Index

Abstract

This study assessed the effects of different contents of two refined phosphogypsums (PG1, PG2) on standard consistency, setting times, compressive strength and sulfate resistance of cement. PG2 possesses higher water-soluble impurities such as P2O5, F content than PG1. The results were compared to those of cement containing natural gypsum (NG). The results showed that water demand of cement reduces when the content of the gypsums increases, especially with PG1 and NG. The phosphogypsums prolong the setting times of cement. The higher the gypsum content is, the more the 3-d compressive strength of cement can obtains. The low impurity phosphogypsum (PG1) accelerates the 3-d strength but lowers the 91-d strength of cement compared to the high impurity phosphogypsum (PG2). The strength of cement containing NG is always between those of the phosphogypsums at the ages of 3 and 91 days. There is a suitable content of gypsum to produce the highest strength of the cement at the ages of 28 and 91 days. Deterioration of the cement in sodium sulfate solution (50g/l) is more severely when increasing gypsum content. The sulfate resistance of cement containing PG1 is lower than that of cement containing PG2 or NG.

Keywords: Portland cement,Natural gypsum,Phosphogypsum ,Setting times,Compressive strength,Sulfate resistance

Received:16 November 2018 / Revised: 21 December 2018 / Accepted:29 January 2019/ Published:13 March 2019

Contribution/ Originality

This study is one of very few studies in Vietnam to investigatewhich have investigated the potential for using refined phosphogypsum to replace natural gypsum in cement for environmental protection in Vietnam. This study has reported that phosphogypsum content and impurities of P2O5 and F- significantly affect the properties of Portland cement.


1. INTRODUCTION

Phosphogypsum (PG) is a by-product from manufacturing process of phosphoric acid in phosphate fertilizer plants and chemical industries. About 5.4 tons of PG is generated when each ton of phosphoric acid is produced by the wet process. PG consists of 65÷70% gypsum, 25÷30% water and 5÷10% impurities, i.e. phosphoric acid and its salts, hydrofluoric acid and its compounds, R2O3 (Al2O3+ Fe2O3), quartz, apatite, alkali, organic matter, and others. The P2O5 and F impurities are found in three different forms: on the surface of gypsum crystals as water-soluble compounds (H3PO4, Ca (H2PO4)2·H2O, H2SiF6), substituted in the lattice of gypsum crystals (effectively solid solutions of CaHPO4·2H2O, SrSO4 or Na2SiF6) and as insoluble compounds, i.e. apatite and quartz. It has been estimated that approximately 100÷280 million tons of phosphogypsum has been disposed annually but only 15% is refined and reused (Leškevičienė and Nizevičienė, 2010). Phosphogypsum is one of the mineral wastes that is accumulated in large a mounts all over the world causing serious environmental problems. Using refined PG in the manufacture of building materials is economical and ecological.

The most volume of PG can be utilized in the construction industry. Refined PG can be used as an ingredient of plaster (Singh, 2002; Singh, 2005) or as a replacement of natural gypsum (NG) which plays the role of a set retarder in Portland cement (Tabikh and Miller, 1971; Mehta and Brady, 1977; Olmez and Erdem, 1989; Akın and Sert, 2004; Bhadauria and Thakare, 2006; Chandara et al., 2009; Islam et al., 2017). Basically, the role of PG in Portland cement is the same as NG. However, the impurities such as P2O5; F-… in PG will effect on properties of cement, cement mortar and concrete. Akın and Sert (2004) studied the effect of PG or natural gypsum content on properties of Portland cement. The results showed that increasing content of the gypsums will increase the consistency and setting times of the cement. Both initial and final setting times of cement containing PG are significantly longer than that of cement containing NG. It seems that compressive strength at the ages of 7 and 28 days obtains maximum at the content of gypsum about 3-5%. Both the gypsums can induce the similar strength of sample at the same content of gypsum used. Islam et al. investigated the effect of phosphogypsum addition with cement clinker on properties of paste, mortar and concrete (Islam et al., 2017). The results indicated that setting times of cement containing unprocessed PG (field condition with high impurities content) exceed the setting times of the control sample and strongly increase when the content of the PG increases. Meanwhile, when using processed phosphogypsum, the higher the content of the PG, the longer the setting times. The setting times of the sample containing 10% processed phosphogypsum are the same as those of the control sample. The flowability of mortar initially increases and then decreases when the content of the phosphogypsum in cement increases. Compressive strength of mortar and concrete containing processed phosphogypsum is always higher than that of sample containing unprocessed phosphogypsum. Processed sample conducted better or at least similar performance to the control samples (100% clinker) in paste, mortar and concrete at 10% addition level (Islam et al., 2017).

