- Email: [email protected]
Preparation and characterization of one-part non-Portland cement
Available online at www.sciencedirect.com
CERAMICS INTERNATIONAL
Ceramics International ] (]]]]) ]]]–]]] www.elsevier.com/locate/ceramint
Preparation and characterization of one-part non-Portland cement H.A. Abdel Gawwada,n, S. Abd El-Aleemb, A.S. Oudaa a
Housing and Building National Research Center, 87 El Tahreer St. Dokki, P.O. Box 1770, Giza, Egypt b Chemistry Department, Faculty of Science, Fayoum University, Fayoum, Egypt Received 30 June 2015; received in revised form 15 August 2015; accepted 19 August 2015
Abstract This work aims at enhancing the mass production and commercial viability of non-Portland cement (NPC) by preparing one-part-NPC (just add water). NPC is an eco-efficient material compared to Portland cement (PC). It was prepared by mixing of blast-furnace slag (BFS) with 2, 4, 6, 8 and 10% sodium hydroxide (SH) by total weight of BFS, and then mixing it with water. The homogeneous slurry is immediately dried in an oven at 80 1C for 24 h, followed by pulverization to a fixed particle size. Two main compositional factors were examined: first is the increase of SH wt%, while the amount of water to slag (W/BFS) ratio is kept maintaining constant, and the second includes increasing of W/BFS ratio at constant SH dosage. One-part NPC was mixed with water at W/NPC ratio of 0.25, and then cured. A conventional two-part NPC, containing SH solution (liquid part) and BFS (solid part), was prepared for comparison. The results showed that, the amount of NPC hydration products increase with SH wt% and W/BFS ratio. The compressive strength values of hardened cements proved that, the activation of slag continues after the addition of water to NPC. After 90 days of curing, the compressive strength of one-part NPC mixture decreased by 10% compared to that of the two-part-NPC containing the same activator content. Some selected hardened pastes were analyzed using FTIR, TGA/DTG and SEM techniques. The results of the different analyses are in a good harmony with those of mechanical properties and prove that; one-part NPC can be beneficially used as an alternative to PC. & 2015 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: One-part non-Portland cement; Granulated blast-furnace slag; Hydration products; Dissolution/condensation rates
1. Introduction Nowadays, the sustainability of the construction sector must be a priority for the civil engineering scientific and technological community. The development of innovative materials and methods aiming at extending the life-time of both existing and new structures is mandatory. Manufacturing of Portland cement (PC) is one the most contributing industries in the environmental pollution due to CO2, dust, oxinitrides (NOX) and oxisulfides (SOX) emissions [1,2]. Alkali activated aluminosilicate cements are considered more environmental friendly because of their low CO2 emission and energy requirements compared to PC [3]. One of the most common alkali activated aluminosilicate cements containing high calcium ions, is alkali activated slag cement or non-Portland cement (NPC) [4–6]. Conventional NPC is prepared from two n
Corresponding author. Tel.: þ20 1004540743; fax: þ 20 33351564. E-mail address: [email protected] (H.A. Abdel Gawwad).
components: the alkaline solution and an aluminosilicate powder, fly ash, blast-furnace slag (BFS), metakaolin and so on [4,6–10]. The alkali activation of slag mainly produces calcium silicate hydrate (CSH) gel with low calcium to silica (Ca/Si) ratio [11,12] in which silicon is present in one-dimensional chains with some substitution of Al and Mg for Si and Ca [13]. The alkali activation of BFS is occurred in three stages; the first is the dissolution process in which the alkali activator cleavages the Al–O–Al and Si–O–Si series in BFS network. The second step includes the condensation of broken activated species to form oligomers. The final step is the polycondensation of oligomers into polymeric structure [14]. However, the dissolution and polycondensation probably occur simultaneously when the solid aluminosilicate material mixed with liquid activator [15]. The alkaline solutions (NaOH, KOH and Na2SiO3 with SiO2/Na2O ¼ 1) are categorized as corrosive products, which must be handled with gloves, masks and glasses. They are not user friendly and would be difficult to use for bulk production
http://dx.doi.org/10.1016/j.ceramint.2015.08.096 0272-8842/& 2015 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
2
without appropriate safety procedures. There is a great need to produce one-part-NPC, like PC, which reacts with the addition of a small amount of water. The dry mixes based on dry NaOH/KOH with aluminosilicate or calcium aluminosilicate materials are corrosive in nature and may not be used [16]. As a matter of fact, very few studies have been conducted regarding preparation and properties of one-part aluminosilicate mixture to enhance the mass production and commercial viability of NPC. Koloušek et al. [17] prepared one-part geopolymer, with low strength (1 MPa after 7 days), by calcination of kaolinite or halloysite together with dry sodium hydroxide. Peng et al. [18] also prepared one-part geopolymer by calcination of low quality kaolin at different temperatures in the presence of sodium hydroxide (SH) or sodium carbonate (SC). Hajimohammadi et al. [19] developed one-part geopolymer by mixing geothermal silica and solid sodium aluminate (SA). Feng et al. [20] prepared one-part geopolymer from thermal activation of albite. Nematollahi et al. [21] developed one-part geopolymer mixtures by mixing fly ash (FA) with sodium silicate. In similar trend, Yang and Song [22] synthesized one-part geopolymer by mixing sodium silicate with BFS or FA or combination of them. Abdollahnejad et al. [23] studied the durability performance of one-part geopolymers concerning water absorption, penetration of chloride, carbonation resistance and resistance to acid attack. In this work an innovative experimental method was used to produce one-part NPC by controlling the BFS dissolution/condensation processes. The BFS was dry-mixed with SH, and then the amount of water lower than that required for the workable activated BFS paste formation was added, where, water plays an important role in the dissolution/polycondensation process. The higher the liquid to solid ratio, the faster the dissolution and polycondensation rates occur until the optimum ratio is reached, after which the addition of water could accelerate the dissolution
rate, while that of polycondensation is hindered [24]. The addition of too much water hinders the condensation rate whereas too low amount of water also leads to the same result. Water supports the required medium for the aluminosilicates dissolution, forming Al3 þ and Si4 þ , which condense to form aluminosilicate geopolymer. As the amount of mixing water decreases, the dissolution rate also decreases also. Thereby, there is no sufficient amount of activated species to be condensed. Hence, BFS was mixed with SH solution at W/BFS ratio, which makes activated BFS paste in a non-workable form. The BFS dissolution starts after mixing with alkaline solution. The removal of water was performed by drying BFS paste in an oven drier at 80 1C for 24 h, yielding solid material, which is crushed to produce powder with a fixed particle size. The individual effects of SH content and W/BFS ratio on the strength properties of hardened NPC-pastes were studied.
2. Materials and methods 2.1. Materials The materials used in this work are granulated blast-furnace slag (GBFS) and sodium hydroxide (SH). BFS was provided by Iron and Steel Company, Helwan, Giza Governorate. SH with 99.99 percent purity was obtained from SHIDO Company, UK. The glassy amorphous structure of BFS was confirmed by XRD as shown in Fig. 1. The chemical oxide analysis of BFS as determined by XRF is listed in Table 1. Its particle size distribution indicates that, it has a major particle size of o 90µm (Fig. 2).
2.2. Preparation of one-part non-Portland cement (NPC) The details of mix proportions are given in Table 2. The NPC-1, NPC-2, NPC-3, NPC-4 and NPC-5 mixes were designed by varying SH content such as 2, 4, 6, 8 and 10 wt% of total weight of BFS at constant W/BFS ratio of 0.10. In order to evaluate the effect of water content on the performance of NPC, NPC-6 and NPC-7, the mixes were designed by varying W/BFS ratio (0.15 and 0.20, respectively), while SH content was kept constant at 6 wt %. BFS was ground and dry-mixed with SH; after that the water was added and mixed for 5 min, using mechanical mixer. The homogenous slurry was then poured in stainless steel container and sealed with aluminum sheets to avoid moisture loss. All specimens were dried at 80 1C for 24 h. The hardened samples were pulverized and sieved through 90 mm sieve to obtain cement powder at a fixed particle size.
Fig. 1. XRD pattern of BFS.
