An efficient methodology for utilization of K-feldspar and phosphogypsum with reduced energy consumption and CO2 emissions☆

Zhixi Gan ,Zheng Cui,Hairong Yue ,*,Siyang Tang ,Changjun Liu ,Chun Li,Bin Liang ,Heping Xie

1 Multi-phases Mass Transfer and Reaction Engineering Laboratory,College of Chemical Engineering,Sichuan University,Chengdu 610065,China

2 Center of CCUS and CO2 Mineralization and Utilization,Sichuan University,Chengdu 610065,China

1.Introduction

The issues of sustainably utilizing natural mineral resources,dealing with industrial waste with low energy consumption and CO2emissions bring challenges to sustainable developmentin energy and the environment[1].The present process for the utilization of natural K-feldspar and the treatment of industrial waste phosphogypsum consume enormous energy and discharge huge amount of CO2,which hinders the industrial applications.K-feldspar is a kind of insoluble potash ore,with a very stable property[2–5].The present technologies using K-feldspar for the soluble potash salts production are at an elevated temperature(e.g.＞1500 °C)to destroy the stable structure of K-feldspar,which is a highly energy-consumption process and not economically practicable[6–10].Phosphogypsum,an industrial waste from phosphorus chemical industry,seriously influences the ecological environment and restricts the development of phosphorus chemicals[11].At present,the large scale utilizations of the phosphogypsum are concentrated in the three aspects:the preparation of building materials[12,13],soil improvement and regeneration via decomposition to useful chemicals[14].Although gypsum decomposition is technically possible[15–17],these processes are all with high energy consumption,low decomposition rate and low economic feasibility[18].These energy-intensive industries will definitely lead to high CO2emissions.In the past couple of decades,technologies of CO2capture and storage(CCS)for CO2emission reduction[19]have been intensively investigated for reduction the industrial CO2emissions[20–22].Although CCS has well-documented advantages,it is not economically favorable due to the high-energy consumptions involved and the low value-added mineralization products,which hinder the industrial applications[23–26].Therefore,it is crucial for the researchers to develop the simple and cheap methodologies for CO2emission reduction,particularly for CO2capture using the industrial waste in the industrial process.

Ourprevious work has reported an ion-exchange method of calcium salts with K-feldspar to deal with phosphogypsum,extract soluble potash salts,as well as mineralize CO2using the slag for emission reduction.The industrial solid waste phosphogypsum and natural potassium feldspar were reacted thermally at the high temperature of 1200°C,then the reacted slag powder could be reacted with CO2for mineralization[35].The slurry after the mineralization was separated to potassiumrich liquid and CaCO3-rich slag.This process generated potassium sulfate,silicon aluminate and CaCO3.However,the conversion of the phosphogypsum was much low and the sulfur in the phosphogypsum could notbe recovered in the process.In addition,CaCl2(a waste deposited in soda production using Solvay method),was used as the promoter for the thermal activation of K-feldspar at ca.900°C to mineralize ofCO2and extract soluble potassium[27].However,the KCl product was accompanied with the CaCl2reactant,which resulted in large amounts of energy consumption for the separation of KCl and CaCl2.

Herein,we propose a novel methodology via coupling the extraction soluble potassium salts from K-feldspar,reduction of phosphogypsum for recovery of sulfur dioxide,and reduction of CO2emissions using the formed calcium compounds.Phosphogypsum in this process could be reduced with C to SO2and the formed CaO could facilitate the ionexchange of the K-feldspar and CO2mineralization for emissions.The coupling thermal reactions and mineralization were systematically investigated to obtain a high SO2recovery ratio,K2O extraction ratio and CO2mineralization ratio.The reaction mechanisms,energy consumptions and CO2emissions were also investigated and discussed on the base of the experiment and characterization results.

2.Experiments

2.1.Materials

K-feldspar ore used in the experiments was mined from Baoxin,China.The raw coal was purchased from Hongda Corp,China,and the chemical compositions were shown in Table S1.All chemicals(Kelong Chemicals Corp,China),including ethanol,calcium sulfate anhydrous,sodium carbonate anhydrous,sodium thiosulfate,potassium dichromate,potassium iodide,sulfuric acid,soluble starch,iodine,phenolphthalein,sodium tetraphenylboron,magnesiumchloride hexahydrate,EDTA,sodium hydroxide and boric acid,were analytical-grade and used as received without any further purification.The deionized water used to prepare all solution was purified through an ultrapure water instrument.

2.1.1.Ore pretreatment

The K-feldspar ore was crushed and wet-milled with a discharge size of 1–3 mm,then it was dried in a oven at 500 °C for 5 h.Finally,the ore powder was sieved and selected with an average particle size of 109–150 μm as the experiment material.

2.1.2.Component analysis

The component analysis of K-feldspar ore powder was obtained by chemical analysis and energy dispersive spectrometer(EDS)analysis.In chemical analysis,the ore powder was fused by alkali fusion,and the operational procedures were as follows.(I)Boric acid and sodium carbonate anhydrous were mixed with the mole ratio of 1:2.(II)～0.3 g of K-feldsparore powderand 3 g of the flux were added in a platinum crucible,and the ore sample was set in the middle layer between the bottom and the cover layers of the flux.(III)The mixture was calcined at 850°C for 30 min in a muffle furnace,soaked in deionized water with agitation at 70–80 °C and then leached for 2 h.(IV)Finally,the content of the K2O in the filtrate was determined by potassium tetraphenyl borate gravimetric method(GB 8574-88).The results of chemical compositions were shown in Table 1.Estimated relative accuracy based on repeated measurements was～0.5%.

