Highly Active Calcium-Silicate Phases for Application in Endodontics

Introduction Mineral trioxide aggregate (MTA) is one of the most commonly used materials in endodontics. Given its shortcomings, there is an intensive search for new materials. Calcium-silicate phase (CS phase) is a new material synthesized by the method based on a combination of sol-gel process and self-propagation synthesis which can significantly improve setting time through accelerated hydration. The aim of this study was to explain the mechanisms of hardening of CS phase in aqueous medium as similar mechanism is expected in contact with body fluids. Materials and Methods CS phases Ca3SiO5 (C3S) and 2β-CaSiO4 (β-C2S) were synthesized from CaCl2·5H2O (Merck, Germany). To investigate the process of hydration, CS were mixed with water and kept at 37 °C for 28 days in closed polyethylene containers. Analysis of the composition of samples before and after immersion in water for 1, 3, 7 and 28 days was performed using X ray diffractometry and IR. Results XRD patterns for hydrated samples during 1, 3, 7 and 28 days showed that the amount of hydrated tobermorite phases relative to the amount of CS phase changes with the time of hydration. After 1 day, in addition to the hydrated phases, a significant amount of untransformed β-C2S and C3S were found while after 28 days hydrated CS phase was completely transformed to tobermorite with a small amount of portlandite. Conclusion Using combined method of sol-gel and self-propagating waves at high temperature very active nanostructured silicate phases were obtained. Hydration process of CS phases was analyzed using XRD and FTIR, the mechanism of hydration was proposed and it was pointed to the difficulties in determining the exact reaction as well as the problem of determining the exact structure of tobermorite.


INTRODUCTION
Mineral trioxide aggregate (MTA) is one of the most commonly used materials in endodontic practice for the treatment of various root canal perforations, pulpotomy and vital pulp treatment, as an apical barrier in teeth with necrotic pulp etc.After application of MTA for direct pulp capping in permanent teeth, the appearance of calcified bridge was observed after one week only [1,2].Several in vitro and in vivo studies have demonstrated that sealing abilities of MTA are superior to those of amalgam, IRM and super EBA [1,36].However, several important dis advantages of MTA have been pointed: long setting time, weak rheological properties and weakening of the material structure during hydration in acidic environment.Also, particle size, powder to liquid ratio, temperature and pres ence of air in the mixture may influence the physical prop erties of MTA [6].Therefore, search for new material with similar composition which shows higher degree of activity of calciumsilicate (CS) and better mechanical properties has been continued.
In the current study, calciumsilicate phases (CS phas es) were synthesized by the method, based on the com bination of solgel process and selfpropagating synthesis, which can significantly improve setting time of obtained phases through their accelerated hydration.The process of their hydration, as the most responsible for their behavior, was particularly analyzed.
The aim of this study is to understand mechanisms of cement hardening in water medium, as similar mechan ism is expected in contact with body fluids.

Synthesis and characterization methods of calcium-silicate phases
The CS phases, Ca 3 SiO 5 (C 3 S) and 2βCaSiO 4 (βC 2 S), were synthesized from CaCl 2 •5H 2 O (Merck, Germany) and silica sol was obtained by hydrothermal treatment [7].The chemicals were used in stoichiometric quanti ties, corresponding to the ratio C 3 S: βC 2 S = 2:1, to obtain silicate active phase.Al(C 2 H 3 O 2 ) was added to the mixture to provide production of small amount (~3%) of active C 3 A phase.
Firstly, the mixture of silica sol and CaCl 2 •5H 2 O was dried at 80 °C to obtain gel, and then heated at 150 °C to remove water among the silica particles.In order to start the combustion reaction, ammonium nitrate (NH 4 NO 3 ), as an oxidation agent, and citric acid (C 6 H 8 O 7 CH 2 O), as a fuel, were added.During the strong reaction of self propagating combustion the gel swelled into the foam and black ashes were obtained as a product of autoignition.After this treatment at high temperatures the sample was quickly cooled using copper plates, in order to obtain high reactivity and low crystallinity of βC 2 S and C 3 S phases.Finally, the resulting black powder was calcined in air at 650°C for 4 h to remove carbon residues and obtain de sired products with small crystallite sizes.
In order to investigate the process of their hydration, CS was mixed with water (watertopowder ratio about 1:2) and compacted using a stainless steel plunger.Ce ment was allowed to set up to 28 days at 37°C in sealed polyethylene containers.Phase compositions of the sam ples before immersing in water and 1, 3, 7 and 28 days after immersion were analyzed using Xray diffractom etry, XRD (Philips PW 1050, Almelo), with Ni-filtered CuKa 1.2 radiation.The patterns were registered in the 2q range 967° with a scanning step size of 0.02°.FTIR analy sis (Nicollet 380 FTIR, Termo Electron Corporation) was done for the samples before soaking and for the samples soaked for 7 and 28 days.
XRD patterns for samples hydrated for 1, 3, 7 and 28 days (Graph 1), show that the quantity of hydrated phase tobermorite relative to sum of CS phases changed with hydration time.For the samples hydrated 1 day, beside the presence of hydrated phase of tobermorite [planes (006), (112), (202) and (213)] and portlandite [planes (011), (012) and (001)] which showed the progress of hydration; significant amounts of βC 2 S and C 3 S remained untrans formed.For the samples hydrated 3 days, a few peaks of low intensity indicated that only slight amounts of CS phases were present.For the samples hydrated 7 days traces of C 3 S were present, while the peaks of the main hydrated phase, tobermorite, were more pronounced.Finally, for the samples hydrated 28 days, CS were completely transformed to tobermorite, with small amounts of portlandite.
