- Email: [email protected]
Stability and decomposition mechanism of ettringite in presence of ammonium sulfate solution
Construction and Building Materials 124 (2016) 786–793
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Stability and decomposition mechanism of ettringite in presence of ammonium sulfate solution q Xuebing Wang a, Zhihua Pan a,⇑, Xiaodong Shen a, Weiqing Liu b a b
College of Materials Science and Engineering, Nanjing Tech University, Nanjing 210009, China College of Civil Engineering, Nanjing Tech University, Nanjing 210009, China
h i g h l i g h t s Decomposition of ettringite can be occurred when the concentration of (NH4)2SO4 reached to a certain extent. Main products of decompostion of ettringite were gypsum and Al(OH)3.
a r t i c l e
i n f o
Article history: Received 1 February 2016 Received in revised form 20 July 2016 Accepted 28 July 2016
Keywords: Ettringite (NH4)2SO4 Stability Mechanism
a b s t r a c t The formation and transformation characteristics of ettringite play an extremely important effect on the accelerated hardening and expansion of concrete as well as its long term durability. In alkaline condition, ettringite can be existed stably; but when the alkalinity exceeds a certain value, ettringite can be decomposed into calcium monosulfoaluminate (AFm). In this paper, the stability and its decomposition mechanism of ettringite in alkaline (NH4)2SO4 solution were then investigated. The results show that in alkaline environment, decomposition of ettringite can be occurred and main produces were gypsum and Al(OH)3 as long as the concentration of (NH4)2SO4 reached to a certain extent. Based on experiment data, a kind of NH+4 involved decomposition mechanism of ettringite had been proposed by method of quantum mechanics combined with frontier molecular orbital theory and thermodynamic theory. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Ettringite always exists in cement based materials, it begins to form during cement hydration process [1]. Ettringite is also applied to the shrinkage compensation and hardening accelerating and set accelerating of concrete [2]. However, formation of excess ettringite has negative effect on durability of cement based materials. Moreover, the negative effect in front of sulfate attack can become still much more negative because it is more aggressive and faster if the Portland cement is replaced partially by metakaolin due to its large reactive alumina content [3–6]. Therefore, to clarify the stability and transformation of ettringites in a variety of environment conditions plays an important guiding significance for
q Funded by China national 973 program (No. 2011CB013800); Program of Research Innovation for University Graduate Students of Jiangsu Province (KYLX150777); Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). ⇑ Corresponding authors. E-mail addresses: [email protected] (X. Wang), [email protected] (Z. Pan).
http://dx.doi.org/10.1016/j.conbuildmat.2016.07.135 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
understanding mechanical properties and sulfate attack of cement based materials. Ettringite is not only originated from the hydration of cement, but also from sulfate attack of cement based materials [7]. This originated ettringite may face many hostile environments and its stability is maybe changed. Factors affecting the stability of ettringite mainly include temperature, pH and others. J. Skalny [8] found that ettringite was instable and decomposed when temperature was exceeded 60–70 °C. If temperature was reached 80 °C, ettringite was converted into metaettringite by losing part of water [9]; K. Ogawa [10] claimed that ettringite was decomposed at 130–150 °C yielding monosulfate hydrate as the main decomposition product, hemihydrate and anhydrite were also formed. At 260 °C ettringite was completely dehydrated which was investigated by Sytle M. Antao [11]. Grusczscinski [12] and R.A. Livingston [13] found that morphology of ettringite was also changed at elevated temperature. But different surrounding would also affect the critical temperature of decomposition of ettringite. H.F.W. Taylor [14] considered supersaturated OH- and SO2 4 in pore solution can increase the stability of ettringite which still was existed even at 90 °C; stability of ettringite also was changed in the solutions
787
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
with different pH. J. Havlica et al. [15] claimed that ettringite was decomposed at pH below 10.7; J Hill et al. [16] and Z Chang [17] studied combined sulfate and acid attack of concrete with different SCM and found that content of ettringite was less. However, ettringite was also not stable even in condition of high alkaline. C.J. Warren et al. [18] found ettringite can be converted into monosulfate with pH above 13. (NH4)2SO4 existing in sewage facilities and industrial areas [19] always reduce the service performance of cement based materials. Anti-freezing admixtures such as urea and melamine superplasticizer sometimes can also generate ammonia and NH+4 ions. Therefore, cement based materials in service time may be threatened by ammonium. NH+4 can react with OH- so that it can reduce the pH of system, therefore, many reference always classify ammoniate salt as acid attack on concrete and the main purpose of NH+4 is to decrease the pH [20–22] and following induce the instability of many hydration produces such as C-S-H and ettringite. Raoul Jauberthie [23] had investigated that corrosion product was gypsum when cement mortar were immersed in 0.25 mol/l ammonium solutions for a period of 2 years, and pH of surrounding solution was changed from 6 to 8 by leaching of alkali. Michel Mbessa [24] found gypsum was existed on the surface of highstrength concrete when concrete was immersed in 20% ammonium solutions. M.T. Bassuoni [19] considered that ammonium sulfate caused a combined acid and sulfate attack characterized by decomposition/softening associated with expansion, cracking and spelling. U. Schneider et al. [25] claimed that disadvantage of ammonia in cement concrete mainly due to the change of alkaline environment of the presence of ettringite. However, due to unshared pair of electrons, NH3 has strong coordination ability and may react with ettringite especially in high pH value. When the pH > 9.25 NH3(aq) was the dominant species in the NH+4 solution [23], so NH3 may be involved into decomposition of ettringite. In this paper, (i) In terms of the pH distribution over depth, stability of ettringite was investigated to achieve the critical NH+4 concentration of decomposition of ettringite; (ii) To investigate the impact of NH+4 and avoid the impact of acid, products changes of decomposition of ettringite in high alkaline was studied; (iii) Decomposition mechanism of ettringite was thus proposed.
Table 2 Oxide compositions of cement/wt.%. CaO
SiO2
Al2O3
Fe2O3
MgO
SO3
K2O
TiO2
LOI
Total
64.68
18.36
4.33
2.95
0.79
2.60
0.73
0.19
3.70
98.33
distilled water; two solutions were mixed and stirred 2 h, stood after one week and pumping filtered, synthesized ettringite was then dried at 60 °C and stored in a closed container.
2.2.2. Mortar specimen preparation Cement mortars were prepared according to ASTM C305-99 (Practice for Mechanical Mixing of Hydraulic Cement Pastes and Mortars of Plastic Consistency), the sand/binder mass ratio was maintained equal to 3, water/cement ratio being 0.5, and mortar specimen size being in 40 mm 40 mm 160 mm. Mortar specimens were then cured under the condition of ambient temperature 20 ± 2 °C and RH not less than 90% in molds for 24 h, then demoulded and cured in water at 20 ± 1 °C for 28 days.
2.2.3. Sulfate corrosion conditions Mortar specimens curing after 28 days were immersed in 5% Na2SO4 (0.35 mol/L) and 5% (NH4)2SO4 (0.38 mol/L) respectively. After 180 days mortars were removed from different sulfate solutions and then the surface water was dried by cloth, and following dried for 24 h at 60 °C. Mortar specimens were then layered sampled every 0.5 mm from exposed surface by precision lathe, powder samples cutting down were collected respectively, then immersed in ethyl alcohol for 4 h and vacuum dried 24 h at 60 °C and stored in a closed container until analysis. Ettringite was accurately weighed and added in 100 ml 5% (0.380 mol/L) (NH4)2SO4 solution, and different dosages of NaOH solution (pH = 13) according to the test program was dropped into this solution to adjust the pH. After reaching of equilibrium state of system, the correlatively analysis was started.
2. Experimental 2.1. Raw materials (1) Cement: Commercial P.II 52.5 Portland cement, its physical properties and chemical compositions were listed in Tables 1 and 2. (2) Sand: China ISO standard sand. (3) Chemical reagents: Analytical reagent CaO, Al2(SO4)318H2O, CaSO4, (NH4)2SO4, Na2SO4. 2.2. Experimental process 2.2.1. Synthesis of ettringite Ettringite was synthesized by the method of Struble et al. [26]. 13.40 g CaO was weighed and dissolved in 890 ml distilled water; 26.55 g Al2(SO4)318H2O was also weighed and dissolved in 40 ml
2.2.4. Methods of physical-chemical properties X-ray diffraction analysis was carried out on ARL X’ TRA X-ray diffract-meter. Experimental parameters: 2.2 kW Cu target; stepping being 0.02°, scanning speed being 10°/min; TG-DSC analysis was carried out on STA449C thermal gravimetric instrument equipped with DSC 204 differential thermal analyser. Experiment parameters: temperature ranged from 40 °C to 900 °C; weight sensitivity being 0.1 lg; heating rate being 10 °C/min; nitrogen atmosphere. SEM-EDS was carried on JSM-5900 scanning electron microscope equipped with NORAN VANTAGE DSI energy dispersive spectrometer. Experiment parameters: accelerating voltage being 15 kV; resolution being 3 nm; magnification being 18 300,000; FTIR was carried out on Nexus 670 Fourier transform infrared spectrometer. Experiment parameters: resolution being 0.1 cm1; signal to noise being 33,400:1(peak to peak in one minute).
