Dense ceramics by cold reaction sintering using 95 % powdered construction and demolition waste and sodium silicate

Dense ceramics by cold reaction sintering using 95 % powdered construction and demolition waste and sodium silicate Sonia Marín-Cort´ es a,* , Aida Serrano a , Esther Enríquez b , Jos´ e F. Fern´ andez a a Instituto de Cer´ amica y Vidrio, ICV - CSIC, Kelsen 5, Madrid 29049, Spain b Instituto de ´ Optica "Daza de Vald´ es", IO - CSIC, Serrano 121, Madrid 28006, Spain ARTICLE INFO Keywords: Construction and demolition waste Sodium silicate Cold reaction sintering Ceramic Circular economy ABSTRACT This work explores the feasibility of producing ceramic-like materials from recycled construction and demolition waste through cold reaction sintering using sodium silicate as the liquid medium. The recycled material consists of a mixture of ceramics, mortar, and/or plasterboard - common building materials. Samples containing 95 wt% of recycled content were successfully fabricated at 150 ºC, a considerably lower temperature compared to conventional ceramic processes. These samples exhibited dense microstructures with relative densities of ~ 90 %, as well as promising mechanical properties, including Vickers hardness values comparable to conventional bricks. In particular, samples with 30 % non-ceramic material showed more than twice the hardness when mortar was incorporated. Leaching tests confirmed the chemical stability of the material, with the incorporation of Ca 2 ⁺ ions into non-leachable species, suggesting a reaction during CRS that enhances its durability. Additionally, FTIR analysis showed a blueshift in the signature from 973 cm⁻ 1 to 1041 cm⁻ 1 under more aggressive conditions, indicating structural changes and improved leachability. The primary advantage of this process would lie in the high recycled content used to create materials. This proposed new method allows obtaining dense pieces at very low temperatures, which also supposes a reduction of energy consumption, and therefore, an improvement of the sustainability of the production process. Given that this construction and demolition waste are often limited to not high added values, this approach not only supports environmental sustainability and resource conservation but also offers a viable pathway to decarbonizing the ceramic industry. 1. Introduction Climate change is one of the greatest contemporary challenges due to its direct influence on the planet’s future. In a world where resources are increasingly limited and the environment is heavily influenced by pollutants, it is of vital importance to adopt strategies that try to reverse these trends. Among the industrial activities, the ceramic sector plays a relevant role in this scenario with its transition towards net-zero emission strategies. Some examples are the decarbonization of the industry [1] by the responsible and efficient management of its waste [2] and the implementation of advanced technology for specific environmental challenges such as energy conversion and storage [3]. Ceramics are a globally used material in construction. They are present in a variety of applications, ranging from structural materials such as bricks to functional elements like tiles, and other products like sanitary ware, including toilets, sinks, and more. When the building is demolished, ceramic waste becomes part of the so-called construction and demolition waste (CDW) along with other construction materials such as concrete, plastic, plasterboard, glass, wood, etc., including excavating soil. CDW accounts for more than a third of the generated waste in the EU [4]. It is estimated that 77 % of this waste is composed of mineral fraction (bricks and concrete mainly), excluding soil, track ballast, dredging spoils, and asphalt. The breakdown of this percentage shows 31 % of concrete, 8 % of brick and tiles, and 61 % of mixed fraction. Gypsum is not considered in the mineral fraction and accounts for 1.4 % of the total [5]. Due to different aspects such as the lack of management policies and the difficulty of its separation and classification [4,5], CDW has a low recycling rate, and it is accumulated in stockpiles without knowing exactly the type of materials that are present. This practice is known as landfilling and represents one of the major challenges for CDW valorization, which relies on the correct identification and separation of the materials. In this sense, the authors * Corresponding author. E-mail addresses: [email protected] (S. Marín-Cort´ es), [email protected] (A. Serrano), [email protected] (E. Enríquez), [email protected] (J.F. Fern´ andez). Contents lists available at ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat https://doi.org/10.1016/j.conbuildmat.2024.138917 Received 13 August 2024; Received in revised form 1 October 2024; Accepted 23 October 2024 Construction and Building Materials 451 (2024) 138917 Available online 29 October 2024 0950-0618/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/bync-nd/4.0/ ). proposed a new way of classification and valorization of the CDW based on Raman spectroscopy and chemometrics analysis which allows for determining not only the kind of material but also its percentage in a CDW powder sample [4]. This new method facilitates the use of CDW in the industry. The silico-calcium-aluminous characteristics of the mineral fraction of this waste make it an interesting secondary raw material for the manufacture of new recycled construction materials. The ultimate application regarding the valorization of this debris is the alkaliactivated materials (AAMs) formation. These are considered alternative binders to Portland cement-based concrete, pursuing a material with a lower emission footprint [6–8]. The AAMs consist of the polymerization reaction between an aluminosilicate source, in this case, mainly the ceramic debris such as bricks, and an alkaline