The FTIR spectrum of CS phases is shown in Graph 2. The characteristic doublet recorded between 2334 and 2364 cm1 can be assigned to the stretching vibration of OH groups, and explained by the decrease in hydrogen bond length, caused by steric effects [7,8].The band at 2172 cm 1 can be assigned to SiH stretching mode in the small grains, while the band at 1968 cm 1 corresponds to SiO 2 vibrational modes [8].The band at 1651 cm 1 can be ascribed to the liberation mode of OH, while the broad band with a minimum at 1414 cm 1 may be ascribed to vi brations corresponding to partially hydrated C 3 S and C 2 S phases.The bands at 1334 cm 1 region and 871 cm 1  respond to the asymmetric stretch and the outofplane bending of CO, respectively.The band at 1121 cm 1 can be assigned to the vibration of C 2 S units, while small band (shoulder) at 921 cm 1 can be indication of slight hydra tion of C 3 S or C 2 S. The band at 664 cm 1 can be attributed to the SiOSi symmetric vibration and band at 516 cm 1 to the outofplane bending vibration of SiO 4 [9,10].
The FTIR spectra of the samples hydrated for 1 day (S1) and 28 days (S28) show the bands at 3726 and 3718 cm 1 , respectively, indicating the presence of the 1.4 tobermorite which contains poorly resolved band in the range of 2800-3700 cm −1 due to larger content of mo lecular H 2 O, compared to the 1.1 nm tobermorite.The bands at 3726 cm −1 and 3718 cm 1 can be assigned to vibrations which involve less hydrogen bonded protons in water molecules, including intensity corresponding to CaOH groups in the structure.The weak bands at 3740 cm −1 can be also assigned to SiOH vibrations.The bands at 2997, 2938 and 2880 cm 1 (S1), and 2990, 2938 and 2872 cm 1 (S28) confirmed also the presence of the 1.4 nm tobermorite [11].
The characteristic sharp and pronounced band at 2357 cm 1 can be assigned to the combination of bands at 959 cm 1 and broad band at 1407 cm 1 , the band at 2327 cm 1 to the combination of bands at 1407 cm 1 and 878 cm 1 while the band at 2364 cm 1 probably belongs to the combination of bands at 1489 cm 1 and 871 cm 1 .This characteristic doublet recorded between 2327 and 2357 cm 1 (S1) and 2364 and 2320 cm 1 (S28) can be stretch ing vibration of OH groups explained by corresponding decrease in hydrogen bond length, caused by steric effects.The weak bands at 1407 and 1458 cm 1 and its combina tion, and overtone bands observed at 2997, 2938 and 2880 cm 1 (S1) and 2990, 2938 and 2872 cm 1 (S28) and weak vibrations at 1407 and 1414 cm 1 indicate the presence of carbonate species in the samples.The band at 1614 cm 1 belongs to the water bending vibration.The bands at 2159 cm 1 (S1) and 2165 cm 1 (S28) can be assigned to SiH stretching mode in the small grains, gradually exposed to oxidation during combustion process at intermediate temperature during synthesis given phases.This is some kind of fingerprints of the oxidation state of silica during synthesis of these powders.The bands at 1709 cm 1 and 1702 cm 1 correspond probably to water bending vibration inside of gypsum dehydrate present in small amount in samples.The band at 1216 cm 1 (S28) can be assigned to the SiO stretching vibrations in Q3 site, characteristic for presence of the 1.1 nm tobermorite, while the bands at 1061 and 1054 cm 1 belong to the SiO stretching vibra tions in Q2 site of the 1.1 nm tobermorite.This proves possible combination of tobermorite 1.4 nm and 1.1 nm.The bands at 955 cm 1 (S1) and 966 cm 1 (S28) belong to SiO lattice vibrations, while the bands at 878 cm 1 (S1) and 871 cm 1 (S28) belong to the out of plane vibrations of CO.The bands at 664 cm 1 (S1) and 671 cm 1 (S28) can be assigned to SiOSi bending vibrations inside silica chains and ν 4 SO 4 2 bending vibrations.These bands can also be attributed to the SiOSi symmetric vibration, while the bands at 494 cm 1 (S1) and 509 cm 1 (S28) belong to the outofplane bending vibration of SiO 4 .

Hydration process of calcium-silicate pastes
The most important parameter affecting the setting and mechanical properties of CS phases is the rate of their hydration.The hydration of βC 2 S and C 3 S phases is the most important process for setting of these phases during their ageing in water solutions.As it was mentioned in various references [812], the reaction between C 3 S and water is the main factor in the setting and hardening of cement mixtures.During this reaction, CS grains become wet, causing rapid release of Ca 2+ and OH ions from each grain surface.The process of transformation of C 3 S to amorphous calciumsilicatehydrate (CSH), known as tobermorite gel, and calcium hydroxide (Ca(OH) 2 ) can be expressed by the reaction: (1) Similarly, the reaction between βC 2 S and water is ex pressed as: (2) The CSH obtained in both reactions was the high lime endmember of a series of hydrates assigned by Taylor as calciumsilicate hydrate (I) or CSH (1) and it can have variable composition.Bernal was the first one who tentatively assigned the structural formula Ca Difficulties in determination of the exact reaction are joined with a huge problem of determination of the exact structure and corresponding formula of tobermor ite.We assumed the ideal formula of tobermorite is 4CaO•Ca(OH) 2 •6SiO 2 •4H 2 O, although it is wellknown that central part of each tobermorite layer is similar to octahedral layer in a clay mineral, which could be de scribed as a distorted calcium hydroxide sheet divested of all its hydrogen atoms (CaO 2 ).This sheet is flanked on both sides by parallel rows of wollastonitetype chains that are kinked in a plane perpendicular to the plane of the sheets [1316].