Table 1 Physical properties of cement. Specific surface area/m2kg1
Standard consistency water consumption/%
397
27.6
Setting time/min
Flexural strength/MPa
Compressive strength/MPa
Initial
Final
3d
28d
3d
28d
139
187
5.9
8.6
33.1
57.2
788
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
3. Results and discussion 3.1. Stability of ettringite in (NH4)2SO4 solution with different pH value Service environment of cement based material is generally complex. Sometimes with existing of (NH4)2SO4 solution, performance of cement based material may be affected. Stability of ettringite in (NH4)2SO4 solution with different pH value was then studied. Cement mortars that was immersed in 5% Na2SO4 (0.35 mol/L) and 5% (NH4)2SO4 (0.38 mol/L) after 180 days respectively were sampled at different depths by lathe, XRD was following carried out to determine the changes of products [27]. One of characteristic peak’s areas of material was then selected to represent its content, such as diffraction peak of Ca(OH)2 was at 18.04°, ettringite at 9.06° and gypsum at 11.66°, and the result was shown in Fig. 1. As was shown in Fig. 1, the reaction product was mainly ettringite when cement mortar was immersed in Na2SO4 solution; but it was gypsum in (NH4)2SO4 solution; not only that, portlandites in two different solutions were all reduced by the depth, 0.0 10000
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Gypsum/Na2SO4 Gypsum/(NH4)2SO4
8000
Area
6000 4000 2000 1500
Ettringite/Na2SO4 Ettringite/(NH4)2SO4
1200
Area
900 600 300 25000
Area
2000 1500 1000 500 0 0.0
Portlandite/ Na2SO4 Portlandite/ (NH4)2SO4 0.5
1.0
1.5
2.0
2.5
3.0
3.5
Depth/mm Fig. 1. Reaction products distribution in cement mortar over depth.
4.0
and they were still existed when depth of specimen was deeper than 1.5 mm. Existing of crystalline-state portlandite showed that pH of pore solution of mortar deeper than 1.5 mm was still about 12.4 [28]. Because corrosion products always depend on the concentration of sulfate content [29] and pH value [16], ettringite all can be formed firstly in both specimens of which was immersed in different sulfate solutions until sulfate concentration was exceeded the lowest concentration of precipitation of gypsum or until the pH of pore solution was lower than 10.7 [15]. However, in (NH4)2SO4 solution ettringite content of mortar was less than that of in Na2SO4 when depth of specimen were deeper than 1.5 mm and most of gypsum was formed in these region. This illustrates that ettringite was instable when the mortar was immersed in (NH4)2SO4 solution even in the condition of alkaline. Mortars were removed from (NH4)2SO4 solution after 180 days and dried at 60 °C, SEM-EDS was following carried out to observe the morphology of corrosion products at surface layer and the result was shown in Fig. 2. Fig. 2 showed that bar-like reaction products were generated at surface area and even some filled in the cracks of mortar. From EDS of point A, this bar-like corrosion product was gypsum [30]. It was formed and accumulated at corrosion areas. Sizes of gypsum were different, all of which the largest was reached to 5 lm and smallest was less than 0.5 lm. Due to the difficult determination of concentration of NH+4 which penetrated into specimen, following experiments were designed to simulate stability of ettringite and determine the critical concentration of NH+4 in different concentrations of (NH4)2SO4 solutions. 5% (NH4)2SO4 (0.380 mol/L) and NaOH solution (pH = 13) were prepared and the mixed-doses was in accordance to Table 3, then 1.5 g Struble et al.’s ettringite was added in each solution, and pH values of solution were measured periodically to ensure the chemical equilibrium of reaction. The results were then shown in Table 3. As was shown in Table 3, when NaOH solution was added into (NH4)2SO4 solution, pH values of solution tended to equilibrium after one week. Solutes were then pumping filtrated and dried at 60 °C after one week and XRD and FTIR was carried out, the results were shown in Fig. 4. It was showed that only the peaks of gypsum appeared in No.1-3 of Fig. 3 and the peaks of gypsum and ettringite all appeared in No.4, while only the peaks of ettringite appeared in No.5 and 6. It suggests that ettringite had been decomposed in No.1-3 of experiment, but it was still existed stably in No.5-6, and was partially existed in No.4. The results showed that critical concentration of (NH4)2SO4 solution which affected stability of ettringite was between 0.042 mol/L and 0.126 mol/L. Ettringite behaved stable when (NH4)2SO4 solution concentration was less than 0.042 mol/L, however, ettringite behaved instable and was decomposed when concentration was more than 0.126 mol/L.
+A
Fig. 2. SEM-EDS of cement mortar immersion in (NH4)2SO4 solution at 180 day.
789
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793 Table 3 pH values of solutions of different (NH4)2SO4 adding ettringite at different immersion time. No.
(NH4)2SO4/NaOH solution volume ratio
(NH4)2SO4 actual concentration (mol/L)
Immersion time 0h
24 h
48 h
72 h
96 h
120 h
1 2 3 4 5 6
1.5 1 0.5 0.25 0.125 0
0.227 0.189 0.126 0.076 0.042 0
8.78 8.90 9.20 9.81 11.23 13.05
9.15 9.22 9.54 9.94 10.20 12.85
9.16 9.37 9.81 10.16 10.22 12.74
9.01 9.20 9.63 10.02 10.06 12.76
9.06 9.25 9.73 10.06 10.01 12.76
8.97 9.15 9.66 10.02 9.88 12.71
:Ettringite :Gypsum
1 2
2 3 4
Transmission/%
3 1
4
5 6
5 6 5
15
25
35
45
55
65
2θ/°
4000
3400
2800
2200
1600
Wavenumber/cm
(a) XRD
1000
400
-1
(b) FTIR
Fig. 3. XRD and FTIR of reaction product affected on different concentration (NH4)2SO4.
Three types of vibration bands were relevant to the stability of ettringite. With increasing of concentration of (NH4)2SO4 solution feature bands of raw material (ettringite) started weaker, on the contrast, feature bands of gypsum gradually became stronger. No.4 in Fig. 3(b) showed the bands at 3200 cm1–3700 cm1 were attributed to O–H absorption band of ettringite as well as gypsum. This results also showed that critical concentration of (NH4)2SO4 which affected stability of ettringite was between 0.042 mol/L–0.126 mol/L when pH of solution was below to 13.
:Ettringite :Gypsum
Product
3.2. Products analysis of decomposition of ettringite
Raw material 5
15
25
35
45
55
In order to investigate the products by the influence of NH+4 on ettringite, stability of ettringite in (NH4)2SO4 solution with high pH value was studied. Because of cement hydration, the pH of cement paste is about 12.5–13.5 which is equivalent as the saturated
65
2θ/° Fig. 4. XRD of raw material and product.
There were three types of vibration in the FTIR from Fig. 3(b) [31]: (1) Vibration of O–H: Stretching vibration band of free OH in ettringite was at 3633 cm1, and stretching vibration and bending vibration bands of lattice water of gypsum and Al (OH)3 were at 3430 cm1 and 1640 cm1 respectively; (2) Vibration of S-O: S-O vibration bands of ettringite were mainly at 1120 cm1, 620 cm1 and 420 cm1 and asymmetric stretching vibration bands of gypsum were at 1118 cm1-1142 cm1. (3) Vibration of Al-O: Vibration bands of Al-O in ettringite appeared at 550 cm1 and 870 cm1 [32].
+A
Fig. 5. SEM-EDS of synthesized ettringite.
790
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
+C +B (a) Overall figure
(b) Partial figure Fig. 6. SEM-EDS of product.
solution of Ca(OH)2 [33]. For the sake of the products changes of decomposition of ettringite in cement based materials under the condition of NH+4 and high alkaline, experiments were carried out at pH = 13. Struble et al.’s ettringite was accurately weighed and added in 100 ml 5% (0.380 mol/L) (NH4)2SO4 solution, and dropped NaOH solution to regulate pH to 13. After one week, solute was pumping filtered and then dried at 60 °C, reaction products were following detected by XRD analysis and the result was shown in Fig. 4. It can be seen from XRD pattern (Fig. 4), characteristic peaks of ettringite in raw material were disappeared but the peaks of gypsum were appeared in 5% (0.380 mol/L) (NH4)2SO4 solution after one week. The result suggested that ettringite in (NH4)2SO4 solution was instable and decomposed; what more, one of reaction product was gypsum. To observe the morphologies of raw material of ettringite and decomposition products, SEM-EDS were carried out and the results were shown in Figs. 5 and 6. It is observed that ettringite in short bar outline was formed by the method of Struble et al. [26] in raw material, and the size of those ettringite was all less than 5 lm. The reaction products in Fig. 6 were mainly made up of columnar-like and gel-like compounds, EDS data at point B showed that columnar crystal was gypsum, all of which crystal sizes differed greatly and ranged from 5 lm to 100 lm. compound around gypsum was colloid, Al content of which the composition from EDS data was higher and Ca/S was close to 1:1, based on XRD analysis, only peaks of gypsum was appeared and other compounds didn’t detected in the products. Therefore colloids which contained Al element was Al(OH)3 gel.