2.2.Methods

2.2.1.Experimental procedure

The experimental procedures were shown in Fig.1,and the operations were performed as flows.(I)Carbon,calcium sulfate anhydrous and K-feldspar ore powder were mixed in a mortar for 120 min.(II)The mixture was molded into tablets using a tablet machine(TY-30T,Tianjin,China)with a diameter of 13 mm and thickness of 5–7 mm under the pressure of 10 MPa(III)The tablets were pushed into a certain temperature zone of a high-temperature tube furnace that was filling with nitrogen(GSL-1600X,Hefei,China)at 900 °C–1200°C for 0.5–2 h,tail gas in this process was absorbed by iodine standard solution packed in multistage tandem absorption(refer to GB/T601-2002).After the calcinati on,the sample of roasting slag was cooled naturally to the room temperature in a dryer.(IV)The calcined sample was milled to the average particle size bellow150 μm.(V)The calcined sample and deionized water(1:30)were added into a continuous autoclave(CJF-0.25L,Henan,China),and was reacted with CO2with a constant pressure of 0.5 MPa controlled by a counterbalance valve.(VI)The mineralization reaction was carried out at room temperature for 2 h.(VII)The slurry after the mineralization was filtered,to separate potassium-rich liquid and CaCO3–CaSiO3-rich slag.The melting point was measured using an ash melting point meter tester(HR-5A,Hebi,China)for drawing the eutectic surface projection of CaO–CaSO4–KAlSi3O8ternary phase diagram.

Table 1 The original K-feldspar chemical compositions(wt%)

Fig.1.The flow chart of experimental procedure.

2.2.2.Analysis and characterization

The amount of SO2obtained from the step III was determined by iodometry and the analysis procedures were conducted as follows:(1)Residual iodine standard solution consumed by SO2was collected;and(2)Collected liquid was titrated by the known concentration of Na2S2O3standard solution.The recovery ratio of sulfur that was accurate for measurements is～5%.

The amountof CO2fixed into CaCO3was determined by roasting the dried slag obtained from the step(VII)in the experimental procedure.The slag was roasted at 550 °C and 850 °C for 1 h,respectively.The mass loss from 550 °C to 850 °C was resulted from the thermal decomposition of CaCO3,and used to calculate the amount of fixed CO2.According to the weight loss,the CO2mineralization ratio was calculated and the relative accuracy is～2%.

X-ray diffraction(XRD)experiments were performed using the DX-2700,(Dandong,China).The voltage and anode current used were 40 kV and 30 mA,respectively.The CuKα=0.154056 nm and scanning mode with 0.03°interval and 0.05 s of set time were used to collect the XRD pattern of samples.

Thermal analysis experiments were carried out on a simultaneous thermogravimetric and differential thermal analyzer(TG/DTA)(HCT-3,Beijing,China),with a heating rate of 10 °C·min?1,and a nitrogen(N2purity over 99.999%) flow rate of 50 ml·min?1.

The recovery ratio of sulfur:

where the m0,m,m1,m1′and m2′represented the mass of CaSO4,ore,precipitate,and slag roasted at 550 °C and 850 °C,respectively;C1,C2,V1,and V2were the concentration of iodine standard solution,Na2S2O3standard solution and the volume of the I2standard solution,Na2S2O3standard solution,separately.w is the extent of the K2O content in the ore,wt%;0.1314 is the calculation coefficient of the mass of potassium tetra phenyl borate to potassium oxide;M0and M described the relative molecular weight of CaSO4and CO2.

3.Results and Discussion

3.1.The key factors of the reaction activities

Since the coupling reaction contains the reduction of CaSO4with carbon and the ion-exchange of K-feldspar with CaSO4,the activities of two solid–solid reactions are largely depended on the molecule/ion diffusion.Generally,four important factors,i.e.,reaction temperature and time,CaSO4/KAlSi3O8and C/CaSO4ratio,which affect the solid–solid reactions,were studied to evaluate the performance of K-extraction,sulfur recovery and CO2carbonation in this work.

3.1.1.Effect of reaction temperature

In the high-temperature solid–solid reactions,the temperature is one of the most important factors since it restricts the ion diffusion and solid reaction activity.The diffusion inside the solid crystal or between phases is generally the rate determining step,and the reactions were sharply accelerated when the temperature was near the melting point of the reactants[28–31].From the results of reactiontemperature experiments in Fig.2,the recovery ratio of sulfur and the extraction ratio of potassium were shown a growth with the increase of the reaction temperature.The sulfur recovery ratio was more sensitive than the extraction ratio to the effects of the activation temperature.The extraction ratio of potassium increased distinctly at the temperature above 1000°C.The formation of eutectic melt phase at the temperature of 1000°C that is lower than the melting point of the reactants(melting point of CaSO4is about 1350°C,and K-feldspar is over 1500°C),shown in Fig.S2,could greatly promote the mass transfer between the reactants and the reaction rate.Therefore,the ion diffusion and solid reaction activity were enhanced.When the temperature was higher than 1100°C,the extraction ratio of potassium was declined suddenly for two reasons.On the one hand,the reduction reaction of C with CaSO4is generally at about 800°C,which is more competitive than the activation of K-feldspar with CaSO4at the temperature higher than 1100°C and leads to insufficient CaSO4for the activation reaction to extract potassium.On the other hand,a large amount of CaO was generated via the reduction reaction at the temperature higher than 1100°C,which elevated the eutectic temperature of the reactants and hindered the formation of co-melting phase in the activation reaction.In addition,a higher temperature facilitated the reduction of CaSO4and the formed CaO was reacted with CO2,which could enhance the reaction of CO2mineralization.

