Thermodynamic Analysis of the Cu-As-S-(O) System Relevantto Sulfuric Acid Baking of Enargite at 473 K (200 �C)

M. SADEGH SAFARZADEH, JAN D. MILLER, and HSIN H. HUANG

While the growing demand for copper has compelled the industry to adapt new technologies forthe treatment of copper-arsenic (enargite) concentrates, the refractory nature of such concen-trates combined with the troublesome presence of arsenic has created a major metallurgical andenvironmental challenge. Preliminary results of the acid bake-leach process at the University ofUtah have shown some potential advantages for the treatment of enargite concentrates. Whilethe transformation of enargite to copper sulfate, arsenolite, and elemental sulfur has alreadybeen established experimentally, thermodynamic evaluation of the sulfuric acid baking processprovides further understanding which should be useful. In this article, the available thermo-dynamic data for the species involved in the Cu-As-S-O system are compiled. These data wereused to calculate the phase stability (Kellogg) diagrams as well as equilibrium compositions at473 K (200 �C) using the STABCAL and HSC Chemistry� 5.1 software packages. The equi-librium composition calculations indicate that enargite can transform to copper sulfate eitherdirectly or through chalcocite and/or covellite. The major gaseous species during baking werefound to be SO2 and H2O. The results of the thermodynamic calculations were further com-pared with two confirmatory baking experiments involving a high-quality enargite sample. Thecondensed reaction products from sulfuric acid baking based on XRD results include CuSO4,As2O3, CuOÆCuSO4, and S8 under both neutral and oxidative conditions. While all thesecompounds were predicted through equilibrium calculations, some of the predicted compoundswere not detected in the sulfuric acid-baked enargite. None of the calculations indicated anyappreciable amounts of arsenic-bearing gases at the baking temperature of 473 K (200 �C).Consistent with thermodynamic predictions, no H2S gas was detected during the sulfuric acidbaking experiment. Approximately, 80 pct of the baked enargite samples were leached in water.

DOI: 10.1007/s11663-013-9965-y� The Minerals, Metals & Materials Society and ASM International 2013

I. INTRODUCTION

THE elevated amount of arsenic found in some of theimportant copper deposits has rendered these resourcesto be of lower economic value. The increasing demandsfor copper and gold have prompted the metallurgicalindustry to develop sustainable processing options forchallenging Cu-As ores/concentrates, notably for orescontaining enargite (Cu3AsS4). Significant developmentshave been reported in the flotation separation ofenargite-bearing ores with the purpose of producing a‘‘clean’’ copper concentrate with less than 0.5 pct arsenicfor smelting. However, now it seems that the time hascome for the processing of high-arsenic ‘‘dirty’’ copperconcentrates. To that end, there is going to be increasinginterest in the treatment of Cu-As concentrates includ-

ing enargite and tennantite (Cu12As4S13). Not only dothese minerals contain more copper than chalcopyrite(48.4 pct in enargite and 51.5 pct in tennantite), but theirconcentrates also contain more gold than the chalcopy-rite concentrates. There has been a dearth of literatureon the treatment of Cu-As ores until about the year2000, after which research activities have been intensi-fied to design strategies to treat these problematic ores/concentrates. Interestingly, the entire body of enargiteresearch published recently is the continuation of studiesperformed in the 1970s and 1980s. For example,development of alkaline sulfide leaching, pressure leach-ing, and roasting strategies for the treatment of enargiteconcentrates have been considered for some time.A few thermodynamic studies of the Cu-As-S-(O)

system have been reported, with little experimentalvalidation of the theoretical calculations. Five well-known ternary compounds in the Cu-As-S system havebeen reported. These compounds include Cu3AsS4 withits high-temperature form (enargite) and the low-tem-perature form (luzonite), with the transition temperaturebetween the two being in the range from 548 K to 573 K(275 to 300 �C). Other compounds include Cu4As2S5,Cu12As4S13 (tennantite), Cu6As4S9 (sinnerite), andCuAsS (lautite). Not only are these compounds rich incopper, but also they are functional materials with

M. SADEGH SAFARZADEH, Ph.D. Candidate, and JAN D.MILLER, Professor, are with the Department of MetallurgicalEngineering, College of Mines and Earth Sciences, University ofUtah, 135 South 1460 East, Room 412, William C. Browning Building,Salt Lake City, UT 84112-0114. Contact e-mail: jan.miller@utah.eduHSIN H. HUANG, Professor, is with the Department of Metallurgicaland Materials Engineering, Montana Tech of the University ofMontana, 215 ELC Building, 1300 West Park Street, Butte, MT59701-8997.