100
1
Raw material Products 16.8%
90
Ettringite þ 4OH ! 2SO2 4 þ hydroxy Ettringite
0 24.2%
4.5%
TG/%
80
33.4%
-2
2.7%
70 46.9%
-3
4.7%
60
1.5%
50 0
100
200
300
400
500
600
700
Temperature / Fig. 7. TG-DSC of ettringite and products.
800
! Monosulfate þ 2Ca2þ DSC/mW/mg
-1
Raw ettringite and reaction products were dried at 60 °C and TG-DSC were carried out. The result was shown in Fig. 7. As was shown in Fig. 7, ettringite had two endothermic peaks at 120 °C and 253 °C, the former peak was attributed to the dehydration of ettringite into calcium monosulfoaluminate (AFm) at 120 °C and weight loss was 33.42%; AFm was then dehydrated and decomposed at 253 °C, weight loss was 4.7%. Dehydration of reaction products at 100 °C was attributed to the loss of absorption water of Al(OH)3 colloids; bound water of Al(OH)3 was all lost at 257 °C [34]and finally converted into Al2O3. gypsum was decomposed to hemihydrate gypsum at 150 °C; the exothermic peak at 372 °C was attributed to the change from soluble a type and b type hemihydrate gypsum to insoluble type; the reason of weight loss at 700 °C was the decomposition of CaCO3 which was from the absorption of CO2 in the reaction process. Before 900 °C, free and bound water in ettringite, Al(OH)3 and gypsum were all lost. Quality of all type water of ettringite was account for 45.9% [35]; Al(OH)3 was dehydrated and transformed to Al2O3 and the quality was account for 34.6% [36], gypsum was dehydrated and transformed to CaSO4 and the weight loss was 20.9%. since reaction products Al(OH)3 and gypsum were decomposed by ettringite of which Al:Ca ratio was 1:3, therefore the molar ratio of Al(OH)3 and gypsum was nearly 1:3 which on this basis the mass of water loss of products was 23.7%. As can be seen from Fig. 4 before 900 °C, weight loss of ettringite was 46.9% and reaction products were 24.2% which was close to the theoretical calculation. Therefore the main solid products were gypsum and Al(OH)3 when ettringite was immersed in combined alkaline and (NH4)2SO4 solution. However, the result was different with the Ref. [18]. Ref. [18] showed that ettringite was converted into calcium monosulfoaluminate at this pH value and the reaction can be expressed as:
ð1Þ
In this state, high pH induce the ettringite converted into calcium monosulfoaluminate. But in (NH4)2SO4 solutions, the main products were gypsum and Al(OH)3 but not monosulfate. So a kind of NH+4 involved decomposition mechanism of ettringite was proposed.
-4
3.3. Decomposition mechanism of ettringite in the (NH4)2SO4 solution -5 900
Structure of ettringite was determined by Toylor [37] at first time and he had found that its crystal structure was hexagonal prism which the structure unit was (Ca3(Al(OH)6)12H2O)3+, this structure unit was composed of column and channel paralleling
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
Ca-H2O LUMO
1.02eV
0.86eV
NH3 LUMO
Ca-H2O HOMO
NH3 HOMO
(a)
(b)
Fig. 8. Comparing study of HOMO and LUMO of Ca–H2O and NH3.
0.18eV
Ca-H2O-NH3 LUMO
Ca-H2O-NH3 HOMO Fig. 9. Comparing study of HOMO and LUMO of Ca–H2O–NH3.
the C axis of column. Among them, column was assembled by Al 2(OH)36 octahedral and three Ca polyhedron (Ca4H2O), SO4 existing in channel was to balance the charge and ensure the electron neutrality of molecule. Ca4H2O was connected by four Ca–O conjugate hybrid orbitals of which due to sp3 hybridization of Ca. As the second main group element, coordination ability of Ca was poor; Ca–O bonding force mainly was from Coulomb force. NH3 had a strong electronegativity which was due to two unshared electron pairs and can form hydrogen bond with water in Ca–H2O, however,
(a) Caa-H2O
791
[Ca–H2O–NH3]2+ system was not stable and decomposed in the water environment. In order to investigate the decomposition tendency of system, Method of quantum mechanics was used to calculate the charge distribution and bond length of Ca–H2O and Ca–H2O–NH3 respectively and following to propose the trend of this reaction process. Calculation method was based on density function theory methods and hybrid function was B3LYP. In this paper, frontier molecular orbital (FMO) theory which proposed by Kenichi Fukui [38] was used to investigate the reaction possibility between Ca–H2O and NH3 and the stability of [Ca–H2O–NH3]2+. FMO theory simplifies reactivity to interactions between the highest occupied molecular orbital (HOMO) of one species and the lowest unoccupied molecular orbital (LUMO) of the other in the chemical reaction. HOMO and LUMO of Ca–H2O and NH3 and [Ca–H2O–NH3]2+ respectively were shown in Figs. 8 and 9. According to FMO theory, when two molecular were closed to each other, only suitable symmetry and smaller difference of energy level and reasonable of directionality of electron shifted between HOMO of one molecule and LUMO of the other can the reaction happened. In Fig. 8, only in the path of Fig. 8(b) symmetries of HOMO of NH3 and LUMO of Ca–H2O were suitable; difference of energy level between two molecules was 0.863 eV which was less than 6 eV; the directionality of electron shifted from HOMO of NH3 to LUMO of Ca–H2O was also reasonable, so reaction can only happened between HOMO of NH3 and LUMO of Ca–H2O. As was shown in Fig. 9, according to energy levels difference between HOMO and LUMO of [Ca–H2O–NH3]2+, two energy levels were so close that it was easy of electron to transfer from HOMO to LUMO and further known that this structure was not stable and easy to decompose. To investigate the product of [Ca–H2O–NH3]2+, charge distribution and bond length of Ca–H2O and [Ca–H2O–NH3]2+ were studied in Fig. 10. In the case of existing of NH3, H atom’s position of coordinated water in Ca–H2O became asymmetric (Fig. 10), one of hydrogen atom was attracted by a nitrogen atom and their positive charge quantity became less, the corresponding O–H bond was elongated and bond length was increased 59.2%, and thus the distance of N atom in NH3 and H in CaH2O approximated the original bond lengths of N–H in NH3, namely four H atoms around N atom were close to equivalent and had tendency to form NH+4; on the contrary, Ca–O bond became short when NH3 was existed, namely Ca–O bond energy become larger. It was considered that [Ca–H2O–NH3]2+ system decomposed into [Ca–OH]+ and [NH4]+, i.e. in NH3
(b) Caa-H2O-NH3
Fig. 10. Charge distribution and bond length of Ca–H2O and Ca–H2O–NH3.
792
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
environment Ca-H2O polyhedron behaved instable and dissociated. Chemical reaction can be thus expressed as:
NHþ4 þ OH NH3 þ H2 O
ð2Þ
½Ca H2 O2þ þ NH3 ! ½Ca OHþ þ NHþ4
ð3Þ
½Ca OHþ ! Ca2þ þ OH
ð4Þ
NH3 played as a catalytic role in the reaction process. When Ca4H2O structure was dissociated, the remaining part Al(OH)3 6 was then converted into Al(OH)3 in alkaline environment; When the pH > 9.25 NH3(aq) was the dominant species in the solution [23], therefore, the entire chemical reaction of ettringite decomposition process can be expressed as:
NHþ4 þ OH NH3 ðaqÞ þ H2 O;
K1 ¼
½NH3 ½NHþ4 ½OH
ð5Þ
3CaO Al2 O3 3CaSO4 32H2 O þ 6NH3 ðaqÞ þ 3SO2 4 ! 6CaSO4 2H2 O þ 2AlðOHÞ3 þ 6NHþ4 þ 12OH þ 8H2 O K2 ¼
½OH 12 ½NHþ4
ð6Þ
6
ð7Þ
3
6 ½SO2 4 ½NH3
4. Conclusions
H
Dr G2 ¼ RT ln K 2 þ RT ln K 2 KH 2 ¼ exp
DG H 2 RT
!
ð8Þ P
¼ exp
m
H
B B Gm2 ðBÞ
! ð9Þ
RT
where K1 and K2 were equilibrium constant of reaction (5) and reaction (6) respectively; Kh2 was the standard state equilibrium constant of reaction (6);vB was the stoichiometric number of each reactants and products of reaction (6); Gm2h was the standard state free energy of formation for each reactants and products; R was standard state mole gas constant; T was Kelvin temperature. Chemical Eq. (5) was a reversible reaction so standard state equilibrium constant of reaction (5) was equal to K1. Only if DrG2 < 0, the forward reaction of decomposition of ettringite (6) was spontaneous [39]. The thermodynamics data of common ions were from Ref. [30]; But the standard state free energy of formation for ettringite was not accurate from many authors; it was estimated as 15,211 ± 4.2 kJ/mol by Robert B Perkins et al. [40] and 15204.7 ± 23 kJ/mol by Satish C.B. Myneni et al. [41], in this paper, the standard state free energy of formation for ettringite was estimated as 15204.7 ± 23 kJ/mol and the maximum and minimum
+
Concentration of NH4/mol/L
1.2 1.0 0.8 0.6 1 0.4
2 3
0.2
4 5
0.0 8
values were selected to calculate the critical concentration of OH and NH+4 ions. According to these basic data, concentration of OH and NH+4 ions were achieved when DrG2 was equal to 0 and the result was shown in Fig. 11. As was shown in Fig. 11, the critical concentration of the ammonium ion was increased with the pH of solution. Ettringite was not stable and decomposed at the concentration of NH+4 ions above the shadow part in the Fig. 11. This reaction may occur at the concentration of NH+4 ions in the shadow part which was caused by the difference of thermodynamic data of ettringite that from the different authors. It can be also found that concentration of ammonium ions got huge impact on the stability of ettringite when in the strong alkaline environment (pH > 11). Impact of NH+4 through the reaction (6) became less and ettringite can be decomposed even in low concentration of NH+4 in the weak alkaline environment (pH < 11). The points in the Fig. 11 were the concentration data of experiment in Table 3. As were seen in the Fig. 11, the concentration at point 1–3 was above the critical concentration of NH+4 and decomposition of ettringite can be occurred. When concentration of NH+4 ions was exceeded to 0.126 mol/L (point 1–3 in the Fig. 11) ettringite can be decomposed, which was same as the result in Fig. 3.