Fig.2.The effects of calcination temperature on the reactivity.Calcination condition:CaSO4/KAlSi3O8 mass ratio was 1:1;C/CaSO4 mole ratio was 1:1;calcination time was 2 h.Mineralization condition:2 h;25°C;0.5 MPa CO2.

We noted that the CO2mineralization ratio declined suddenly when the temperature was over 1000°C,which may be attributed to the generation of CaSiO3and Ca2SiO4in the high-temperature reaction.[32,33].The previous results indicated that the CaSiO3and Ca2SiO4were potential resources for CO2mineralization and also play an important role on the CO2mineralization ratio[34,35].However,the crystal transition of CaSiO3at a high calcination temperature(e.g.,1100°C)results in a stable structure(Fig.S1)with little reactivity with CO2.Therefore,the activation temperature above 1000°C led to a low CO2mineralization ratio.

3.1.2.Effect of the CaSO4/KAlSi3O8mass ratio

CaSO4acts as the reactant not only in the reduction reaction with carbon but also in the extraction reaction with K-feldspar,the concentration of CaSO4in raw material inevitably affects the co-activation reaction.Therefore,the effect of stoichiometric CaSO4/KAlSi3O8mass ratio(with a constant CaSO4/C mole ratio)was investigated and the results were shown in Fig.3.The results showed that the K-extraction ratio and CO2mineralization ratio were augmented with the increase of the CaSO4/ore mass ratio.At the CaSO4/ore ratio of 4:1,the K-extraction and CO2mineralization ratio reached～87.7%and～14.2%,respectively.Since the dispersion of particles is critical for the solid–solid reaction of the activation reaction,a high CaSO4/KAlSi3O8mass ratio could enhance the dispersion of K-feldspar particles with CaSO4species and lead to a high conversion of K-feldspar.

3.1.3.Effect of the C/CaSO4mole ratio condition

The concentration of the reduct ant plays an important role on the conversion of CaSO4,thus affects the recovery ratio of sulfur,and the amount of formed CaO for the CO2mineralization.The effect of different C/CaSO4mole ratios on the reactivity was shown in Fig.4.The results indicated that the recovery ratio of sulfur and CO2mineralization ratio at the C/CaSO4ratio of 3:1,reached up to～95.6%and～27.7%,respectively,while the extraction ratio of potassium was decreased to～0.5%.The reduction reaction was greatly enhanced at a high mole proportion of carbon,and most of the CaSO4was converted into CaO.The much high melting point of CaO(i.e.,～2600 °C)resulted in the high eutectic temperature between CaO and K-feldspar thereby the mass transfer between the reactants was depressed.Therefore,the ion-exchange reaction of formed CaO with K-feldspar was impeded.

Fig.4.The effect of the C/CaSO4 mole ratio on the reactivity.Calcination condition:CaSO4/KAlSi3O8 mass ratio was 1:1;calcination temperature was 1100°C;calcination time was 2 h.Mineralization condition:2 h;25°C;0.5 MPa CO2.

3.1.4.Effect of the Activation Time

Since the solid phase reaction is rather slow and the reaction time greatly affects conversion and extraction.The influence of activation time on the recovery of sulfur,the extraction of potassium and CO2mineralization was investigated and shown in Fig.5.The extraction ratio of potassium and CO2mineralization ratio increased linearly up to～87.7%and～12.2%with the activation time in the initial 2 h.

In addition,the mineralization reaction is a gas–liquid–solid multiphase reaction.The pressure of CO2and the mineralization temperature are the primary parameters that conspicuously influenced the mineralization rate.The mineralization condition for the experiment was that the roasting slag was stirred in 0.5 MPa of CO2at air temperature for 2 h.The results indicated that all the CaO could be mineralized with CO2and the CO2mineralization ratio of CaO was～100%.Nevertheless,CaSiO3and Ca2Al2SiO7were mineralized difficultly,resulting in the CO2mineralization ratio of CaSiO3of～0.48%.In addition,blank experiments of mineralization using unactivated K-feldspar and CaSO4showed that the CO2mineralization ratio of CaSO4was～0.0%,and that of K-feldspar was＜1.00%at these mineralization reaction conditions.

Fig.5.The Effect of the activation time on the reactivity.Calcination condition:CaSO4/KAlSi3O8 mass ratio was 4:1;C/CaSO4 mole ratio was 1:1;calcination temperature was 1100 °C.Mineralization condition:2 h;25 °C;0.5 MPa CO2.

