Ri

LF

a

ARR1AA

KPCRFI

1

da[pcOsndfisOri

pcartd

0d

Thermochimica Acta 526 (2011) 163– 168

Contents lists available at SciVerse ScienceDirect

Thermochimica Acta

jo ur n al homepage: www.elsev ier .com/ locate / tca

eaction mechanism and influence factors analysis for calcium sulfide generationn the process of phosphogypsum decomposition

iping Ma ∗, Xuekui Niu, Juan Hou, Shaocong Zheng, Wenjuan Xuaculty of Environmental Science and Engineering, Kunming University of Science and Technology, Kunming 650093, China

r t i c l e i n f o

rticle history:eceived 14 July 2011eceived in revised form4 September 2011ccepted 20 September 2011vailable online 29 September 2011

a b s t r a c t

FactSage6.1 software simulation and experiments had been used to analysis the reaction mechanism andinfluence factors for CaS generation during the process of phosphogypsum decomposition. Thermody-namic calculation showed that the reaction for CaS generation was very complex and CaS was generatedmainly through solid–solid reaction and gas–solid reaction. The proper CO and CO2 have benefit forimproving the decomposition effects of phosphogypsum and reducing the generation of CaS at 1100 ◦C.Using high sulfur concentration coal as reducer, the proper reaction conditions to control the generation

eywords:hosphogypsumalcium sulfideeaction mechanismactSage6.1

of CaS were: the coal particle size was between 60 mesh and 100 mesh, reaction temperature was above1100 ◦C and the heating rate was 5 ◦C/min. Experimental and theoretical calculation indicated that theconcentration of CaS was only ten percents in the solid product at 1100 ◦C, which is favorable for thefurther cement producing using solid production.

Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved.

nfluence factors

. Introduction

Phosphogypsum (PG) is a by-product of wet phosphoric aciduring which phosphate rock is reacted with sulfuric acid. It islmost over 40 million tons of PG produced in China each year1]. Which has become the primary barrier for the development ofhosphate fertilizer industry because of its complicated chemicalomposition, large-tonnage output, and environment harm [2–7].ne of recycling methods for PG is thermal decomposition to reuse

ulfur and produce cement [8–11]. Because of its complex compo-ents, many middle-products are produced during the process ofecomposition [12,13]. Especially the generation of calcium sul-de (CaS), which has disadvantageous effect on the recovery ofulfur dioxide (SO2) and the further reuse of solid products [14–18].h and Wheelock [19] and Talukdar et al. [20] had done some

esearches on explaining the generation of CaS. It is important tonvestigate the mechanism of CaS generation for the reuse of PG.

In the previously research of our group, the decompositionrocess of phosphogypsum in a nitrogen atmosphere at differentonditions had been studied with high sulfur concentration coals reducer [21,22]. In this study, chemical thermodynamic equilib-

ium calculation by the Equilibrium model of FactSage6.1 was usedo elucidate the possible reaction mechanism and phase transformuring the decomposition process of PG at different conditions. As

∗ Corresponding author. Tel.: +86 871 5170905; fax: +86 871 5170906.E-mail address: lpma2522@hotmail.com (L. Ma).

040-6031/$ – see front matter. Crown Copyright © 2011 Published by Elsevier B.V. All rioi:10.1016/j.tca.2011.09.013

comparing, experiments were taken as the same conditions as insimulation calculation for the decomposition of PG. The purposeof this study is to prove the reaction mechanism of CaS generationand its effect on the decomposition of PG.

2. Mechanism analysis

Because of the complex components of PG, many reactions couldtake place during the process of PG decomposition. The followingreactions about CaS generation and transformation could take placewith coal in nitrogen atmosphere:

CaSO4 + 4C → CaS + 4CO (1)

CaSO4 + 2C → CaS + 2CO2 (2)

4CO + CaSO4 → CaS + 4CO2 (3)

4CaO + 3S2(g) → 4CaS + 2SO2 (4)

3CaSO4 + CaS → 4CaO + 4SO2 (5)

CaSO4 + 3CaS → 4CaO + 4S (6)

CaS + 3CaSO4 + 4CO2 → 4CaCO3 + 4SO2 (7)

CaS + 2SO2 → CaSO4 + 2S (8)

CaS + 3SO3 → CaO + 4SO2 (9)

The beginning temperature of the possible reactions takingplace was listed in Table 1. Thermodynamic results calculated by

ghts reserved.

164 L. Ma et al. / Thermochimica Acta 526 (2011) 163– 168

Table 1�rG and �rH of reactions at possible beginning temperatures.