9
10
11
12
13
Initial pH Fig. 11. the critical initial concentration of NH+4 ions.
14
Under the condition of alkaline, when the concentration of NH+4 ion reaches a critical concentration, decomposition of ettringite can be occurred and the products are gypsum and Al(OH)3. The critical NH+4 concentration of decomposition of ettringite is increased with the pH of solution. Decomposition mechanism of ettringite has been proposed by method of quantum mechanics and thermodynamic theory. NH+4 was involved in the decomposition of ettringite in the (NH4)2SO4 solution and it not only can reduce the pH in the solution, but also the coordination ability of NH3 can lead to decomposition of ettringite especially in high pH solution.
References [1] M. Santhanam, M.D. Cohen, J. Olek, Mechanism of sulfate attack: a fresh look Part 1: summary of experimental results, Cem. Concr. Res. 32 (6) (2002) 915– 921. [2] E. Sakai, Y. Nikaido, T. Itoh, M. Daimon, Ettringite formation and microstructure of rapid hardening cement, Cem. Concr. Res. 34 (9) (2004) 1669–1673. [3] R. Talero, Expansive synergic effect of ettringite from pozzolan and from OPC, co-precipitating in a common plaster-bearing solution. Part I: by cement pastes and mortars, Constr. Build. Mater. 24 (9) (2010) 1779–1789. [4] R. Talero, Kinetic and morphological differentiation of ettringites in plain and blended Portland cements with metakaolin and the ASTM C 452–68 test. Part II: morphological differentiation by SEM and XRD analysis, Mater. Construct. 59 (293) (2009) 35–51. [5] R. Talero, Expansive synergic effect of ettringite from pozzolan and from OPC, co-precipitating in a common plaster-bearing solution. Part II: II: fundamentals, explanation and justification, Constr. Build. Mater. 25 (3) (2011) 1139–1158. [6] R. Talero, Performance of metakaolin and Portland cements in ettringite formation as determined by ASTM C 452–68: kinetic and morphological differences, Cem. Concr. Res. 35 (7) (2005) 1269–1284. [7] J. Peng, Z. Lou, Study on the mechanism of ettringite formation, J. Chin. Ceram. Soc. 28 (6) (2000) 511–515. [8] J. Skalny, V. Johansen, N. Thaulow, DEF: as a form of sulfate attack, Mater. Construct. 46 (2) (1996) 5–29. [9] Q. Zhou, E.E. Lachowski, F.P. Glasser, Metaettringite, a decomposition product of ettringite, Cem. Concr. Res. 34 (4) (2004) 703–710. hydration ettringite formation, and its expansion [10] K. Ogawa, D. Roy, C4A3S mechanism: I. Expansion; Ettringite stability, Cem. Concr. Res. 11 (S5–6) (1981) 741–750. [11] S.M. Antao, DTA, TG and XRD studies of sturmanite ettringite, Can. Mineral. 40 (5) (2002) 1403–1409. [12] E. Grusczscinski, P.W. Brown, J.V. Bothe, The formation of ettringite at elevated temperature, Cem. Concr. Res. 23 (4) (1993) 981–987.
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793 [13] S.K. Deb, M.H. Manghnani, K. Ross, et al., Raman scattering and X-ray diffraction study of the thermal decomposition of an ettringite-group crystal, Phys. Chem. Miner. 30 (1) (2003) 31–38. [14] H.F.W. Taylor, C. Famy, K.L. Scrivener, Delayed ettringite formation, Cem. Concr. Res. 31 (2001) 683–693. [15] J. Havlica, S. Sahu, Mechanism of ettringite and monosulphate formation, Cem. Concr. Res. 22 (4) (1992) 671–677. [16] J. Hill, E.A. Byars, J.H. Sharp, C.J. Lynsdale, et al., An experimental study of combined acid and sulfate attack of concrete, Cem. Concr. Compos. 25 (8) (2003) 997–1003. [17] Z. Chang, X. Song, R. Munn, et al., Using limestone aggregates and different cements for enhancing resistance of concrete to sulphuric acid attack, Cem. Concr. Res. 35 (8) (2005) 1486–1494. [18] C.J. Warren, E.J. Tteardon, The solubility of ettringite at 25 °C, Cem. Concr. Res. 24 (8) (1994) 1515–1524. [19] M.T. Bassuoni, M.L. Nehdi, Resistance of self-consolidating concrete to ammonium sulphate attack, Mater. Struct. 45 (7) (2012) 977–994. [20] A. Neville, Properties of Concrete, Longman, New York, 1997. [21] F.M. Lea, The Chemistry of Cement and Concrete, 3rd ed., Chemical Publishing Co., Inc, New York, 1971. [22] H.F.W. Taylor, Cement Chemistry, 2nd ed., Thomas Telford, New York, 1997. [23] R. Jauberthie, F. Rendell, Physicochemical study of the alteration surface of concrete exposed to ammonium salts, Cem. Concr. Res. 33 (1) (2003) 85–91. [24] M. Mbessa, J. Péra, Durability of high-strength concrete in ammonium sulfate solution, Cem. Concr. Res. 31 (8) (2001) 1227–1231. [25] U. Schneider, S.W. Chen, Modeling and empirical formulas for chemical corrosion and stress corrosion of cementitious materials, Mater. Struct. 31 (10) (1998) 662–668. [26] L.J. Struble, Synthesis and characterization of ettringite and related phases, 8th International Congress on the Chemistry of Cement, 1986. Rio de Janeiro. [27] Y. Fu, P. Gu, P. Xie, et al., A Kinetic study of delayed ettringite formation in hydrated Portland cement paste, Cem. Concr. Res. 25 (1) (1995) 63–70.
793
[28] T. Gutberlet, H. Hilbig, R.E. Beddoe, Acid attack on hydrated cement – effect of mineral acids on the degradation process, Cem. Concr. Res. 74 (2015) 35–43. [29] A. Neville, The confused world of sulfate attack on concrete, Cem. Concr. Res. 34 (8) (2004) 1275–1296. [30] J.G. Speight, Lange’s Handbook of Chemistry, 6th Ed., McGraw-Hill, New York, 1999. [31] C.W. Hargis, A. Telesca, P.J.M. Monteiro, Calcium sulfoaluminate (Ye’elimite) hydration in the presence of gypsum, calcite, and vaterite, Cem. Concr. Res. 65 (2014) 15–20. [32] H. Böke, S. Akkurt, Ettringite formation in historic bath brick-lime plasters, Cem. Concr. Res. 33 (9) (2003) 1457–1464. [33] W. Shen, Cement Technology, Wuhan University of Technology Press, Wuhan, 1991. [34] T. Sato, Thermal decomposition of aluminium hydroxides to aluminas, Thermochim. Acta 88 (1) (1985) 69–84. [35] A. Jiménez, M. Prieto, Thermal stability of ettringite exposed to atmosphere: implications for the uptake of harmful ions by cement, Environ. Sci. Technol. 49 (13) (2015) 7957–7964. [36] P.M. Carmona-Quiroga, M.T. Blanco-Varela, Ettringite decomposition in the presence of barium carbonate, Cem. Concr. Res. 52 (10) (2013) 140–148. [37] A.E. Moore, H.F.W. Taylor, Crystal structure of ettringite, Acta Crystallogr. 26 (4) (1970) 386–393. [38] F. Kenichi, Y. Teijiro, S. Haruo, A molecular orbital theory of reactivity in aromatic hydrocarbons, J. Chem. Phys. 20 (4) (1952) 722–725. [39] P. Atkins, J.D. Paula, Physical Chemistry, 9th ed., W. H. Freeman and Company, New York, 2010. [40] R.B. Perkins, C.D. Palmer, Solubility of ettringite (Ca6[Al(OH)6]2(SO4)326H2O) at 5–75 °C, Geochim. Cosmochim. Acta 63 (S13–14) (1999) 1969–1980. [41] S.C.B. Myneni, S.J. Traina, T.J. Logan, Ettringite solubility and geochemistry of the Ca(OH)2-Al2(SO4), 3–H2O system at 1 atm pressure and 298 K, Chem. Geol. 148 (1) (1998) 1–19.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Stability and decomposition mechanism of ettringite in presence of ammonium sulfate solution q Xuebing Wang a, Zhihua Pan a,⇑, Xiaodong Shen a, Weiqing Liu b a b
College of Materials Science and Engineering, Nanjing Tech University, Nanjing 210009, China College of Civil Engineering, Nanjing Tech University, Nanjing 210009, China
h i g h l i g h t s Decomposition of ettringite can be occurred when the concentration of (NH4)2SO4 reached to a certain extent. Main products of decompostion of ettringite were gypsum and Al(OH)3.
