The Pozzolanic Activity of the Sediment Treated by the Flash Calcination Method

9 The dredged sediment has been positioned for years as alternative materials in the construction field. However, it 10 is often necessary to apply a treatment to improve their reactivity and performance. This article aims to study the 11 pozzolanic reactivity of fluvial sediment treated by flash calcination method at different temperatures 650 °C, 750 12 °C, and 800 °C. The physico-chemical, mineralogical, and environmental characteristics were studied for treated 13 (flash-calcined sediment) and raw sediment. The pozzolanic reactivity of the flash-calcined sediments was estimated 14 with Frattini ’s test , isothermal calorimetry test, lime consumption analysis and compressive strength then compared 15 to metakaolin which is considered as the reference. The results of the compressive strength of mortars show the 16 detrimental effect of raw sediment on the development of resistance. Contrary to the raw sediment, the treatment 17 of the sediments by flash calcination activates the pozzolanic reactivity of the clay phases and considerably improves 18 the contribution of the sediments to the development of resistance and the porous structure. Moreover, the 19 sediment calcined at 750 °C gives better properties than those obtained at 650 °C and 800 °C. The result 20 demonstrates the feasibility of using calcined sediments as a pozzolanic mineral addition in a cementitious material. The originality of this research is the use of an innovative calcination technique to treat the sediment. The raw sediment is a polluted sediment and has harmful effects on the hydration of the cement. However, Flash calcination 27 significantly improved the hydration behavior of sediment compared to raw sediment. The different methods were 28 used to demonstrate the beneficial effects provided by sediments calcined at different temperatures and to choose 29 an optimal calcination temperature.


Introduction 31
Nowadays, concrete is the most used material in the construction field and public works [1]. However, the 32 production of this material has a great impact on the environment due to the requirement of significant extraction 33 of raw materials. With a global cement production of 4.6 billion tonnes in 2016 and an average annual growth of 34 6.9% [2], the global cement production process releases between 652 kg and 894 kg of CO2 for each tonne of cement 35 produced [3]. Above 7% of the worldwide CO2 emissions are related to the cement manufacturing [2]. To achieve 36 the goal of reducing CO2 emissions by 2050 and to attend carbon neutrality, the use of mineral additions to replace 37 part of the cement is considered one of the most promising solutions [4]. To this day, metakaolin, slag, fly ash, silica 38 fume, and limestone fillers are the most widely used mineral additions in the construction industry. Thanks to their 39 pozzolanic reactivity, the durability properties and the mechanical performances of concrete have improved [5]. 40 However, one of the main disadvantages of large scale application of metakaolin, is the limititation of the high purity 41 kaolinite deposits [6]. In addition, environmental restrictions are increasingly strict (2006/21/EC) for an ecological 42 transition and the protection of biodiversity. 43 Each year, the dredging operations for maintenance at ports and inland waterways generate around 300 million 44 tonnes of sediment in Europe [6], with 56 million m 3 are in France. Several studies have demonstrated the potential 45 for sediment recovery in various fields: road technology [7,8] as a mineral addition in mortar and concrete [4,9], 46 [10,11] and as a raw material in the manufacture of cement [12,13]. Nevertheless, most studies of valorization of 47 sediments as an addition are based on the traditional calcination method to activate their pozzolanic reactivity. This 48 method requires significant energy consumption and arises a significant ecological problem [14]. 49 Snellings et al [15] studied the effect of the flash calcination of dredged sediment, from the port of Antwerp 50 (Belgium), at different temperatures (820 °C, 865 °C and 905 °C ) on their physico-chemical properties and on their 51 pozzolanic reactivity. The treatment reduced the total organic carbon content by 85% or more. The clay minerals in 52 the sediment were completely dehydroxylated. The amorphous phase formed during calcination constitutes the 53 major part of the calcined material. In addition, the result of the calorimetric analysis showed that the calcined 54 sediments had pozzolanic reactivity more than that of siliceous fly ash, however less than that of metakaolin. 55 Snellings et al [6], showed also in another study that increasing the amount of cement replacement from 20 wt% to 56 40 wt.% by calcined sediment slowed down the strength development. Whereas, at 91 days of hydration, the mortar 57 with 20 wt.% calcined sediment substitution gives similar strength as the reference mortar. 58 Amar et al [16] compared the pozzolanic reactivity of sediment treated by flash calcination at 820 °C and by 59 traditional calcination method. The results showed that the flash calcination has strongly modified the physico-60 chemical and mineralogical properties and improved the pozzolanic reactivity. The study showed also that the 61 activity index (XA) of mortars containing 25% wt of flash sediments as cement subsitution, is higher than the one 62 containing traditional calcined sediment at 28 days. 63 This study focuses on optimizing the flash calcination process of sediment on an semi-industrial scale. First, the 64 sediment was calcined at three different temperatures (650 °C, 750 °C and 800 °C). Then, the pozzolanic reactivity 65 of the calcined sediments and the raw sediment was examined by several methods in order to assess the 66 performance of the method and the effect of the sediments on the hydration of the cementitious material. Finally, 67 these results were used to choose an optimal calcination temperature. 68