reagent that favors the formation of strong, insoluble binding phases [9–12]. When the precursors contain little or no calcium, they are called geopolymers. Different authors have tailored the Ca content to the aluminosilicate nature of the CDW by the addition of Ca sources such as CaSO 4 ⋅2 H 2 O, quick lime (CaO), and hydrated lime [13], slag [14] or waste concrete [15]. However, AAMs formation normally requires long curing times (days to weeks) where the complexity of the systems hinders the ability to meet all the performance requirements that a Portland cement exhibits, requiring an intensive tailoring to obtain explicit technical properties for a specific desired application [16]. Another promising way for valorizing CDW is by using it as a secondary raw material in the production of new ceramic tiles through conventional methods. In this context, various researchers have investigated the partial replacement of virgin raw materials with different types of waste in tile manufacturing [17–23]. However, the proportion of CDW used as a secondary raw material remains low, and most of these studies have been conducted on a laboratory scale. The employment of 100 % demolition debris and other waste materials has been proposed by reducing their particle size, but the final product exhibits preferential leaching of alkaline-earth cations despite the high densification of the sintered ceramics [24]. Moreover, in the ICEBERG project intermediate report, the manufacturing of new recycled ceramic tiles containing the majority of CDW in their composition at the laboratory scale was explored [25]. The manufacture of ceramics for construction applications (i.e., brick, tiles, and sanitary materials) is a high energy demand practice, where temperatures between 900 and 1300 ºC are required. In the context of the current energy crisis and emissions, the sector would benefit from its transition to alternative low-temperature sintering techniques, such as the cold sintering process (CSP), hot pressing or flash sintering, among others [26]. Specifically, CSP is a non-conventional sintering route where the use of an external pressure normally up to 600 MPa and notably lower temperatures up to 300 ºC, together with a liquid medium promotes the densification of the system. Besides advanced ceramics where this technique has been largely explored, CSP has been proven on natural materials such as in Mars regolith simulant [27] and diatomaceous earths [28] or in recycled materials such as the sintering of mussel shells for bioceramic products [29], with good results. Several studies deal with CDW for obtaining construction materials, using calcium carbonate [30–32] and pozzolanic wastes such as fly ash with a second drying step of 12 h [33]. This former study by Nishikawa et al. consists of a hybrid route combining geopolymerization and CSP, called cold reaction sintering (CRS). In such a study, the mechanical properties are enhanced by implementing warm pressing of the precursors to the conventional slurry geopolymerization route. However, to our knowledge, this non-conventional approach has not been explored for ceramic building materials to date, which have a less reactive and pozzolanic nature. One of the main limitations of external pressure sintering techniques, such as CSP, CRS, or hot pressing among others, is the possibility of upscaling the process since the defects and the limited particle size development during the sintering compromises the strength of the pieces. Nonetheless, there have been advances in the matter as proposed by Jabr et al. [34], where the possibility of enhancing mechanical properties by tailoring sintering parameters, such as heating rates and load transfer misalignments, is investigated. Moreover, the possibility of obtaining complex shapes by CSP has been explored, opening new possibilities for this technique [35]. This scenario encourages research and development of new practices aimed at producing environmentally friendly ceramic materials for construction applications. This work explores for the first time the production of dense recycled ceramic materials by CRS at 150 ºC using sodium silicate as the liquid medium. To this end, the potential use of building demolition waste to produce a material containing 95 wt% recycled debris has been investigated. The studied compositions consisted of common building materials such as ceramic, mortar, and plasterboard. Based on the available literature, Ca-bearing materials’ influence in the CRS parameters was explored by the testing of 0–50 wt% of mortar or plasterboard to a ceramic-based mixture. Morphological, structural, and mechanical properties were examined along with chemical stability, obtaining characteristics comparable to those of a conventional brick, demonstrating the feasibility of CRS to develop new ceramic materials. Finally, an experimental comparison with geopolymer conventional tests was conducted. 2. Materials and methods 2.1. Materials Common building materials were used to prepare the samples to simulate CDW. These building materials come from unused materials purchased from common suppliers of building materials in Spain and are extensively characterized in reference [36]. In this study, these materials are considered hypothetical recycled materials, therefore for the sake of simplicity, they will be referred to as recycled materials in the manuscript. The ceramic proportion (named CER) combines 81 wt% of red brick, 14 wt% of roof tile, and 5 wt% of red wall tile. This mixture is representative of the common ceramic content in a single-family house of the Spanish south regions. The non-ceramic recycled material consists of cement-based materials (named MOR) from a sidewalk block and a gypsum-based material (named GYP) from a plasterboard panel. In all cases, the prepared compositions contain at least 95 wt% of recycled materials and up to 5 wt% of