a r t i c l e
i n f o
Article history: Received 1 February 2016 Received in revised form 20 July 2016 Accepted 28 July 2016
Keywords: Ettringite (NH4)2SO4 Stability Mechanism
a b s t r a c t The formation and transformation characteristics of ettringite play an extremely important effect on the accelerated hardening and expansion of concrete as well as its long term durability. In alkaline condition, ettringite can be existed stably; but when the alkalinity exceeds a certain value, ettringite can be decomposed into calcium monosulfoaluminate (AFm). In this paper, the stability and its decomposition mechanism of ettringite in alkaline (NH4)2SO4 solution were then investigated. The results show that in alkaline environment, decomposition of ettringite can be occurred and main produces were gypsum and Al(OH)3 as long as the concentration of (NH4)2SO4 reached to a certain extent. Based on experiment data, a kind of NH+4 involved decomposition mechanism of ettringite had been proposed by method of quantum mechanics combined with frontier molecular orbital theory and thermodynamic theory. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Ettringite always exists in cement based materials, it begins to form during cement hydration process [1]. Ettringite is also applied to the shrinkage compensation and hardening accelerating and set accelerating of concrete [2]. However, formation of excess ettringite has negative effect on durability of cement based materials. Moreover, the negative effect in front of sulfate attack can become still much more negative because it is more aggressive and faster if the Portland cement is replaced partially by metakaolin due to its large reactive alumina content [3–6]. Therefore, to clarify the stability and transformation of ettringites in a variety of environment conditions plays an important guiding significance for
q Funded by China national 973 program (No. 2011CB013800); Program of Research Innovation for University Graduate Students of Jiangsu Province (KYLX150777); Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). ⇑ Corresponding authors. E-mail addresses: [email protected] (X. Wang), [email protected] (Z. Pan).
http://dx.doi.org/10.1016/j.conbuildmat.2016.07.135 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
understanding mechanical properties and sulfate attack of cement based materials. Ettringite is not only originated from the hydration of cement, but also from sulfate attack of cement based materials [7]. This originated ettringite may face many hostile environments and its stability is maybe changed. Factors affecting the stability of ettringite mainly include temperature, pH and others. J. Skalny [8] found that ettringite was instable and decomposed when temperature was exceeded 60–70 °C. If temperature was reached 80 °C, ettringite was converted into metaettringite by losing part of water [9]; K. Ogawa [10] claimed that ettringite was decomposed at 130–150 °C yielding monosulfate hydrate as the main decomposition product, hemihydrate and anhydrite were also formed. At 260 °C ettringite was completely dehydrated which was investigated by Sytle M. Antao [11]. Grusczscinski [12] and R.A. Livingston [13] found that morphology of ettringite was also changed at elevated temperature. But different surrounding would also affect the critical temperature of decomposition of ettringite. H.F.W. Taylor [14] considered supersaturated OH- and SO2 4 in pore solution can increase the stability of ettringite which still was existed even at 90 °C; stability of ettringite also was changed in the solutions
787
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
with different pH. J. Havlica et al. [15] claimed that ettringite was decomposed at pH below 10.7; J Hill et al. [16] and Z Chang [17] studied combined sulfate and acid attack of concrete with different SCM and found that content of ettringite was less. However, ettringite was also not stable even in condition of high alkaline. C.J. Warren et al. [18] found ettringite can be converted into monosulfate with pH above 13. (NH4)2SO4 existing in sewage facilities and industrial areas [19] always reduce the service performance of cement based materials. Anti-freezing admixtures such as urea and melamine superplasticizer sometimes can also generate ammonia and NH+4 ions. Therefore, cement based materials in service time may be threatened by ammonium. NH+4 can react with OH- so that it can reduce the pH of system, therefore, many reference always classify ammoniate salt as acid attack on concrete and the main purpose of NH+4 is to decrease the pH [20–22] and following induce the instability of many hydration produces such as C-S-H and ettringite. Raoul Jauberthie [23] had investigated that corrosion product was gypsum when cement mortar were immersed in 0.25 mol/l ammonium solutions for a period of 2 years, and pH of surrounding solution was changed from 6 to 8 by leaching of alkali. Michel Mbessa [24] found gypsum was existed on the surface of highstrength concrete when concrete was immersed in 20% ammonium solutions. M.T. Bassuoni [19] considered that ammonium sulfate caused a combined acid and sulfate attack characterized by decomposition/softening associated with expansion, cracking and spelling. U. Schneider et al. [25] claimed that disadvantage of ammonia in cement concrete mainly due to the change of alkaline environment of the presence of ettringite. However, due to unshared pair of electrons, NH3 has strong coordination ability and may react with ettringite especially in high pH value. When the pH > 9.25 NH3(aq) was the dominant species in the NH+4 solution [23], so NH3 may be involved into decomposition of ettringite. In this paper, (i) In terms of the pH distribution over depth, stability of ettringite was investigated to achieve the critical NH+4 concentration of decomposition of ettringite; (ii) To investigate the impact of NH+4 and avoid the impact of acid, products changes of decomposition of ettringite in high alkaline was studied; (iii) Decomposition mechanism of ettringite was thus proposed.
Table 2 Oxide compositions of cement/wt.%. CaO
SiO2
Al2O3
Fe2O3
MgO
SO3
K2O
TiO2
LOI
Total
64.68
18.36
4.33
2.95
0.79
2.60
0.73
0.19
3.70
98.33
distilled water; two solutions were mixed and stirred 2 h, stood after one week and pumping filtered, synthesized ettringite was then dried at 60 °C and stored in a closed container.
2.2.2. Mortar specimen preparation Cement mortars were prepared according to ASTM C305-99 (Practice for Mechanical Mixing of Hydraulic Cement Pastes and Mortars of Plastic Consistency), the sand/binder mass ratio was maintained equal to 3, water/cement ratio being 0.5, and mortar specimen size being in 40 mm 40 mm 160 mm. Mortar specimens were then cured under the condition of ambient temperature 20 ± 2 °C and RH not less than 90% in molds for 24 h, then demoulded and cured in water at 20 ± 1 °C for 28 days.
2.2.3. Sulfate corrosion conditions Mortar specimens curing after 28 days were immersed in 5% Na2SO4 (0.35 mol/L) and 5% (NH4)2SO4 (0.38 mol/L) respectively. After 180 days mortars were removed from different sulfate solutions and then the surface water was dried by cloth, and following dried for 24 h at 60 °C. Mortar specimens were then layered sampled every 0.5 mm from exposed surface by precision lathe, powder samples cutting down were collected respectively, then immersed in ethyl alcohol for 4 h and vacuum dried 24 h at 60 °C and stored in a closed container until analysis. Ettringite was accurately weighed and added in 100 ml 5% (0.380 mol/L) (NH4)2SO4 solution, and different dosages of NaOH solution (pH = 13) according to the test program was dropped into this solution to adjust the pH. After reaching of equilibrium state of system, the correlatively analysis was started.
2. Experimental 2.1. Raw materials (1) Cement: Commercial P.II 52.5 Portland cement, its physical properties and chemical compositions were listed in Tables 1 and 2. (2) Sand: China ISO standard sand. (3) Chemical reagents: Analytical reagent CaO, Al2(SO4)318H2O, CaSO4, (NH4)2SO4, Na2SO4. 2.2. Experimental process 2.2.1. Synthesis of ettringite Ettringite was synthesized by the method of Struble et al. [26]. 13.40 g CaO was weighed and dissolved in 890 ml distilled water; 26.55 g Al2(SO4)318H2O was also weighed and dissolved in 40 ml
2.2.4. Methods of physical-chemical properties X-ray diffraction analysis was carried out on ARL X’ TRA X-ray diffract-meter. Experimental parameters: 2.2 kW Cu target; stepping being 0.02°, scanning speed being 10°/min; TG-DSC analysis was carried out on STA449C thermal gravimetric instrument equipped with DSC 204 differential thermal analyser. Experiment parameters: temperature ranged from 40 °C to 900 °C; weight sensitivity being 0.1 lg; heating rate being 10 °C/min; nitrogen atmosphere. SEM-EDS was carried on JSM-5900 scanning electron microscope equipped with NORAN VANTAGE DSI energy dispersive spectrometer. Experiment parameters: accelerating voltage being 15 kV; resolution being 3 nm; magnification being 18 300,000; FTIR was carried out on Nexus 670 Fourier transform infrared spectrometer. Experiment parameters: resolution being 0.1 cm1; signal to noise being 33,400:1(peak to peak in one minute).