3.2.Characterizations of XRD

The XRD characterization could determine the changes of the substance composition before and after the calcination reactions and mineralization reactions.We have estimated the temperature of the reduction reaction and the activation reaction and the respective products.

The XRD patterns of calcined samples at calcination temperature of 900 °C,1000 °C,1100 °C,and 1200 °C were shown in Fig.6(a).The peaks of CaSO4weakened and the peaks of CaS(PDF#08-0464)appeared at the calcination temperature of 900°C.At the temperatures above 1000°C,the peaks of KAlSi3O8weakened gradually and the peaks of CaS disappeared,while the peaks of CaO(PDF#37-1497),CaSiO3(PDF#45-0156)and Ca2Al2SiO7(PDF#35-0755)strengthened.On the contrary,the formation of CaOwas plenty when the temperature was between 1000 and 1100°C.Almost all the peaks of CaSO4and KAlSi3O8disappeared as the temperature was over1100°C.Meanwhile,the characteristic peaks of CaSiO3,Ca2SiO4and Ca2Al2SiO7strengthened.The XRD results indicated that only the reduction reaction was occurred between the reduction reaction and activation reaction in the calcinati on process below 900°C and the reduction product was CaS.When the temperature was over 1000°C,both of the reduction reaction and the activation reaction were enhanced.The rate of reduction reaction was higher than that of activation reaction.This could be explained by the fact that the peak strength of CaO or CaSiO3increased faster than that of Ca2Al2SiO7.The XRD spectra of raw and roasted K-feldspar at temperatures of 1000 °C,1100 °C,1150 °C,and 1200 °C were illustrated in Fig.6(b).Only the microcline(PDF#84-0708)and quartz(PDF#46-1045)were detected in the raw K-feldspar.There was no change of diffraction peaks until the calcination temperature reached 1100°C.Butthe peaks ofKAlSi3O8gradually weakened and the peaks ofKAlSi2O6(PDF#38-1423)appeared at the temperature above 1100°C.When calcination temperature was at 1200°C,only the peaks of KAlSi2O6and SiO2were shown in the roasted feldspar sample.

From the above XRD analyses,the temperature of K-feldspar decomposition reaction was over 1200°C,while the decomposition temperature of CaSO4was～1350 °C.The reaction of CaO with KAlSi3O8and that of CaSO4with KAlSi3O8could not occur when the temperature was 1000 °C.However,the activation reaction worked at 1000 °C,itwas because the reduction and the activation were occurred synchronously.Since the solid melting point of multicomponent mineral is lower than that of some components in materials(e.g.,CaO,KAlSi3O8),and the appearance of new melts in calcination reaction proceeding further reduce the melting point between these minerals[36–39],we selected the different ratios of the mixture of CaO,CaSO4and K-feldspar,and measured their softening temperature by a pyramid method on an intelligent ash-melting point measuring device.The melting points of ternary mixture,i.e.,CaO,CaSO4and K-feldspar,were shown in Fig.7.Since the fusing of the reactants is conductive to the mass transform between the reactants in solid–solid reaction,the activation reaction could happen between CaO,CaSO4and KAlSi3O8at 1000°C.

Fig.6.The XRD patterns of the samples after activation.(a)The XRD analysis of the calcined samples generated by the coupling reactions between the reduction and activation at the temperatures of 900 °C,1000 °C,1100 °C,1200 °C;(b)The XRD analysis of raw K-feldspar and K-feldspar was roasted at temperatures of 1000 °C,1100 °C,1150 °C,1200 °C;(c)The XRD analysis of the calcination samples at the different CaSO4/KAlSi3O8 mass ratios(C/CaSO4 mole ratio was 1:1 and the calcination temperature was 1100°C)of 1:1,2:1,3:1,4:1;(d)The XRD analysis of the calcination samples at the different C/CaSO4 mole ratios(the CaSO4/KAlSi3O8 mass ratio was 1:1 and the calcination temperature was 1100°C)of 1:1,2:1,3:1.

The XRD patterns of the calcination samples(C/CaSO4mole ratio was 1:1 and the calcination temperature was 1100°C)at the CaSO4/KAlSi3O8mass ratio of 1:1,2:1,3:1,and 4:1 were shown in Fig.6(c).With the increase of mass ratio,the peaks of CaSO4and KAlSi3O8could hardly be detected,and the peaks of CaSiO3weakened.However,the diffraction peaks ofCa2SiO4and Ca2Al2SiO7enhanced.The XRD patterns of the calcination samples(the CaSO4/KAlSi3O8mass ratio was 1:1 and the calcination temperature was 1100°C)at the C/CaSO4mole ratio of 1:2,1:1,2:1,3:1,were illustrated in Fig.6(d).The peaks of Ca2Al2SiO7weakened but the peaks of CaSiO3and KAlSi3O8grown with the rising of mole ratio from 1:1 to 2:1.Besides,CaAlSi2O6was detected in the XRD analysis,the peaks of CaAlSi2O6enhanced while the peaks of KAlSi3O8greatly weakened at the mole ratio above 2:1.The results indicated that the increase of the CaSO4/KAlSi3O8mass ratio could improve extraction ratio of potassium efficiently.The rate of solid–solid reaction is largely depended on the ion diffusion and solid reaction activity.Therefore,the increase of CaSO4/KAlSi3O8ratio could enhance the contact area between the reactants and thus enhance the rate of activation reaction.On the other hand,the stoichiometric ratio of C with CaSO4had a tremendous impact on the reduction reaction.As the mole ratio of C/CaSO4grows,the reduction reaction was greatly enhanced.A large amount of CaSO4was transformed into CaO,which hindered the activation reaction.