Table 1 Chemical composition of BFS, wt%. Oxide, %
SiO2
Al2O3
Fe2O3
CaO
MgO
Na2O
K2O
SO3
TiO2
P2O5
L.O.I
Total
GGBFS
37.81
13.14
0.23
38.70
7.11
1.03
0.19
1.19
0.40
0.17
–
99.97
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
3
Table 2 Mix composition of NPC at constant drying temperature and curing time. Sample designation SH, wt.,% W/BFS ratio
NPC-1
NPC-2
NPC-3
NPC-4
NPC-5
NPC-6
NPC-7
2 0.10
4 0.10
6 0.10
8 0.10
10 0.10
6 0.15
6 0.20
2.3. Preparation pastes
Fig. 2. Particle size distribution of BFS.
The paste was prepared by the addition of water to one-part NPC, at W/NPC ratio of 0.25 and mixed for 5 min using mechanical mixer. The conventional two-part NPC (NPC-8 mix as counterpart to one-part NPC-3, NPC-6 and NPC-7 mixes) was prepared by mixing BFS with SH solution at W/BFS ratio of 0.25. The fresh pastes of one-part and two-part NPC were poured to one inch cubic molds, mechanically vibrated for one min., and then the surface was smoothed by a thin edged trowel. After molding the specimens were cured in a chamber with 100% relative humidity (RH) for 24 h at 3772 1C. The pastes were then demoulded and cured in a humidifier at the same condition for 90 days. 2.4. Methods of investigations The compressive strength of the hardened specimens was measured after residence times of 1, 3, 7, 28 and 90 days according to ASTM C109M-12 [25]. Thermogravimetric analysis (TGA) was carried out by heating the sample in nitrogen atmosphere up to 1000 1C with a heating rate of 20 1C/min using a DT-50 Thermal Analyzer (Schimadzu CoKyoto, Japan). The infrared (IR) spectral analysis was recorded from KBr discussing Genesis-II FT-IR spectrometer in the range of 400–4000 cm‐1. 3. Results and discussion 3.1. Compressive strength
Fig. 3. Compressive strength of one-part NPC-1, NPC-2, NPC-3, NPC-4 and NPC-5 mixtures.
Fig. 4. Compressive strength of one-part NPC-3, NPC-6 and NPC-7 as well as conventional two-part NPC-8 mixtures.
Fig. 3 shows the compressive strength (FCompressive) of the hardened cement pastes of NPC-1, NPC-2, NPC-3, NPC-4 and NPC-5 one-part NPC mixtures. It is clear that, the amount of SH is strongly affected the FCompressive of the hardened cement pastes. Increasing the SH content, at constant W/BFS ratio, increases the compressive strength at all curing ages until the optimum at 6 wt% SH is reached, and then the FCompressive is dropped. This may be attributed to that, as the percentage of SH increases up to 6 wt%, the dissolution of slag network increases to form Si4 þ , Al3 þ , Ca2 þ and Mg2 þ , which participate in the geopolymerization process. The higher SH content of 8 and 10 wt% accelerates the dissolution rate, while the remaining free sodium oxide (Na2O) in cement matrix minimizes the binding capacity of CSH, resulting in adhesion and strength losses [26,27]. Fig. 4 represents the FCompressive values of hardened cement pastes made from conventional two-part NPC (NPC-8) mixture as well as NPC-6, NPC-7 and NPC-3 one-part NPC mixtures. It is obvious that, at all hydration ages, the FCompressive values of NPC-3
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
4
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
Fig. 5. FTIR spectra of one-part NPCs prepared at different SH content.
one-part NPC are higher than those of NPC-6, NPC-7. This indicates that, the amount of water added during preparation of one-part NPC strongly affects its performance. W/BFS ratio plays an important role in the dissolution and polycondensation during NPC preparation. The relatively higher W/BFS ratio means higher dissolution and polycondensation rates [16], where the amount of hydration products of NPC6 and NPC-7 are higher than those formed in NPC-3. In other word, the increase of the hydration products during one-part NPC preparation reflects the decrease of activation continuance feasibility after the addition of water to onepart NPC, which in turn negatively affects the FCompressive. On the other hand, the 90 days compressive strength value of two-part NPC mixture (NPC-8), as counterpart to NPC-3, NPC-6 and NPC7 one-part mixtures, is higher than those of NPC-3, NPC-6 and NPC-7 one-part mixture by 10%, 22% and 35%, respectively. This is attributed to the fact that some of the SH is consumed during the preparation of one-part NPC and this consumption increases with W/BFS ratio, leading to the increase of compressive strength gap between one-part NPC and conventional two-part NPC mixtures. Indeed, the FCompressive results are in a good harmony with those of TG/DTG, FTIR and SEM investigations.
3.2. FTIR spectroscopy
Fig. 6. FTIR spectra of one-part NPCs prepared at different W/BFS ratio.
Fig. 7. FTIR spectra of one-pat NPCP-3 cement powder and cement pastes of the same mix at 1 day (NPC-3/1) and 28 days (NPCP-3/28) of curing as well as conventional two-part NPC-8 cement paste at 28 days of curing (NPC-8/28).
Fig. 5 represents the FTIR spectroscopy of BFS as well as one-part NPC-1, NPC-3 and NPC-5. It can be noted that, FTIR gives good interpretation about the effect of SH content on the properties of one-part NPCs. The absorption bands at 458– 464 cm 1 are related to O–Si–O bending vibration. The absorption bands at 640–659 cm 1 are due to symmetric stretching vibration of Si–O–Al bridges. The other bands at 1606–1625 cm 1, 3419–3444 cm 1 are assigned to bending H–O–H vibration and stretching vibration of O–H group. The intensities of these bands increase with SH content. This demonstrates that, the amount of hydration products, mainly CSH, sodium calcium aluminosilicate hydrate: NCASH formed in NPC increase with SH wt%. The strongest absorption band related to asymmetric stretching vibration of Si–O–T (where T is Si or Al) is observed at 937–964 cm 1. Its intensity increases with SH dosage. This is attributed to the fact that the dissolution rate of BFS increases as the amount of SH increases, forming more hydration products. Finally, the characteristics of absorption bands of Ht, calcium carbonate or sodium carbonate are observed at 1413–1427 cm 1. Fig. 6 illustrates the FTIR spectra of one-part NPC mixes prepared at varying W/BFS ratio while the amount of SH is kept constant (one-part NPC-3, NPC-6 and NPC-7). It is observed that, the main band corresponds to a symmetric stretching vibration of Si–O–T shifts to lower wave number with the W/BFS ratio. This fact should be explained by the changes occurred in the silicate network as a result of the formation of hydration products containing sodium cations at non-bridging oxygen (Na, Ca-orthosialate hydrate, Na, Cacyclo-orthosialate-disiloxo) or increasing Al substitution in silicate network [19,28]. Also, the absorption bands related to bending H–O–H vibration and stretching vibration of O–H
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
Fig. 8. (a) TG and (b) DTG of one-part NPCs prepared at different SH content.
group increase with W/BFS ratio, suggesting the formation of more hydration products. Fig. 7 shows FTIR spectra of one-part NPC-3 cement powder and cement pastes of the same mixture (one-part NPC-3/1) at 28 days (one-part NPC-3/28) and 1 day of curing as well as hardened cement pastes of two-part NPCP-8 at 28 days (two-part NPC-8/28). FTIR results indicate that, the absorption band related to asymmetric stretching vibration of Si–O–T at 958 cm 1 in case of one-part NPC-3 powder is shifted to 936 and 933 cm 1 in case one-part NPC-3/1 and NPC-3/28 pastes. Also, high full width at half-maximum (FWHM) of this band in case of one-part NPC-3/28 paste is lower than that of one-part NPC-3/1 paste or one-part NPC-3 powder. Both, shifting to lower wavenumber or the decrease in FWHM are ascribed to the increase of sample crystallinity, which is caused by structure ordering [29]. In addition, the intensity of absorption bands associated with H–O–H vibration at 3401–3421 cm 1 and O–H stretching vibration at 1608– 1626 cm 1 obtained for one-part NPC-3/28 cement paste are higher than that of one-part NPC-3/1 paste or one-part NPC-3 cement powder. This is due to the fact that amount of CSH formed in one-part NPC-3/28 is greater than that of the onepart NPCP-3/1 or NPC-3 cement powder, suggesting the continuation of activation and formation of hydration products
5
Fig. 9. (a) TG and (b) DTG of one-part NPCs prepared at different W/BFS ratio.
after the addition of water to one-part NPC powder. There is no significant difference in the intensity of absorption bands obtained for conventional two-part NPC-8/28 and one-part NPC-3/28 pastes. This is relevant enough for FCompressive results.