Table 1 Physical properties of cement. Specific surface area/m2kg1
Standard consistency water consumption/%
397
27.6
Setting time/min
Flexural strength/MPa
Compressive strength/MPa
Initial
Final
3d
28d
3d
28d
139
187
5.9
8.6
33.1
57.2
788
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
3. Results and discussion 3.1. Stability of ettringite in (NH4)2SO4 solution with different pH value Service environment of cement based material is generally complex. Sometimes with existing of (NH4)2SO4 solution, performance of cement based material may be affected. Stability of ettringite in (NH4)2SO4 solution with different pH value was then studied. Cement mortars that was immersed in 5% Na2SO4 (0.35 mol/L) and 5% (NH4)2SO4 (0.38 mol/L) after 180 days respectively were sampled at different depths by lathe, XRD was following carried out to determine the changes of products [27]. One of characteristic peak’s areas of material was then selected to represent its content, such as diffraction peak of Ca(OH)2 was at 18.04°, ettringite at 9.06° and gypsum at 11.66°, and the result was shown in Fig. 1. As was shown in Fig. 1, the reaction product was mainly ettringite when cement mortar was immersed in Na2SO4 solution; but it was gypsum in (NH4)2SO4 solution; not only that, portlandites in two different solutions were all reduced by the depth, 0.0 10000
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Gypsum/Na2SO4 Gypsum/(NH4)2SO4
8000
Area
6000 4000 2000 1500
Ettringite/Na2SO4 Ettringite/(NH4)2SO4
1200
Area
900 600 300 25000
Area
2000 1500 1000 500 0 0.0
Portlandite/ Na2SO4 Portlandite/ (NH4)2SO4 0.5
1.0
1.5
2.0
2.5
3.0
3.5
Depth/mm Fig. 1. Reaction products distribution in cement mortar over depth.
4.0
and they were still existed when depth of specimen was deeper than 1.5 mm. Existing of crystalline-state portlandite showed that pH of pore solution of mortar deeper than 1.5 mm was still about 12.4 [28]. Because corrosion products always depend on the concentration of sulfate content [29] and pH value [16], ettringite all can be formed firstly in both specimens of which was immersed in different sulfate solutions until sulfate concentration was exceeded the lowest concentration of precipitation of gypsum or until the pH of pore solution was lower than 10.7 [15]. However, in (NH4)2SO4 solution ettringite content of mortar was less than that of in Na2SO4 when depth of specimen were deeper than 1.5 mm and most of gypsum was formed in these region. This illustrates that ettringite was instable when the mortar was immersed in (NH4)2SO4 solution even in the condition of alkaline. Mortars were removed from (NH4)2SO4 solution after 180 days and dried at 60 °C, SEM-EDS was following carried out to observe the morphology of corrosion products at surface layer and the result was shown in Fig. 2. Fig. 2 showed that bar-like reaction products were generated at surface area and even some filled in the cracks of mortar. From EDS of point A, this bar-like corrosion product was gypsum [30]. It was formed and accumulated at corrosion areas. Sizes of gypsum were different, all of which the largest was reached to 5 lm and smallest was less than 0.5 lm. Due to the difficult determination of concentration of NH+4 which penetrated into specimen, following experiments were designed to simulate stability of ettringite and determine the critical concentration of NH+4 in different concentrations of (NH4)2SO4 solutions. 5% (NH4)2SO4 (0.380 mol/L) and NaOH solution (pH = 13) were prepared and the mixed-doses was in accordance to Table 3, then 1.5 g Struble et al.’s ettringite was added in each solution, and pH values of solution were measured periodically to ensure the chemical equilibrium of reaction. The results were then shown in Table 3. As was shown in Table 3, when NaOH solution was added into (NH4)2SO4 solution, pH values of solution tended to equilibrium after one week. Solutes were then pumping filtrated and dried at 60 °C after one week and XRD and FTIR was carried out, the results were shown in Fig. 4. It was showed that only the peaks of gypsum appeared in No.1-3 of Fig. 3 and the peaks of gypsum and ettringite all appeared in No.4, while only the peaks of ettringite appeared in No.5 and 6. It suggests that ettringite had been decomposed in No.1-3 of experiment, but it was still existed stably in No.5-6, and was partially existed in No.4. The results showed that critical concentration of (NH4)2SO4 solution which affected stability of ettringite was between 0.042 mol/L and 0.126 mol/L. Ettringite behaved stable when (NH4)2SO4 solution concentration was less than 0.042 mol/L, however, ettringite behaved instable and was decomposed when concentration was more than 0.126 mol/L.
+A
Fig. 2. SEM-EDS of cement mortar immersion in (NH4)2SO4 solution at 180 day.
789
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793 Table 3 pH values of solutions of different (NH4)2SO4 adding ettringite at different immersion time. No.
(NH4)2SO4/NaOH solution volume ratio
(NH4)2SO4 actual concentration (mol/L)
Immersion time 0h
24 h
48 h
72 h
96 h
120 h
1 2 3 4 5 6
1.5 1 0.5 0.25 0.125 0
0.227 0.189 0.126 0.076 0.042 0
8.78 8.90 9.20 9.81 11.23 13.05
9.15 9.22 9.54 9.94 10.20 12.85
9.16 9.37 9.81 10.16 10.22 12.74
9.01 9.20 9.63 10.02 10.06 12.76
9.06 9.25 9.73 10.06 10.01 12.76
8.97 9.15 9.66 10.02 9.88 12.71
:Ettringite :Gypsum
1 2
2 3 4
Transmission/%
3 1
4
5 6
5 6 5
15
25
35
45
55
65
2θ/°
4000
3400
2800
2200
1600
Wavenumber/cm
(a) XRD
1000
400
-1
(b) FTIR
Fig. 3. XRD and FTIR of reaction product affected on different concentration (NH4)2SO4.
Three types of vibration bands were relevant to the stability of ettringite. With increasing of concentration of (NH4)2SO4 solution feature bands of raw material (ettringite) started weaker, on the contrast, feature bands of gypsum gradually became stronger. No.4 in Fig. 3(b) showed the bands at 3200 cm1–3700 cm1 were attributed to O–H absorption band of ettringite as well as gypsum. This results also showed that critical concentration of (NH4)2SO4 which affected stability of ettringite was between 0.042 mol/L–0.126 mol/L when pH of solution was below to 13.
:Ettringite :Gypsum
Product
3.2. Products analysis of decomposition of ettringite
Raw material 5
15
25
35
45
55
In order to investigate the products by the influence of NH+4 on ettringite, stability of ettringite in (NH4)2SO4 solution with high pH value was studied. Because of cement hydration, the pH of cement paste is about 12.5–13.5 which is equivalent as the saturated
65
2θ/° Fig. 4. XRD of raw material and product.
There were three types of vibration in the FTIR from Fig. 3(b) [31]: (1) Vibration of O–H: Stretching vibration band of free OH in ettringite was at 3633 cm1, and stretching vibration and bending vibration bands of lattice water of gypsum and Al (OH)3 were at 3430 cm1 and 1640 cm1 respectively; (2) Vibration of S-O: S-O vibration bands of ettringite were mainly at 1120 cm1, 620 cm1 and 420 cm1 and asymmetric stretching vibration bands of gypsum were at 1118 cm1-1142 cm1. (3) Vibration of Al-O: Vibration bands of Al-O in ettringite appeared at 550 cm1 and 870 cm1 [32].
+A
Fig. 5. SEM-EDS of synthesized ettringite.
790
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
+C +B (a) Overall figure
(b) Partial figure Fig. 6. SEM-EDS of product.
solution of Ca(OH)2 [33]. For the sake of the products changes of decomposition of ettringite in cement based materials under the condition of NH+4 and high alkaline, experiments were carried out at pH = 13. Struble et al.’s ettringite was accurately weighed and added in 100 ml 5% (0.380 mol/L) (NH4)2SO4 solution, and dropped NaOH solution to regulate pH to 13. After one week, solute was pumping filtered and then dried at 60 °C, reaction products were following detected by XRD analysis and the result was shown in Fig. 4. It can be seen from XRD pattern (Fig. 4), characteristic peaks of ettringite in raw material were disappeared but the peaks of gypsum were appeared in 5% (0.380 mol/L) (NH4)2SO4 solution after one week. The result suggested that ettringite in (NH4)2SO4 solution was instable and decomposed; what more, one of reaction product was gypsum. To observe the morphologies of raw material of ettringite and decomposition products, SEM-EDS were carried out and the results were shown in Figs. 5 and 6. It is observed that ettringite in short bar outline was formed by the method of Struble et al. [26] in raw material, and the size of those ettringite was all less than 5 lm. The reaction products in Fig. 6 were mainly made up of columnar-like and gel-like compounds, EDS data at point B showed that columnar crystal was gypsum, all of which crystal sizes differed greatly and ranged from 5 lm to 100 lm. compound around gypsum was colloid, Al content of which the composition from EDS data was higher and Ca/S was close to 1:1, based on XRD analysis, only peaks of gypsum was appeared and other compounds didn’t detected in the products. Therefore colloids which contained Al element was Al(OH)3 gel.
100
1
Raw material Products 16.8%
90
Ettringite þ 4OH ! 2SO2 4 þ hydroxy Ettringite
0 24.2%
4.5%
TG/%
80
33.4%
-2
2.7%
70 46.9%
-3
4.7%
60
1.5%
50 0
100
200
300
400
500
600
700
Temperature / Fig. 7. TG-DSC of ettringite and products.