In Vietnam, PG is being disposed and stored into open areas with total about six million tons. It is estimated that annual production of PG from phosphoric acid and fertilizer plants in Vietnam in 2020 is approximately three and a half million tons. However, only about one million tons of PG are refined and reused in the construction industry. Meanwhile, Vietnam imports about four million tons of natural gypsum for producing of ninety million tons of cement in a year. Therefore, using PG to replace natural gypsum as a set retarder in Portland cement is needed to solve the environmental and disposal problems, and to get economical benefits.

This study investigated the effects of content of two types PG with different impurities content on the properties such as standard consistancy, setting times, compressive strength and sulfate resistance of Portland cement. The results were compared with those of sample containing natural gypsum. The study results are expected to promote the using of PG to replace natural gypsum in cement industry in Vietnam.

2. MATERIALS AND METHODS

2.1. Materials

A Portland cement clinker, natural gypsum and two types of phosphogypsum (PG1 and PG2) were used in this study. Chemical compositions of the clinker can be found in Table 1. PG1 and PG2 are samples taken from different stages of refinement process of Dinh Vu gypsum JSC. In Vietnam. PG2 is in powder form taken after water washing and drying. Meanwhile, PG1 is a commercial product which is completely processed and granulated to change the water-soluble impurities such as P2O5, F-… into water-insoluble. Therefore, the difference in chemical compositions of PG1 and PG2 are mainly in the water-soluble content of P2O5 and F- Table 2.

Table-1. Chemical compositions of clinker.

Materials
CaO
SiO2
Al2O3
Fe2O3
MgO
L.O.I
Clinker (wt%)
64.6
21.9
4.62
3.39
1.18
0.85

L.O.I-  Loss on Ignition.

2.2. Experimental Methods

Fifteen Portland cement samples were produced by grinding clinker and the 3 gypsums in a batch vibration ball mill with the same material volume and grinding period. The gypsum contents in the cement samples are calculated based on the SO3 content. The fineness of the samples is about 3010-3360 cm2/g Table 3.

Table-2. Chemical compositions of gypsums.

Materials
CaO
SO3
H2O
SiO2
R2O3
P2O5(*)
F- (*)
L.O.I
I.C
NG (wt%)
31.92
45.49
18.99
-
-
-
-
-
1.52
PG1 (wt%)
29.68
40.73
16.44
9.21
0.62
0.04
0.01
18.93
10.22
PG2 (wt%)
29.40
39.35
16.47
9.16
0.68
0.24
0.11
18.83
10.34

L.O.I-  Loss on Ignition.
I.C- Insoluble Compound.

(*)- Water-soluble content.

Table-3. SO3, gypsum content and fineness of different cement samples.

N°
SO3content (%)
Gypsum content (%)
Fineness of cement (cm2/g)
NG
PG1
PG2
NG
PG1
PG2
1
1.32
2.9
3.24
3.35
3010
3160
3200
2
1.57
3.55
3.85
3.99
3180
3210
3360
3
1.82
4.0
4.47
4.6
3150
3210
3250
4
2.32
5.1
5.70
5.9
3200
3310
3150
5
2.82
6.2
6.90
7.2
3180
3220
3280

The standard consistency, setting times of the cement samples were tested with a Vicat needle as specified in the Vietnam Standard (2015) conforming to ISO 9597:2008. For compressive strength of cement, mortar samples (40x40x160mm3) were prepared with cement:sand:water ratio of 1:3:0.5 in accordance with the VS (2011) conforming to ISO 679:2009.

The effects of different gypsum on sulfate resistance of the cement sample were tested in accordance with the VS (2007). 50x50x285mm3 samples were immersed in sodium sulfate 50g/l solution with the volume of the solution and mortar of 4 at 27±1°C. The weight and length changes of 3 specimens in the solution were measured after 1, 2, 3, 4 and 8 weeks. The solution was renewed after the length change measurement.

3. RESULT AND DISCUSSIONS

3.1. Standard Consistency and Setting Times

Standard consistency and setting times of cements containing different content of NG, PG1 and PG2 are showed in Table 4 and Figure 1, 2, 3. The results indicate that increasing NG or PG1 in cement results in reducing of the standard consistency of the cement. And the effect of NG on the standard consistency of the cement is more significantly than that of PG1 Figure 1. Meanwhile, the water demand of cement containing PG2 does not significantly change and is clearly lower than that of cement containing NG or PG1.