Flash calcination method 69
The Flash calcination method is a technique of heat treatment of finely ground materials under high temperature 70 in the short term. It has been applied to the activation of clays [17,18] In addition, under the thermal effect, the other clay minerals will be dehydroxylated according to the following 76 equations (Table 1):  77 The above reactions are influenced by several parameters such as the kaolinite structure, the rate of heating, 78 and the particle size [22]. 79

Pozzolanic reaction in the cementitious material 80
A constituent having pozzolanic properties, contains silicate and aluminate phases allowing hydrate production 81 of ternary system CaO-SiO2-Al2O3-H2O in the presence of lime and water. The pozzolanic addition used as a substitute 82 for the clinker, will react with portlandite (CH) to form the hydrated calcium silicate "pozzolanic" (pozzolanic C-S-H 83 (II)). This hydrate generally has a C/S lower ratio than that of C-S-H formed from the hydration of cement without 84 pouzzolanic addition. Unlike C-S-H (II), the C/S ratio of C-S-H (I) is independent of temperature and the Ca 2+ 85 concentration of the solution in pores. The pozzolanic reaction can be described by the following equations [23]: 86  Dissolution of SiO2: 87 88  Pozzolanic C-S-H precipitation: 89 xCa 2+ + H2SiO4 2-+ 2(x-1) OH -+ y H2O  CxSHy (11) In general, the pozzolanic C-S-H stoichiometry is C1.1SH3.9. In addition, the presence of a large amount of 90 aluminate can allow the formation of other phases such as C2ASH8 (stratlingite); C4AH13 or hydrogrenat (C3AH6). 91

2.1
Presentation of the kiln and the Flash calcination process 93 The flash calcination tests are carried out in a flash furnace available in research center of IMT Nord-Europe 94 (France). The product output can reach 20 kg of material per hour. A propane gas burner providing a power between 95 10 -50 kW allows materials to be calcined up to a temperature of 850 °C. Fig. 1 illustrates the schematic of the flash 96 calcination furnace used in this study. 97 The principle and operation of the flash oven was early described by Teklay et al. [17]. However, several 98 parameters are adjustable such as burner power, air-flow, material flow and the calcination temperature. These 99 parameters should be adapted regarding the material properties. The chemical composition of the materials was determined using X-ray fluorescence analysis (XRF) according 130 to  with an S4 POINEER equipped with a 4-kW generator and a rhodium anode. 131  The mineralogical composition of the materials is identified by X-ray diffraction (XDR) analysis using a Bruker 132 apparatus equipped with a D2 diffractometer with a Cu anode (λ = 1.5406 Å). To specify the nature of the 133 argillaceous phases of the sediment, a specific analysis was carried out on the fraction less than 2 µm of the 134 sediments [8,11,30]. For the study of clay species, the angular range explored is between 4° and 35° 2θ with 135 a step size reduced to 0.008 ° 2θ and a step time of 0.84 s/step. 136  Thermogravimetric analysis (TGA) is performed using the Netzsch STA 409 device. About 100 mg of the finely 137 ground material was placed in an alumina crucible and heated from 40 °C to 1000 °C with a heating rate of 5° 138 C/min. 139 Dehydroxylation of kaolinite occurs at the temperature range of 400 °C to 600 °C during calcination [31]. The 140 amount of kaolinite could be determined from the mass loss of water (MH2O) on the TGA curve according to the 141 following equation (Eq (12) For Mkaolinite and nH2O values, the stoichiometry of kaolinite in Eq(2) was retained for the dehydroxylation. 150 The degree of dehydroxylation of kaolinite DTG was made by comparing the amount of kaolinite in the calcined 151 materials and on the raw material according to the following Eq(13): 152 were added. The chemical analysis is performed by inductively coupled plasma optical emission spectrometer 160 (IPC -OES 5100 Agilent Technologies). Anionic elements were analyzed by chromatography and without 161 acidification. The leaching limit values for inert waste (IW) and non-hazardous waste (NHW) specified in 162 Directive 1999/31/EC were used to verify material compliance. 163

