted that the proposed process is still a long way from producing poly-mineralic compositions similar to OPC. However, we have shown that it is possible to produce single clinker minerals by the molten salt method and such minerals may be added to cements in order to enhance their performance or to provide specific properties. Furthermore, preparation in molten salts has the potential to produce minerals over a wide range of particle sizes, including on the nano-scale.
References
1. Maries A. et al.: 'A sustainability analysis of a potential low-energy route to cement production by synthesis in molten salts.' Paper 405, Proc. 13th Int. Congr. Chem. Cement, Madrid, July 2011.
2. Gartner E.: 'Are there any practical alternatives to the manufacture of Portland cement clinker?' Private Communication, 2011.
3. Hammond G.; Jones C.: 'Inventory of carbon and energy (ver. 2.0).' Sustainable Energy Research Team, Dept. of Mechanical Engineering, University of Bath, UK, 2011.
4. Du Y.; Inman D.; Morgan H.: 'Making ceramic powders from molten salts.' Materials World, 4, 458, 1996.
5. Photiadis G.M. et al.: 'Low energy synthesis of cement compounds in a molten salt.' Adv. Appl. Ceramics 110 (3), 137, 2011.
6. Gartner E. 'Industrially interesting approaches to 'low-CO2' cements.' Cem. Conc. Res., 34, 1489, 2006.