Only twothirds of the tetrahedrons of each chain are linked directly to the central CaO 2 sheet by sharing the oxygen atoms.The other third, held away from the CaO 2 sheet, can be observed as "bridging" tetrahedrons.The corresponding composite 2:1 sheets, so far described, have the composition Ca 4 SiO 4 O 18 .The remaining or interlayer calcium atoms and water molecules are placed between these sheets.The distribution of hydrogen is uncertain.There are probably more SiOH groups than it is implied by the formula Ca 4 (SiO 4 O 18 H 2 )•Ca•4H 2 O, with corresponding replacement of interlayer water by hydroxyl groups [15,16,17].Following this, when CS paste is prepared, C 3 S phase probably reacts very quickly and water becomes suddenly saturated with Ca(OH) 2 (reaction occurs within minutes).The same happens with βC 2 S, but more slowly.As the result of hydration, Ca(OH) 2 and a member of the tobermorite (G, tobermorite 11) subgroup are formed [15].If, however, the silicates are hydrated in a large ex cess of water, so that the concentration of the lime in the solution is below saturation, the CSH that forms is not a member of the tobermorite (G) series and, of course, no Ca(OH) 2 is produced in the reaction.The CaO/SiO 2 ratio in the CSH depends on the final concentration of the lime in the solution [the lower limit of the ratio (at a low lime concentration) is around 0.8, and the upper limit is about 1.5 (at or near lime saturation)] [16].
Taylor and Howison [17] measured densities of CSH (I), in the range of CaO/SiO 2 ratio from 0.81 to 1.50, and found no systematic trend of density increase with the ratio.They have also advanced a hypothesis for the explanation of constant density together with constant unit cell dimension.They suggested that, as the CaO/SiO 2 ratio increases, calcium replaces silicon in the lattice.The "bridging" SiO 4 tetrahedrons, which are not attached to the central CaO 2 sheet, are frequently gradually removed or, more correctly, only SiO 2 is removed because two of the oxygen ions in each tetrahedron are common to adjacent groups.The calcium ion enters the lattice, but not directly into the place of SiO 2 , i.e., not into the layer, but between layers.Since the calcium ion has two positive charges and SiO 2 is neutral, two hydrogen ions must also be removed to preserve charge balance.The overall replacement is SiO 2 H 2 by Ca.Such replacement would have relatively small effect on cell dimensions.It would also explain very slow equilibration between CSH (I) hydrates and Ca(OH) 2 solutions.If the increase of the lime content would involve merely uptake of lime between the layers, one would not expect such a slow equilibration.Finally, the hypothesis predicts that the upper limit of the CaO/SiO 2 ratio would be 1.75, corresponding to the removal of all "bridging" tetrahedrons.Although the upper limit of the ratio for the CSH (I) subgroup is around 1.5, the upper limit for the tobermorite (G) subgroup is 1.75, as was stated be fore, and the upper limit for another subgroup, CSH (II), is also 1.75.It is possible that either the tobermorite (G) or CSH (II) or both of these subgroups form a continuous series with CSH (I).Brunauer and Greenberg have modi fied the TaylorHowison hypothesis with the suggestion that SiO 2 H 2 is replaced, not by Ca alone, but by Ca plus a water molecule.Such substitution would lead to a far better constancy of the cell dimension than the Taylor Howison substitution; it would explain found compos itions of tobermorite better; and it would also explain the constancy of the c spacing, mentioned earlier [18].The modified TaylorHowison hypothesis corresponds quite well for the compositions, structures, densities, and some other properties of the hydrates of the CSH (I) subgroup, possibly even of the entire tobermorite group.Besides structural variations, the CSH (I) hydrates exhibit a wide range of specific surface areas.Such differences in the structure and surface area are, doubtless, responsible for differences in the ratio between compositions of various kind of CSH, formed during the hydration process [16,17,18].
Otherwise, it is wellknown that tobermorite (G) hy drates are poorly crystallized.Their Xray diffraction patterns consist of three lines: a very strong and very broad reflection, with a maximum at 0.305 nm, and two weaker broad lines at 0.279 and 0.182 nm.Heller & Tay lor found that these spacings have indices of (110), (200), and (020) planes, respectively, and they reported the same spacings at 0.307, 0.281, and 0.183 nm for less dried CSH (I) samples prepared hydrothermally [19].The three lines of tobermorite (G) hydrates correspond to the three of the strongest lines of all tobermorites.All three are hk0 spacings, i.e., spacings within the layers.The (020) spacing represents half the distance between neighboring Si atoms in the chains of the SiO 4 tetrahedrons of the 2:1 layers.Twice this distance, 0.364 nm is the length of the ortho rhombic pseudo cell.Twice the (200) spacing, 0.559 nm, is the length of the pseudo cell.However, tobermorite (G) exhibits no basal spacing, so the c length can be obtained only by using slightly modified approach [19,20].
According to this approach, it can be concluded that in the tobermorite (G) hydrate of composition Ca 3 Si 2 O 7 • H 2 O, one molecule of water is between layers.After drying, the molecule contains only the hydroxylic water attached to the SiO 4 tetrahedrons in layers, and this water is more firmly held than the interlayer water.On the other hand, tobermorite is often finely divided in the c direction.This dimension of crystallite is colloidal.The dimension of the particles in the b direction, i.e., along silicate chains, is quite large; it is in the order of a micron or several mi crons.Thus, it seems that tobermorite is colloid in the c direction but not in the b direction.It is interesting that interlayer water is lost continuously, as water is usually lost by colloids, but the water content of the silicate chains remains almost constant independently on the conditions of its setting [16,21].