Methods for evaluating the pozzolanic reactivity of materials 164
The evaluation of pozzolanic reactivity of calcination products can be carried out using various chemical and 165 mechanical methods. The principle consists in determining the content of Ca(OH)2 consumed by reactive 166 components in pozzolanic materials [33]. 167

a. Frattini's test (NF EN 196-5 [34]) 168
The Frattini's test was used in previous studies [11,30,35] to assess the pozzolanic reactivity of materials. The 169 essay consists to react 16 grs of CEM I cement and 4 grs of material in 100 ml of distilled water. Then, the mixture 170 was kept tightly at 40 °C for at least 8 days or 15 days. The samples were filtered under vacuum and the filtrates 171 were subsequently analyzed. First, the OHions content was determined using the HCl 0.1N hydrochloric acid . After 172 adjusting the pH to 12.5, the Ca 2+ ion content was determined by the use of an EDTA complexometric test. The Ca 2+ 173 content (expressed as CaO equivalent) function of OHcontent are shown relative to the limit of solubility of Ca(OH)2. 174 A position under the curve of Ca(OH)2 solubility suggests a consumption of CaO by pozzolanic reaction [30]. 175

b. Thermogravimetric analysis of mixture with Ca(OH)2 176
In this method, the consumption of lime by the pozzolanic reaction in the lime paste is evaluated by 177 thermogravimetric analysis (TGA) for the differents materials (flash calcinated sediments as well as metakaolin and 178 the raw sediment ) . The Ca(OH)2/mineral addition mass ratio is 4 and the amount of water was adjusted to have a 179 good consistency of the pastes. Table 3 shows the composition of the different lime pastes studied. 180 The lime pastes were stored hermetically at 20 °C until the test was measured at 7, 14, and 28 days. Before 181 testing, the hydration of the samples should be stopped. Therefore, the pastes were immersed in acetone solution 182 for 4 days, then filtered under vacuum before the thermogravimetrie analysis. 183

Determination of the amount of chemically bound water and the Ca(OH)2 content by the TGA curve 184
The quantity of chemically bound water (wt %) could be calculated from the TGA curve according to the following 185 Eq (14): 186 The amount of Ca(OH)2 reacted was estimated using the following Eq (16) and Eq (17):  200  201 Ca(OH)2 reacted = Ca(OH)2 total -Ca(OH)2 non reacted (16) 202