Manuscript submitted April 30, 2013.Article published online October 23, 2013.

568—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

semiconducting and photoelectric properties.[1] Babanlyet al.[1] estimated the standard Gibbs free energies of theformation of these compounds from electrochemicalmeasurements using solid electrolytes. These values aretabulated in Table I.

The recent reviews on the treatment of enargiteconcentrates[2,3] indicate that both conventional hydro-metallurgical and pyrometallurgical options fail to meetthe required techno-economical and environmental cri-teria for the processing of enargite concentrates. Whilethe only viable hydrometallurgical technique for theleaching of enargite appears to be pressure leaching (withall its complexities), generally the roasting of enargitedoes not expel all of the arsenic from the concentrate,contaminating smelting and electrorefining streams.

In keeping with the goal of effective copper recoveryand minimum arsenic emission problems, a pyro-hydro-metallurgical treatment option was developed at theUniversity of Utah, which renders the enargite watersoluble while keeping most of the arsenic in solution.The process includes the baking of enargite concentratewith sulfuric acid at a low temperature [473 K (200 �C)],which transforms the enargite to copper sulfate andarsenic trioxide, followed by water leaching, whichreleases copper, arsenic, and iron into the solution. Thedetails of the process can be found elsewhere.[4–6]

A sound understanding of the thermodynamics of thesulfuric acid baking reaction is required to explain theobserved experimental results. Thermodynamic infor-mation for Cu-As-S system in general, and for enargitein particular, has not been given much attention inthe literature. Only two authors have independentlyreported the thermodynamic values for enargite. Thefirst and the most reliable information was published byCraig and Barton,[7] who estimated the values based onthe studies of Wernik and Benson[8] and Maske andSkinner,[9] followed by the study of Seal II et al.[10] whomeasured the heat capacity of enargite experimentally.Most researchers use the information from either of thementioned articles to estimate the free energy offormation for enargite. It should be noted that tennan-tite (Cu12As4S13) and sinnerite (Cu6As4S9) are the othercopper-arsenic sulfosalts that have been mentioned inthe literature, with the former one being the second mostabundant copper-arsenic feedstock in the copper indus-try. Listed in Table II is a summary of the thermody-namic information for enargite, tennantite, and sinneritein chronological order. Taking into account the accu-racy of the thermodynamic information, the dataprovided by Craig and Barton[7] will be used in all ourcalculations.

These data were incorporated into the database ofHSC Chemistry� 5.1[12] and also STABCAL[13] softwarefor subsequent calculations. In view of the possiblereactions that might occur during the sulfuric acidbaking process, it is appropriate to see if the proposedreactions are thermodynamically favorable. To that end,a series of sulfuric acid baking reactions were proposedbased on preliminary experimental results and alsobased on the results reported by Prater et al.[14] Thestandard free energies of formation for the compoundsinvolved in the reactions were obtained from HSCChemistry� 5.1 software. Then, the standard freeenergies of reactions (DG�r ) at 298 K and 473 K (25 �Cand 200 �C) (the latter one being the typical bakingtemperature) were calculated. The results are listed inTable III (Reactions [1] to [21]).Reactions [2] and [3] represent the oxidation of

enargite in an oxygen atmosphere. These reactions arehighly exothermic. In terms of acid baking reactions, themost favorable reaction appears to be Reaction [4],where in the presence of oxygen, enargite converts intocopper sulfate, arsenic trioxide, water, and sulfurdioxide. While oxygen favors the baking reaction to agreat extent, the reaction can proceed in the absence ofoxygen, as observed in Reactions [1] and [5]. In this case,the reaction is not favorable at room temperature, andtherefore increasing the temperature is necessary.ConsideringH2S (g) as a reactionproduct,Reactions [7],

[11], and [14]maybewritten.However, it is noted thatnoneof these reactions is expected either at room temperature orat higher temperatures. According to Reactions [9] and[10], in the absenceofoxygen, theproductionofS2 (g) is notfavorable but in the presence of oxygen, it is highlyfavorable (Reaction [12]). The generation of SO3 (g) as agaseous reaction product may be favored only in thepresence of oxygen (Reactions [13] to [15]).The key role of SO3 (g) in the sulfation roasting of

metal sulfides has been well established.[15–17] One wayto increase the partial pressure of SO3 (g) in the roasteris through the introduction of sulfuric acid solution thatwill be decomposed at high temperatures and willgenerate SO3 (g) according to the following reaction:

H2SO4 ¼ H2Oþ SO3 gð Þ ½22�

Reaction [22] is highly endothermic (DH�;298Kð25�CÞ

r ¼132:39 kJ/mol, DH�;473Kð200

�CÞr ¼ 129:35 kJ/mol).