3.3. TG/DTG analysis Fig. 8(a) and (b) shows TG/DTG curves of NPC-1, NPC-3 and NPC-5 one-part cement powder (prepared at varying SH content and constant W/BFS ratio). Evidently, there are three main endothermic peaks related to CSH, Ht, and carbonate containing phases. The weight loss (WL) at 50–200 1C is mainly due to the removal of free water and the dehydration of CSH [30,31]. The WL between 200 and 400 1C [32] is attributed to the Ht decomposition. Other peaks of WL located in the range of 600–800 1C [33] are due to the dissociation of sodium or/and calcium carbonate. The amounts of CSH and Ht formed in one-part NPC-5 cement are higher than those formed in one-part NPC-1 and NPC-3 powder. This proves that, the rate of dissolution and formation of hydration products during one-part NPC powder preparation, increase with the SH content.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
6
Fig. 10. (a) TG and (b) DTG of one-pat NPCP-3 cement powder and cement pastes of the same mix at 1 day (NPC-3/1) and 28 days (NPCP-3/28) of curing as well as conventional two-part NPC-8 cement paste at 28 days of curing (NPC-8/28). Table 3 Weight losses calculated from TG curves. Weight loss, %
Sample designation 0 Day
NPC-1 NPC-3 NPC-5 NPC-6 NPC-7 NPC-8
1 Day
28 Day
Δm
CSH
Δm
CSH
Δm
CSH
2.91 4.76 6.32 5.92 5.16 –
1.03 2.43 4.73 2.93 3.11 –
– 7.39 – – – –
– 4.18 – – – –
– 14.39 – – – 14.58
– 8.29 – – – 8.72
Fig. 9(a) and (b) represents TG/DTG thermograms obtained for one-part NPC-3, NPC-6 and NPC-7 (prepared at varying W/BFS ratio while SH content kept constant). It shown that, the amount of hydration products formed in one-part NPC-7 is higher than that formed in one-part NPC-3 and NPC-6
powders, indicating that, the rate of BFS dissolution/condensation rate is accelerated with the increase of W/BFS ratio. This negatively affects the continuation of slag activation after addition of water to one-part NPC powder. Also, the probability of carbonation during the preparation of one-part NPC
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
US
7
RS
US RS 20 µm
20 µm
RS RS
US
RS
US 20 µm
20 µm
Fig. 11. SEM micrographs (a) NPC-1, (b) NPC-3, (c) NPC-5 and (d) NPC-7 cement powder.
decreases with the decrease of W/BFS. This is evidenced by the relatively more intense endothermic peaks of calcium or sodium carbonate in case of one-part NPC-7 compared to one-part NPC-3 powder. This is compatible with the finding of Otsuki et al. [34], which stated that the cement with relatively low water to cement ratio is more resistive to carbonation. Fig. 10(a) and (b) shows the TG/DTG thermograms of onepart NPC-3 powder and pastes of the same mixture at 1 day (one-part NPC-3/1) and 28 days (one-part NPC-3/28) of curing as well as the hardened cement pastes of two-part NPCP-8 at 28 days (two-part NPC-8/28). Obviously, the weight loss of onepart NPC-3/28 paste is higher than that of NPC-3/1 and NPC-3 powders, due to the continuance activation of slag after water addition to cement powder. This can be observed from the increase in the intensity of CSH and Ht peaks with curing time. In contrast, NPC-1/28 paste shows the lowest amounts of CSH and Ht. The endothermic peak related to carbonate containing phases at the range of 600–800 1C is observed in one-part NPC3 powder mixture disappears in cement pastes of the same mixture at 1 and 28 days of curing. This proves that, the carbonate containing phases in cement powder are mainly sodium carbonate (Na2CO3). After the addition of water to cement powder, Na2CO3 is consumed as a result of activation and dissolution of the unreacted BFS that exists in one-part cement powder. Finally, there is a marginal decrease in the weight loss of one-part NPC-3/28 cement paste compared to the conventional two-part NPC-8/28. This is compatible with FTIR and FCompressive results.
The total weight loss (WL), including chemically combined water content (Wn, %) is often used as a measure of the hydration progress of cement clinker phases [35]. The TG weight losses (Δm) of different one-part NPC-cement powders and the resulting one-part NPC-3 cement paste at different curing times as well as conventional two-part NPC-8 cement paste at 28 days of hydration are summarized in Table 3. It can be seen that the increase of SH and water contents significantly increases the formation of hydration products during preparation of one-part NPC powder. The increase in the WL of onepart NPC-3 paste with curing time proves the continuation of slag activation after water addition to one-part NPC-3 powder. The value of WL related to CSH and Δm of conventional twopart NPC-8 paste is near and close to that of the one-part NPC3 at 28 days of curing, confirming the FCompressive results. 3.4. Scanning electron microscopy (SEM). Fig. 11 displays the SEM-micrographs of cement powder prepared from different mixes. The micrograph of cement powder derived from NPC-1 presents high amount of unreacted slag (US). The amount of reacted slag (RS) increases with sodium hydroxide (SH) content and W/BFS ratio. This can be observed from the microstructure of NPC-3, NPC-5 and NPC-7 powders. This proves that as the amount of SH and W/BFS ratio increases, the dissolution rate also increases, forming more hydration products. After the addition of water to cement powder the activation reaction is continued to dissolve the remainder
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
8
50 µm
50 µm
50 µm Fig. 12. SEM micrographs (a) NPC-3 cement powder, (b) one-part NPC-3 cement paste at 28 days (NPC-3/28) and (c) cement pastes of two-part NPC-8 mix at 28 days of curing.
slag, resulting in Si4 þ , Al3 þ and Ca2 þ formation followed by their condensation, yielding a more homogeneous and compact microstructure. The hardened one-part NPC-3 paste cured at 28 days (Fig. 12b) seems to be more homogeneous with dense microstructure compared to one-part NPC-3 powder (Fig. 12a), suggesting the continuation of slag activation after addition water to cement powder. Clearly, there is no great difference between the microstructures of one-part NPC-3 paste and two-part NPC-8 paste (Fig. 12c) in the homogeneity and compaction, confirming the FTIR, TG/DTG and FCompressive results.
2.
3.
4. Conclusions 4. According to the results gathered throughout this work, it can be concluded that 1. A suitable non-Portland cement (NPC) powder which can be used as an alternative for Portland cement was prepared in this study. Ground granulated blast-furnace slag (BFS) was dry-mixed with NaOH at W/BFS ratios, which make the paste in non-workable form and is dried in an oven to yield solidified product. The solidified material was crushed to produce powder with a fixed particle size. In order to
5.
examine the continuation probability of cement powder activation, the water was added to it at W/NPC of 0.25. The cement powder performance is strongly affected by NaOH dosages and W/BFS ratio; where the addition of too high or too low NaOH content potentially reduced the continuation probability activation after the addition of water to NPC powder. Also, as the amount of W/BFS ratio decreases, the resulting cement powder shows a good performance for the activation continuance. The amount of dissolved calcium aluminosilicate materials increases with NaOH content and W/BFS ratio as confirmed by TG/DTG, FTIR and SEM techniques. The hardened cement paste with highest compressive strength (FCompressive) was obtained in case of the activation of NPC powder that was prepared by the addition of 6 wt% NaOH to BFS at W/BFS ratio of 0.1. The compressive strength gap between one-part and conventional two-part NPC cement pastes increases with the amount of water used in one-part NPC powder preparation. Hence, the FCompressive of conventional two-part NPC-8 mixture, as counter-part to one-part NPC-3, NPC-6 and NPC-7 mixtures was higher than that of one-part NPC-3, NPC-6 and NPC-7 mixtures by 10%, 22% and 35%, respectively.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
6. The increase in the TG/DTG-peak intensity of CSH as well as the bonding-peak intensity of Si–O–Al, H–O–H and O–H in FTIR spectroscopy proves that, after the addition of water to one-part NPC powder, the activation process continued. Also, the formation of dense and compact microstructure as seen in SEM micrographs is relevant enough for the same behavior.