Determination of the quantity of hydrated phases (C-S-H, CASH) from the DTG derived curve 203
The hydrated phases as C-S-H, calcium aluminate hydrate (CAH) or calcium silico-aluminate hydrate (CASH) can 204 be calculated from the DTG derived curve [36,37]. However, it is difficult to distinguish two phases CAH and CASH. 205 The spectra obtained from the curves derived from dTG(%)/dt = f(T °C) were deconvolved in order to determine the 206 area of each hydrate decomposition according to the Lorentzian's area deconvolution method with the amplitude 207 a0, center a1 and width a2 according to the following Eq (18)  The pozzolanic reaction is an exothermic reaction. Isothermal calorimetry analysis is used to follow the evolution 216 of heat from exothermic hydration reactions in different cement paste mixes. Cement paste mixtures with a 10% wt 217 substitution rate of the addition, are made with a water/binder ratio equal to 0. resistance. An increase in the time difference could be due to the effect of the pozzolanic reaction. 250 e. Measurement of dynamic modulus of elasticity Edyn : 251 The dynamic modulus of elasticity of the mortars was determined by the equipment GrindoSonic Mk5 252 "Industrial". It is a non-destructive test that assesses microcracking, homogeneity, and compactness [40]. The 253 dynamic modulus of elasticity is measured according to the following Eq (20): 254

Leach analysis of mortars 265
Leach analysis was performed on particles (0-4 mm) from the mortars after 90 days to measure the content of 266 metallic trace elements and anionic elements. The analysis process is identical to that used in the material 267 characterization part. 268

3.1
The physical characteristics of materials 270 Table 5 shows the physical characreristics of the materials used in the study. The RS sediment has a high content 271 of organic matter (16.1%). Flash calcination significantly removes organic matter from products, especially in SF 750 272 and SF 800 where a reduction of 88% and 93% is respectively observed. 273 The increase in density was also observed for the calcination products. This is due to the elimination of organic 274 matter (density 1 g/cm 3 ) and the decomposition of limestone (density 2.6-2.8 g/m 3 ) to form calcium oxide (density 275 approximately 3.3 g/cm 3 ). However, a decrease in the density of SF 800 upon increasing the calcination temperature 276 could be due to the recrystallization of the mineral phases [44]. The flash calcination also induces a slight increase 277 in particle size without a reduction in specific surfaces. This is due to an agglomeration of clay particles without 278 sintering during calcination [31]. The increase in particle size in the calcined sediment was also observed in previous 279 studies [11,15,45]. Concerning the reactivity of the addition, the latter increases when the BET surface increases, 280 in contrary to the water demand. However, an increase in calcination temperature to 800 °C led to a decrease in the 281 specific surface. This could be explained by the melting of certain particles that contain elements such as Na, K, and 282 Fe reacting like fluxes and sintering during calcination at high temperatures [15,31]. 283 From the TGA -DTG analysis of the sediments (Fig.3 and Fig.4), several operations can be carried out in order to 284 follow the phase transformation: 285  Calcite content (CaCO3) 286 The CaCO3 content can be calculated from the TGA analysis according to the following Eq (21) and Eq (22) Table 6 shows the content as well as the degree of decarbonation of calcite in the sediments. 293 The presence of CaCO3 and CaO in the calcined products could have impacts on the hydration of calcined 294 sediments based mixtures. In fact, mineral additions with a high CaCO3 content promote hydration of C3S [46], but 295 are very unfavorable for the C3A phases hydration [47]. 296 On contact with water, CaO turns into Ca(OH)2 according to the following reaction Eq (23) [48]: 297 CaO + H2O  Ca(OH)2 + 15.5 kcal (23) This reaction is exothermic and this could cause a strong degradation due to the heat generated in the core of 298 the structure. In addition, an increase in the Ca(OH)2 volume could be the cause of cracking which influenced the 299 durability and the mechanical performance of the structure. 300  Degree of dehydroxylation of kaolinite 301 The content and degree of dehydroxylation of kaolinite (Table 7) can be estimated from the results of TGA 302 analysis according to Eq (12) and Eq (13). 303 The result shows that the kaolinite was transformed into metakaolin Eq(2) from calcination at 650 °C, however 304 the degree of dehydroxylation is relatively low (31%). The increase in calcination temperature significantly improves 305 the dehydroxylation of kaolinite. The degree of dehydroxylation of kaonilite is an important index for evaluating the 306 quality of calcination because the metakaolin formation promotes the pozzolanic reaction. 307

Mineralogical and chemical characterizations 308
The XRD analysis shows the presence of the main mineral phases in the RS, SF 650, SF 750, and SF 800 309 sediments (Fig.5). The results indicate that : 310  The quartz is the major crystalline phase identified in the RS sediment as well as in the calcined 311 sediments.