DG�;473Kð200�CÞ

r is 52.77 kJ/mol, with DG�r starting tobecome negative at 803 K (530 �C). Therefore, theminimum temperature of 803 K (530 �C) is necessaryto get the benefit of Reaction [22]. This means that SO3

(g) generated from the sulfuric acid decomposition maynot contribute to the sulfuric acid baking reaction.However, there is another source of SO3 (g) according toReaction [23]:

2SO2 gð Þ þO2 gð Þ ¼ 2SO3 gð Þ ½23�

This reaction is highly exothermic (DH�;473Kð200�CÞ

r ¼�198:34 kJ/mol, DG�;473Kð200

�CÞr ¼ �108:74 kJ/mol)

below 773 K (500 �C). On the other hand, this reaction

Table I. Standard Gibbs Free Energies of Formation for the

Compounds in the Cu-As-S System[1]

Compound DG�f (kJ/mol)

Cu3AsS4 �179.2 ± 0.6Cu6As4S9 �429.4 ± 1.2Cu4As2S5 �257.8 ± 0.8Cu3AsS3 �170.2 ± 0.6CuAsS �65.5 ± 0.3

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—569

is very slow at low temperatures, requiring catalysis toachieve acceptable rates of SO3 (g) generation from SO2

(g), and O2 (g) mixtures.[16] This situation is furtherdemonstrated by looking at the equilibrium compositionof sulfuric acid as a function of temperature (Figure 1),using the equilibrium module of the HSC Chemistry�

5.1 software. While no significant amount of SO3 (g) isgenerated below 573 K (300 �C), the maximum amountof SO3 (g) is found in the temperature range from 723 Kto 773 K (450 �C to 500 �C), after which SO3 (g) startsto decompose into SO2 (g) and O2 (g).Therefore, it is less likely that SO3 (g) can play a role in

the sulfuric acid baking of enargite through this mecha-nism. Having said that, the effect of SO3 (g) in thesulfation of enargitemay not be ruled out, as described byReactions [16] to [21]. In fact, these reactions are all highlyfavorable, considering the variety of reaction products.More importantly, Reactions [18] to [20] could resemblethe baking of enargite with oleum, which is obtained bydissolving different amounts of SO3 (g) in 100 pct sulfuricacid. If sufficient amounts of H2S gas are generated in thebaking reaction, then it can be used to produce elementalsulfur through the Claus process (Reaction [24]).[18]

2H2S gð Þ þ SO2 gð Þ ¼ 2H2O gð Þ þ 1:5S2 gð Þ ½24�

In view of the foregoing, the purpose of this article isto evaluate these enargite and associated reactions usingphase stability diagrams and also equilibrium composi-tion calculations. Finally, the results from two sulfuricacid baking experiments under neutral and oxidativeatmospheres are reported to evaluate the credibility ofthe thermodynamic analysis.

II. METHODS AND MATERIALS

Thermodynamic calculations were performed usingHSC Chemistry� 5.1 and also STABCAL software witha careful collection of the most reliable thermodynamicinformation for the species involved in the reactions.Most of the phase stability diagrams were constructedusing STABCAL software unless otherwise stated. Allthe equilibrium calculations were performed using thefree energy minimization module of the HSC Chemis-try� 5.1 software. Both these softwares are available atthe University of Utah. All calculations were performedin the absence and the presence of oxygen.To confirm the results of the thermodynamic calcula-

tions, two baking experiments were performed. A high-quality enargite specimen from Butte, Montana, USAwas purchased. Then, it was crushed and groundmanually using a ceramic mortar and pestle, to avoidiron contamination from ball milling. The groundenargite was characterized using X-ray diffraction(XRD) and electron microprobe analysis (EMPA) meth-ods. For EMPA and also optical microscopy (OM), theenargite powder was mounted in epoxy, polished to anextremely fine surface, and carbon coated. An opticalimage of the polished section is shown in Figure 2, whichshows no impurity minerals are observed within theenargite particles.

Table

II.