Reaction T (◦C) �rG (kJ/mol) �rH (kJ/mol)

(1) 500 −38.135 513.247(2) 200 −0.093 173.735(3) 100 −177.036 −172.356(4) 100 −289.971 −331.996(5) 1200 −30.996 936.345(6) 1500 209.079 316.825(7) 1500 145.735 163.917

FcBtace

s(ia1thr

3

3

sFKncwPip(ms

3

(gd

atan5p

cc2

(8) 100 −224.302 −364.913(9) 100 −237.927 −161.925

actSage6.1 show that all reactions could be occurred theoreti-ally except reactions (6) and (7). CaS is produced by reaction (2).esides, an important reaction path is CaSO4 reacting with CO (reac-ion (3)), in which CO is produced by vaporization reaction of coalnd reaction (1). During the decomposition process of PG with coal,oal is very easily vaporized to produce CO therefore reaction (3) isasy take place to generate CaS.

The elimination process of CaS is mainly achieved througholid–solid reaction and gas–solid reaction such as reactions5)–(9). Table 1 indicates that the gas–solid reactions for CaS elim-nation (reactions (6)–(9)) are exothermic reactions and occurredt very low temperature except reaction (5), which is occurred at200 ◦C and reaction heat is 936.345 kJ/mol. At the high tempera-ure, reactions (8) and (9) are limited while reaction (5) could beappened mostly. This indicates that reaction (5) will be the keyeaction of CaS elimination under high temperature.

. Experimental

.1. Sample preparation

PG is a chemical compound, which mainly consists of calciumulfate dihydrate (CaSO4·2H2O) and some impurities such as P2O5,−, organic substances and some trace metals (e.g. Mn, Al, Fe, Na,a). The samples of PG used in this study were collected from Yun-an Natural Gas and Chemical Engineering Company; the mainhemical compounds were listed in Table 2, the size of PG samplesas about 0.074 mm. After drying in tube furnace at 60 ◦C for 12 h,

G was used in experiment. The high sulfur concentration coal usedn the experiment come from Yunnan Chuxiong; the main com-ounds were detected by standard of Chinese coal classificationGB476-1979) and the results were listed in Table 3. After treat-

ent, coal sample also was crushed into particles with differentizes (200, 160, 100, and 60 mesh) before experiment.

.2. Experiment methods and equipments

The equipments used in this study were: KTL1600 tube furnaceNanjing University Experiment Factory), KM9106 complex fuelas analyzer (KANE Company), D/max-3BPEX-P96 powder X-rayiffract meter (Rigaku).

Decomposition reaction of PG was carried out in a tube furnacet nitrogen atmosphere under different coal particle size, reac-ion temperature and heating rate. Temperature programming wasdopted for heating process, heating rate was 8 ◦C/min from begin-ing to 800 ◦C firstly, above that, using heating rate of 3 ◦C/min,◦C/min, 8 ◦C/min, and 10 ◦C/min, respectively. The reaction tem-erature was controlled at 1000 ◦C, 1100 ◦C, and 1200 ◦C.

The solid products were analyzed by XRD to investigate the

omposition. The contents of CaS and CaSO4 were examined byhemical methods according to GB/T 16489-1996 and GB/T 5484-000. Therefore the mass fraction of CaO could be calculated.

Fig. 1. Result of chemical thermodynamic equilibrium calculation at different molarratio of CaSO4/C.

3.3. Theoretical calculation

FactSage software, which contains FACT-Win and ChemSage isa powerful calculation software and it could be used to calculatethe chemical thermodynamic equilibrium under the limitation of avariety of conditions. In this study, the equilibrium model of Fact-Sage6.1 was used to investigate the influence for CaS generationby different effect factors such as reaction atmosphere (N2, CO andCO2), molar ratio of CaSO4/C (ε) and reaction temperature. Accord-ing to the composition of original PG, initial input componentsfor the reaction equilibrium calculations were CaSO4, SiO2, Al2O3,Fe2O3, and the mass was 7.2420 g, 0.943 g, 0.0236 g and 0.0132 grespectively. The inputting composition of high-sulfur coal was C,H, O, N, S and the mass was according to the requirement of ε. Thereaction pressure was 1.01 × 105 Pa.

Reactive degrees of PG decomposition were expressed by thedecomposition rate (ϕ) and the desulfurization rate (�).