References [1] P. Duxson, J.L. Provis, G.C. Lukey, The role of inorganic polymer technology in the development of “green concrete”, Cem. Concr. Res. 37 (2007) 1590–1597. [2] C.J. Shi, A. Fernandez-Jimenez, A. Palomo, New cements for the 21st century: the pursuit of an alternative to Portland cement, Cem. Concr. Res. 40 (2011) 750–763. [3] P. Duxson, A. Fernandez-Jimenez, J.L. Provis, Geopolymer technology: the current state of the art, J. Mater. Sci. 42 (2007) 2917–2933. [4] A.O. Purdon, The action of alkalis on blast-furnace slag, J. Soc. Chem. Ind. – Trans. Commun. 59 (1940) 191–202. [5] I. Teoreanu, A. Volceanov, S. Stoleriu, Non-Portland cements and derived materials, Cem. Concr. Res. 27 (6) (2005) 650–660. [6] Anne E. Oberlink, Non-Portland Cement Activation of Blast Furnace Slag (Master's theses), University of Kentucky, UK, 2010. [7] H. El-Didamony, A.A. Amer, H. Abd El-Aziz, Properties and durability of alkali-activated slag pastes immersed in sea water, Ceram. Int. 38 (2012) 3773–3780. [8] H. El-Didamony, A.A. Amer, T.M. El-Sokkary, H. Abd-El-Aziz, Effect of substitution of granulated slag by air-cooled slag on the properties of alkali activated slag, Ceram. Int. 39 (2013) 171–181. [9] C. Bilim, O. Karahan, C. Duran Atiş, S. İlkentapar, Effects of chemical admixtures and curing conditions on some properties of alkali-activated cement less slag mixtures, KSCE J. Civ. Eng. 19 (3) (2015) 733–741. [10] C. Cartwright, F. Rajabipour, A. Radlińska, Shrinkage characteristics of alkali-activated slag cements, J. Mater. Civ. Eng. 27 (7) (2015) B4014007. [11] I.G. Richardson, A.R. Brough, G.W. Groves, C.M. Dobson, The characterization of hardened alkali activated blast-furnace slag pastes and the nature of calcium silicate hydrate (C–S–H) paste, Cem. Concr. Res. 24 (1994) 813–829. [12] S.D. Wang, K.L. Scrivener, Hydration products of alkali-activated slag cement, Cem. Concr. Res. 25 (1995) 561–571. [13] J.L. Provis, J.S.J. van Deventer, Introduction to geopolymer, in: J. L. Provis, J.S.J. van Deventer (Eds.), Geopolymers: Structures, Processing, Properties And Industrial Applications, Woodhead Publishing, Abingdon UK, 2009, pp. 1–11. [14] J. Davidovits, Calcium based geopolymer, in: J. Davidovits (Ed.), Geopolymer Chemistry and Applications, 3rd ed., Geopolymer Institute, Saint Quentin, France, 2011, pp. 201–244. [15] W. Hongling, L. Haihong, Y. Fengyuan, Synthesis and mechanical properties of metakaolinite-based geopolymer, Colloids Surf. A: Physicochem. Eng. Asp. 268 (2005) 1–6. [16] J. Davidovits, Geopolymer chemistry and applications, in: J. Davidovits (Ed.), Development of user – friendly systems, 3rd ed., Geopolymer Institute, Saint Quentin, France, 2011, pp. 433–443.
9
[17] D. Koloušek, J. Brus, M. Urbanova, J. Andertova, V. Hulinsky, J. Vorel, Preparation, structure and hydrothermal stability of alternative (sodium silicate-free) geopolymers, J. Mater. Sci. 42 (22) (2007) 9267–9275. [18] M.X. Peng, Z.H. Wang, S.H. Shen, Q.G. Xiao, Synthesis, characterization and mechanisms of one-part geopolymeric cement by calcining lowquality kaolin with alkali, Mater. Struct./Mater. et Constr. 48 (3) (2014) 699–708. [19] A. Hajimohammadi, J.L. Provis, J.S.J. Deventer, One-part geopolymer mixes from geothermal silica and sodium aluminate, Ind. Eng. Chem. Res. 47 (23) (2008) 9396–9405. [20] D. Feng, J.L. Provis, J.S.J. Deventer, Thermal activation of albite for the synthesis of one-part mix geopolymers, J. Am. Ceram. Soc. 95 (2) (2012) 565–572. [21] B. Nematollahi, J. Sanjayan, F.U.A. Shaikh, Synthesis of heat and ambient cured one-part geopolymer mixes with different grades of sodium silicate, Ceram. Int. 41 (4) (2015) 5696–5704. [22] K.H. Yang, J.K. Song, Workability loss and compressive strength development of cementless mortars activated by combination of sodium silicate and sodium hydroxide, J. Mater. Civ. Eng. 21 (3) (2009) 119–127. [23] Z. Abdollahnejad, F. Pacheco-Torgal, J.B. Aguiar, C. Jesus, Durability performance of fly ash based one-part geopolymer mortars, Key Eng. Mater. 634 (2015) 113–120. [24] Z. Zuhua, Y. Xiao, Z. Huaju, C. Yue, Role of water in the synthesis of calcined kaolin-based geopolymer, Appl. Clay Sci. 43 (2009) 218–223. [25] ASTM C109M, Standard test method for compressive strength of hydraulic cement mortars, 2012. [26] J.L. Provis, J.S.J. van Deventer, Geopolymer synthesis and kinetics, in: J. L. Provis, C.A. Rees (Eds.), Geopolymers: Structures, Processing, Properties and Industrial Applications, Woodhead Publishing, Abingdon UK, 2009, pp. 118–136. [27] T. Bakharev, J.G. Sanjayanand, Y.B. Cheng, Sulfate attack on alkali activated slag concrete, J. Cem. Concr. Res. 32 (2002) 211–216. [28] J.R. Sweet, W.B. White, Study of sodium silicate glasses and liquids by infrared reflectance spectroscopy, Phys. Chem. Glas. 10 (6) (1969) 246–251. [29] W. Mozgawa, J. Deja, Spectroscopic studies of alkaline activated slag geopolymers, J. Mol. Struct. 924–926 (2009) 434–441. [30] P.C. Hewlett, Lea's Chemistry of Cement and Concrete, Elsevier Science & Technology Books, 2004. [31] I. Ismail, S.A. Bernal, J.L. Provis, S. Hamdan, J.S.J. van Deventer, Microstructural changes in alkali activated fly ash/slag geopolymers with sulfate exposure, Mater. Struct. 46 (2013) 361–373. [32] S.D.S. Wang, K.L.K. Scrivener, P. Pratt, Factors affecting the strength of alkali-activated slag, Cem. Concr. Res. 24 (1994) 1033–1043. [33] F. Jin, K. Gu, A. Al-Tabbaa, Strength and drying shrinkage of reactive MgO modified alkali-activated slag paste, Constr. Build. Mater. 51 (2014) 395–404. [34] N. Otsuki, S. Ronaldo Gallardo, T. Annaka, S. Takaki, T. Nishida, Field survey for carbonation depth of reinforced concrete buildings in the Philippines, in: Proceedings of the 37th Conference on Our World in Concrete & Structures, Singapore, 2012. [35] M. Ben Haha, G.L. Saout, F. Winnefeld, B. Lothenbach, Influence of activator type on hydration kinetics, hydrate assemblage and microstructural development of alkali-activated blast-furnace slags, Cem. Concr. Res. 41 (2011) 301–310.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
CERAMICS INTERNATIONAL
Ceramics International ] (]]]]) ]]]–]]] www.elsevier.com/locate/ceramint
Preparation and characterization of one-part non-Portland cement H.A. Abdel Gawwada,n, S. Abd El-Aleemb, A.S. Oudaa a
Housing and Building National Research Center, 87 El Tahreer St. Dokki, P.O. Box 1770, Giza, Egypt b Chemistry Department, Faculty of Science, Fayoum University, Fayoum, Egypt Received 30 June 2015; received in revised form 15 August 2015; accepted 19 August 2015
Abstract This work aims at enhancing the mass production and commercial viability of non-Portland cement (NPC) by preparing one-part-NPC (just add water). NPC is an eco-efficient material compared to Portland cement (PC). It was prepared by mixing of blast-furnace slag (BFS) with 2, 4, 6, 8 and 10% sodium hydroxide (SH) by total weight of BFS, and then mixing it with water. The homogeneous slurry is immediately dried in an oven at 80 1C for 24 h, followed by pulverization to a fixed particle size. Two main compositional factors were examined: first is the increase of SH wt%, while the amount of water to slag (W/BFS) ratio is kept maintaining constant, and the second includes increasing of W/BFS ratio at constant SH dosage. One-part NPC was mixed with water at W/NPC ratio of 0.25, and then cured. A conventional two-part NPC, containing SH solution (liquid part) and BFS (solid part), was prepared for comparison. The results showed that, the amount of NPC hydration products increase with SH wt% and W/BFS ratio. The compressive strength values of hardened cements proved that, the activation of slag continues after the addition of water to NPC. After 90 days of curing, the compressive strength of one-part NPC mixture decreased by 10% compared to that of the two-part-NPC containing the same activator content. Some selected hardened pastes were analyzed using FTIR, TGA/DTG and SEM techniques. The results of the different analyses are in a good harmony with those of mechanical properties and prove that; one-part NPC can be beneficially used as an alternative to PC. & 2015 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: One-part non-Portland cement; Granulated blast-furnace slag; Hydration products; Dissolution/condensation rates