Summary

ofThermodynamic

Inform

ationforEnargite

Mineral

Thermodynamic

Function

Tem

perature

Range[K

(�C)]

Reference

DS�;298Kð25� CÞ

m

(Jmol�

1K�1)

DH�;298Kð25� CÞ

m

(kJmol�

1)

C� p;mðTÞ¼

aþ10�3bT�105cT�2

(Jmol�

1K�1)

DG

�(T)

ab

c

Enargite(C

u3AsS

4)

——

——

—�104,458+

56.16T(calmol�

1)

298to

873(25to

600)

[7]

Tennantite

(Cu12As 4S13)

——

——

—�355,468+

172.28T(calmol�

1)

298to

873(25to

600)

[7]

Sinnerite(C

u6As 4S9)

——

——

—�236,340+

136.83T(calmol�

1)

298to

762(25to

489)

[7]

Enargite(C

u3AsS

4)

——

——

—�437,160+

235.0T(J

mol�

1)

298to

873(25to

600)

[11]

Enargite(C

u3AsS

4)

257.7

37.4

196.7

±1.2

49.9

±1.6

19.18±0.84

—298to

944(25to

671)

[10]

570—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

Table

III.

ProposedReactionsfortheSulfuricAcidBakingofEnargite

ReactionNumber

Reaction

DG�;298Kð25� CÞ

r(kJ/mol)

DG�;473Kð200� CÞ

r(kJ/mol)

12Cu3AsS

4+

31H

2SO

4=

6CuSO

4+

As 2O

3+

31H

2O+

33SO

2(g)

6.621

�947.815

22Cu3AsS

4+

15.5O

2(g)=

6CuSO

4+

As 2O

3+

2SO

2(g)

�4726.514

�4269.112

32Cu3AsS

4+

13.5O

2(g)=

6CuSO

4+

As 2O

3+

2S

�4126.328

�3667.262

44Cu3AsS

4+

31H

2SO

4+

15.5O

2(g)=

12CuSO

4+

2As 2O

3+

31H

2O+

35SO

2(g)

�4719.89

�5216.93

52Cu3AsS

4+

27H

2SO

4=

6CuSO

4+

As 2O

3+

27H

2O+

27SO

2(g)+

2S

�3.920

�774.520

62Cu3AsS

4+

2H

2SO

4+

12.5O

2(g)=

6CuSO

4+

As 2O

3+

2SO

2(g)+

2H

2O+

2S

�3820.964

�3452.985

72Cu3AsS

4+

6.75H

2SO

4=

6CuSO

4+

As 2O

3+

6.75H

2S(g)+

2S

305.673

237.877

82Cu3AsS

4+

9H

2SO

4=

6CuSO

4+

As 2O

3+

9H

2O+

11S

�51.356

5.310

92Cu3AsS

4+

9H

2SO

4=

6CuSO

4+

As 2O

3+

9H

2O+

5.5S2(g)

386.974

295.622

10

2Cu3AsS

4+

10.333H

2SO

4=

6CuSO

4+

As 2O

3+

10.333H

2O+

5.167S2(g)+

2SO

2(g)

363.757

220.077

11

2Cu3AsS

4+

8.333H

2SO

4=

6CuSO

4+

As 2O

3+

6.333H

2O+

4.167S2(g)+

2H

2S(g)

386.334

293.967

12

2Cu3AsS

4+

2H

2SO

4+

10.5O

2(g)=

6CuSO

4+

As 2O

3+

2H

2O+

2S2(g)

�3061.386

�2745.567

13

2Cu3AsS

4+

2H

2SO

4+

13.5O

2(g)=

6CuSO

4+

As 2O

3+

2SO

3(g)+

2H

2O+

2S

�3962.802

�3561.728

14

2Cu3AsS

4+

8.25H

2SO

4=

6CuSO

4+

As 2O

3+

8.25H

2S(g)+

2SO

3(g)

548.538

395.093

15

2Cu3AsS

4+

11H

2SO

4=

6CuSO

4+

As 2O

3+

11H

2O+

5.5S2(g)+

2SO

3(g)

550.500

401.156

16

2Cu3AsS

4+

31SO

3(g)=

6CuSO

4+

As 2O

3+

33SO

2(g)

�2528.031

�2583.597

17

2Cu3AsS

4+

27SO

3(g)=

6CuSO

4+

As 2O

3+

27SO

2(g)+

2S

�2211.521

�2199.233

18

2Cu3AsS

4+

25SO

3(g)+

2H

2O

=6CuSO

4+

As 2O

3+

25SO

2(g)+

2H

2S(g)