The adhesion of tobermorite particles to each other and to other bodies is responsible for the strengths of hardened pastes of tricalcium and dicalcium silicates; it is also the most important factor in the strengths of hardened portland cement pastes.Compressive strength of hardened portland cement pastes exceed their ten sile strength by an order of magnitude, and compressive strengths of hardened CS pastes are about the same with those of hardened portland cement pastes.Although no data have been published on tensile strengths of CS pastes, it seems likely that they are lower than those of portland cement pastes because CS pastes are dispersed easily by ultrasonic vibrations contrary to portland cement pastes [22,23].

CONCLUSION
The advantages of synthesis of active silicate phases by combined sol gel and hightemperature selfpropagating wave method, by the first time applied in this paper, are clearly shown.The obtained nanostructured silicate phases are very active.They show significant increase of setting properties of active silicate phase, as the most im portant phase in any endodontic mixture.
The process of hydration of CS phases is carefully analyzed by XRD and FTIR, from the aspect of structural changes inside the mixture during its wetting for 1, 3, 7 and 28 days.After 3 days, βC 2 S and C 3 S phases were mostly transformed to tobermorite, while after 7 day they were completely transformed.The mechanism of hydra tion was proposed and deep insight into difficulties in determination of the exact reaction, joined with a problem of determination of the exact structure of tobermorite, was given.

UVOD
Mi ne ral tri ok sid ni agre gat (MTA) je je dan od naj če šće ko ri šće nih ma te ri ja la u en do don ci ji, u le če nju per fo ra ci je ka na la, pul po to mi ji, te ra pi ji vi tal ne pul pe, za po sti za nje apek sne ba ri je re kod ne kro tič nih zu ba itd.Na kon pri me ne MTA za di rekt no pre kri va nje pul pe kod stal nih zu ba po ja va kal ci fi ko va nog mo sta uoče na je na kon sa mo ne de lju da na [1,2].Ne ko li ko stu di ja in vi tro i in vi vo po ka za lo je da MTA obez be đu je mno go bo lje zap ti va nje u od no su na amal gam, IRM i Su per EBA ma te ri ja le [1,36].Me đu tim, po sto ji ne ko li ko va žnih ne do sta ta ka MTA: du go vre me ve zi va nja, sla ba re o lo ška svoj stva i sla blje nje struk tu re ma te ri ja la to kom hi dra ta ci je u ki se loj sre di ni.Ta ko đe, ve li či na če sti ca, od nos pra ha pre ma teč no sti, tem pe ra tu ra i pri su stvo va zdu ha mo gu uti ca ti na fi zič ke oso bi ne MTA [6].Sto ga po tra ga za no vim ma te ri ja lom sa slič nim svoj stvi ma ko ji po ka zu je bo lju ak tiv nost kal ci jumsi li ka ta (CS) i bo lja me ha nič ka svoj stva ni je za vr še na.
U na šoj stu di ji je sin te ti sa na kal ci jumsi li kat na fa za (CS fa za) pre ma me to di za sno va noj na kom bi na ci ji solgel pro ce sa i sa mo pre no si voj sin te zi, što mo že zna čaj no po bolj ša ti vre me ve zi va nja do bi je nih fa za kroz nji ho vu ubr za nu hi dra ta ci ju.U stu di ji je po seb no ana li zi ran pro ces hi dra ta ci je, kao naj od go vor ni ji pro ces za nji ho ve oso bi ne.
Cilj ove stu di je je bio da ob ja sni me ha ni zme otvrd nja va nja ce men ta u vo de nom me di ju mu bu du ći da se sli čan me ha ni zam oče ku je u kon tak tu s te le snim teč no sti ma.

Sinteza kalcijum-silikatne faze i karakterizacija metode
CS fa ze Ca 3 SiO 5 (C 3 S) i 2βCa SiO 4 (βC 2 S) su sin te ti sa ne iz CaCl 2 •5H 2 O (Merck, Ne mač ka), a si li kasol je do bi jen hi dro ter mič kim tret ma nom [7].He mij ske sup stan ce su ko ri šće ne po ste hi o me trij skim raz me ra ma (od nos C 3 S i βC 2 S bio je 2:1), da bi se do bi le si li kat ne ak tiv ne fa ze.Al(C 2 H 3 O 2 ) je do dat u ovu me ša vi nu da bi se omo gu ći lo do bi ja nje ma le ko li či ne (~3%) ak tiv ne fa ze C 3 A. Naj pre je sme sa si li kaso la i CaCl 2 •5H 2 O osu še na na 80°C da bi se do bio gel, a za tim je za gre ja na na 150°C da bi se uklo ni la vo da iz me đu si li kače sti ca.Da bi za po če la re ak ci ja sa go re va nja, do da ti su amo ni jumni trat (NH 4 NO 3 ) kao ok si dans i li mun ska ki se li na (C 6 H 8 O 7 CH 2 O) kao go ri vo.To kom ja ke re ak ci je sa mo pre no si vog sa go re va nja gel se pre tvo rio u pe nu a cr ni pe peo je do bi jen kao pro iz vod sa mo za pa lje nja.Na kon ovog tret ma na na vi so kim tem pe ra tu ra ma uzo rak je br zo ohla đen po mo ću ba kar nih plo ča, ka ko bi se do bi li vi so ka re ak tiv nost i ni ski kri sta li tet βC 2 S i C 3 S fa ze.Na kra ju, do bi je ni cr ni prah je kal ci ni san na va zdu hu i 650°C če ti ri sa ta da bi se uklo nio osta tak uglje ni ka i do bi li že lje ni pro iz vo di s ma lim ve li či na ma kri sta li ta.