1. Introduction Nowadays, the sustainability of the construction sector must be a priority for the civil engineering scientific and technological community. The development of innovative materials and methods aiming at extending the life-time of both existing and new structures is mandatory. Manufacturing of Portland cement (PC) is one the most contributing industries in the environmental pollution due to CO2, dust, oxinitrides (NOX) and oxisulfides (SOX) emissions [1,2]. Alkali activated aluminosilicate cements are considered more environmental friendly because of their low CO2 emission and energy requirements compared to PC [3]. One of the most common alkali activated aluminosilicate cements containing high calcium ions, is alkali activated slag cement or non-Portland cement (NPC) [4–6]. Conventional NPC is prepared from two n
Corresponding author. Tel.: þ20 1004540743; fax: þ 20 33351564. E-mail address: [email protected] (H.A. Abdel Gawwad).
components: the alkaline solution and an aluminosilicate powder, fly ash, blast-furnace slag (BFS), metakaolin and so on [4,6–10]. The alkali activation of slag mainly produces calcium silicate hydrate (CSH) gel with low calcium to silica (Ca/Si) ratio [11,12] in which silicon is present in one-dimensional chains with some substitution of Al and Mg for Si and Ca [13]. The alkali activation of BFS is occurred in three stages; the first is the dissolution process in which the alkali activator cleavages the Al–O–Al and Si–O–Si series in BFS network. The second step includes the condensation of broken activated species to form oligomers. The final step is the polycondensation of oligomers into polymeric structure [14]. However, the dissolution and polycondensation probably occur simultaneously when the solid aluminosilicate material mixed with liquid activator [15]. The alkaline solutions (NaOH, KOH and Na2SiO3 with SiO2/Na2O ¼ 1) are categorized as corrosive products, which must be handled with gloves, masks and glasses. They are not user friendly and would be difficult to use for bulk production
http://dx.doi.org/10.1016/j.ceramint.2015.08.096 0272-8842/& 2015 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
2
without appropriate safety procedures. There is a great need to produce one-part-NPC, like PC, which reacts with the addition of a small amount of water. The dry mixes based on dry NaOH/KOH with aluminosilicate or calcium aluminosilicate materials are corrosive in nature and may not be used [16]. As a matter of fact, very few studies have been conducted regarding preparation and properties of one-part aluminosilicate mixture to enhance the mass production and commercial viability of NPC. Koloušek et al. [17] prepared one-part geopolymer, with low strength (1 MPa after 7 days), by calcination of kaolinite or halloysite together with dry sodium hydroxide. Peng et al. [18] also prepared one-part geopolymer by calcination of low quality kaolin at different temperatures in the presence of sodium hydroxide (SH) or sodium carbonate (SC). Hajimohammadi et al. [19] developed one-part geopolymer by mixing geothermal silica and solid sodium aluminate (SA). Feng et al. [20] prepared one-part geopolymer from thermal activation of albite. Nematollahi et al. [21] developed one-part geopolymer mixtures by mixing fly ash (FA) with sodium silicate. In similar trend, Yang and Song [22] synthesized one-part geopolymer by mixing sodium silicate with BFS or FA or combination of them. Abdollahnejad et al. [23] studied the durability performance of one-part geopolymers concerning water absorption, penetration of chloride, carbonation resistance and resistance to acid attack. In this work an innovative experimental method was used to produce one-part NPC by controlling the BFS dissolution/condensation processes. The BFS was dry-mixed with SH, and then the amount of water lower than that required for the workable activated BFS paste formation was added, where, water plays an important role in the dissolution/polycondensation process. The higher the liquid to solid ratio, the faster the dissolution and polycondensation rates occur until the optimum ratio is reached, after which the addition of water could accelerate the dissolution
rate, while that of polycondensation is hindered [24]. The addition of too much water hinders the condensation rate whereas too low amount of water also leads to the same result. Water supports the required medium for the aluminosilicates dissolution, forming Al3 þ and Si4 þ , which condense to form aluminosilicate geopolymer. As the amount of mixing water decreases, the dissolution rate also decreases also. Thereby, there is no sufficient amount of activated species to be condensed. Hence, BFS was mixed with SH solution at W/BFS ratio, which makes activated BFS paste in a non-workable form. The BFS dissolution starts after mixing with alkaline solution. The removal of water was performed by drying BFS paste in an oven drier at 80 1C for 24 h, yielding solid material, which is crushed to produce powder with a fixed particle size. The individual effects of SH content and W/BFS ratio on the strength properties of hardened NPC-pastes were studied.
2. Materials and methods 2.1. Materials The materials used in this work are granulated blast-furnace slag (GBFS) and sodium hydroxide (SH). BFS was provided by Iron and Steel Company, Helwan, Giza Governorate. SH with 99.99 percent purity was obtained from SHIDO Company, UK. The glassy amorphous structure of BFS was confirmed by XRD as shown in Fig. 1. The chemical oxide analysis of BFS as determined by XRF is listed in Table 1. Its particle size distribution indicates that, it has a major particle size of o 90µm (Fig. 2).
2.2. Preparation of one-part non-Portland cement (NPC) The details of mix proportions are given in Table 2. The NPC-1, NPC-2, NPC-3, NPC-4 and NPC-5 mixes were designed by varying SH content such as 2, 4, 6, 8 and 10 wt% of total weight of BFS at constant W/BFS ratio of 0.10. In order to evaluate the effect of water content on the performance of NPC, NPC-6 and NPC-7, the mixes were designed by varying W/BFS ratio (0.15 and 0.20, respectively), while SH content was kept constant at 6 wt %. BFS was ground and dry-mixed with SH; after that the water was added and mixed for 5 min, using mechanical mixer. The homogenous slurry was then poured in stainless steel container and sealed with aluminum sheets to avoid moisture loss. All specimens were dried at 80 1C for 24 h. The hardened samples were pulverized and sieved through 90 mm sieve to obtain cement powder at a fixed particle size.
Fig. 1. XRD pattern of BFS.