312
 The presence of calcite is well identified in the RS sediment. An increase in calcination temperature 313 led to a decrease in the calcite content in the calcined products due to the decomposition of this phase. 314  The formation of anhydrite phase in the calcined sediments has been well identified. This could be 315 explained by the reaction between the CaO released from the decomposition of CaCO3 with the sulfate to form the 316 anhydrous calcium sulfate (CaSO4) [22]. 317 The result of the analysis on oriented slides showed that the clay phases of the RS sediment consist mainly of 318 kaolinite and illite (Fig.6). After calcination, the kaolinite transforms into metakaolin due to a dehyroxylation 319 reaction. The degree of dehydroxylation at 800 °C is the highest and this seems in accordance with the result 320 calculated from the TGA analysis (Table 7). Despite calcination temperatures are higher than the dehydroxylation 321 temperature of kaolinite but the calcination time is very short, this is explained by the incomplete reaction during 322 calcination of the sediment at 650 °C,750 °C and 800 °C. 323 The chemical composition of sediments as well as OPC cement and MK is given in Table 8. 324  The major oxides of materials are SiO2, Al2O3, CaO, Fe2O3. 325  The chemical composition of calcined sediments is relatively similar. Therefore the increase in the 326 calcination temperature did not change the composition in the calcined products. Fig.7 shows the 327 content of major oxides in the materials in the CaO-SiO2-Al2O3 ternary diagram. 328 The result of the leaching analysis shows the presence of metallic trace elements as well as anionic elements in 329 sediments (Table 9 and Table 10). The content of certain metallic trace elements in the RS sediment exceeds the 330 value specified for inert waste (IW), in particular, the content of zinc. 331 By comparing the calcined sediments to the raw sediment, it can be seen that the content of some metals such 332 as Cu, Ni and Zn has decreased. This could have beneficial impacts on the hydration of the cementitious matrix 333 incorporating calcined sediments, in particular for zinc. The presence of Zn decreases the compressive strength and 334 reduces the density of the material [49]. In fact Zn 2+ ions combined with OHions to form an impermeable Zn(OH)2 335 film which inhibits the hydration of C3S and C3A with a more pronounced effect on C3S. In addition, Zn(OH) 2and 336 Zn(OH) 3prevent adsorption to the electronegative surface of C-S-H at high pH. 337 The sulfates content in the RS sediment is higher than the limit value for IW. In addition, it can also be seen from 338 the results that the RS sediment has a higher sulfate content compared to the calcined sediments, where the sulfate 339 content decreases with the increase in temperature. The presence of sulfate under certain conditions can cause 340 degradation of materials, such as: significant swelling, cracking, and a decrease in mechanical strength. 341 The presence of chlorides accelerates the setting and hardening of the cementitious matrix. Chlorides can also 342 combine with C3A in the cement to form chloro-aluminate causing a swelling. The chloride content in the sediments 343 seems to remain unchanged during calcination. Similar results were found in the study of Ruben et al [15]. 344

The Frattini test 346
The Frattini's test was performed on samples after 15 days of hydration. The result (Fig.8) shows that, after 15 347 days, the calcined sediments and the MK have pozzolanic reactivity while the RS sediment does not exhibit 348 pozzolanic reactivity. This is due to the formation of metakaolin in the calcined sediments during the dehydroxylation 349 of kaolinite. This transformation was demonstrated in the XRD analysis (Fig.5) and the analysis on oriented slides 350 (Fig.6). In addition, Fig.8 shows that for MK, the pozzolanic reaction could be clearly detected after 8 days. 351