Table-4. Standard consistency and setting times of different cement samples.

N°
SO3content (%)
Standard consistency (%)
Initial setting time (min)
Final setting time (min)
NG
PG1
PG2
NG
PG1
PG2
NG
PG1
PG2
1
1.32
30.0
29.5
26.0
120
85
135
215
165
270
2
1.57
29.5
29.0
25.0
115
110
130
210
220
280
3
1.82
28.5
29.0
25.5
125
120
145
195
230
285
4
2.32
26.5
28.5
26.5
120
155
145
200
265
285
5
2.82
25.5
27.0
25.5
115
205
155
210
310
290

Figure-1. Effects of gypsum on standard consistency.

Figure-2. Effects of gypsum on initial setting time

Figure-3. Effects of gypsum on final setting time.

Generally, the setting times of the cement increase when PG content in cement increases. But changing NG content in range of this study does not change both the setting times Table 4 and Figure 2, 3. For PG1, the initial and final setting times of cement containing 3.24% PG1 (1.32% SO3) are 85 and 165 min, clearly shorter than those of cement containing NG and PG2, respectively. However, the setting times of the sample containing 6.9% PG1 (2.82% SO3) increase up to 205 and 310 min, significantly exceed those of the samples containing NG or PG2. The cement containing 3.35% PG2 (1.32% SO3) possesses the initial and final setting times of 135 and 270 min, respectively. When PG2 increases from 3.35% (1.32% SO3) to 7.2% (2.82% SO3), both the setting times of cement increase only 20 minutes Table 4.

It should be noted that the water-soluble form of P2O5 and F- in PG2 are higher than those in PG1 and NG. It means that water-soluble P2O5 and F- in phosphogypsum induce longer setting times of cement. The water demand of cement reduces or the workability of cementitious mixture increases when the content of gypsum or impurities in cement increases. These results are in good agreement with those of previous studies (Akın and Sert, 2004; Islam et al., 2017) where the setting times of cement containing high water-soluble P2O5 and F- impurities are longer. For instance, Sadiqul Islam showed that the water demand of the mortar initially decreases when the content of the phosphogypsum in cement increases (Islam et al., 2017).

3.2. Compressive Strength

Compressive strength at the ages of 3, 28 and 91 days of mortar can be found in Table 5 and Figure 4, 5, 6. The 3-d compressive strength results at Figure 4 show that increasing the gypsum content in cement will accelerate the early strength of the mortar. Comparing to natural gypsum, the low impurities content in phosphogypsum (PG1) enhances the strength of the mortar at the early age but it will be reduced by phosphogypsum with high impurities content PG2, Figure 4. In contrast, compressive strength of samples containing PG2 exceeds that of samples containing NG and PG1 possesses the lowest 91-d compressive strength Figure 6. There is an optimum gypsum content in cement which produces the highest strength of the mortar at the ages of 28 and 91 days Figure 5 and 6. These results are also similar to those reported by Akın and Sert (2004).

Table-5. Compressive strength of different cement samples.

N°
SO3content (%)
3-d compressive strength (MPa)
28-d compressive strength (MPa)
91-d compressive strength (MPa)
NG
PG1
PG2
NG
PG1
PG2
NG
PG1
PG2
1
1.32
19.0
27.6
18.0
42.9
41.8
44.0
51.5
49.0
56.2
2
1.57
21.3
31.6
20.2
43.5
43.5
44.3
55.2
52.8
57.0
3
1.82
24.9
32.5
21.8
43.7
45.1
45.0
55.5
53.4
60.7
4
2.32
25.2
33.0
22.6
44.6
44.6
45.1
54.1
50.7
58.3
5
2.82
27.7
34.4
23.5
40.2
44.1
44.2
52.5
47.1
57.7

 

Figure-4. Effects of gypsum on 3-d compressive strength.

Figure-5. Effects of gypsum on 28-d compressive strength.

Figure-6. Effects of gypsum on 91-d compressive strength.