Table 1 Chemical composition of BFS, wt%. Oxide, %
SiO2
Al2O3
Fe2O3
CaO
MgO
Na2O
K2O
SO3
TiO2
P2O5
L.O.I
Total
GGBFS
37.81
13.14
0.23
38.70
7.11
1.03
0.19
1.19
0.40
0.17
–
99.97
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
3
Table 2 Mix composition of NPC at constant drying temperature and curing time. Sample designation SH, wt.,% W/BFS ratio
NPC-1
NPC-2
NPC-3
NPC-4
NPC-5
NPC-6
NPC-7
2 0.10
4 0.10
6 0.10
8 0.10
10 0.10
6 0.15
6 0.20
2.3. Preparation pastes
Fig. 2. Particle size distribution of BFS.
The paste was prepared by the addition of water to one-part NPC, at W/NPC ratio of 0.25 and mixed for 5 min using mechanical mixer. The conventional two-part NPC (NPC-8 mix as counterpart to one-part NPC-3, NPC-6 and NPC-7 mixes) was prepared by mixing BFS with SH solution at W/BFS ratio of 0.25. The fresh pastes of one-part and two-part NPC were poured to one inch cubic molds, mechanically vibrated for one min., and then the surface was smoothed by a thin edged trowel. After molding the specimens were cured in a chamber with 100% relative humidity (RH) for 24 h at 3772 1C. The pastes were then demoulded and cured in a humidifier at the same condition for 90 days. 2.4. Methods of investigations The compressive strength of the hardened specimens was measured after residence times of 1, 3, 7, 28 and 90 days according to ASTM C109M-12 [25]. Thermogravimetric analysis (TGA) was carried out by heating the sample in nitrogen atmosphere up to 1000 1C with a heating rate of 20 1C/min using a DT-50 Thermal Analyzer (Schimadzu CoKyoto, Japan). The infrared (IR) spectral analysis was recorded from KBr discussing Genesis-II FT-IR spectrometer in the range of 400–4000 cm‐1. 3. Results and discussion 3.1. Compressive strength
Fig. 3. Compressive strength of one-part NPC-1, NPC-2, NPC-3, NPC-4 and NPC-5 mixtures.
Fig. 4. Compressive strength of one-part NPC-3, NPC-6 and NPC-7 as well as conventional two-part NPC-8 mixtures.
Fig. 3 shows the compressive strength (FCompressive) of the hardened cement pastes of NPC-1, NPC-2, NPC-3, NPC-4 and NPC-5 one-part NPC mixtures. It is clear that, the amount of SH is strongly affected the FCompressive of the hardened cement pastes. Increasing the SH content, at constant W/BFS ratio, increases the compressive strength at all curing ages until the optimum at 6 wt% SH is reached, and then the FCompressive is dropped. This may be attributed to that, as the percentage of SH increases up to 6 wt%, the dissolution of slag network increases to form Si4 þ , Al3 þ , Ca2 þ and Mg2 þ , which participate in the geopolymerization process. The higher SH content of 8 and 10 wt% accelerates the dissolution rate, while the remaining free sodium oxide (Na2O) in cement matrix minimizes the binding capacity of CSH, resulting in adhesion and strength losses [26,27]. Fig. 4 represents the FCompressive values of hardened cement pastes made from conventional two-part NPC (NPC-8) mixture as well as NPC-6, NPC-7 and NPC-3 one-part NPC mixtures. It is obvious that, at all hydration ages, the FCompressive values of NPC-3
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
4
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
Fig. 5. FTIR spectra of one-part NPCs prepared at different SH content.
one-part NPC are higher than those of NPC-6, NPC-7. This indicates that, the amount of water added during preparation of one-part NPC strongly affects its performance. W/BFS ratio plays an important role in the dissolution and polycondensation during NPC preparation. The relatively higher W/BFS ratio means higher dissolution and polycondensation rates [16], where the amount of hydration products of NPC6 and NPC-7 are higher than those formed in NPC-3. In other word, the increase of the hydration products during one-part NPC preparation reflects the decrease of activation continuance feasibility after the addition of water to onepart NPC, which in turn negatively affects the FCompressive. On the other hand, the 90 days compressive strength value of two-part NPC mixture (NPC-8), as counterpart to NPC-3, NPC-6 and NPC7 one-part mixtures, is higher than those of NPC-3, NPC-6 and NPC-7 one-part mixture by 10%, 22% and 35%, respectively. This is attributed to the fact that some of the SH is consumed during the preparation of one-part NPC and this consumption increases with W/BFS ratio, leading to the increase of compressive strength gap between one-part NPC and conventional two-part NPC mixtures. Indeed, the FCompressive results are in a good harmony with those of TG/DTG, FTIR and SEM investigations.
3.2. FTIR spectroscopy
Fig. 6. FTIR spectra of one-part NPCs prepared at different W/BFS ratio.
Fig. 7. FTIR spectra of one-pat NPCP-3 cement powder and cement pastes of the same mix at 1 day (NPC-3/1) and 28 days (NPCP-3/28) of curing as well as conventional two-part NPC-8 cement paste at 28 days of curing (NPC-8/28).
Fig. 5 represents the FTIR spectroscopy of BFS as well as one-part NPC-1, NPC-3 and NPC-5. It can be noted that, FTIR gives good interpretation about the effect of SH content on the properties of one-part NPCs. The absorption bands at 458– 464 cm 1 are related to O–Si–O bending vibration. The absorption bands at 640–659 cm 1 are due to symmetric stretching vibration of Si–O–Al bridges. The other bands at 1606–1625 cm 1, 3419–3444 cm 1 are assigned to bending H–O–H vibration and stretching vibration of O–H group. The intensities of these bands increase with SH content. This demonstrates that, the amount of hydration products, mainly CSH, sodium calcium aluminosilicate hydrate: NCASH formed in NPC increase with SH wt%. The strongest absorption band related to asymmetric stretching vibration of Si–O–T (where T is Si or Al) is observed at 937–964 cm 1. Its intensity increases with SH dosage. This is attributed to the fact that the dissolution rate of BFS increases as the amount of SH increases, forming more hydration products. Finally, the characteristics of absorption bands of Ht, calcium carbonate or sodium carbonate are observed at 1413–1427 cm 1. Fig. 6 illustrates the FTIR spectra of one-part NPC mixes prepared at varying W/BFS ratio while the amount of SH is kept constant (one-part NPC-3, NPC-6 and NPC-7). It is observed that, the main band corresponds to a symmetric stretching vibration of Si–O–T shifts to lower wave number with the W/BFS ratio. This fact should be explained by the changes occurred in the silicate network as a result of the formation of hydration products containing sodium cations at non-bridging oxygen (Na, Ca-orthosialate hydrate, Na, Cacyclo-orthosialate-disiloxo) or increasing Al substitution in silicate network [19,28]. Also, the absorption bands related to bending H–O–H vibration and stretching vibration of O–H
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
Fig. 8. (a) TG and (b) DTG of one-part NPCs prepared at different SH content.
group increase with W/BFS ratio, suggesting the formation of more hydration products. Fig. 7 shows FTIR spectra of one-part NPC-3 cement powder and cement pastes of the same mixture (one-part NPC-3/1) at 28 days (one-part NPC-3/28) and 1 day of curing as well as hardened cement pastes of two-part NPCP-8 at 28 days (two-part NPC-8/28). FTIR results indicate that, the absorption band related to asymmetric stretching vibration of Si–O–T at 958 cm 1 in case of one-part NPC-3 powder is shifted to 936 and 933 cm 1 in case one-part NPC-3/1 and NPC-3/28 pastes. Also, high full width at half-maximum (FWHM) of this band in case of one-part NPC-3/28 paste is lower than that of one-part NPC-3/1 paste or one-part NPC-3 powder. Both, shifting to lower wavenumber or the decrease in FWHM are ascribed to the increase of sample crystallinity, which is caused by structure ordering [29]. In addition, the intensity of absorption bands associated with H–O–H vibration at 3401–3421 cm 1 and O–H stretching vibration at 1608– 1626 cm 1 obtained for one-part NPC-3/28 cement paste are higher than that of one-part NPC-3/1 paste or one-part NPC-3 cement powder. This is due to the fact that amount of CSH formed in one-part NPC-3/28 is greater than that of the onepart NPCP-3/1 or NPC-3 cement powder, suggesting the continuation of activation and formation of hydration products
5
Fig. 9. (a) TG and (b) DTG of one-part NPCs prepared at different W/BFS ratio.
after the addition of water to one-part NPC powder. There is no significant difference in the intensity of absorption bands obtained for conventional two-part NPC-8/28 and one-part NPC-3/28 pastes. This is relevant enough for FCompressive results.