Heat hydration of cement pastes 352
The heat of hydration of cement pastes is shown in the Fig. 9. The first exothermic peak in the acceleration period 353 (point 2) is mainly related to the hydration of C3S and C3A to form C-S-H and ettringite respectively. In the 354 deceleration period, the second exothermic peak is observed, which corresponds to the additional dissolution of C3A protons that provokes the acidifying of the matrix according to the following reaction Eq (24): 362 In addition, Pollard et al. [52] showed that the formation of a physical barrier altered the growth and/or the 363 morphology of the crystals and prevented the normal hydration of the cement. Indeed, the reduction of Ca 2+ ions 364 retarded the nucleation of Ca(OH)2 and generated an unstable C-S-H gel with a low Ca/Si ratio which slowly 365 transformed into a stable C-S-H gel [53]. Agglomeration of sediment may be the cause of delay in hydration because 366 it trapped water in the mix and reduced the water available for hydration [54]. Removal of organic matter from 367 calcined sediment is beneficial by reducing the delay in hydration of the mixture. Metakaolin has the highest 368 pozzolanic reactivity while the three calcined sediments have relatively identical hydration kinetics. This result 369 appears to be consistent with that of Frattini's test. 370 The incorporation of the calcined sediment reduces the heat of hydration of the mixtures. This result is contrary 371 to that reported in the study by Benzerzour et al [9]. However, a similar result was obtained in the study of Safhi et 372 al [55]. This contradiction could be explained by the coarser size (d50 about 15 µm) of the sediments in this study 373 and the study conducted by Safhi et al [55] compared to that in the study of Benzerzour et al [9] (d50 = 5 µm). 374 Therefore, this could lead to higher hydration kinetics. 375

Determination of pozzolanic reactivity by thermal analysis 376
a.
Quantification of chemically bound water and Ca(OH)2 from TGA analysis 377 Fig.10 shows, an example for the method of calculating the content of bound water in the (Ca(OH)2+SF 750) 378 mixture by TGA analysis. 379 From the TGA analysis results of (Ca(OH)2+SF 750) mixture, the following conclusions can be extract : 380  The content of Ca(OH)2 in the mixture decreases with time, due to the pozzolanic reaction between the 381 sediment and the portlandite. 382  An increase in the bound water content in the mixture corresponds to hydrates formed over the time of 383 hydration process. This is consistent with the observed Ca(OH)2 consumption. 384  The CaCO3 content in the mixture at each test time (7, 14 and 28 days) is constant. This stability means that 385 the mixture was well homogenized and that the used sample of the mixture is well representative. 386 The amount of chemically bound water of the mixtures during hydration calculated using Eq (14) is shown in 387 Fig.11. The (Ca(OH)2+MK) mixture showed the highest amount of bound water at all measured times. This is related 388 to the amount of hydrates formed in the paste over the time of hydration. At 7 days, the amount of water bound in 389 mixtures containing sediment is relatively the same. This could be due to the slow pozzolanic reactivity of the 390 calcined sediments. Indeed, after 28 days of hydration, the content of bound water increases in all mixtures. Pastes 391 containing calcined sediment exhibited a higher amount of bound water than the paste incorporating raw sediment. 392 Among the calcined products, the mixture with SF 750 sediment leads to a higher bounded water content at short 393 term as well as at long term. This result is in accordance with previous tests on the pozzolanic reactivity of sediments. 394

b. Ca(OH)2 amount in lime paste 395
The pozzolanic reaction is a reaction that consumes Ca(OH)2, therefore this pozzolanic reaction can be evaluated 396 by measuring the Ca(OH)2 over time of hydration. Fig.12 shows the evolution of Ca(OH)2 in the different mixtures. It 397 can be seen from the results that the MK consumes a greater quantity of Ca(OH)2 than the sediments. Among the 398 sediments, SF 750 seems to be the most reactive when comparing the quantity of Ca(OH)2 consumed at short term 399 as well as at long term. In addition, at short term, the Ca(OH)2 content is relatively stable. This is consistent with the 400 result of the amount of water chemically bound. In fact, the more portlandite is consumed, the more hydrates are 401 formed. 402 In order to demonstrate the consumption of Ca(OH)2 in the mixtures, XRD analysis was performed on the samples 403 after 28 days of hydration. The result has been presented in Fig.13. We can confirm that: 404  The Ca(OH)2 content in the (MK + Ca(OH)2 mixture after 28 days of hydration is the lowest. This means that 405 the Ca(OH)2 content consumed in the mixture is the highest. This result is consistent with the result 406 measured by the TGA analysis (Fig.12). 407  The Ca(OH)2 content in the mixtures contained in the calcined sediments is relatively similar and lower than 408 the (RS + Ca(OH)2) mixture. This result is also consistent with the result found by the TGA analysis (Fig.12). 409