ϕ = 1.889mCaS + 1.889mCaS

mCaSO4

� = 2.125mSO2

mCaSO4

where mCaS = theoretical yield of CaS, g; mCaO = theoretical yield ofCaO, g;

4. Results and discussion

4.1. Theoretical calculations and analysis

4.1.1. Effect of the molar ratio of CaSO4/CFig. 1 gives the results calculated by chemical thermodynamic

equilibrium of FactSage at different molar ratio of CaSO4/C. It isclear that both decomposition rate and desulphurization rate arevery high when ε is between 1.5 and 2.2. That means the reaction forthe generation of CaS is hardly to take place under weak reductioncondition. While at high ε, it is very difficult for CaSO4 to decomposecompletely at lacking reducing agent. The decomposition productsaccording to the same calculation at the same conditions are shownin Fig. 2. The mass of CaS is decreased with the increasing of ε,however, the mass of CaO and SO2 are increased with the increasingof ε. These results confirm the above analysis and are consistentwith the results by the research of Ma et al. [14].

4.1.2. Effect of reaction atmosphereFigs. 3 and 4 are the calculation results of PG decomposition at

different amounts of CO and CO2 under 1100 ◦C, which indicate that

L. Ma et al. / Thermochimica Acta 526 (2011) 163– 168 165

Table 2Chemical composition of PG (wt.%).

Composition SO3 CaO SiO2 Hydrotropic F Total F Total P2O5 Hydrotropic P2O5

Content 40.86 29.82 9.43 0.12 0.52 1.17 0.87Composition Fe2O3 Al2O3 Na2O K2O Free water Crystal water Acid-insoluble materialContent 0.132 0.236 0.043 0.086 5.38 4.27 8.42

Table 3The components of coal sample.

Elementary analysis (wt.%)/.daf Proximate analysis (wt.%)/.daf

C H O N S V FC A

rtϕawaeddg

F

Ft

are showed in Figs. 7 and 8 at different temperatures in nitrogen

81.17 4.43 11.22 1.56

eaction atmospheres have an important role on PG decomposi-ion. With the increasing of the mass of CO, the decomposition rate

is increased, while the desulfurization rate � is increased firstlynd then decreased rapidly. However, both ϕ and � are increasedith the increasing of additive amount of CO2. When the additive

mount of CO2 is more than 1.0 g, both ϕ and � reaches to the high-st point. Fig. 5 shows the change of decomposition products at

ifferent additive amount of CO. It is clear that CaS is easily pro-uced in high concentration of CO, in contrary, CaO would be lessenerated at high concentration of CO. Contrarily, it is benefit for

ig. 2. Decomposition products of PG calculation at different molar ratio of CaSO4/C.

ig. 3. Result of chemical thermodynamic equilibrium calculation at different addi-ive amounts of CO (CaSO4/C = 2).

1.62 10.42 63.90 25.68

the generation of CaO and restraining the generation of CaS whenadded CO2 (in Fig. 6).

4.1.3. Effect of reaction temperatureThe chemical thermodynamic equilibrium calculations results

atmosphere. It indicates that with the increasing of temperature,the ϕ and � are increased. When the reaction temperature ishigher than 1150 ◦C, the decomposition reaction almost achieved to

Fig. 4. Result of chemical thermodynamic equilibrium calculation at different addi-tive amounts of CO2 (CaSO4/C = 2).

Fig. 5. Decomposition products of PG at different additive amounts of CO(CaSO4/C = 2).

166 L. Ma et al. / Thermochimica Acta 526 (2011) 163– 168

Fig. 6. Decomposition products of PG at different additive amounts of CO2

(CaSO4/C = 2).

Fig. 7. Result of chemical thermodynamic equilibrium calculation at different tem-p

bttaTtw

Fig. 8. Decomposition products of PG at different temperatures under N2

eratures under N2 (CaSO4/C = 2).

alance. The change of decomposition products also explains thathe suitable decomposition temperature was 1000–1150 ◦C, underhis condition the mass of SO2 and CaO begin to quickly increasend CaS is less generated with the increasing of temperature (Fig. 8).hese means the decomposition reactions will complete when the

emperature was high than 1000 ◦C and the middle product CaSould be restrained.

Fig. 9. Concentration curve of the instantaneous release of SO

(CaSO4/C = 2).

4.2. Experiment results and analysis

4.2.1. Effect of the particle sizeFig. 9 shows the concentration transformation of gas products

SO2 and CO with the temperature at different coal particle size.With the coal particle size increasing, the concentration of SO2increase at high temperature, the proper particle size is between60 mesh and 100 mesh for the product SO2. While for the produc-ing of CO, the proper particle size is 160 mesh. Comparing withthe solid products analysis (Fig. 10), the characteristics absorptionband of CaS is very weak with the increasing of high-sulfur coalparticle size. The coal particle size between 60 mesh and 100 meshis suitable for getting high recovery of CaO. In order to achievethe high recovery of SO2 and CaO, the coal particle size must becontrolled.