The XRD patterns of the mineralization samples were shown in Fig.8(a)–(c).The peaks of CaSiO3and Ca2Al2SiO7did not change significantly compared with that of roasted slag.The large increase of CO2mineralization at 1000°C may be attributed to the large amounts of CaO generated at the calcination temperature of 1000°C.Besides,the mineralization ratio of CO2was decreased seriously over 1100°C,this was because a part of CaO was transformed to CaSiO3and Ca2SiO4,which could not be mineralized to produce CaCO3in this mineralization condition.

3.3.The Physico-chemical properties of the reactant for the solid–solid reaction

The reduction of C with CaSO4and the exchange of K-feldspar with CaSO4are typical thermal solid–solid reactions.The diffusion inside the solid crystal or between phases is generally the limiting step in the solid–solid reaction[40,41].Our previous work has demonstrated that the activation process was satisfied with Ginstling–Brounshtein kinetics process[42–44].Molten liquid emerged with the proceeding of the solid–solid reaction when the reaction temperature was near to the melting point.The emergence of liquid phase could not only change the environment of the reaction,but also greatly promote mass transfer between the reactants.Therefore,the factors that affect the molecular diffusion played important roles in the solid–solid reactions,such as the reaction temperature,reaction time and the raw material proportion.

Fig.7.The melting points of ternary mixture among the formed CaO,CaSO4 and K-feldspar.

Based on the experimental and characterization results,the temperature was one of the key factors on the reduction and activation[45,46].The activation reaction hardly happened unless the reactants formed into melting phase or partial melting phase at or nearly at the melting point temperature.It was because that the molecular diffusion and molecular activities accelerated,which enhanced the mass transfer and reaction rate significantly.Moreover,it could be confirmed from the results of eutectic point temperature of the mixture among CaO,CaSO4and K-feldspar,in the calcination process,that the formation of CaOfrom reduction of CaSO4with C blended with raw materials further reduced the eutectic temperature,which was more beneficial to enhance the mass transfer,and thus reduced the calcination temperature and energy consumption.The parameters of the reaction time,the mass proportion between the raw materials,etc.were important for the exchange of the K+with Ca2+and the formation of the CaO.Thus adjusting those parameters could improve the extraction of K,the recovery of sulfur and the mineralization of CO2.Reducing to the appropriate eutectic temperature by adjusting the generation of CaO consequently decrease the calcination temperature of activation reaction.

3.4.The pathway of thermal solid–solid co-reaction

Based on the XRD results,it could be induced that thermal solid–solid co-reactions were the ion exchange reaction between KAlSi3O8and CaSO4and reduction of CaSO4with C.The process of mineralization was possibly the reaction of CO2with CaO,CaSiO3,Ca2Al2SiO7,etc.

To clarify the pathway of thermal solid–solid co-reaction,an experiment was designed to calculate the balance of the concentration of Ca2+in calcination process and the consequence was shown in Table 2 and Fig.9.The theoretical value of Ca2Al2SiO7and CaAl2Si2O8wasbased on the theory of the ion-exchange between K+(2 molecules)and Ca2+(1 molecule).

The results confirmed that the K+exchanged with Ca2+for the formation of K2SO4,and the rest molecular skeleton after ion exchange with electric neutrality for the formation of Ca2Al2SiO7and CaAl2Si2O8.Meanwhile,the Ca2+provided by CaOin molten phase also participated in the ion-exchange activation reaction,owing to the reductive of the solid melting point between the multicomponent minerals in the calcination process.The reaction mechanism model was proposed in Fig.10.In the mineralization reaction,CaSiO3(calcined at a high temperature,e.g.,1100°C),Ca2Al2SiO7and CaAl2Si2O8were reacted little with CO2,due to the stable structure.Therefore,CaO played the primary role on the CO2mineralization.

In addition,CaSO4was involved in both of reduction reaction and activation reaction.The recovery of sulfur and Kextraction ratio presented a changing-trend of contradictions when the ratio of C,CaSO4and K-feldspar was constant.The quantitative results showed a reciprocal relationship between the reduction of CaSO4and the exchange reaction of K-feldspar with CaSO4,which should be balanced for an optimal sulfur recovery and K extraction ratio.

The proposed whole process could be described as:

when reaction temperature below 1000°C,CaSO4was just converted into CaS according to Eq.(1):

when the calcination temperature over 1000°C,both of reduction reaction and ion exchange reaction were preceded,the equations were:

When the reaction temperature reached at 1100°C,CaSiO3converted into Ca2SiO4according to Eq.(8):

3.5.Evaluation of CO2 emission and energy consumption

Fig.8.The XRD patterns of the samples after mineralization.(a)The XRD analysis results of the mineralization samples at different calcination temperatures of 900 °C,1000 °C,1100 °C,1200 °C;(b)The XRD analysis of the mineralization samples calcinated at different CaSO4/KAlSi3O8 mass ratios(C/CaSO4 mole ratio was 1:1 and the calcination temperature was 1100 °C)of 1:1,2:1,3:1,4:1;(c)The XRD analysis results of the mineralization samples calcinated at the different C/CaSO4 mole ratios(the CaSO4/KAlSi3O8 mass ratio was 1:1 and the calcination temperature was 1100°C)of 1:1,2:1,3:1.