virgin kaolinitic clay (named CLAY). Chemical analysis of each material was assessed by X-ray fluorescence (XRF) spectroscopy in terms of equivalent oxides, whose results are provided in Table 1. The ceramic materials show a composition typically observed for materials between the ternary diagrams SiO 2 - Al 2 O 3 -K 2 O. A representative Fe 2 O 3 content is also identified, which is usually found in red ceramics. The plasterboard material (GYP) is related to gypsum in line with the CaO and SO 3 content shown in XRF. The sidewalk block corresponds to a mortar material (MOR), which is composed of cement, sand, gravel, and water. Due to the age of the material, it has been fully carbonated into CaCO 3 as previously analyzed [36], where further characterization of these materials can be found. Exemplary photographs of the employed recycled materials are shown in Fig. 1. The recycled materials were comminuted separately according to the flow chart shown in Fig. 2. First, each material was cut into small pieces (<1 cm) using a circular table saw. In an industrial environment, this step would be done using a jaw crusher or similar. Then, they were pulverized separately using a WC ring mill (Siebtechnick TS250, 60 s of stage, without solvent). Finally, a porcelain jar mill was used (Heramika, alumina balls of 2 cm, 20 min of stage, with 37 wt% of deionized water). The slip was dried overnight at 80 ºC and sieved with a 63 µm mesh. The particle size was assessed by diffraction method with a He-Ne laser on an ultrasonically dispersed powder (Malvern Mastersizer S, λ =632.8 nm). In all the materials, the obtained particle average size was ~2 µm. S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 2 Different compositions were produced by weighting the conditioned materials and mixing them with a high-energy dry ball mill (SPEX SamplePrep Mixer/Mill 8000D, 10 min of stage). The final compositions were sieved under 100 µm mesh. The weight percentage of each material is shown in Table 2. The chemical analysis of some samples, i.e., CER, CER_30GYP, and CER_30MOR, is shown in Table 1 to serve as an example. The labeling of the mixtures describes the CER combination with the percentage of non-ceramic waste (MOR or GYP) (e.g., CER_30GYP for the mixture containing 70:30 of the ceramic mixture and plasterboard). 2.2. Samples preparation The mixture compositions were processed by CRS using a heated press (D-7290, Burkle) in an air atmosphere. CRS pellets were prepared by mixing 0.2 g of each powder with a 20–25 wt% sodium silicate by manual mixing in an agate mortar for 2 min. The sodium silicate corresponds to a commercial solution with a density of 1.135 g/cm 3 and chemical composition assessed by XRF of Na 2 O 16.5 %, SiO 2 64.1 %, and K 2 O 5.7 %. Loss on ignition after 120 min for 1000 ºC was 13.4 %. The moistened mixture was then placed into a cylindrical 8.3 mm inner diameter die. Subsequently, the sample was sintered by uniaxial pressure to 595 MPa, 150 ºC for 90 min. Optimization of the CRS parameters Table 1 Chemical composition of the employed materials in terms of equivalent oxides obtained by XRF spectroscopy. CER, CER_30GYP, and CER_30MOR analysis are included as examples of compositions explained below. Recycled materials Raw material Compositions Ceramic Non-ceramic Brick Roof tile Red body tile GYP MOR CLAY CER CER_30 GYP CER_30MOR SiO 2 61.1 64.3 64.5 7.7 42.4 58.5 61.7 46.2 56.1 Al 2 O 3 17.4 17.0 14.7 2.7 3.7 20.0 17.2 13.2 13.5 Fe 2 O 3 7.2 6.5 4.6 0.7 1.3 7.2 7.0 5.2 5.4 K 2 O 5.4 4.1 3.6 0.5 1.3 1.8 5.1 3.6 3.9 CaO 3.0 3.2 8.5 31.7 25.4 0.4 3.3 11.3 9.5 MgO 2.0 2.6 1.8 0.4 2.5 0.5 2.1 1.5 2.1 TiO 2 0.9 0.9 0.8 0.0 0.0 1.3 0.9 0.7 0.7 Na 2 O 0.3 0.3 0.6 0.0 0.3 0.0 0.3 0.2 0.3 SO 3 0.0 0.0 0.1 35.3 0.3 0.0 0.0 10.1 0.1 Fig. 1. Photographs of the building materials used: red brick, red body tile, and roof tile as ceramic materials (labeled as CER) plasterboard panel (GYP), and sidewalk block (MOR) as non-ceramic materials. The images are not shown to scale and are intended for illustrative purposes only. Fig. 2. Flow chart of the material size reduction and composition preparation along with photographs of a selected material as an example: red brick comminution. S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 3 was conducted from 100 to 750 MPa, RT to 200 ºC, and 5–180 min. The conditions were set based on the maximum density obtained in balance with the minimum energy required (shorter processing time and working temperature). Thereafter, the sintered pellets were left to naturally cold down inside the die until 80 ºC was reached under pressure and then removed. It should be mentioned that all the prepared samples were not visually soluble in water after the CRS process, and they maintained their integrity through normal manipulation and characterization activities. 2.3. Samples characterization The chemical composition of the materials was analyzed by XRF spectroscopy. For that, an IQ + semi-quantitative analysis curve was employed using a MagiX spectrometer (Philips). Li 2 B 4 O 7 was used as an additive. Loss of ignition of the materials was calculated at 1000 ºC for 1 h. The pellet densification was evaluated by Archimedes’ method and geometric density measurement. The theoretical density was obtained by He-Pycnometry of the powder using an Accupyc II 1340 Pycnometer (Micromeritics) to calculate the relative density. For geopolymer’s test samples, the density was assessed only geometrically. X-ray diffraction (XRD) was carried out to study the mineralogical phases using a D8 Advanced diffractometer (Bruker), equipped with Cu K α monochrome radiation (λ =1.5406 Å) in the range of 10–70◦2θ and at a speed rate of 1◦/min. The mechanical properties of the densified pellets were studied by Vickers hardness (Hv) indentation tests using a Universal Indentation and Scratch Tester Model APEX-1 (Bruker), making a statistic