�1945.722

�1967.025

19

2Cu3AsS

4+

6.75SO

3(g)+

6.75H

2O

=6CuSO

4+

As 2O

3+

6.75H

2S(g)+

2S

�246.227

�118.301

20

2Cu3AsS

4+

7.25SO

3(g)+

5.25H

2O

=6CuSO

4+

As 2O

3+

5.25H

2S(g)+

2S2(g)

�207.056

�90.799

21

2Cu3AsS

4+

23SO

3(g)=

6CuSO

4+

As 2O

3+

21SO

2(g)+

2S2(g)

�1735.617

�1709.301

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—571

The XRD spectrum of the enargite powder obtainedusing a Rigaku D-Max 2000 model confirmed thatenargite is the only major mineral present in the sample(Figure 3).

EMPA analysis, using a Cameca-SX50 machine, fromfour different locations on the sample indicated that thesample composition is uniform and very close to theideal composition of enargite (48.41 pct Cu, 19.02 pctAs, and 32.57 pct S) with very good repeatability. Aswith most of the enargite specimens, there is someantimony in this enargite sample, ~2 pct. The detailedresults are listed in Table IV.

A �150+106 lm portion of the ground enargite wasprepared by screening using standard sieves. This sizefraction was used in the baking experiments. The samplecontained ~41.5 pct Cu and ~13.86 pct As, based on wet

chemical assay. The first baking experiment was per-formed in a muffle furnace (Barnstead Thermolyne 1300model) to examine the effect of oxygen on the oxidationof enargite. The second experiment was performed in astandard tube furnace (Thermolyne 21100 model),connected to a nitrogen gas cylinder equipped with flowmeters. A Pyrex glass tube (76 9 5 cm) was used toallow for observation during the reaction. In bothfurnaces, the temperature was controlled within ±1 �C.Analytic grade sulfuric acid (Mallinckrodt Chemicals,95 to 98 pct) was used in the baking experiments. Bothexperiments were performed at a constant temperatureof 473 K (200 �C).A porcelain crucible (diameter 9 height: 42 9

35 mm) was used for the muffle furnace experiment.The distance between the thermocouple and the surfaceof material was about 5 cm. Approximately 2 g ofenargite powder was weighed and mixed with ~3.3 g ofsulfuric acid and thoroughly mixed in the crucible. Thecrucible was weighed and transferred to the mufflefurnace, which was preheated to the desired tempera-ture. There was no atmosphere control, and the bakingwas done open to the ambient atmosphere. After7 hours of baking, the crucible was taken out of themuffle furnace and cooled down in the ambient atmo-sphere to room temperature and weighed out. Thebaking time of 7 hours and the baking temperature of473 K (200 �C) were selected based on our experimentalresults published earlier,[4] where it was established thatunder similar experimental conditions, the extractions ofcopper and arsenic upon water leaching did not changeby baking the enargite concentrate for more than6 hours. The baked material was removed from thecrucible and ground using a mortar and pestle. A samplewas taken for XRD.In the tube furnace experiment, a porcelain boat

(length 9 height 9 width: 100 9 15 9 10 mm) wasemployed. The thermocouple tip was placed right abovethe material’s surface in the boat. Approximately 2 g ofenargite powder was weighed and mixed with ~3.3 g ofsulfuric acid and thoroughly mixed in the boat. Then,the boat was weighed and transferred to the tube

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 50 100 150 200 250 300 350 400 450 500 550 600

Temperature (°C)

Eq

uili

bri

um

am

ou

nt

of

spec

ies

(km

ol)

H2SO4

H2O (g)

SO3 (g)

H2SO4 (g)

SO2 (g)

O2 (g)

Fig. 1—The distribution of species at equilibrium as a function of temperature representing the heating of sulfuric acid under neutral conditions(input: 1 kmol sulfuric acid, total pressure: 1 bar).

Fig. 2—Polished surface of the high-quality enargite particles(1009).

572—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

furnace, which was preheated to the desired tempera-ture. The experiment was performed under a flow ofnitrogen (flow rate ~340 mL/min) to provide an oxygen-free atmosphere. Lead acetate test papers (FisherScientific) were used inside the tube to detect for theformation of H2S gas. At the end of the experiment, theelemental sulfur formed inside the tube was rinsed withcarbon disulfide (HPLC grade, Alfa Aesar). After7 hours of baking, the sample was cooled down toroom temperature under the flow of nitrogen to preventany oxidation reactions at high temperatures outside thefurnace. The baked material was removed from the boatand ground using a pestle and mortar. A sample wastaken for XRD.