800
! Monosulfate þ 2Ca2þ DSC/mW/mg
-1
Raw ettringite and reaction products were dried at 60 °C and TG-DSC were carried out. The result was shown in Fig. 7. As was shown in Fig. 7, ettringite had two endothermic peaks at 120 °C and 253 °C, the former peak was attributed to the dehydration of ettringite into calcium monosulfoaluminate (AFm) at 120 °C and weight loss was 33.42%; AFm was then dehydrated and decomposed at 253 °C, weight loss was 4.7%. Dehydration of reaction products at 100 °C was attributed to the loss of absorption water of Al(OH)3 colloids; bound water of Al(OH)3 was all lost at 257 °C [34]and finally converted into Al2O3. gypsum was decomposed to hemihydrate gypsum at 150 °C; the exothermic peak at 372 °C was attributed to the change from soluble a type and b type hemihydrate gypsum to insoluble type; the reason of weight loss at 700 °C was the decomposition of CaCO3 which was from the absorption of CO2 in the reaction process. Before 900 °C, free and bound water in ettringite, Al(OH)3 and gypsum were all lost. Quality of all type water of ettringite was account for 45.9% [35]; Al(OH)3 was dehydrated and transformed to Al2O3 and the quality was account for 34.6% [36], gypsum was dehydrated and transformed to CaSO4 and the weight loss was 20.9%. since reaction products Al(OH)3 and gypsum were decomposed by ettringite of which Al:Ca ratio was 1:3, therefore the molar ratio of Al(OH)3 and gypsum was nearly 1:3 which on this basis the mass of water loss of products was 23.7%. As can be seen from Fig. 4 before 900 °C, weight loss of ettringite was 46.9% and reaction products were 24.2% which was close to the theoretical calculation. Therefore the main solid products were gypsum and Al(OH)3 when ettringite was immersed in combined alkaline and (NH4)2SO4 solution. However, the result was different with the Ref. [18]. Ref. [18] showed that ettringite was converted into calcium monosulfoaluminate at this pH value and the reaction can be expressed as:
ð1Þ
In this state, high pH induce the ettringite converted into calcium monosulfoaluminate. But in (NH4)2SO4 solutions, the main products were gypsum and Al(OH)3 but not monosulfate. So a kind of NH+4 involved decomposition mechanism of ettringite was proposed.
-4
3.3. Decomposition mechanism of ettringite in the (NH4)2SO4 solution -5 900
Structure of ettringite was determined by Toylor [37] at first time and he had found that its crystal structure was hexagonal prism which the structure unit was (Ca3(Al(OH)6)12H2O)3+, this structure unit was composed of column and channel paralleling
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
Ca-H2O LUMO
1.02eV
0.86eV
NH3 LUMO
Ca-H2O HOMO
NH3 HOMO
(a)
(b)
Fig. 8. Comparing study of HOMO and LUMO of Ca–H2O and NH3.
0.18eV
Ca-H2O-NH3 LUMO
Ca-H2O-NH3 HOMO Fig. 9. Comparing study of HOMO and LUMO of Ca–H2O–NH3.
the C axis of column. Among them, column was assembled by Al 2(OH)36 octahedral and three Ca polyhedron (Ca4H2O), SO4 existing in channel was to balance the charge and ensure the electron neutrality of molecule. Ca4H2O was connected by four Ca–O conjugate hybrid orbitals of which due to sp3 hybridization of Ca. As the second main group element, coordination ability of Ca was poor; Ca–O bonding force mainly was from Coulomb force. NH3 had a strong electronegativity which was due to two unshared electron pairs and can form hydrogen bond with water in Ca–H2O, however,
(a) Caa-H2O
791
[Ca–H2O–NH3]2+ system was not stable and decomposed in the water environment. In order to investigate the decomposition tendency of system, Method of quantum mechanics was used to calculate the charge distribution and bond length of Ca–H2O and Ca–H2O–NH3 respectively and following to propose the trend of this reaction process. Calculation method was based on density function theory methods and hybrid function was B3LYP. In this paper, frontier molecular orbital (FMO) theory which proposed by Kenichi Fukui [38] was used to investigate the reaction possibility between Ca–H2O and NH3 and the stability of [Ca–H2O–NH3]2+. FMO theory simplifies reactivity to interactions between the highest occupied molecular orbital (HOMO) of one species and the lowest unoccupied molecular orbital (LUMO) of the other in the chemical reaction. HOMO and LUMO of Ca–H2O and NH3 and [Ca–H2O–NH3]2+ respectively were shown in Figs. 8 and 9. According to FMO theory, when two molecular were closed to each other, only suitable symmetry and smaller difference of energy level and reasonable of directionality of electron shifted between HOMO of one molecule and LUMO of the other can the reaction happened. In Fig. 8, only in the path of Fig. 8(b) symmetries of HOMO of NH3 and LUMO of Ca–H2O were suitable; difference of energy level between two molecules was 0.863 eV which was less than 6 eV; the directionality of electron shifted from HOMO of NH3 to LUMO of Ca–H2O was also reasonable, so reaction can only happened between HOMO of NH3 and LUMO of Ca–H2O. As was shown in Fig. 9, according to energy levels difference between HOMO and LUMO of [Ca–H2O–NH3]2+, two energy levels were so close that it was easy of electron to transfer from HOMO to LUMO and further known that this structure was not stable and easy to decompose. To investigate the product of [Ca–H2O–NH3]2+, charge distribution and bond length of Ca–H2O and [Ca–H2O–NH3]2+ were studied in Fig. 10. In the case of existing of NH3, H atom’s position of coordinated water in Ca–H2O became asymmetric (Fig. 10), one of hydrogen atom was attracted by a nitrogen atom and their positive charge quantity became less, the corresponding O–H bond was elongated and bond length was increased 59.2%, and thus the distance of N atom in NH3 and H in CaH2O approximated the original bond lengths of N–H in NH3, namely four H atoms around N atom were close to equivalent and had tendency to form NH+4; on the contrary, Ca–O bond became short when NH3 was existed, namely Ca–O bond energy become larger. It was considered that [Ca–H2O–NH3]2+ system decomposed into [Ca–OH]+ and [NH4]+, i.e. in NH3
(b) Caa-H2O-NH3
Fig. 10. Charge distribution and bond length of Ca–H2O and Ca–H2O–NH3.
792
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793
environment Ca-H2O polyhedron behaved instable and dissociated. Chemical reaction can be thus expressed as:
NHþ4 þ OH NH3 þ H2 O
ð2Þ
½Ca H2 O2þ þ NH3 ! ½Ca OHþ þ NHþ4
ð3Þ
½Ca OHþ ! Ca2þ þ OH
ð4Þ
NH3 played as a catalytic role in the reaction process. When Ca4H2O structure was dissociated, the remaining part Al(OH)3 6 was then converted into Al(OH)3 in alkaline environment; When the pH > 9.25 NH3(aq) was the dominant species in the solution [23], therefore, the entire chemical reaction of ettringite decomposition process can be expressed as:
NHþ4 þ OH NH3 ðaqÞ þ H2 O;
K1 ¼
½NH3 ½NHþ4 ½OH
ð5Þ
3CaO Al2 O3 3CaSO4 32H2 O þ 6NH3 ðaqÞ þ 3SO2 4 ! 6CaSO4 2H2 O þ 2AlðOHÞ3 þ 6NHþ4 þ 12OH þ 8H2 O K2 ¼
½OH 12 ½NHþ4
ð6Þ
6
ð7Þ
3
6 ½SO2 4 ½NH3
4. Conclusions
H
Dr G2 ¼ RT ln K 2 þ RT ln K 2 KH 2 ¼ exp
DG H 2 RT
!
ð8Þ P
¼ exp
m
H
B B Gm2 ðBÞ
! ð9Þ
RT
where K1 and K2 were equilibrium constant of reaction (5) and reaction (6) respectively; Kh2 was the standard state equilibrium constant of reaction (6);vB was the stoichiometric number of each reactants and products of reaction (6); Gm2h was the standard state free energy of formation for each reactants and products; R was standard state mole gas constant; T was Kelvin temperature. Chemical Eq. (5) was a reversible reaction so standard state equilibrium constant of reaction (5) was equal to K1. Only if DrG2 < 0, the forward reaction of decomposition of ettringite (6) was spontaneous [39]. The thermodynamics data of common ions were from Ref. [30]; But the standard state free energy of formation for ettringite was not accurate from many authors; it was estimated as 15,211 ± 4.2 kJ/mol by Robert B Perkins et al. [40] and 15204.7 ± 23 kJ/mol by Satish C.B. Myneni et al. [41], in this paper, the standard state free energy of formation for ettringite was estimated as 15204.7 ± 23 kJ/mol and the maximum and minimum
+
Concentration of NH4/mol/L
1.2 1.0 0.8 0.6 1 0.4
2 3
0.2
4 5
0.0 8
values were selected to calculate the critical concentration of OH and NH+4 ions. According to these basic data, concentration of OH and NH+4 ions were achieved when DrG2 was equal to 0 and the result was shown in Fig. 11. As was shown in Fig. 11, the critical concentration of the ammonium ion was increased with the pH of solution. Ettringite was not stable and decomposed at the concentration of NH+4 ions above the shadow part in the Fig. 11. This reaction may occur at the concentration of NH+4 ions in the shadow part which was caused by the difference of thermodynamic data of ettringite that from the different authors. It can be also found that concentration of ammonium ions got huge impact on the stability of ettringite when in the strong alkaline environment (pH > 11). Impact of NH+4 through the reaction (6) became less and ettringite can be decomposed even in low concentration of NH+4 in the weak alkaline environment (pH < 11). The points in the Fig. 11 were the concentration data of experiment in Table 3. As were seen in the Fig. 11, the concentration at point 1–3 was above the critical concentration of NH+4 and decomposition of ettringite can be occurred. When concentration of NH+4 ions was exceeded to 0.126 mol/L (point 1–3 in the Fig. 11) ettringite can be decomposed, which was same as the result in Fig. 3.