of at least 9 micro-indentations per sample. A force of 500 mN and a diamond indenter with a 100 nm radius tip were used. The Hv was then obtained based on the equations given by Oliver et al. in 1992 [37], dividing the maximum load (P max ) by the function area (A⋅h c ) addressed by the multiplication of the area that the indentator does (A) by the contact depth (h c ), as follows: Hv =P max /A⋅h c (1) The leaching test was performed by following the NEN 7341 availability test. For this assessment, CER_30GYP and its starting powder (pwd CER_30GYP) were reduced to a powder of about 100–200 µm and immersed in deionized water for 3 h at RT while magnetic stirring. The pH was maintained constant during extraction by the addition of nitric acid or sodium hydroxide when needed. The liquid/solid ratio was 50 L/ kg. The resulting extraction was then chemically analyzed using inductively coupled plasma optical emission spectroscopy (ICP-OES) (Agilent 720) for Na, Si, and Ca determination. The same procedure was performed for geopolymer experiments, using the CER_30MOR composition instead. For comparative purposes, the production of a geopolymer-like procedure was performed for CER_30MOR composition, which was mixed with sodium silicate at a liquid/solid mass ratio of 0.45. Then, the mixture was treated at different pressing and curing conditions as described below. The CRS is developed at the pressing-temperature stage where time and temperature are changed. Pressing time was selected as the standard pressing time for a laboratory press of 10 s, signaled as 0, and 90 min, signaled as 90. The pressing temperature was selected between room temperature, RT, and 150 ºC signaled as cold temperature, CT. In all cases, the pressure value was 595 MPa. In the second stage, to replicate the curing conditions usually used for geopolymerization, the atmosphere and the curing time at ambient pressure and 40 ºC temperature were studied. The curing atmosphere saturated with water was achieved by the introduction of the samples in a sealed container, signaled as wet atmosphere, W, or in dry condition, D. For example, 90CT7W corresponds to a sample pressed for 90 min at 150 ºC and cured in saturated water atmosphere for 7 days. Fourier-transform infrared spectroscopy (FTIR) was performed for geopolymerization tests on the sample surface using a Spectrum 100 Optica FTIR spectrometer (Perkin Elmer) in attenuated total reflectance (ATR) mode for the range of 4000–300 cm −1 . The approximate penetration distance (pd) is 0.5 – 5.5 µm for the studied range, calculated from Eq. (2): pd =λ/(2 π n₁√(sin2θ− (n₂/n₁)2)) (2) where λ is the wavelength of the infrared light (2500–33333 nm), n 1 is the refractive index of the ATR crystal (in this case diamond =2.4), n 2 is the refractive index of the sample (~ 1.4 for geopolymers [38]) and θ is the angle of incidence (45º in this case). The microstructural evolution and grain morphology of the starting powder were studied by field emission scanning electron microscopy with energy-dispersive X-ray spectroscopy (FESEM-EDS) with an S-4700 Hitachi instrument at 20 kV on the fresh fracture surface of each sample. The particle distribution has been evaluated by particle counting (sample =300) using ImageJ image processing software. 3. Results and discussion 3.1. Material characteristics: density, hardness, and phase evolution The effect of the non-ceramic material on the relative density of the samples is studied. This parameter is chosen for two key aims. On the one hand, the inherent nature of CDW makes them highly heterogeneous, making it unrealistic to consider them as a single source of one type of material such as ceramic. On the other hand, these materials are a source of divalent calcium cations that come from their composition, such as in the case of plasterboard, CaSO 4 ⋅nH 2 O, and mortar, CaCO 3 . These cations may be available in the medium to react with the solvent used during the CRS. In fact, some studies that explore the production of alkali-activated materials based on CDW have regarded previously this strategy by the addition of CaSO 4 ⋅2 H 2 O, quick lime (CaO), and hydrated lime [13], slag [14] or waste concrete [15] as calcium sources. Therefore, considering these premises, samples containing from 0 to 50 wt% of non-ceramic Ca-bearing materials are produced under the same sintering conditions (i.e., 150 ºC, 90 min, and 595 MPa). The results of the obtained relative density are shown in Fig. 3a. All samples containing non-ceramic materials exhibit relative density values above 80 %, indicating their densification. Higher values than the comparable sample using only ceramic materials (CER), which has a 76 % relative density, are obtained. Moreover, the relative density increases with the content of secondary material until it reaches 30 wt% with values up to 90 %, from which the density begins to decline for compositions Table 2 Weight percentage of the studied compositions in terms of recycled and raw materials. (wt%) Composition name Recycled material Raw material Ceramic (CER) Non-ceramic material (MOR/GYP) CLAY Brick Roof tile Red body tile CER 81 14 5 0 0 CER_10MOR 69.2 12.0 4.3 9.5 MOR 5 CER_10GYP 69.2 12.0 4.3 9.5 GYP 5 CER_20MOR 61.6 10.6 3.8 19.0 MOR 5 CER_20MOR 61.6 10.6 3.8 19.0 GYP 5 CER_30MOR 53.9 9.3 3.3 28.5 MOR 5 CER_30GYP 53.9 9.3 3.3 28.5 GYP 5 CER_15MOR15GYP 53.9 9.3 3.3 14.2 MOR + 14.2 GYP 5 CER_40MOR 46.1 8.0 2.8 38.1 MOR 5 CER_40GYP 46.1 8.0 2.8 38.1 GYP 5 CER_50MOR 38.5 6.6 2.4 47.5 MOR 5 CER_50GYP 38.5 6.6 2.4 47.5 GYP 5 S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 4 prepared with MOR while there is a decrease in slope for the samples prepared with GYP. This fact could be indicating that the incorporation of this secondary material, up to a specified percentage, plays a key role in the CRS of the material. As it is intended with its incorporation into the