Da bi smo is tra ži li pro ces hi dra ta ci je, CS su po me ša ni s vo dom (od nos vo de i pra ha bio je 1:2) i po ti snu ti ko ri ste ći na bi jač od ner đa ju ćeg če li ka.Ce ment se ve zi vao to kom 28 da na na 37°C u za tvo re nim po li e ti len skim kon tej ne ri ma.Ana li za sa sta va uzo ra ka pre po ta pa nja u vo du i je dan dan, tri, se dam i 28 da na na kon po ta pa nja su ana li zi ra ni di frak to me tri jom ra di o lo ških zra ka, XRD (Phi lips PW 1050, Al me lo), sa Nifil tri ra nim Cu Ka 1.2 zra če njem.Obra sci su re gi stro va na u ra spo nu 2q 967° s ve li či nom ko ra ka od 0,02°.FTIR ana li za (Ni col let 380 FTIR, Ter mo Elec tron Cor po ra tion) ura đe na je za uzo rak pre po ta pa nja i, kao i za uzor ke po to plje ne u vo du, 7 i 28 da na.
XRD obra sci za uzor ke hi dra ti sa ne je dan dan, tri, se dam i 28 da na (Gra fi kon 1) po ka zu ju da se ko li či na hi dra ti sa ne fa ze to ber mo ri ta u od no su na ko li či nu fa ze CS me nja s vre me nom hi dra ta ci je.Kod uzor ka hi dra ti sa nog je dan dan, po red hi dra ti sa ne fa ze to ber mo ri ta [rav ni (006), (112), (202) i ( 213)] i por tlan di ta [(rav ni (011), (012) i (001)], ko ji su po ka za li na pre dak hi dra ta ci je, utvr đe ne su i zna čaj ne ko li či ne βC 2 S i C 3 S ko je su osta le ne tran sfor mi sa ne.Kod uzo ra ka hi dra ti sa nih tri da na ne ko li ko vr ho va ni skog in ten zi te ta uka zu ju na to da je sa mo ne znat na ko li či na CS fa ze bi la pri sut na, dok je kod uzo ra ka hi dra ti sa nih se dam da na C 3 S na đen u tra go vi ma, gde su vr ho vi glav ne hi dra ti sa ne fa ze, to ber mo ri ta, bi li iz ra že ni ji.Ko nač no, kod uzo ra ka hi dra ti sa nih 28 da na, CS fa za je bi la pot pu no tran sfor mi sa na u to ber mo rit uz ma le ko li či ne por tlan di ta.
FTIR spek tar CS fa za je pri ka zan na gra fi ko nu 2. Ti pi čan du blet za be le žen iz me đu 2334 i 2364 cm 1 mo že se pri pi sa ti is te žu ćoj vi bra ci ji OHgru pa i ob ja sni ti pa dom du ži ne vo do nič ne ve ze uzro ko va ne ster nim efek tom [7,8].Li ni ja na 2172 cm 1 mo že se pri pi sa ti is te žu ćem mo du su u ma la zr na, dok li ni ja na 1968 cm 1 od go va ra SiO 2 vi bra cij skom mo du su [8].Li ni ja na 1651 cm 1 mo že se pri pi sa ti oslo ba đa ju ćem mo du su OH, dok ši ro ka li ni ja s mi ni mu mom na 1414 cm 1 mo že od go va ra ti vi bra ci ja ma de li mič no hi dra ti sa ne C 3 S i C 2 S fa ze.Li ni je na 1334 cm 1 i 871 cm 1 od go va ra ju asi me trič nom is te za nju i sa vi ja nju van rav ni CO gru pe.Li ni ja na 1121 cm 1 mo že se pri pi sa ti vi bra ci ji C 2 S je di ni ca, dok ma la li ni ja (ra me) na 921 cm 1 mo že bi ti po ka za telj bla ge hi dra ta ci je C 3 S ili C 2 S. Li ni ja na 664 cm 1 mo že se pri pi sa ti SiOSi si me trič noj vi bra ci ji, dok se li ni ja na 516 cm 1 pri pi su je sa vi je noj vi bra ci ji van rav ni SiO 4 [9,10].
FTIR spek tri uzo ra ka hi dra ti sa nih 1 dan (S1) i 28 da na (S 28) po ka zu ju li ni je na 3726 i 3718 cm 1 uka zu ju ći na pri su stvo 1,4 to ber mo ri ta ko ji sa dr ži sla bo raz dvo je ne li ni je u ra spo nu od 2800 do 3700 cm 1 zbog ve ćeg sa dr ža ja mo le ku lar ne H 2 O u od no su na 1,1 nm to ber mo ri ta.Li ni je na 3726 cm 1 i 3718 cm 1 mo gu se pri pi sa ti vi bra ci ja ma ko je uklju ču ju pro to ne vo do ni ka ve za ne u mo le ku le vo de, uklju ču ju ći in ten zi tet ko ji od go va ra CaOH gru pa ma.Sla be li ni je na 3740 cm 1 mo gu se ta ko đe pri pi sa ti SiOH vi bra ci ja ma.Li ni je na 2997, 2938 i 2880 cm 1 (S1) i 2990, 2938 i 2872 cm 1 (S28) ta ko đe do ka zu ju pri su stvo to ber mo ri ta od 1,4 nm [11].