3.3. Sulfate Resistance

Results of weight and length changes of samples containing different contents of gypsums in the sodium sulfate solution (50g/l) up to 8 weeks are given in Figure 7 to Figure 9. It shows clearly that the longer the sample is in the solution, the more the weight increases and the larger the length changes. The increasing gypsum content in cement induces lower durability of the sample in the solution, i.e. absorbing more Na2SO4 and showing larger expansion. It should be noted that the expansion of the samples containing 1.32 SO3 of phosphogypsums begins to exceed that of the sample containing 1.82 SO3 of phosphogypsums after 8 weeks in the solution Figure 8 and 9.

Comparing the results of the samples at the same SO3 content between different gypsums (i.e. NG, PG1 and PG2 in Figure 7 to Figure 9) indicates that there is no significant difference in weight change between the three gypsums. The expansion of the samples containing PG1 is largest Figure 8. The samples containing low contents of NG (i.e. 1.32% and 1.82% SO3) have the length change slightly higher than the samples containing low contents of PG2, respectively. Meanwhile, the length change of the sample containing high content of NG (i.e. 2.82% SO3) is lower than that of the sample containing high content of PG2.

Figure-7. Effects of NG content on weight and length change of mortar sample.

Figure-8. Effects of PG1 content on weight and length change of mortar sample.

Figure-9. Effects of PG2 content on weight and length change of mortar sample.

4. CONCLUSIONS

The following conclusions can be drawn from the results of this study:

Funding: This study as part of the research project “Studying the effects of phosphogypsum from fertilizer plants to replace natural gypsum on properties of cement and concrete”, B2017-XDA-14 that received the financial support from the Vietnam Ministry of Education and Training.

Competing Interests: The authors declare that they have no competing interests.

Contributors/Acknowledgement: The authors would like to thank the Vietnam Ministry of Education and Training for the financial support.

REFERENCES

Akın, A.I. and Y. Sert, 2004. Utilization of weathered phosphogypsum as set retarder in Portland cement. Cement and Concrete Research, 34(4): 677-680.Available at: https://doi.org/10.1016/j.cemconres.2003.10.017 .

Bhadauria, S.S. and R.B. Thakare, 2006. Utilisation of phospho gypsum in cement mortar and concrete. Conference on Our Word in Concrete & Structure, Singapore.

Chandara, C., K.A.M. Azizli, Z.A. Ahmad and E. Sakai, 2009. Use of waste gypsum to replace natural gypsum as set retarders in portland cement. Waste Management, 29(5): 1675-1679.Available at: https://doi.org/10.1016/j.wasman.2008.11.014 .

Islam, G.S., F.H. Chowdhury, M.T. Raihan, S.K.S. Amit and M.R. Islam, 2017. Effect of phosphogypsum on the properties of Portland cement. Procedia Engineering, 171: 744-751.

Leškevičienė, V. and D. Nizevičienė, 2010. Anhydrite binder calcined from phosphogypsum. Ceramics-Silikaty, 54(2): 152-159.

Mehta, P. and J.R. Brady, 1977. Utilization of phosphogypsum in Portland cement industry. Cement and Concrete Research, 7(5): 537-544.Available at: https://doi.org/10.1016/0008-8846(77)90115-6 .

Olmez, H. and E. Erdem, 1989. The effects of phosphogypsum on the setting and mechanical properties of Portland cement and trass cement. Cement and Concrete Research, 19(3): 377-384.Available at: https://doi.org/10.1016/0008-8846(89)90027-6 .

Singh, M., 2002. Treating waste phosphogypsum for cement and plaster manufacture. Cement and Concrete Research, 32(7): 1033-1038.Available at: https://doi.org/10.1016/s0008-8846(02)00723-8 .

Singh, M., 2005. Role of phosphogypsum impurities on strength and microstructure of selenite plaster. Construction and Building Materials, 19(6): 480-486.Available at: https://doi.org/10.1016/j.conbuildmat.2004.07.010 .

Tabikh, A. and F. Miller, 1971. The nature of phosphogypsum impurities and their influence on cement hydration. Cement and Concrete Research, 1(6): 663-678.Available at: https://doi.org/10.1016/0008-8846(71)90020-2 .

Vietnam Standard, 2015. TCVN 6017:2015 cement – test methods – determination of setting time and soundness (in Vietnamese).

VS, 2007. TCVN 7713:2007 cement – determination of length change of hydraulic cement mortars exposed to a sulfate solution (in Vietnamese).

VS, 2011. TCVN 6016:2011 cement – test methods – determination of strength (in Vietnamese).

Views and opinions expressed in this article are the views and opinions of the author(s), International Journal of Sustainable Energy and Environmental Research shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content.