3.3. TG/DTG analysis Fig. 8(a) and (b) shows TG/DTG curves of NPC-1, NPC-3 and NPC-5 one-part cement powder (prepared at varying SH content and constant W/BFS ratio). Evidently, there are three main endothermic peaks related to CSH, Ht, and carbonate containing phases. The weight loss (WL) at 50–200 1C is mainly due to the removal of free water and the dehydration of CSH [30,31]. The WL between 200 and 400 1C [32] is attributed to the Ht decomposition. Other peaks of WL located in the range of 600–800 1C [33] are due to the dissociation of sodium or/and calcium carbonate. The amounts of CSH and Ht formed in one-part NPC-5 cement are higher than those formed in one-part NPC-1 and NPC-3 powder. This proves that, the rate of dissolution and formation of hydration products during one-part NPC powder preparation, increase with the SH content.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
6
Fig. 10. (a) TG and (b) DTG of one-pat NPCP-3 cement powder and cement pastes of the same mix at 1 day (NPC-3/1) and 28 days (NPCP-3/28) of curing as well as conventional two-part NPC-8 cement paste at 28 days of curing (NPC-8/28). Table 3 Weight losses calculated from TG curves. Weight loss, %
Sample designation 0 Day
NPC-1 NPC-3 NPC-5 NPC-6 NPC-7 NPC-8
1 Day
28 Day
Δm
CSH
Δm
CSH
Δm
CSH
2.91 4.76 6.32 5.92 5.16 –
1.03 2.43 4.73 2.93 3.11 –
– 7.39 – – – –
– 4.18 – – – –
– 14.39 – – – 14.58
– 8.29 – – – 8.72
Fig. 9(a) and (b) represents TG/DTG thermograms obtained for one-part NPC-3, NPC-6 and NPC-7 (prepared at varying W/BFS ratio while SH content kept constant). It shown that, the amount of hydration products formed in one-part NPC-7 is higher than that formed in one-part NPC-3 and NPC-6
powders, indicating that, the rate of BFS dissolution/condensation rate is accelerated with the increase of W/BFS ratio. This negatively affects the continuation of slag activation after addition of water to one-part NPC powder. Also, the probability of carbonation during the preparation of one-part NPC
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
US
7
RS
US RS 20 µm
20 µm
RS RS
US
RS
US 20 µm
20 µm
Fig. 11. SEM micrographs (a) NPC-1, (b) NPC-3, (c) NPC-5 and (d) NPC-7 cement powder.
decreases with the decrease of W/BFS. This is evidenced by the relatively more intense endothermic peaks of calcium or sodium carbonate in case of one-part NPC-7 compared to one-part NPC-3 powder. This is compatible with the finding of Otsuki et al. [34], which stated that the cement with relatively low water to cement ratio is more resistive to carbonation. Fig. 10(a) and (b) shows the TG/DTG thermograms of onepart NPC-3 powder and pastes of the same mixture at 1 day (one-part NPC-3/1) and 28 days (one-part NPC-3/28) of curing as well as the hardened cement pastes of two-part NPCP-8 at 28 days (two-part NPC-8/28). Obviously, the weight loss of onepart NPC-3/28 paste is higher than that of NPC-3/1 and NPC-3 powders, due to the continuance activation of slag after water addition to cement powder. This can be observed from the increase in the intensity of CSH and Ht peaks with curing time. In contrast, NPC-1/28 paste shows the lowest amounts of CSH and Ht. The endothermic peak related to carbonate containing phases at the range of 600–800 1C is observed in one-part NPC3 powder mixture disappears in cement pastes of the same mixture at 1 and 28 days of curing. This proves that, the carbonate containing phases in cement powder are mainly sodium carbonate (Na2CO3). After the addition of water to cement powder, Na2CO3 is consumed as a result of activation and dissolution of the unreacted BFS that exists in one-part cement powder. Finally, there is a marginal decrease in the weight loss of one-part NPC-3/28 cement paste compared to the conventional two-part NPC-8/28. This is compatible with FTIR and FCompressive results.
The total weight loss (WL), including chemically combined water content (Wn, %) is often used as a measure of the hydration progress of cement clinker phases [35]. The TG weight losses (Δm) of different one-part NPC-cement powders and the resulting one-part NPC-3 cement paste at different curing times as well as conventional two-part NPC-8 cement paste at 28 days of hydration are summarized in Table 3. It can be seen that the increase of SH and water contents significantly increases the formation of hydration products during preparation of one-part NPC powder. The increase in the WL of onepart NPC-3 paste with curing time proves the continuation of slag activation after water addition to one-part NPC-3 powder. The value of WL related to CSH and Δm of conventional twopart NPC-8 paste is near and close to that of the one-part NPC3 at 28 days of curing, confirming the FCompressive results. 3.4. Scanning electron microscopy (SEM). Fig. 11 displays the SEM-micrographs of cement powder prepared from different mixes. The micrograph of cement powder derived from NPC-1 presents high amount of unreacted slag (US). The amount of reacted slag (RS) increases with sodium hydroxide (SH) content and W/BFS ratio. This can be observed from the microstructure of NPC-3, NPC-5 and NPC-7 powders. This proves that as the amount of SH and W/BFS ratio increases, the dissolution rate also increases, forming more hydration products. After the addition of water to cement powder the activation reaction is continued to dissolve the remainder
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
8
50 µm
50 µm
50 µm Fig. 12. SEM micrographs (a) NPC-3 cement powder, (b) one-part NPC-3 cement paste at 28 days (NPC-3/28) and (c) cement pastes of two-part NPC-8 mix at 28 days of curing.
slag, resulting in Si4 þ , Al3 þ and Ca2 þ formation followed by their condensation, yielding a more homogeneous and compact microstructure. The hardened one-part NPC-3 paste cured at 28 days (Fig. 12b) seems to be more homogeneous with dense microstructure compared to one-part NPC-3 powder (Fig. 12a), suggesting the continuation of slag activation after addition water to cement powder. Clearly, there is no great difference between the microstructures of one-part NPC-3 paste and two-part NPC-8 paste (Fig. 12c) in the homogeneity and compaction, confirming the FTIR, TG/DTG and FCompressive results.
2.
3.
4. Conclusions 4. According to the results gathered throughout this work, it can be concluded that 1. A suitable non-Portland cement (NPC) powder which can be used as an alternative for Portland cement was prepared in this study. Ground granulated blast-furnace slag (BFS) was dry-mixed with NaOH at W/BFS ratios, which make the paste in non-workable form and is dried in an oven to yield solidified product. The solidified material was crushed to produce powder with a fixed particle size. In order to
5.
examine the continuation probability of cement powder activation, the water was added to it at W/NPC of 0.25. The cement powder performance is strongly affected by NaOH dosages and W/BFS ratio; where the addition of too high or too low NaOH content potentially reduced the continuation probability activation after the addition of water to NPC powder. Also, as the amount of W/BFS ratio decreases, the resulting cement powder shows a good performance for the activation continuance. The amount of dissolved calcium aluminosilicate materials increases with NaOH content and W/BFS ratio as confirmed by TG/DTG, FTIR and SEM techniques. The hardened cement paste with highest compressive strength (FCompressive) was obtained in case of the activation of NPC powder that was prepared by the addition of 6 wt% NaOH to BFS at W/BFS ratio of 0.1. The compressive strength gap between one-part and conventional two-part NPC cement pastes increases with the amount of water used in one-part NPC powder preparation. Hence, the FCompressive of conventional two-part NPC-8 mixture, as counter-part to one-part NPC-3, NPC-6 and NPC-7 mixtures was higher than that of one-part NPC-3, NPC-6 and NPC-7 mixtures by 10%, 22% and 35%, respectively.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096
H.A. Abdel Gawwad et al. / Ceramics International ] (]]]]) ]]]–]]]
6. The increase in the TG/DTG-peak intensity of CSH as well as the bonding-peak intensity of Si–O–Al, H–O–H and O–H in FTIR spectroscopy proves that, after the addition of water to one-part NPC powder, the activation process continued. Also, the formation of dense and compact microstructure as seen in SEM micrographs is relevant enough for the same behavior.