In the process of PG decomposition under different high-sulfurcoal particle size, excepting reactions (1) and (2) which could gener-ate CaS, another important way to produce CaS may be the reactionof CaSO4 reacting with CO which produced by vaporization of coal.With the temperature increasing, the small coal particle size couldproduce high concentration of CO at initial stage. Therefore thedecomposition reaction could be proceeded in intense reducingcondition and produce more CaS. With the coal particle size increas-

ing, the coal is difficult to vaporize and reaction (3) could not takeplace therefore less CaS is generated.

2 and CO at different coal particle size (a) –SO2; (b) –CO.

L. Ma et al. / Thermochimica Acta 526 (2011) 163– 168 167

Fig. 10. Powder X-ray diffraction pattern of the solid products at different coalparticle size.

Fa

4

uiwwt(CoC

ig. 11. Powder X-ray diffraction pattern of the solid products at different temper-tures.

.2.2. Effect of the reaction temperatureFig. 11 gives the XRD analysis for decomposition solid prod-

cts at different temperature, the size of coal was 100 mesh. Its clear that the characteristic peak of CaS and CaSO4 are weaken

ith the increasing of reaction temperature, which is consistentith the result of chemical thermodynamic equilibrium calcula-

ion (Fig. 8). At 1200 ◦C, the peak of CaSO4 is almost disappeared

Fig. 11). Both theoretical calculation and experiment prove thataSO4 is completely decomposed and generated the target productf CaO at 1200 ◦C. In order to achieve high concentration of SO2 andaO, the suitable decomposition temperature is 1200 ◦C. From the

Fig. 13. Concentration curve of the instantaneous release of

Fig. 12. Powder X-ray diffraction pattern of the solid products at different heatingrates.

considering of industrial application, this temperature is very highwhich causes the higher energy consumption and limits its practicalapplication and development of this technology. Both theoreticalcalculations and experiments results show that the reaction tem-perature takes an important role on the generation of CaS duringthe process of PG decomposition.

4.2.3. Effect of the heating rateFig. 12 shows the XRD analysis of PG decomposition products at

different heating rate at 1100 ◦C. When heating rate is 5 ◦C/min,the characteristic peak of CaS is lower than at other conditions(3 ◦C/min and 8 ◦C/min). The gas products (SO2 and CO) concen-tration changes with temperature at different heating rates areshowed in Fig. 13. At lower heating rate such as 3 ◦C/min, thereare lower concentration of CO because of lower gasification of coal(Fig. 13b), CaS may be generated mainly through the solid–solidreaction (reactions (1) and (2)). At high heating rate, more CO isproduced which makes reaction (3) taken place easily.

From Fig. 13 it is cleared that the release rate of SO2 is slowwith the increasing of temperature at early reaction stage, at thiscondition, CaSO4 is discomposed to produce CaS and less SO2 isproduced. When reaction temperature higher than 800 ◦C, the con-centration of SO2 is increased suddenly. Comparing with theorycalculation result in Fig. 8, at the same condition, the generationof CaS decreases abruptly and CaO increases quickly, which means

reaction (5) took place. With the increasing of heating rate, the gen-eration of both SO2 and CO are increased, the higher heating ratespur the reactions take place speedily.

SO2and CO at different heating rates (a) –SO2; (b) –CO.

1 ica A

5

tpsrCgpctd6r

A

H2(

R

[

[

[

[

[

[

[

[

[

[

[

[

68 L. Ma et al. / Thermochim

. Conclusion

Both theoretical calculation and experiment results also showhat the mechanism of CaS generation is very complex during therocess of PG decomposition. The theoretical calculation demon-trated that the producing of CaS mainly through solid–solideaction and gas–solid reaction. The proper additive amounts ofO and CO2 benefit for the decomposition. Analysis for solid andas products also indicates that during the process of PG decom-osition, high concentration of SO2 and high concentration of CaOould be got by controlling the coal particle size, reaction tempera-ure and heating rate. 1100 ◦C could be the best temperature for PGecomposition, at the same time, the coal particle size is between0 mesh and 100mesh, using heating rate of 5 ◦C/min are the propereaction conditions to control the generate of CaS.

cknowledgements

Financial support for this project was provided by Nationaligh Technology Research and Development Plan (863 of China,007AA06Z321) and the National Science Foundation of ChinaNSFC21176108).

eferences

[1] R. Li, Comprehensive utilization of ardealite, Yunnan Chemical Technology 34(2007) 74–80.