Table 2 Calculating the balance of Ca2+.

The CO2emission in this process mainly originated from the carbon combustion for energy supply and the reduction of CaSO4with carbon.Two typical technologies,i.e.,reduction of CaSO4with carbon[47]and thermal decomposition of K-feldspar,[48]were introduced and estimated for comparison of CO2emission with this coupling technology.The reaction conditions of typical technologies for the SO2production is at 1200°C for 1.5 h with a C/CaSO4mole ratio of 3:1 and sulfur recovery of～60%,while for the K2SO4is at 1050 °C for 2.5 h with a K-feldspar/CaSO4/CaCO3mass ratio of 1:1:3.4 and potassium recovery of～93%.

3.5.1.Mass balance

The CO2emission in this technology for generation of 1 t K2SO4and 1628 m3SO2is ca.0.8477 t CO2(0.7503 t CO2was mineralized).

Fig.9.The contrast between the experiment extraction ratio of potassium and theoretical extraction ratio of potassium came from different samples.

However,the comparative technologies generate much more CO2with a total CO2emission of 9.165 t CO2(1.598 t for SO2and 7.567 t CO2for K2SO4).

3.5.2.Energy balance

The calculation of energy balance is based on the generation of 1 t K2SO4and 1628 m3SO2with a technological sulfur recovery of 59.91%and potassium recovery of 85.6%.The reactions involved in calcination step are:

The energy consumption was calculated according to the reactions and heat of the reactants.The heat energy consumption could be calculated according to the energy changes that reactants were heated from room temperature to reaction temperature and the phase changes in the heat process.

Fig.10.The schematic model of the calcination process.

Table 3 Calculated energy consumption for three processes

Because of the absence of phase changes in the high-temperature reactions:

The reaction energy consumption was calculated according to the Hess'Law for the high-temperature reactions at room pressure and the mineralization reaction at a constant volume:

The thermodynamic parameters are introduced from the handbook,“Thermo chemical Data of Pure Substances”,written by Ihsan Barin.The energy consumption is supplied in the form of the combustion of anthracite(98,300 kg CO2per tera joule,IPCC,2006).The calculation results were shown in Table 3.

The high-temperature reactions were in a rotary kiln reactor(Ф3.0× 30.0 m with a filling rate of～10%and reactants of 23447 kg)and the designed flowchart in commercial production was shown in Fig.11.The symbol Qsis denoted for the heat loss of rotary kiln,while Qgfor heat in the tail gas,Qtfor heat of reactants,Qffor reaction heat,Qhfor the recovered heat and Qyfor the heat in the mineral slag.

The general thermal efficiency η1in industrial production is ca.40%:

The theoretical energy consumption in the rotary kiln reactor was 2.523 × 107kJ((1-η2)Qt+Qf)and theoretical anthracite was 2.480 t.Since the heat loss in rotary kiln,tail gas and the mineral slag,the practical energy consumption was 9.114 × 107kJ(Qs+Qg+Qf+(1-η2)Qt).Because the heat in the tail gas(40%of the total energy,3.884×107kJ)was much higher than the heat for mineralization reaction,which could be reused for CO2mineralization.Therefore,the anthracite consumption in this process was ca.8.959 t.

Fig.11.Designed flowchart in commercial production.

Table 4 The energy consumption and CO2 emissions for production every 1628 m3 SO2 and 1.000 t K2SO4

On the other hand,the traditional technologies using the reduction of CaSO4for 1628 m3SO2production and decomposition of K-feldspar for 1.00 t K2SO4production required 5.793×107kJ and 6.791×107kJ,respectively.The thermal efficiency and recovery of heat in the tail gas were also 40%and 50%.The anthracite consumption was 5.695 t and 6.676 t,respectively.

3.5.3.CO2emission reduction

According to the emission factor of 98300 kg CO2per terajoule published by the IPCC,2006,the calculated theoretical and practical CO2emissions for production of 1628 m3SO2and 1 t K2SO4were shown in Table 4.In this technology,theoretical and practical CO2emissions were 3.328 t CO2(0.8477 t for sulfur reduction CO2and 2.480 t for heat)and 9.807 t CO2(0.8477 t for sulfur reduction 8.959 t for heat).However,the theoretical and practical CO2emissions in the contrastive technologies were 13.41 t and 21.54 t,respectively.Therefore,this technology has greatly reduced the energy consumption and CO2emissions,with theoretical energy consumption reduction of 41.54%,CO2emission reduction of 75.18%,and practical energy consumption reduction of 27.57%and CO2emission reduction of 54.47%.