c. Evaluation of hydrates formed from the DTG curve 410
The relative amount of hydrates formed over hydration time of the pastes was estimated from the DTG curves 411 by the deconvolution method based on Eq (16). Fig.14 and Fig.15 show the relative amounts of the C-S-H and other 412 phases such as CASH produced in pastes over time of hydration. By comparing the relative amount of hydrates 413 formed (C-S-H and CASH), it can be concluded that the MK is the most reactive comparing to the other additions. 414 This result appears to be in accordance with previous results, that demonstrated the higher reactivity of MK with 415 respect to sediments. However, we note that the relative amount of C-S-H at 28 days in all mixtures are relatively 416 the same. In addition, the CASH amount formed from the (Ca(OH)2+MK) mixture is much higher than the other 417 mixtures incorporating sediments. This can be explained by the higher Al2O3 content of MK (31.18% wt) compared 418 to sediment (12 wt.% in SF 650, SF 750, SF 800, and 9.64 wt.% in RS). 419 The reaction for C-S-H formation occurs relatively slowly in mixtures containing sediments for up to 14 days. This 420 is well consistent with the quantification result of the Ca(OH)2 amount consumed and the content of the chemically 421 bound water by the TGA analysis (Fig. 11 performance. An increase in the calcination temperature seems to improve the mechanical performances 434 of mortars by increasing the specific surface of the sediments. However, calcination at 800 °C reduces the 435 specific surface of the SF 800 sediment compared to SF 750 and this explains the reduction in the strength 436 of MSF 800 mortars compared to MSF 750. 437 Fig.17 and Fig.18 show the resistance measured on the specimen and the resistance calculated according to 438 Féret's law (with the activity coefficient k = 0) of the mortars over time of hydration. 439 The result shows that: 440  The compressive strength of MRS mortar is always below the theoretical strength curve. This shows a 441 detrimental retardation effect of RS sediment on resistance development. Indeed, the delay effect of RS 442 sediment was also shown in the isothermal calorimetry test (Fig.9). 443  The difference in resistance of MSF 650, MSF 750 and MSF 800 mortars is almost zero after 1 day of 444 hydration. This means that these calcined sediments are inerts and do not contribute to the development 445 of resistance. However, after 2 days of hydration, the difference is positive. This difference seems to get 446 bigger and bigger with the time of hydration. This means that these sediments would have a chemical effect 447 such as the pozzolanic reaction contributing to the development of resistance. 448  The compressive strength of MMK mortar is always above the theoretical strength curve. An increase in 449 this difference in hydration time has been observed. In addition, this difference is also greater than the 450 difference in the case of mortars containing calcined sediment. This means the MK would have a higher 451 pozzolanic reactivity than the calcined sediments. This result is consistent with the result shown in the 452 analysis of isothermal calorimetry (Fig.9) and lime consumption (Fig.12). 453

3.5
Dynamic modulus of mortars 454 Fig.19 shows the evolution of the dynamic modulus of mortars calculated according to Eq (20) from 28 days to 455 90 days. Stabilization of the dynamic modulus after 56 days indicates that all of the portlandite has been consumed 456 by the pozzolanic reaction [55]. Indeed, the consumption of portlandite makes it possible to form a more dense and 457 coherent microstructure. In addition, a relatively large increase in dynamic modulus of sediment containing mortars 458 between 28 and 56 days compared to MMK mortar indicates the long-term pozzolanic reactivity of the sediments. 459