composition, these non-ceramic materials would be acting as a possible source of divalent calcium that reacts with the liquid medium (sodium silicate aqueous solution), probably producing hydration products like those that can be observed in cementitious hydration [39]. Likewise, the possibility of finding both secondary materials in the same proportion is also studied, that is, 15 wt% each (CER_15GYP15MOR). In this sample, a decrease in relative density at 86 % is observed compared to the samples having 30 wt% of only a non-ceramic material (i.e., CER_30GYP and CER_30MOR). However, density values are acceptable considering the physical integrity of sintered parts and that these 3 types of materials, so different from each other, are used together. In Fig. 3b, X-ray diffractograms of the CER_30GYP and CER_30MOR composition are shown, labeling with “pwd” the starting powder before the CRS process. The principal reflections are labeled on the figure. First, the principal difference due to the composition of each sample before the sintering is the presence of gypsum and bassanite in the case of CER_30GYP composition and calcite for the CER_30MOR composition. The absence of portlandite reflections in mortar-bearing composition shows the age of the material, where carbonation is completely developed. The rest of the main phases found correspond to those that come from the ceramic mixture: quartz (which is also found in the mortar material), hematite, microcline, and plagioclase-type feldspars such as anorthite. Once the CRS is completed to the CER_30MOR composition, there are no significant changes in the reflections because all phases are stable at the sintering temperatures. There is also no development of new phases. In the case of the CER_30GYP composition, it can be seen how some phases reacted after the CRS. These phases correspond to hydrated calcium sulfates such as gypsum, which transforms into anhydrite due to heating the samples to 150 ºC. Besides, some of the reflections are developed in terms of relative intensity when compared to the quartz presence. This could be indicative of the development of phases related to different silicate compounds that are being favored in the process. However, the use of this technique is limited in the detection of products that may be formed during the sintering process with a very low concentration and amorphous materials. Since the final purpose of this investigation is to propose a material to be used as a construction material, structural (like a brick) or functional (like a tile), the study of mechanical properties is of vital importance in the characterization. Fig. 3c displays mechanical properties, through Vickers hardness (Hv), of CRS samples with 30 wt% of content of secondary material (non-ceramic material), corresponding with the samples labeled by a hollow symbol in Fig. 3a and showed in the photograph provided in Fig. 3d. This type of test is not very common in ceramic construction materials, being the flexural (for tiles) or compression test (for bricks) the most popular for them. However, due to the limitation in dimensions of the sintered samples, the mechanical properties will be addressed by comparing the Hv between them, which is a common approach in cold sintered samples [40,41] and provides a reliable method of comparing the resistance between samples. For that, a conventional commercial brick was tested by the same conditions and indenter to have a value with which to compare the produced samples. This value, resulting in 0.98 GPa is denoted in the Fig. 3c by a grey Fig. 3. a) Influence of non-ceramic material (mortar and plasterboard) presence in the relative density of the CRS samples at the same sintering conditions of temperature (150 ºC), pressure (595 MPa), time (90 min), and solvent (23 wt% of sodium silicate aqueous solution). b) XRD diffractograms of CER_30GYP and CRS sintering process. Legend: G: gypsum, B: bassanite, Q: quartz, Ay: anhydrite, Mc: microcline, At: anorthite H: hematite, and C: calcite. c) Hv values of selected samples prepared with a 30 wt% of content of non-ceramic material and denoted with hollowed symbols in Fig. 1a. Hv value measured for a conventional brick is shown with a dashed grey line as reference. d) Photographs of the sintered samples studied in Fig. 3c. e) Leaching test results by means of equivalent oxides obtained by ICP-OES. S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 5 dash line as the reference threshold. For the case of CRS samples, as we expect according to the density results, all the compositions that have non-ceramic material show considerably high hardness values. These values are comparable to those obtained for the reference commercial brick. Interestingly, in the case of the CER_30MOR composition, a hardness value double that of the other compositions is obtained. This increase of mechanical properties as a 30 wt% of the content of MOR is correlated with the Ca-based phases that react during the CRS, which will be evaluated below. In further studies, these hardness values should serve as an indicator for the process evaluation and be complemented with a deeper mechanical characterization, specifically studying their compressive and/or flexural strength for their use as building materials. Finally, in Fig. 3e the leaching test results in terms of leached equivalent oxides are presented. In the case of SiO 2 and Na 2 O, the main components of the dosage liquid transient phase, the leaching for CER_30GYP is high compared with that of the starting powder. However, considering the Na + dosage, it can be calculated that approximately 75 wt% of sodium silicate has been consumed during the densification mechanism. Moreover, considering the CaO leached amount, a clear reduction after the CRS happens. This result is in concordance with the consumption of the Ca-bearing material during the CRS, in this case, the plasterboard content. Likewise, the former leaching of the unreacted sodium silicate could be explained by the consumption of the Ca 2+ , being possible to reduce by the controlled dosage of these species. These conclusions should be completed with further studies, however this leaching test, although preliminary, would indicate the reaction and densification of the material beyond a simple mechanical rearrangement of the particles. 