4.Conclusions

The co-reaction of K-feldspar and phosphogypsum is an efficient methodology for extraction of soluble potassium salts and recovery of SO2with reduced CO2emissions and energy consumption.The thermal solid–solid reaction is the key step in this process,CaSO4was reduced by carbon to recover the sulfur and K-feldspar was reacted with CaSO4and CaO to exchange K+with Ca2+and release K+from the insoluble ore.Under the optimal conditions,the highest recovery ratio of sulfur,K extraction ratio and the CO2mineralization ratio could reach up to～66.1%,～87.7%and～14.2%,respectively.On the base of characterization and reaction mechanism discussion,the high efficiency and low energy consumption could be attributed to the partial melting on the intra surface of the particles for the enhancement of the solid–solid reactions and diffusion.Furthermore,the co-reaction process significantly facilitated the formation of CaO and silicates,which enhanced the mineralization reaction for CO2emission reduction.

Appendix A.Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.cjche.2016.04.001.

[1]H.Xie,H.Yue,J.Zhu,et al.,Scientific and engineering progress in CO2mineralization using industrial waste and natural minerals,Engineering 1(1)(2015)150–157.

[2]K.Huang,X.Meng,G.Wang,Research progress of extracting potassium from potassium feldspar,Phosphate Comp.Fertil.5(2011)008.

[3]M.Xu,L.Zhang,B.Wang,et al.,Preparation of 13X molecular sieve from potassium feldspar and its performance characterization,Ind.Catal.22(7)(2014)521–524.

[4]M.V.Twigg,M.S.Spencer,Deactivation of supported copper metal catalysts for hydrogenation reactions,Appl.Catal.A Gen.212(1)(2001)161–174.

[5]M.Sorai,T.Ohsumi,M.Ishikawa,Nanoscale surface observation of feldspar dissolved under supercritical CO2–water–mineral system,Energy 30(11)(2005)2334–2343.

[6]M.Qiu,H.Ma,Y.Nie,et al.,Experimental study on synthesis of Tobermorite by decomposing potassium feldspar,Geoscience 3(2005)004.

[7]J.A.Stewart,Potassium sources,use,and potential,Potassium Agric.(1985)83–98(potassiuminagri).

[8]D.Zhenwu,Extraction of potassium from K-feldspar with Fluosilicic acid using hydrothermal decomposition method,J.Sichuan Univ.(Eng.Sci.Ed.)46(2015)151–154.

[9]N.Kampman,M.Bickle,J.Becker,et al.,Feldspar dissolution kinetics and Gibbs free energy dependence in a CO2-enriched groundwater system,Green River,Utah,Earth Planet.Sci.Lett.284(3)(2009)473–488.

[10]X.Bi,D.H.Cornell,R.Hu,REE composition of primary and altered feldspar from the mineralized alteration zone of alkaline intrusive rocks,Western Yunnan Province,China,Ore Geol.Rev.19(1)(2002)69–78.

[11]X.Wang,Z.Zhang,X.Yang,et al.,Analysis on new approaches for utilization of phosphogypsum in China,Mod.Chem.Ind.5(2011)002.

[12]Y.Shen,J.Qian,J.Chai,et al.,Calcium sulphoaluminate cements made with phosphogypsum:Production issues and material properties,Cem.Concr.Compos.48(2014)67–74.

[13]J.Zhou,H.Gao,Z.Shu,et al.,Utilization of waste phosphogypsum to prepare nonfired bricks by a novel hydration–recrystallization process,Constr.Build.Mater.34(2012)114–119.

[14]B.Yan,L.Ma,J.Ma,et al.,Mechanism analysis of Ca,S transformation in phosphogypsum decomposition with Fe catalyst,Ind.Eng.Chem.Res.53(2014)7648–7654.

[15]L.Ma,Y.Du,X.Niu,et al.,Thermal and kinetic analysis of the process of thermochemical decomposition of phosphogypsum with CO and additives,Ind.Eng.Chem.Res.51(2012)6680–6685.

[16]X.Yan,L.Ma,B.Zhu,et al.,Reaction mechanism process analysis with phosphogypsum decomposition in multiatmosphere control,Ind.Eng.Chem.Res.53(50)(2014)19453–19459.

[17]X.Y. Han, H.K. Chen, B.Q. Lin, Thermogravimetric research of reductiondecomposition of CaSO4in the atmosphere of H2,Coal Convers.23(2)(2000)72–75.

[18]S.Schuyten,S.Guerrero,J.T.Miller,et al.,Characterization and oxidation states of Cu and Pd in Pd–CuO/ZnO/ZrO2catalysts for hydrogen production by methanol partial oxidation,Appl.Catal.A Gen.352(1)(2009)133–144.

[19]J.Gibbins,H.Chalmers,Carbon capture and storage,Energ Policy 36(12)(2008)4317–4322.

[20]J.A.Bergerson,D.W.Keith,The truth about dirty oil:Is CCS the answer?Environ.Sci.Technol.44(16)(2010)6010–6015.

[21]J.H.Han,I.B.Lee,Multiperiod stochastic optimization model for carbon capture and storage infrastructure under uncertainty in CO2 emissions,product prices,and operating costs,Ind.Eng.Chem.Res.51(35)(2012)11445–11457.

[22]M.Yang,Y.Song,L.Jiang,et al.,CO2hydrate formation and dissociation in cooled porous media:A potential technology for CO2capture and storage,Environ.Sci.Technol.47(17)(2013)9739–9746.