Ti pič na oštra i iz ra že na li ni ja na 2357 cm 1 mo že bi ti do de lje na kom bi na ci ji li ni ja na 959 cm 1 i ši ro ke li ni je na 1407 cm 1 , li ni je na 2327 cm 1 i kom bi na ci ji li ni ja na 1407 cm 1 i 878 cm 1 , dok li ni ja na 2364 cm 1 ve ro vat no pri pa da kom bi na ci ji li ni ja na 1489 cm 1 i 871 cm 1 .Ti pič ni du blet za be le žen iz me đu 2327 i 2357 cm 1 (S1) i 2364 i 2320 cm 1 (S28) mo že bi ti is te žu ća vi bra ci ja OHgru pe, a ob ja šnja va se od go va ra jućim sma nje njem u du ži ni vo do nič ne ve ze iza zva ne ster nim efek ti ma.Sla be li ni je na 1407 i 1458 cm 1 i nje ne kom bi na ci je, kao i do dat ne li ni je na 2997, 2938 i 2880 cm 1 (S1) i 2990, 2938 i 2872 cm 1 (S28) i sla be vi bra ci je na 1407 i 1414 cm 1 uka zu ju na pri su stvo uglje ni ka u uzor ci ma.Li ni ja na 1614 cm 1 pri pa da sa vi ja ju ćim vi bra ci ja ma mo le ku la vo de.Li ni je na 2159 i 2880 cm 1 (S1) i 2165 cm 1 (S28) mo gu po ti ca ti od is teg nu tih mo du sa ma lih zr na SiH, po ste pe no iz lo že nih ok si da ci ji to kom pro ce sa sa go re va nja na sred njoj tem pe ra tu ri to kom sin te ze da tih fa za.Ovo je ne ka vr sta oti sa ka pr sti ju ok si da cij skog sta nja si li ci jumdi ok si da to kom sin te ze ovog pra ha.Li ni je na 1709 cm 1 i 1702 cm 1 od go va ra ju ve ro vat no sa vi ja ju ćim vi bra ci ja ma vo de unu tar de hi dri ra nog gip sa pri sut nog u ma loj ko li či ni u uzor ci ma.Li ni ja na 1216 cm 1 (S28) mo že po ti ca ti od SiO is te žu ćih vi bra ci ja na Q3 me stu, ti pič nog za pri su stvo to ber mo ri ta ve li či ne 1,1 nm, dok li ni je na 1061 i 1054 cm 1 pri pa da ju SiO is te žu ćim vi bra ci ja ma na Q2 me stu to ber mo ri ta od 1,1 nm.Ovo do ka zu je mo guću kom bi na ci ju to ber mo ri ta ve li či ne 1,4 i 1,1 nm.Li ni je na 955 cm 1 (S1) i 966 cm 1 (S28) pri pa da ju vi bra ci ja ma SiO re šet ke, dok li ni je na 878 cm 1 (S1) i 871 cm 1 (S28) pri pa da ju vi bra ci ja ma van rav ni CO gru pe.Li ni je na 664 cm 1 (S1) i 671 cm 1 (S28) mo gu se pri pi sa ti SiOSi sa vi ja ju ćim vi bra ci ja ma unu tar si li ka la na ca i ν 4 SO 4 2 sa vi ja ju ćih vi bra ci ja.Ove li ni je se ta ko đe mo gu pri pi sa ti SiOSi si me trič nim vi bra ci ja ma, dok tra ke na 494 cm 1 (S1) i 509 cm 1 (S28) pri pa da ju sa vi ja ju ćim vi bra ci ja ma van rav ni gru pe SiO 4 .Ko nač no, li ni je na 406 cm 1 (S1) i 414 cm 1 (S28) od go va ra ju de for ma ci ja ma SiO 4 te tra e da ra.

Proces hidratacije kalcijum-silikatne paste
Naj va žni ji pa ra me tar ko ji uti če na ve zi va nje i me ha nič ka svoj stva CS fa za je pro ce nat nji ho ve hi dra ta ci je.Hi dra ta ci ja βC 2 S i C 3 S fa za je naj va žni ji pro ces za ve zi va nje ovih fa za to kom nji ho vog sta re nja u vo de nim ras tvo ri ma.Kao što je po me nu to u raz li či tim stu di ja ma [812], re ak ci ja iz me đu C 3 S i vo de je glav ni fak tor u od re đi va nju ve zi va nja i stvrd nja va nja sme se ce men ta.To kom ove re ak ci je zr na CS se na vla že i uzro ku ju br zo oslo ba đa nje Ca 2+ i OH jo na s po vr ši ne sva kog zr na.Pro ces tran sfor ma ci je C 3 S u amorf ni kal ci jumsi li kahi drat (CSH), po znat kao to ber mo ritgel i kal ci jumhi drok sid (Ca(OH) 2 ), mo že se pri ka za ti po moću re ak ci je: (1) Slič no to me, re ak ci ja iz me đu βC 2 S i vo de je iz ra že na kao: (2) CSH do bi jen u obe re ak ci je po sled nji je član u se ri ji hi dra ta ko je je Tej lor (Taylor) opi sao kao kal ci jumsi li kahi drat (I) ili CSH (1) i on mo že bi ti pro men lji vog sa sta va.Ber nal (Ber nal) je pr vi do de lio struk tur nu for mu lu Ca 2 [SiO 2 (OH) 2 ] 2 [Ca(OH) 2 ] ovom je di nje nju [814].
Te škoće u od re đi va nju tač ne re ak ci je su po ve za ne s pro ble mom utvr đi va nja tač ne struk tu re i od go va ra juće for mu le to ber mo ri ta.Pret po sta vi li smo da je ide al na for mu la to ber mo ri ta 4CaO•Ca(OH) 2 •6SiO 2 •4H 2 O iako je do bro po zna to da je cen tral ni deo sva kog slo ja to ber mo ri ta sli čan ok ta e dar skom slo ju u mi ne ra lu gli ne, ko ji bi se mo gao opi sa ti kao is kri vljen kal ci jumhi drok sid ni sloj li šen svih ato ma vo do ni ka (CaO 2 ).Ovaj sloj je okru žen na obe stra ne pa ra lel nim re do vi ma la na ca ti pa vo la sto ni ta, ko je su pri gnje če ne u rav ni uprav noj na rav ni slo je va [1316].