3.2. Microstructural characterization FESEM micrographs of CER, CER_30GYP, and CER_30MOR compositions for the starting powder and after CRS on the fresh fracture surface are shown in Fig. 4. Firstly, it can be observed in all cases that the starting powder has an irregular morphology and a bimodal particle size distribution with particles of approximately 4 µm and 300 nm on average. The logarithmic distributions of the counted diameters in the ranges of 0–1µm and 1–10 µm are found for the starting powders of samples CER, CER_30GYP, and CER_30MOR in Fig. 4b, e, and h, respectively. The mean and standard error of the mean in brackets are depicted in the figures. This bimodal particle size distribution increases the compaction capability of the system, improving the densification of material that can be initiated in the compact previously to the CRS. Specifically, compositions containing non-ceramic material present a larger number of submicronic particles with a smaller size on average (∅=30 nm), covering the larger ones than for the CER composition (see Fig. 4a, d, and g). These submicronic particles are assigned to the non-ceramic material. The lower hardness of the non-ceramic material during the grinding stages enables a more pronounced reduction of said material because the particles from the ceramic materials intensify the microgrinding. This reduced size confers a greater specific surface area to the mixture that will improve their reactivity, favoring the CRS and improving the density of pieces (see Fig. 3). Comparing the starting powders with their appearance after CRS, a clear compaction of the particles is identified in the case of the samples containing non-ceramic material (see Fig. 4f and i). In both samples a continuous nanostructured region is formed between microparticles, indicated on micrographs f and i with the number 1, which could be related to the CRS process between Ca-rich nanoparticles and the sodium silicate. In this case, the nanostructured region presents a more rounded morphology compared to the starting nanoparticles. In addition, larger microparticles are observed (designated in micrograph e with the number 2) that could be assigned mainly to less reactive particles related to quartz. The microparticles appear embedded into the nanostructured matrix formed by the submicronic particles after the cold reaction sintering. For the case of the sample processed by CRS from CER composition, no significant changes are denoted with the starting powder showing the relevant role that Ca-rich phases have in the formation reaction of the nanostructure that consolidates the ceramic material. Finally, the sample CER_30MOR was polished and chemically etched (HF, 5 vol%) to reveal the microstructure. Fig. 5a shows the Fig. 4. FESEM micrographs of CER, CER_30GYP, and CER_30MOR compositions for the starting powder (a, d, and g, respectively) along with their particle size distribution (b, e, and h, respectively) and on fresh fracture surface of samples (c, f, and i, respectively) prepared after CRS at 150 ºC, for 90 min and under 595 MPa with a sodium silicate aqueous solution as solvent. S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 6 characteristic microstructure of CRS sintered samples that consist of a nanostructured matrix, where larger microparticles are embedded. When the nanostructured matrix is observed at higher magnifications, corresponding to the white box in Fig. 5a (Fig. 5b), the elimination of the glassy phase with the chemical etching provokes the removal of submicron grains, typical phenomenon of clay-based ceramics. The fingerprints of the submicron grains pull out present grain boundaries with low curvature, indicating that the equilibrium between them is reached. In addition, the formation of sintering necks between coalescing particles (pointed with an arrow), evidences the occurrence of mass transport in crystalline phases during the CRS. This process could happen in the particles with the highest chemical potential, obtained from the microgrinding. However, the reaction of the nanostructured matrix and the micrometer-sized particles has also occurred, as shown by the fresh fracture of the ceramic materials obtained by CRS (see Fig. 4), which occurs preferably in the nanostructured matrix and not at its interface with the micrometer particles. EDX analysis of the a) polished surface and b) polished and chemically etched (HF 5 % 3 s) sample CER_30MOR after CRS for 90 min at 150 ºC and under 595 MPa are shown in Fig. 6. Areas or points where the analysis has been developed are highlighted with numbers in the micrographs, related with the numbers of the equivalent oxides shown in the left tables. On the one hand, it is observed that the chemical attack of the sample has been able to remove preferentially the nanostructure matrix and microparticles are thus revealed. Performing the EDX analysis on the unetched sample (points 1–4 Fig. 6a) it is observed a similar composition for the different samples. Mainly, this nanostructure matrix contains a high concentration of SiO 2 (>50 %). This cation is part of species such as glass matrix in porcelain ceramics, feldspars, or hydration products such as C-H-S gel. On the other hand, if the EDX analysis is carried out once this layer has been eliminated by chemical attack (Fig. 6b), the different particles of different materials that were embedded in this matrix are observed. Mainly, quartz particles can be observed (EDX points 2, and 3) due to their high hardness and chemical stability, which means that they remain less altered during the comminution process or after CRS. The rest of the points analyzed (EDX points 1, 4, and 5) show equivalent oxides comparable to the phase studied in the previous figure, confirming that this matrix covers all the particles. 