The baked and ground enargite samples from bothexperiments were subjected to two two-step leachingexperiments. In the first leaching step, ~2 g portions ofthe baked and ground enargite samples were leached inabout 160 mL DI water. The water leaching experi-ments were performed in a 500 mL Erlenmeyer flask.The samples were leached for 45 minutes at 343 K(70 �C) using a stirring speed of 400 rpm. At the end ofeach leaching experiment, the suspension was filtered,and solid residues were rinsed with DI water, and driedin a furnace at 343 K (70 �C) over night. The residueswere then weighed. The dried leach residues were thentaken to a second leach step in 0.5 M sulfuric acid toensure complete dissolution of soluble components. Allother leaching conditions were the same as the first

water leaching step. The leach residues from the secondleaching experiments were rinsed with DI water, dried,weighed, and sampled for XRD.

III. RESULTS AND DISCUSSION

A. Phase Stability and Equilibrium CompositionDiagrams

The phase stability diagrams for the Cu-As-S systemat 473 K (200 �C) are constructed under neutral condi-tions. While enargite, tennantite, and sinnerite are thethree copper-arsenic sulfosalts considered in construct-ing these diagrams, the phase sinnerite has not beenreported to exist in copper-arsenic feedstocks orotherwise formed as an intermediate phase during thethermal treatment of enargite in neutral and oxidativeatmospheres. This could be due to the rapid formationand transition of sinnerite that does not allow foridentification at room temperature.

1. Cu-As-S systemThe phase stability diagram for the Cu-As-S system at

473 K (200 �C) (Figure 4(a)) indicates that enargitecould be at equilibrium with sinnerite, tennantite,covellite (CuS), and chalcocite (Cu2S), depending onthe partial pressure of S2 (g) and As4 (g). The relativeamount of these phases as a function of temperature canbe calculated by a free energy minimization program

0

200

400

600

800

1000

1200

1400

1600

0 10 20 30 40 50 60 70

2θ°

Inte

nsi

ty (

cou

nts

)

*

*

*

*

*

* *

*

*: Enargite

Fig. 3—XRD spectrum of the high-quality enargite specimen (scan rate 2 deg/min).

Table IV. EMPA Results for the High-quality Enargite Sample Used in the Current Study (Accelerating Voltage: 15 keV,

Current: 30 nA, Beam Size: 10 lm)

Spot No.

Wt pct Elements

TotalsCu As Sb S

1 47.5 18.8 0.55 32.1 99.12 47.9 18.2 1.50 31.6 99.13 47.8 17.1 2.57 31.6 98.94 47.5 16.8 3.57 31.5 99.2Average 47.67 17.72 2.047 31.70 99.07

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—573

(a)

0

10

20

30

40

50

60

70

80

100 200 300 400 500 600 700 800 900 1000

Temperature (°C)

Eq

uili

bri

um

Co

mp

osi

tio

n (

mo

l.%)

S2 (g)

Cu2S

Cu3AsS4

CuS

AsS

As2S3

As2S3 (g)

S5 (g)

S3 (g)

(b)

(c)

Fig. 4—(a) Predominance area (Kellogg) diagram for the system Cu-As-S at 473 K (200 �C) (S2 vs As4), (b) equilibrium composition as a func-tion of temperature for enargite in a neutral atmosphere (input: 1 kmol enargite, total pressure: 1 bar), and (c) predominance area (Kellogg) dia-gram for the system Cu-As-S at 473 K (200 �C) (S2 vs As2S3). Mineral key: CuS = covellite, Cu2S = chalcocite, Cu3AsS4 = enargite,Cu12As4S13 = tennantite, Cu6As4S9 = sinnerite, and Cu3As = domeykite.

574—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

such as SOLGASMIX[19] that is used to calculate theequilibrium composition for multiphase systems.[17] Theresult of such a calculation for enargite is shown inFigure 4(b). According to Figure 4(b), major decompo-sition of enargite starts at around 673 K (400 �C) withthe most dominant gas species being S2 and As2S3.These results indicate that a more accurate phasestability diagram should be built considering thesegaseous species as the major gaseous products ofroasting in neutral atmospheres. The resulting phasestability diagram is shown in Figure 4(c), which indi-cates a greater stability domain for covellite than forchalcocite, being opposite to the trend observed inFigure 4(a). It is postulated from Figure 4(b) that at473 K (200 �C), enargite converts into CuS, AsS, andCu2S as the major conversion products.