Sa mo dve trećine te tra e da ra sva kog lan ca je di rekt no po ve za no s cen tral nim CaO 2 slo jem de leći ato me ki se o ni ka.Pre o sta la tre ći na ko ja je uda lje na od CaO 2 slo ja mo že se po sma tra ti kao "pre mošćuju ći" te tra e dri.Opi sa ni kom po zit ni slo je vi 2:1 ima ju sa stav Ca 4 SiO 4 O 18 .Pre o sta li ili me đu sloj či ne ato mi kal ci ju ma i mo le ku li vo de ko ji su po sta vlje ni iz me đu ovih slo je va.Ras po red vo do ni ka je ne iz ve stan.Ve ro vat no ima vi še Si OH gru pa ne go što je pred sta vlje no for mu lom Ca 4 (SiO 4 O 18 H 2 )•Ca•4H 2 O, uz od go va ra juću za me nu vo de iz me đu slo ja hi drok sil ni ma gru pa ma [15,16,17].U skla du s tim, ka da se pri pre ma CS pa sta, C 3 S fa za ve ro vat no re a gu je ve o ma br zo i vo da po sta je iz ne na da za sićena sa Ca(OH) 2 (re ak ci ja se de ša va u ne ko li ko mi nu ta).Isto se de ša va i sa βC 2 S, ali spo ri je.Kao re zul tat hi dra ta ci je do la zi do stva ra nja Ca(OH) 2 i jed nog čla na pod gru pe to ber mo ri ta (G, to ber mo ri te 11) [15].Ako su, me đu tim, si li ka ti hi dra ti sa ni u ve li kom vi šku vo de, ta ko da je kon cen tra ci ja kre ča u ras tvo ru is pod za sićenja, CSH ko ji se for mi ra ni je član (G) se ri je to ber mo ri ta i on da ne do la zi do for mi ra nja Ca(OH) 2 .CaO/SiO 2 od nos u CSH za vi si od ko nač ne kon cen tra ci je kre ča u ras tvo ru [do nja gra ni ca od no sa (pri ni skoj kon cen tra ci ji kre ča) je oko 0,8, a gor nja gra ni ca je oko 1,5 (na ili bli zu za sićenja kre ča)] [16].
Tej lor i Ho vi son (Ho wi son) [17] su me ri li gu sti nu CSH (I) u op se gu od no sa od 0,81 do 1,50 CaO/SiO 2 , ali ni su na šli si ste mat ski trend po većanja gu sti ne s po ra stom od no sa.Oni su ta ko đe una pre di li hi po te zu za ob ja šnje nje kon stant ne gu sti ne za jed no s kon stant nom di men zi jom je di nič ne ćeli je.Uka za li su na to da kod ra stu ćeg od no sa CaO/SiO 2 kal ci jum za me nju je si li ci jum u re šet ki."Pre mošćuju ći" SiO 4 te tra e dri ko ji ni su ve za ni za cen tral ni CaO 2 sloj po ste pe no se ukla nja ju ili, tač ni je, sa mo SiO 2 se ukla nja, za to što su dva jo na ki se o ni ka u sva kom te tra e dru za jed nič ki za su sed ne gru pe.Jon kal ci ju ma ula zi u re šet ku, ali ne di rekt no na me sto SiO 2 , od no sno ne u sloj, već iz me đu slo je va.Po što jon kal ci ju ma ima dvo stru ko po zi tiv no na e lek tri sa nje a SiO 2 je ne u tra lan, dva jo na vo do ni ka mo ra ju ta ko đe bi ti uklo nje na da bi se oču va lo kon stant no na e lek tri sa nje.Ukup na za me na je SiO 2 H 2 s kal ci ju mom.Ta kva za me na će ima ti re la tiv no ma li uti caj na ve li či nu ćeli ja.To bi ta ko đe ob ja sni lo spo ru ekvi li bra ci ju iz me đu CSH (I) hi dra ta i Ca(OH) 2 ras tvo ra.Ako bi po većanje sa dr ža ja kre ča pod ra zu me va lo sa mo ap sorp ci ju kre ča iz me đu slo je va, ne bi se oče ki va la ta ko spo ra ekvi li bra ci ja.Ko nač no, hi po te za pred vi đa da će gor nja gra ni ca od no sa CaO/SiO 2 bi ti 1,75, od go va ra ju ći ukla nja nju svih "pre mošćuju ćih" te tra he da ra.Iako je gor nja gra ni ca od no sa za CSH (I) pod gru pu oko 1,5, gor nja gra ni ca za to ber mo rit (G) pod gru pu je 1,75, kao i za dru gu pod gru pu, CSH (II).Mo guće je da to ber mo rit (G) ili CSH (II) ili obe pod gru pe for mi ra ju kon ti nu i ra nu se ri ju sa CSH (I).Bru na u er i Grin berg su mo di fi ko va li Tej lorHo vi so no vu hi po te zu uz na po me nu da je SiO 2 H 2 za me njen ne sa mo kal ci ju mom, već i mo le ku li ma vo de.Ta kva za me na će do ve sti do da le ko bo lje kon stant no sti ve li či ne ćeli ja ne go zamena po Tej lorHo vi so nu, a ta ko đe bi bo lje ob ja sni la pro na đen sa stav to ber mo ri ta, što bi da lje ob ja sni lo kon stant nost ra ni je po me nu tog c raz ma ka [18].Mo di fi ko va na Tej lorHo vi so no va hi po te za ob ja šnja va do bro sa stav, struk tu ru, gu sti nu i ne ke dru ge oso bi ne hi dra ta iz CSH (I) pod gru pe, mo žda čak ce le to ber mo rit ne gru pe.Po red struk tur nih va ri ja ci ja, CSH (I) hi dra ti po ka zu ju ši rok spek tar spe ci fič nih po vr ši na.Ta kve raz li ke u struk tu ri i po vr ši ni su, ne sum nji vo, od go vor ne za raz li ke u od no su iz me đu sa sta va raz li či tih vr sta CSH for mi ra nih to kom pro ce sa hi dra ta ci je [16,17,18].