3.3. Cold reaction sintering vs. geopolymerization process Based on the aluminosilicate nature of CDW materials and the silicate alkaline solution here used, it is sensible to wonder about the effect of CRS on the system and the difference it presents with geopolymerization processes. Therefore, it is worth analyzing how each of the stages of this process affects the evolution of the material. For that, Fig. 5. (a) FESEM micrograph of CER_30MOR after the CRS (150 ºC for 90 min under 595 MPa) polished and chemically etched with HF (5 %vol). (b) Magnified image of the area marked in (a). Fig. 6. EDX analysis for the sample CER_30MOR obtained by CRS (90 min, 150 ºC and 595 MPa) at the a) polished and b) chemically etched surfaces. S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 7 an experiment in two steps is carried out: stage of pressing or CRS (stage 1) plus a subsequent stage of curing which mimics that carried out in the field of geopolymers (stage 2), as explained above. In Fig. 7a, the FTIR spectra are presented in the range of 1500–600 cm −1 for the explored conditions, ordered from highest to lowest energy contributed to the system, this being in the form of pressing temperature and time for the stage 1 together with curing time and atmosphere for the stage 2. The most significant difference in these absorption spectra is the blueshift of the absorption band centered at 1000 cm −1 , which is enlarged in Fig. 7b. This broad absorption band is directly related to a geopolymerization fingerprint, corresponding to the asymmetric vibration of Si-O-T (being T=Si or Al) [42]. The broad aspect of the absorption band, which in these samples ranges between 1090 and 972 cm −1 , highlights the coexistence of several types of gels such as C-S-H (970 cm −1 ) and C-(A)-S-H (1090, 1085, and 1005 cm −1 ) [43]. The higher the incorporation of Al cations in the Si-O-T bond the higher the strength of the bond that it is evidenced by the blueshift of the absorption band. Meanwhile, the formation of C-S-H is promoted by the reaction of the alkaline silicate with the Ca-based materials (in this case gypsum waste), and the formation of C-(A)-S-H requires the reaction of the alkaline silicate with the Ca-based material and the ceramic composition that it is the aluminum cation source. The aluminum cations probably come from the glassy phase of ceramic waste. This fact implies that the reaction with ceramic waste is critical to obtain a gel phase with strengthened bonds. Firstly, regarding the pressing conditions, it is observed that the pressing time (10 s vs. 90 min) at RT does not have a significative influence on the shift of absorption spectra. 0RT1D and 90RT1D present comparable absorption bands in both samples. The curing in the water atmosphere has a more relevant change as evidenced by 90RT7W case, where the 7 days of curing in the water atmosphere produce a blueshift of this absorption band from 973 to 1029 cm −1 . However, by increasing the temperature at which the pressing is carried out from RT to 150 ºC, a significant blueshift up to 1041 cm −1 is observed. It should be noted that the absorption band displacement is comparable between the samples 90CT7W and 90CT, which indicates that once the CRS is finished, the systems in no further changing and chemically comparable materials are obtained, avoiding long curing periods. In addition to the relevant blueshift in 90CT and 90CT7W samples, the shape variation in the absorption peak is a possible sign of the precipitation of new phases that are not typical components of C-S-H or C-(A)-S-H [44] when compared with the samples in which the pressing is done at RT. Therefore, the combination of pressing and cold temperature, the CRS process, accelerates the reaction of the alkaline silicate with the materials that provide calcium cations and ceramic residues. In this way, the CRS contributes to the densification of the new ceramic piece by combining the partial Fig. 7. FTIR spectra of CER_30GYP composition under several experimental conditions according to Table 2 in the a) range of 1550 – 670 cm −1 and b) enlarged 1150 – 860 cm −1 . c) Leaching test results for pwd CER_30GYP, 0RT7W, 90RT7W and 90CT7W samples. S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 8 reaction mechanisms of the CDWs with the alkaline silicate and generating new chemical species with a greater bond strength than the one’s processes based on geopolymerization reactions. Once corroborated the little influence of the curing conditions in comparison with the pressure conditions, the leaching of 3 selected cured in a water atmosphere for 7 days samples: 0RT7W, 90RT7W, and 90CT7W. Comparatively, leaching of pwd CER_30GYP is performed. The leaching of the cations Na, Ca, and Si (mg/L) is shown in Fig. 7c. First, regarding the Na leaching it can be seen the preferable leaching when RT is employed. The one that leaches the least is the pwd CER_30GYP, due to the absence of sodium silicate in that sample. In the case of Ca release, its leaching is reduced in all the samples and the cold reaction sintering one presents the greatest reduction. The leaching of Si and Na represents the most significant trend, in all the samples both cations leached more than the starting powder due to the addition of sodium silicate. In the samples pressed at RT, both cations leached to a higher extent than the sample with CRS. This fact is preceded by FTIR analysis in which the reaction of alkaline silicate with CDW produces a strengthened bond of the gel phase and thereby, these reacted phases are