2. Cu-As-S-O systemThe direct application of Kellogg diagrams to explain

the enargite-sulfuric acid system is not possible becausesuch diagrams are built for solid–gas systems. However,the study of the Cu-As-S-O system can provide someuseful information for the sulfuric acid baking reaction.Considering this fact, constructing Kellogg diagrams fora quaternary system such as Cu-As-S-O at 473 K(200 �C) is complex. In fact, such a diagram would bean isothermal section of the Cu-As-S-O system at aconstant molar ratio of Cu/As. Therefore, the equilib-rium compositions in the Cu-As-S-O system will bediscussed. Figure 5 shows the equilibrium compositionof enargite in oxidative atmospheres. When the molarratio of oxygen/enargite is 2.5, enargite can completelyreact at temperatures less than 501 K (228 �C) to giveCuS, SO2 (g), As2O3 (A)-Arsenolite, As4O6, CuSO4, andCu2S. While the formation of copper sulfate is favoredat temperatures below ~463 K (190 �C), the formationof solid arsenic oxides (As2O3 (A) and As4O6) isexpected at temperatures up to ~598 K (325 �C). Highertemperatures result in increased formation of chalcocite

and a reduction in the formation of covellite. At thisspecific oxygen potential, the only arsenic-bearing gasevolved from the oxidation process is As4O6 (g), whichstarts to form in significant amounts at ~523 K(250 �C).To examine the effect of oxygen potential on the

oxidation of enargite, the equilibrium state was estab-lished as a function of the oxygen potential at a constanttemperature of 473 K (200 �C) (Figure 6). According toFigure 6, complete reaction of enargite at 473 K(200 �C) would require an oxygen/enargite molar ratioof at least 2.5, after which copper sulfate starts to form.Chalcocite starts to take over covellite at oxygen/enargite molar ratios greater than 3. The major arsenicspecies in the product are As2O3 (A) and As4O6 up to anoxygen/enargite molar ratio of greater than ~2.5, afterwhich they decrease owing to the formation of the solidphase Cu(AsO2)2. The major gaseous reaction product isSO2 which continues to form up to an oxygen/enargitemolar ratio of ~8, after which SO3 starts to predomi-nate. Based on the information obtained from Figures 5and 6, enargite can theoretically be completely oxidizedto produce a variety of reaction products, depending onthe conditions, specifically temperature and oxygenpotential. Of course, these diagrams provide no infor-mation regarding the kinetics of enargite oxidation.It is evident that effective conversion of enargite to

copper sulfate requires a very high oxygen/enargitemolar ratio. Also the formation of copper sulfate andarsenic oxides is highly favorable and can be expectedunder certain conditions for the treatment of enargiteconcentrates. To that end, it is worthwhile to examinethe effect of sulfuric acid on the oxidation of enargite asdiscussed in the next section.

3. Equilibrium distribution of the species for theenargite-sulfuric acid systemAs can be expected, the equilibrium state in the

enargite-sulfuric acid system would be more complex

0

10

20

30

40

50

60

70

80

90

100

25 100 175 250 325 400 475 550 625 700

Temperature (°C)

Eq

uili

bri

um

Co

mp

osi

tio

n (

mo

l.%) CuS

SO2 (g)

Cu2S

As2O3 (A)

As4O6

As4O6 (g)

CuSO4

Cu(AsO2)2

Cu3AsS4

AsS

As2S2

As2S3

Fig. 5—The equilibrium composition as a function of temperature for enargite in oxidative atmospheres (input: 2 kmol enargite, 5 kmol O2, andtotal pressure: 1 bar).

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—575

than the enargite-oxygen system. The equilibrium dis-tribution of species as a function of temperature in theabsence of oxygen is shown in Figure 7. The sulfuricacid to enargite molar ratio is 8. While the main gaseousreaction products are SO2 and H2O, hydrated coppersulfates tend to form at low temperatures [298 K to573 K (25 �C to 300 �C)] along with anhydrous coppersulfate, which continues to form even at higher temper-atures. As2O3 (A) and As4O6 are the major condensedarsenic species, which their formation is facilitated ataround 473 K (200 �C). Chalcocite and covellite con-tinue to be major reaction products with the formertaking over at temperatures greater than ~443 K(170 �C). At temperatures greater than ~423 K(150 �C), Cu(AsO2)2 is a major reaction product.