References [1] P. Duxson, J.L. Provis, G.C. Lukey, The role of inorganic polymer technology in the development of “green concrete”, Cem. Concr. Res. 37 (2007) 1590–1597. [2] C.J. Shi, A. Fernandez-Jimenez, A. Palomo, New cements for the 21st century: the pursuit of an alternative to Portland cement, Cem. Concr. Res. 40 (2011) 750–763. [3] P. Duxson, A. Fernandez-Jimenez, J.L. Provis, Geopolymer technology: the current state of the art, J. Mater. Sci. 42 (2007) 2917–2933. [4] A.O. Purdon, The action of alkalis on blast-furnace slag, J. Soc. Chem. Ind. – Trans. Commun. 59 (1940) 191–202. [5] I. Teoreanu, A. Volceanov, S. Stoleriu, Non-Portland cements and derived materials, Cem. Concr. Res. 27 (6) (2005) 650–660. [6] Anne E. Oberlink, Non-Portland Cement Activation of Blast Furnace Slag (Master's theses), University of Kentucky, UK, 2010. [7] H. El-Didamony, A.A. Amer, H. Abd El-Aziz, Properties and durability of alkali-activated slag pastes immersed in sea water, Ceram. Int. 38 (2012) 3773–3780. [8] H. El-Didamony, A.A. Amer, T.M. El-Sokkary, H. Abd-El-Aziz, Effect of substitution of granulated slag by air-cooled slag on the properties of alkali activated slag, Ceram. Int. 39 (2013) 171–181. [9] C. Bilim, O. Karahan, C. Duran Atiş, S. İlkentapar, Effects of chemical admixtures and curing conditions on some properties of alkali-activated cement less slag mixtures, KSCE J. Civ. Eng. 19 (3) (2015) 733–741. [10] C. Cartwright, F. Rajabipour, A. Radlińska, Shrinkage characteristics of alkali-activated slag cements, J. Mater. Civ. Eng. 27 (7) (2015) B4014007. [11] I.G. Richardson, A.R. Brough, G.W. Groves, C.M. Dobson, The characterization of hardened alkali activated blast-furnace slag pastes and the nature of calcium silicate hydrate (C–S–H) paste, Cem. Concr. Res. 24 (1994) 813–829. [12] S.D. Wang, K.L. Scrivener, Hydration products of alkali-activated slag cement, Cem. Concr. Res. 25 (1995) 561–571. [13] J.L. Provis, J.S.J. van Deventer, Introduction to geopolymer, in: J. L. Provis, J.S.J. van Deventer (Eds.), Geopolymers: Structures, Processing, Properties And Industrial Applications, Woodhead Publishing, Abingdon UK, 2009, pp. 1–11. [14] J. Davidovits, Calcium based geopolymer, in: J. Davidovits (Ed.), Geopolymer Chemistry and Applications, 3rd ed., Geopolymer Institute, Saint Quentin, France, 2011, pp. 201–244. [15] W. Hongling, L. Haihong, Y. Fengyuan, Synthesis and mechanical properties of metakaolinite-based geopolymer, Colloids Surf. A: Physicochem. Eng. Asp. 268 (2005) 1–6. [16] J. Davidovits, Geopolymer chemistry and applications, in: J. Davidovits (Ed.), Development of user – friendly systems, 3rd ed., Geopolymer Institute, Saint Quentin, France, 2011, pp. 433–443.
9
[17] D. Koloušek, J. Brus, M. Urbanova, J. Andertova, V. Hulinsky, J. Vorel, Preparation, structure and hydrothermal stability of alternative (sodium silicate-free) geopolymers, J. Mater. Sci. 42 (22) (2007) 9267–9275. [18] M.X. Peng, Z.H. Wang, S.H. Shen, Q.G. Xiao, Synthesis, characterization and mechanisms of one-part geopolymeric cement by calcining lowquality kaolin with alkali, Mater. Struct./Mater. et Constr. 48 (3) (2014) 699–708. [19] A. Hajimohammadi, J.L. Provis, J.S.J. Deventer, One-part geopolymer mixes from geothermal silica and sodium aluminate, Ind. Eng. Chem. Res. 47 (23) (2008) 9396–9405. [20] D. Feng, J.L. Provis, J.S.J. Deventer, Thermal activation of albite for the synthesis of one-part mix geopolymers, J. Am. Ceram. Soc. 95 (2) (2012) 565–572. [21] B. Nematollahi, J. Sanjayan, F.U.A. Shaikh, Synthesis of heat and ambient cured one-part geopolymer mixes with different grades of sodium silicate, Ceram. Int. 41 (4) (2015) 5696–5704. [22] K.H. Yang, J.K. Song, Workability loss and compressive strength development of cementless mortars activated by combination of sodium silicate and sodium hydroxide, J. Mater. Civ. Eng. 21 (3) (2009) 119–127. [23] Z. Abdollahnejad, F. Pacheco-Torgal, J.B. Aguiar, C. Jesus, Durability performance of fly ash based one-part geopolymer mortars, Key Eng. Mater. 634 (2015) 113–120. [24] Z. Zuhua, Y. Xiao, Z. Huaju, C. Yue, Role of water in the synthesis of calcined kaolin-based geopolymer, Appl. Clay Sci. 43 (2009) 218–223. [25] ASTM C109M, Standard test method for compressive strength of hydraulic cement mortars, 2012. [26] J.L. Provis, J.S.J. van Deventer, Geopolymer synthesis and kinetics, in: J. L. Provis, C.A. Rees (Eds.), Geopolymers: Structures, Processing, Properties and Industrial Applications, Woodhead Publishing, Abingdon UK, 2009, pp. 118–136. [27] T. Bakharev, J.G. Sanjayanand, Y.B. Cheng, Sulfate attack on alkali activated slag concrete, J. Cem. Concr. Res. 32 (2002) 211–216. [28] J.R. Sweet, W.B. White, Study of sodium silicate glasses and liquids by infrared reflectance spectroscopy, Phys. Chem. Glas. 10 (6) (1969) 246–251. [29] W. Mozgawa, J. Deja, Spectroscopic studies of alkaline activated slag geopolymers, J. Mol. Struct. 924–926 (2009) 434–441. [30] P.C. Hewlett, Lea's Chemistry of Cement and Concrete, Elsevier Science & Technology Books, 2004. [31] I. Ismail, S.A. Bernal, J.L. Provis, S. Hamdan, J.S.J. van Deventer, Microstructural changes in alkali activated fly ash/slag geopolymers with sulfate exposure, Mater. Struct. 46 (2013) 361–373. [32] S.D.S. Wang, K.L.K. Scrivener, P. Pratt, Factors affecting the strength of alkali-activated slag, Cem. Concr. Res. 24 (1994) 1033–1043. [33] F. Jin, K. Gu, A. Al-Tabbaa, Strength and drying shrinkage of reactive MgO modified alkali-activated slag paste, Constr. Build. Mater. 51 (2014) 395–404. [34] N. Otsuki, S. Ronaldo Gallardo, T. Annaka, S. Takaki, T. Nishida, Field survey for carbonation depth of reinforced concrete buildings in the Philippines, in: Proceedings of the 37th Conference on Our World in Concrete & Structures, Singapore, 2012. [35] M. Ben Haha, G.L. Saout, F. Winnefeld, B. Lothenbach, Influence of activator type on hydration kinetics, hydrate assemblage and microstructural development of alkali-activated blast-furnace slags, Cem. Concr. Res. 41 (2011) 301–310.
Please cite this article as: H.A. Abdel Gawwad, et al., Preparation and characterization of one-part non-Portland cement, Ceramics International (2015), http: //dx.doi.org/10.1016/j.ceramint.2015.08.096