less leached. It is worth mentioning that all produced samples showed a density value of 2.1 g/cm 3 , except for 90CT, which showed 2.3 g/cm 3 . In addition of the densification and mechanical response, the ceramic obtained by CRS has a higher chemical stability, which suggests a larger reaction between the different components. Given the evidence shown such as this blueshift observed by FTIR, the greater leaching stability, and the characteristic morphology observed by FESEM, it can be concluded that the CRS plays a categorical role in the final characteristics of the material, notably differentiating it from what is known as a conventionally alkaline-activated materials. Finally, the obtention of a dense ceramic piece using a majority composition of CDW has been demonstrated based on the present study. However, the addition of a liquid medium could be a reason for concern when thinking about scaling the proposed methodology, although it should presumably be low attending the average ceramic tile price. However, it is important to note that other advantages and drawbacks must be considered. On the one hand, the substantial amount of recycled raw material replacing virgin material, along with savings in transportation and lower sintering temperatures, would more than justify the cost of adding the alkaline medium in the process. On the other hand, the CDW conditioning and the initial investment for the industrial equipment adaptation (high pressures and alkaline-resistant equipment) also play a relevant role in the economic analysis of the proposed methodology. A deep study of these and further details must be conducted in the following works to assess the economic profitability of the process. 4. Conclusions This study explores the production of a ceramic-type construction material using 95 wt% recycled content from construction and demolition waste through CRS, at a low sintering temperature of 150 ºC using sodium silicate. The material consists of a mix of non-ceramic and ceramic components commonly found in conventional buildings. The main conclusions are: •The process produces dense samples with relative densities around 90 % employing 28.5 wt% of non-ceramic material for CER_30MOR and CER_30GYP compositions, identifying the Ca 2+ bearing materials have a critical influence on the obtained relative density of the samples. •Studying the microstructural characteristics of the samples, the small particle size showed to enhance the effective consolidation during CRS, resulting in a nanostructured matrix that accommodates larger particles. •In terms of mechanical properties, the Vickers hardness values of samples containing 30 % non-ceramic material are comparable to those of conventional bricks, with the sample containing mortar showing more than twice the hardness of commercial traditional bricks. •Leaching tests indicate significant chemical stability, with the complete consumption of Ca 2 ⁺ into non-leachable species suggesting its reaction during the process. •Comparative experiments with standard geopolymerization procedures show that CRS conditions—such as increased time, temperature, or wet atmosphere—shift the FTIR signature from approximately 973 cm⁻ 1 to 1041 cm⁻ 1 . This shift, along with the observed leachability behavior, underscores the critical role of CRS in defining the final properties of materials, distinguishing them from conventional alkaline-activated materials. Overall, this research highlights the potential of using CRS processes to produce recycled materials for construction, paving the way for further exploration in this field where many questions remain unanswered. CRediT authorship contribution statement Aida Serrano: Writing – review & editing, Validation, Supervision, Investigation, Funding acquisition, Conceptualization, Formal analysis. Sonia Marín-Cort´ es: Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Jos´ e F. Fern´ andez: Writing – review & editing, Validation, Supervision, Project administration, Investigation, Funding acquisition, Formal analysis, Conceptualization. Esther Enríquez: Writing – review & editing, Validation, Supervision, Investigation, Conceptualization, Formal analysis. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors gratefully acknowledge the financial support of the European Union’s Horizon 2020 research and innovation programs ICEBERG (No. 869336) for their financial support. This work was supported by the Spanish Ministry for Science and Innovation (MCIN) under the project PID2020–114192RB-C41 and PID2021–124585NB-C33 funded by MCIN/AEI/10.13039/501100011033 and by ‘‘ERDF A way of making Europe’’ and by Project TED2021–130957B-C51 funded by MCIN/AEI/10.13039/501100011033 and by the ‘‘European Union NextGenerationEU/PRTR’’. A.S. acknowledges financial support from grant RYC2021–031236-I funded by MCIN/AEI/10.13039/ 501100011033 and by the “European Union NextGenerationEU/PRTR”. Data availability Data will be made available on request. References [1] D.D. Furszyfer Del Rio, B.K. Sovacool, A.M. Foley, S. Griffiths, M. Bazilian, J. Kim, D. Rooney, Decarbonizing the ceramics industry: A systematic and critical review of policy options, developments and sociotechnical systems, Renew. Sustain. Energy Rev. 157 (2022) 112081, https://doi.org/10.1016/J.RSER.2022.112081. [2] G. Boschi, G. Bonvicini, G. Masi, M.C. Bignozzi, Recycling insight into the ceramic tile manufacturing industry, Open Ceram. 16 (2023) 100471, https://doi.org/ 10.1016/j.oceram.2023.100471. [3] O. Guillon, Ceramic materials for energy conversion and storage: A perspective, Int. J. Ceram. Eng. Sci. 3 (2021) 100–104, https://doi.org/10.1002/CES2.10086. [4] S. Marín-Cort´ es, M. Fern´ andez-´ Alvarez, A. Moure, J.F. Fern´ andez, E. Enríquez, Chemometric-driven quantification of construction and demolition waste using S. Marín-Cort´ es et al. Construction and Building Materials 451 (2024) 138917 9