U su prot nom, do bro je po zna to da su to ber mo rit ni (G) hi dra ti lo še kri sta li zi ra ni.Di frak ci jom ra di o lo ških zra ka do bi ja ju se tri li ni je: jed na s ve o ma ja kom i ši ro kom re flek si jom, sa mak si mu mom na 0,305 nm, i dve mno go sla bi je ši ro ke li ni je na 0,279 i 0,182 nm.He ler (Hel ler) i Tej lor [19] su ot kri li da ovi raz ma ci ima ju in dek se u rav ni ma (110), (200) i (020).Oni su ta ko đe usta no vi li ra sto ja nja od 0,307, 0,281 i 0,183 nm za sla bi je osu še ne uzor ke CSH (I) pri pre mlje ne hi dro ter mal nom me to dom.Tri li ni je to ber mo rit nih (G) hi dra ta su naj ja če li ni je svih to ber mo ri ta.Sve tri su (hk0) raz ma ci, od no sno raz ma ci unu tar slo je va.Raz mak (020) pred sta vlja po lo vi nu ra sto ja nja iz me đu su sed nih Si ato ma u lan ci ma SiO 4 te tra e da ra u 2:1 slo je vi ma.Dvo stru ko ovo ra sto ja nje, 0,364 nm, je ste du ži na or to rom bič ne pse u doćeli je.Dvo stru ki raz mak (200), 0,559 nm, je ste du ži na pse u doćeli je.Ipak, to ber mo rit (G) ne po ka zu je osnov ni raz mak, pa se c du ži na mo že do bi ti sa mo po moću ne znat no iz me nje nog pri stu pa [19,20].
Pre ma ta kvom pri stu pu, mo že se za klju či ti da je u sa sta vu to ber mo rit nih (G) hi dra ta Ca 3 Si 2 O 7 • H 2 O sa mo je dan mo le kul vo de iz me đu slo je va.Po sle su še nja, mo le kul sa dr ži sa mo hi drok sil nu vo du ve za nu za SiO 4 te tra e dre u slo je vi ma, i ova vo da je mno go čvr šće ve za na ne go vo da iz me đu slo ja.S dru ge stra ne, to ber mo rit je ve o ma če sto po de ljen u c prav cu.Ova di men zi ja kri sta li ta je ko lo id na.Di men zi ja če sti ca u b prav cu, tj.duž si li kat nih la na ca, pri lič no je ve li ka, re da mi kro me tra ili vi še mi kro me ta ra.Sto ga se či ni da je to ber mo rit ko loid u c prav cu, ali ne i u b prav cu.Za ni mlji vo je da se me đu sloj vo de stal no gu bi, ka ko se vo da obič no gu bi kod ko lo i da, ali sa dr žaj vo de si li kat nih la na ca osta je sko ro kon stan tan ne za vi sno od uslo va ve zi va nja [16,21].
Pri a nja nje to ber mo rit nih če sti ca me đu sob no kao i sa dru gim je od go vor no za sna žnu ve zu u stvrd nu toj pa sti tri kal ci jum si li ka ta i di kal ci jumsi li ka ta.To je ta ko đe naj va žni ji fak tor sna ge stvrd nu tih Por tland ce men ta.Ot por nost na pri ti sak Por tland ce men ta pre va zi la ze nji ho ve za te zne čvr stoće za je dan red ve li či ne, a čvr stoća očvr slog CS je ot pri li ke ista kao i kod Por tland ce men ta.Iako ne po sto je pu bli ko va ne za te zne čvr stoće CS pa sti, pret po sta vlja se da su ve ro vat no ni že ne go kod Por tland ce men ta, jer CS pa ste se la ko ras pr šu ju po mo ću ul tra zvuč nih vi bra ci ja za raz li ku od Por tland ce men ta [22,23].

ZAKLJUČAK
Pred no sti sin te ze ak tiv nih si li kat nih fa za kom bi no va nom sol gel i me to dom sa mo pro pa gi ra ju ćih ta la sa na vi so koj tem pe ra tu ri se pr vi put pri me nju ju u ovom ra du.Do bi je ne su ve o ma ak tiv ne na no struk tur ne si li kat ne fa ze.One su po sti gle po bolj ša no ve zi va nje ak tiv ne si li kat ne fa ze, kao naj va žni je fa ze u sva koj sme si en do dont skog ma te ri ja la.
Pro ces hi dra ta ci je CS fa za je pa žlji vo ana li zi ran po mo ću XRD i FTIR, sa aspek ta struk tur nih pro me na unu tar sme se to kom kva še nja od jed nog, tri, se dam i 28 da na.Po sle tri da na βC 2 S i C 3 S fa ze su uglav nom pre tvo re ne u to ber mo rit, a na kon se dam da na su pot pu no bi le tran sfor mi sa ne.Pred lo žen je me ha ni zam hi dra ta ci je i uka za no na te škoće u utvr đi va nju tač ne re ak ci je, kao i na pro blem utvr đi va nja tač ne struk tu re to ber mo ri ta.

ZAHVALNICA
Ovu stu di ju je fi nan sij ski po dr ža lo Mi ni star stvo za obra zo va nje, na u ku i teh no lo ški raz voj Re pu bli ke Sr bi je (Pro je kat br.172026).