[23]W.Bao,H.Li,Y.Zhang,Selective leaching of steelmaking slag for indirect CO2mineral sequestration,Ind.Eng.Chem.Res.49(5)(2010)2055–2063.

[24]E.Eikeland,A.B.Blichfeld,C.Tyrsted,et al.,Optimized carbonation of magnesium silicate mineral for CO2storage,ACS Appl.Mater.Interfaces 7(9)(2015)5258–5264.

[25]W.J.J.Huijgen,G.J.Witkamp,R.N.J.Comans,Mineral CO2sequestration by steel slag carbonation,Environ.Sci.Technol.39(24)(2005)9676–9682.

[26]O.Rahmani,R.Junin,M.Tyrer,et al.,Mineral carbonation of red gypsum for CO2sequestration,Energy Fuel 28(9)(2014)5953–5958.

[27]L.Ye,H.Yue,Y.Wang,et al.,CO2mineralization of activated K-feldspar+CaCl2slag to fix carbon and produce soluble potash salt,Ind.Eng.Chem.Res.53(26)(2014)10557–10565.

[28]E.A.Dorko,R.W.Crossley,Solid state reaction kinetics.III.Calculation of rate constants of decomposition for a melting system undergoing volume and surface changes,J.Phys.Chem.76(16)(1972)2253–2256.

[29]S.Tamhankar,L.Doraiswamy,Solid–solid reactions.Diffusion and reaction in pellet pellet systems,Ind.Eng.Chem.Fundam.17(1978)84–89.

[30]K.C.Chou,A kinetic model for oxidation of Si–Al–O–N materials,J.Am.Ceram.Soc.89(5)(2006)1568–1576.

[31]C.Li,W.Qiu,X.Kang,et al.,Kinetics of synthesis of Ba 1.0 Co 0.7 Fe 0.2 Nb 0.1 O 3–δ through solid–solid reaction,Acta Phys.-Chim.Sin.24(5)(2008)767–771.

[32]L.Yang,M.Fang,J.Liu,et al.,Effects of nonionic surfactants on the luminescent properties of CaSiO3:pb,Mn phosphors,J.Mater.Sci.Eng.2(2007)005.

[33]W.J.J.Huijgen,G.J.Witkamp,R.N.J.Comans,Mechanisms of aqueous wollastonite carbonation as a possible CO2sequestration process,Chem.Eng.Sci.61(13)(2006)4242–4251.

[34]C.Wang,H.Yue,C.Li,et al.,Mineralization of CO2using natural K-feldspar and industrial solid waste to produce soluble potassium,Ind.Eng.Chem.Res.53(19)(2014)7971–7978.

[35]W.J.J.Huijgen,G.J.Ruijg,R.N.J.Comans,et al.,Energy consumption and net CO2sequestration of aqueous mineral carbonation,Ind.Eng.Chem.Res.45(26)(2006)9184–9194.

[36]W.Wong-Ng,L.P.Cook,BaO–1/2Y2O3–CuO x eutectic melting in air,J.Am.Ceram.Soc.77(7)(1994)1883–1888.

[37]L.Chudinovskikh,R.Boehler,Eutectic melting in the system Fe–S to 44 GPa,Earth Planet.Sci.Lett.257(1)(2007)97–103.

[38]A.Zerr,G.Serghiou,R.Boehler,Melting of CaSiO3perovskite to 430 kbar and first insitu measurements of lower mantle eutectic temperatures,Geophys.Res.Lett.24(8)(1997)909–912.

[39]J.F.Carpenter,B.S.Chang,W.Garzon-Rodriguez,et al.,Rational design of stable lyophilized protein formulations:Theory and practice[M]//rational design of stable protein formulations,Springer,US,2002 109–133.

[40]J.R.Bielenberg,H.J.Viljoen,Chemo-mechanical interaction in solid–solid reactions,AIChE J.45(5)(1999)1072–1084.

[41]C.Ghoroi,A.K.Suresh,Solid–solid reaction kinetics:Formation of tricalcium aluminate,AIChE J.53(2)(2007)502–513.

[42]A.M.Ginstling,B.I.Brounshtein,Concerning the diffusion kinetics of reactions in spherical particles,J.Appl.Chem.USSR 23(12)(1950)1327–1338.

[43]G.W.Brindley,R.Hayami,Kinetics and mechanism of formation of forsterite(Mg2SiO4)by solid state reaction of MgO and SiO2,Philos.Mag.12(117)(1965)505–514.

[44]B.M.Mohamed,J.H.Sharp,Kinetics and mechanism of formation of tricalcium aluminate,Ca3Al2O6,Thermochim.Acta 388(1)(2002)105–114.

[45]J.?esták,G.Berggren,Study of the kinetics of the mechanism of solid-state reactions at increasing temperatures,Thermochim.Acta 3(1)(1971)1–12.

[46]H.E.Kissinger,Reaction kinetics in differential thermal analysis,Anal.Chem.29(11)(1957)1702–1706.

[47]H.L.Lu,Research on preparation of Portland cement from desulphurization gypsum,Nanjing Univ.Sci.Technol.(2013).

[48]L.H.Qiu,Z.M.Jin,Experimental study of the thermal decomposition of potassium feldspar for producing potassium sulfate,Fertil.Ind.27(2000)19–21.