3.6
Porosity of mortars 460 The mercury porosity of mortars measured at 56 days is shown in Table 11  reduced the porosity of the mortar. The pozzolanic reaction appears to overcome the replacement dilution 470 effect. In addition, these mineral additions appear to reduce the size of the pores towards the finer pores. 471 According to Zengfeng et al [57], the presence of the sediment allowed a finer distribution of the porosity, 472 most often associated with a better texture of C-S-H. 473

3.7
Leach analysis result of mortars at 90 days of curing 474 Table 12 shows the content of metallic trace elements in the mortars after 90 days of hydration. In addition, the 475 content of anionic elements in the mortars was shown in Table 13. A considerable reduction in the sulfate content 476 was observed. The result showed a stabilization of all the elements compared to the leaching limits of inert waste. 477

3.8
Analysis of the quality of the exhaust fumes 478 The quality of the gases is an important indicator to assess the deasibility of the calcination process.The gases 479 emitted during calcination of the sediment were analyzed by the company Socor air. The results of certain elements 480 in the flue gases are presented in Table 14. 481 The results in Table 14 leads to tell that the gas content (O2, CO2) stabilizes during calcination. An increase in the 482 calcination temperature allows the reduction of the CO gas emissions by reducing it by 97.6% for the temperature 483 of 750 °C comparing to 650 °C . The similar trend with NOx gases was observed with increasing calcination 484 temperature. 485

Conclusion 486
The objective of this article is to study the influence of the calcination temperature of sediment on their 487 properties. This consists to evaluate several criteria such as the activation of pozzolanic reactivity, mechanical 488 performance, energy consumption, and the quality of the gases emitted. Besides, the effects of the use of calcined 489 sediments on the kinetics of hydration were investigated. It can be concluded from this study that: 490  Flash calcination method removes the fraction of organic matter already present in the raw sediment. This 491 reduces the delay in hydration related to the presence of organic material. 492  The calcination temperature modifies the physical properties comparing to the raw sediment. Indeed, 493 calcined sediments have a greater BET specific surface and density than the raw sediment. The transformation of 494 kaolinite to metakaolin and the formation of anhydrite in calcined sediments were observed by XRD analysis. 495  The flash calcination also active the pozzolanic reactivity of the sediment. The calcined sediments showed 496 a strong ability to react with portlandite to form hydrates compared to the raw sediment. 497  The heat of hydration of binders containing calcined sediment is greater than that of the binder with raw 498 sediment. However, the calcined sediment reduces the heat of hydration of the binder compared to the reference 499 cement. This could have a positive effect in the case of concrete and a negative effect in the development of strength 500 at an early age. 501  The incorporation of the calcined sediment significantly improved the mechanical performance of the 502 mortars compared to the raw sediment. Besides, the reducing in the hydration delay with calcined sediments 503 improves the development of mechanical strength at young ages comparing to the raw sediment. MSF mortars have 504 similar mechanical strength as the reference mortar at long-term. 505  The incorporation of calcined sediments reduces the porosity of the mortar due to the pozzolanic reaction. 506  An increase in the calcination temperature from 650 °C to 750 °C improves the quality of the gases emitted 507 by comparing the content of the elements in the emitted gases. An increase in the calcination temperature allows 508 the CO gas content to be greatly reduced (97.6% reduction). 509  The content of some metallic trace elements was reduced in calcined sediment, especially zinc content. In 510 addition, the content of flourures, chlorides and sulfates tends to decrease with increasing calcination temperature. 511

Declaration of competing interest 512
The authors declare that they no known competing financial interests or personal relationships that could have 513 appeared to influence the work reported in this paper. 514 Acknowlegments 515 The authors wish to acknowledge the SEDICIM project and the FEDER funds. 516 Data Availability 517 The datasets generated during and/or analysed during the current study are available from the corresponding 518 author on reasonable request. 519 Reference 520 1. Bordy, A.: Influence des conditions thermo-hydriques de conservation sur l ' hydratation de matériaux 521 cimentaires à base d ' une fine recyclée. 1-155 (2016)