[2] M. Singh, Effect of phosphatic and fluoride impurities of phosphogypsum onthe properties of selenite plaster, Cement and Concrete Research 33 (2003)1363–1369.

[3] L. Reijnders, Disposal, uses and treatments of combustion ashes: a review,Resources, Conservation and Recycling 43 (2005) 313–336.

[4] M.M. Makweba, E. Holm, The natural radioactivity of the rock phosphates, phos-

phatic products and their environmental implications, The Science of The TotalEnvironment 133 (1993) 99–110.

[5] M. Azouazi, Y. Ouahidi, S. Fakhi, Y. Andres, J.C. Abbe, M. Benmansour, Naturalradioactivity in phosphates, phosphogypsum and natural waters in Morocco,Journal of Environmental Radioactivity 54 (2001) 231–242.

[

cta 526 (2011) 163– 168

[6] W.C. Burnett, A.W. Elzerman, Nuclide migration and the environmental radio-chemistry of Florida phosphogypsum, Journal of Environmental Radioactivity54 (2001) 27–51.

[7] N. Degirmenci, The using of waste phosphogypsum and natural gypsum inadobe stabilization, Construction and Building Materials 22 (2008) 1220–1224.

[8] N. Degirmenci, Utilization of phosphogypsum as raw and calcined material inmanufacturing of building products, Construction and Building Materials 22(2008) 1857–1862.

[9] L. Reijnders, Cleaner phosphogypsum, coal combustion ashes and waste incin-eration ashes for application in building materials: a review, Building andEnvironment 42 (2007) 1036–1042.

10] C.A. Strydom, J.H. Potgieter, Dehydration behaviour of a natural gypsum and aphosphogypsum during milling, Thermochimica Acta 332 (1999) 89–96.

11] M.A. Taher, Influence of thermally treated phosphogypsum on the properties ofPortland slag cement, Resources, Conservation and Recycling 52 (2007) 28–38.

12] K. Qiu, E.J. Anthony, L. Jia, Oxidation of sulfided limestone under the conditionsof pressurized fluidized bed combustion, Fuel 80 (2001) 549–558.

13] E.M. van der Merwe, C.A. Strydom, J.H. Potgieter, Thermogravimetric analysisof the reaction between carbon and CaSO4·H2O, gypsum and phosphogypsumin an inert atmosphere, Thermochimica Acta 340–341 (1999) 431–437.

14] L.P. Ma, P. Ning, S.C. Zheng, X.K. Niu, W. Zhang, Y.L. Du, Reaction mech-anism and kinetic analysis of the decomposition of phosphogypsum via asolid-state reaction, Industrial and Engineering Chemistry Research 49 (2010)3597–3602.

15] W. Zhang, P. Ning, L.P. Ma, Z.G. Fang, M. Jiang, L. Wang, Pilot study on reduc-tive decomposition of phosphogypsum with high sulfur coal in a circulatingfluidized bed, Modern Chemical Industry 29 (2009) 78–81.

16] L.Z. Ma, P. Ning, S. Qing, Study on influence factors between high sulfur coaland phosphogypsum, Acta Sciertiarum Naturalium Universities Sunyatseni,Chinese 48 (2009) 85–88.

17] S.B. Jagtap, T.D. Wheelock, Regeneration of sulfided calcium-based sorbents bya cyclic process, Energy & Fuels 10 (1996) 821–827.

18] H.P. Xun, J.H. Zhu, X.Y. Cheng, H.Y. Fang, C. Jun, C. Kefa, Experiments and modelof the decomposition of CaSO4 under CO atmosphere, Journal of Fuel Chemistryand Technology 33 (2005) 151–154.

19] J.S. Oh, T.D. Wheelock, Reductive decomposition of calcium sulfate with carbonmonoxide: reaction mechanism, Industrial & Engineering Chemistry Research29 (1990) 544–550.

20] J. Talukdar, P. Basu, J.H. Greenblatt, Reduction of calcium sulfate in a coal-firedcirculating fluidized bed furnace, Fuel 75 (1996) 1115–1123.

21] Z.G. Fang, P. Ning, Y.H. Yang, Investigating the effect factors of deoxidizing phos-

phogypsum by compound reducing agent, Chemical Engineering Development28 (3) (2009) 522–527.

22] L.P. Ma, P. Ning, A equipment for phosphogypsum decomposition withcirculating fluidization bed and heat reusing, Patent (China), ZL2008 20081540.7.