9
10
11
12
13
Initial pH Fig. 11. the critical initial concentration of NH+4 ions.
14
Under the condition of alkaline, when the concentration of NH+4 ion reaches a critical concentration, decomposition of ettringite can be occurred and the products are gypsum and Al(OH)3. The critical NH+4 concentration of decomposition of ettringite is increased with the pH of solution. Decomposition mechanism of ettringite has been proposed by method of quantum mechanics and thermodynamic theory. NH+4 was involved in the decomposition of ettringite in the (NH4)2SO4 solution and it not only can reduce the pH in the solution, but also the coordination ability of NH3 can lead to decomposition of ettringite especially in high pH solution.
References [1] M. Santhanam, M.D. Cohen, J. Olek, Mechanism of sulfate attack: a fresh look Part 1: summary of experimental results, Cem. Concr. Res. 32 (6) (2002) 915– 921. [2] E. Sakai, Y. Nikaido, T. Itoh, M. Daimon, Ettringite formation and microstructure of rapid hardening cement, Cem. Concr. Res. 34 (9) (2004) 1669–1673. [3] R. Talero, Expansive synergic effect of ettringite from pozzolan and from OPC, co-precipitating in a common plaster-bearing solution. Part I: by cement pastes and mortars, Constr. Build. Mater. 24 (9) (2010) 1779–1789. [4] R. Talero, Kinetic and morphological differentiation of ettringites in plain and blended Portland cements with metakaolin and the ASTM C 452–68 test. Part II: morphological differentiation by SEM and XRD analysis, Mater. Construct. 59 (293) (2009) 35–51. [5] R. Talero, Expansive synergic effect of ettringite from pozzolan and from OPC, co-precipitating in a common plaster-bearing solution. Part II: II: fundamentals, explanation and justification, Constr. Build. Mater. 25 (3) (2011) 1139–1158. [6] R. Talero, Performance of metakaolin and Portland cements in ettringite formation as determined by ASTM C 452–68: kinetic and morphological differences, Cem. Concr. Res. 35 (7) (2005) 1269–1284. [7] J. Peng, Z. Lou, Study on the mechanism of ettringite formation, J. Chin. Ceram. Soc. 28 (6) (2000) 511–515. [8] J. Skalny, V. Johansen, N. Thaulow, DEF: as a form of sulfate attack, Mater. Construct. 46 (2) (1996) 5–29. [9] Q. Zhou, E.E. Lachowski, F.P. Glasser, Metaettringite, a decomposition product of ettringite, Cem. Concr. Res. 34 (4) (2004) 703–710. hydration ettringite formation, and its expansion [10] K. Ogawa, D. Roy, C4A3S mechanism: I. Expansion; Ettringite stability, Cem. Concr. Res. 11 (S5–6) (1981) 741–750. [11] S.M. Antao, DTA, TG and XRD studies of sturmanite ettringite, Can. Mineral. 40 (5) (2002) 1403–1409. [12] E. Grusczscinski, P.W. Brown, J.V. Bothe, The formation of ettringite at elevated temperature, Cem. Concr. Res. 23 (4) (1993) 981–987.
X. Wang et al. / Construction and Building Materials 124 (2016) 786–793 [13] S.K. Deb, M.H. Manghnani, K. Ross, et al., Raman scattering and X-ray diffraction study of the thermal decomposition of an ettringite-group crystal, Phys. Chem. Miner. 30 (1) (2003) 31–38. [14] H.F.W. Taylor, C. Famy, K.L. Scrivener, Delayed ettringite formation, Cem. Concr. Res. 31 (2001) 683–693. [15] J. Havlica, S. Sahu, Mechanism of ettringite and monosulphate formation, Cem. Concr. Res. 22 (4) (1992) 671–677. [16] J. Hill, E.A. Byars, J.H. Sharp, C.J. Lynsdale, et al., An experimental study of combined acid and sulfate attack of concrete, Cem. Concr. Compos. 25 (8) (2003) 997–1003. [17] Z. Chang, X. Song, R. Munn, et al., Using limestone aggregates and different cements for enhancing resistance of concrete to sulphuric acid attack, Cem. Concr. Res. 35 (8) (2005) 1486–1494. [18] C.J. Warren, E.J. Tteardon, The solubility of ettringite at 25 °C, Cem. Concr. Res. 24 (8) (1994) 1515–1524. [19] M.T. Bassuoni, M.L. Nehdi, Resistance of self-consolidating concrete to ammonium sulphate attack, Mater. Struct. 45 (7) (2012) 977–994. [20] A. Neville, Properties of Concrete, Longman, New York, 1997. [21] F.M. Lea, The Chemistry of Cement and Concrete, 3rd ed., Chemical Publishing Co., Inc, New York, 1971. [22] H.F.W. Taylor, Cement Chemistry, 2nd ed., Thomas Telford, New York, 1997. [23] R. Jauberthie, F. Rendell, Physicochemical study of the alteration surface of concrete exposed to ammonium salts, Cem. Concr. Res. 33 (1) (2003) 85–91. [24] M. Mbessa, J. Péra, Durability of high-strength concrete in ammonium sulfate solution, Cem. Concr. Res. 31 (8) (2001) 1227–1231. [25] U. Schneider, S.W. Chen, Modeling and empirical formulas for chemical corrosion and stress corrosion of cementitious materials, Mater. Struct. 31 (10) (1998) 662–668. [26] L.J. Struble, Synthesis and characterization of ettringite and related phases, 8th International Congress on the Chemistry of Cement, 1986. Rio de Janeiro. [27] Y. Fu, P. Gu, P. Xie, et al., A Kinetic study of delayed ettringite formation in hydrated Portland cement paste, Cem. Concr. Res. 25 (1) (1995) 63–70.
793
[28] T. Gutberlet, H. Hilbig, R.E. Beddoe, Acid attack on hydrated cement – effect of mineral acids on the degradation process, Cem. Concr. Res. 74 (2015) 35–43. [29] A. Neville, The confused world of sulfate attack on concrete, Cem. Concr. Res. 34 (8) (2004) 1275–1296. [30] J.G. Speight, Lange’s Handbook of Chemistry, 6th Ed., McGraw-Hill, New York, 1999. [31] C.W. Hargis, A. Telesca, P.J.M. Monteiro, Calcium sulfoaluminate (Ye’elimite) hydration in the presence of gypsum, calcite, and vaterite, Cem. Concr. Res. 65 (2014) 15–20. [32] H. Böke, S. Akkurt, Ettringite formation in historic bath brick-lime plasters, Cem. Concr. Res. 33 (9) (2003) 1457–1464. [33] W. Shen, Cement Technology, Wuhan University of Technology Press, Wuhan, 1991. [34] T. Sato, Thermal decomposition of aluminium hydroxides to aluminas, Thermochim. Acta 88 (1) (1985) 69–84. [35] A. Jiménez, M. Prieto, Thermal stability of ettringite exposed to atmosphere: implications for the uptake of harmful ions by cement, Environ. Sci. Technol. 49 (13) (2015) 7957–7964. [36] P.M. Carmona-Quiroga, M.T. Blanco-Varela, Ettringite decomposition in the presence of barium carbonate, Cem. Concr. Res. 52 (10) (2013) 140–148. [37] A.E. Moore, H.F.W. Taylor, Crystal structure of ettringite, Acta Crystallogr. 26 (4) (1970) 386–393. [38] F. Kenichi, Y. Teijiro, S. Haruo, A molecular orbital theory of reactivity in aromatic hydrocarbons, J. Chem. Phys. 20 (4) (1952) 722–725. [39] P. Atkins, J.D. Paula, Physical Chemistry, 9th ed., W. H. Freeman and Company, New York, 2010. [40] R.B. Perkins, C.D. Palmer, Solubility of ettringite (Ca6[Al(OH)6]2(SO4)326H2O) at 5–75 °C, Geochim. Cosmochim. Acta 63 (S13–14) (1999) 1969–1980. [41] S.C.B. Myneni, S.J. Traina, T.J. Logan, Ettringite solubility and geochemistry of the Ca(OH)2-Al2(SO4), 3–H2O system at 1 atm pressure and 298 K, Chem. Geol. 148 (1) (1998) 1–19.