Cu2O starts to form in appreciable amounts at temper-atures greater than ~573 K (300 �C). No significantamount of arsenic gases are generated.The effect of sulfuric acid dosage on the equilibrium

composition at 473 K (200 �C) is illustrated in Figure 8,which shows that complete decomposition of enargiterequires a sulfuric acid/enargite molar ratio of at least 5.The amount of sulfuric acid has a meaningful effect onthe formation of copper sulfate, so that copper sulfatekeeps increasing up to sulfuric acid/enargite molar ratioof 16, while the amounts of SO2 (g) and H2O (g) remainalmost unchanged. At this point, the entire amounts ofchalcocite and covellite are consumed, and coppersulfates predominate. Also, note that the stable formof arsenic becomes As2O4.

0

10

20

30

40

50

60

70

80

90

100

0 2 4 6 8 10 12 14 16 18 20

O2 (g), kmol

Eq

uili

bri

um

Co

mp

osi

tio

n (

mo

l.%) SO2 (g)

CuS

CuSO4

Cu2S

As2O3 (A)

SO3 (g)

Cu3AsS4

As4O6

As2O4

Cu(AsO2)2

As2O5

AsS

As2S3

Fig. 6—The equilibrium composition as a function of oxygen potential for enargite [input: 2 kmol enargite, total pressure: 1 bar, and tempera-ture 473 K (200 �C)].

0

10

20

30

40

50

60

70

80

90

100

25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500

Temperature (°C)

Eq

uili

bri

um

Co

mp

osi

tio

n (

mo

l.%)

SO2 (g)

CuS

H2SO4

CuSO4.3H2O

CuSO4.5H2O

H2O

CuSO4.H2O

As2O3 (A)

As4O6

H2O (g)

CuSO4

Cu(AsO2)2

Cu2SO4

Cu3(AsO4)2

CuO

Cu2O

Cu2S

As4O6 (g)

Cu3AsS4

S

Fig. 7—The equilibrium composition as a function of temperature for enargite in sulfuric acid in the absence of oxygen (input: 1 kmol enargite,8 kmol H2SO4, and total pressure: 1 bar).

576—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

Examination of Figures 5 through 8 indicates thatsulfuric acid is most significant for the decomposition ofenargite to produce water-soluble copper sulfates andarsenic trioxide. More importantly, the SO2 produced inthe reaction can be used in a sulfuric acid plant toregenerate some of the acid required for the acid bakingreaction.

The effect of the addition of oxygen to the enargite-sulfuric acid system can be observed from Figure 9,which shows the equilibrium composition as a functionof temperature. The major difference from the previouscase in the absence of oxygen is that there is no

significant amount of covellite formed and chalcocitepredominates.The effect of oxygen dosage on the oxidation of

enargite in the presence of sulfuric acid at 473 K(200 �C) is shown in Figure 10. It is understood fromFigure 10 that at oxygen/enargite molar ratios up to 4,all the sulfuric acid is consumed. At ratios greater than4, no chalcocite and covellite exist, and these mineralsconvert into copper sulfate. At the ratio of 8, SO3 (g)starts to form along with SO2 (g); however, at oxygen/enargite ratios greater than 9, SO3 (g) takes over as theonly major gaseous species.

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

H2SO4, kmol

Eq

uili

bri

um

Co

mp

osi

tio

n (

mo

l.%)

CuS

As2O3 (A)

Cu3AsS4

As4O6

AsS

As2S3

N2 (g)

SO2 (g)

Cu(AsO2)2

Cu2S

CuSO4

H2O (g)

CuSO4.H2O

As2O4

H2O

H2SO4 (g)

H2SO4

Fig. 8—The equilibrium composition as a function of sulfuric acid amount for enargite [input: 1 kmol enargite, total pressure: 1 bar, and tem-perature 473 K (200 �C)].

0

10

20

30

40

50

60

70

80

90

100

25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500

Temperature (°C)

Equ

ilibr

ium

Com

posi

tion

(mol

.%)

SO2 (g)

CuS

H2SO4

CuSO4.3H2O

CuSO4.5H2O

H2O

CuSO4.H2O

As4O6

As2O3 (A)

H2O (g)

CuSO4

Cu2S

Cu2O

Cu(AsO2)2

Cu2SO4

CuO.CuSO4

As4O6 (g)

CuO

Cu3(AsO4)2

Fig. 9—The equilibrium composition as a function of temperature for enargite in sulfuric acid in the presence of oxygen (input: 1 kmol enargite,8 kmol H2SO4, 1 kmol O2, and total pressure: 1 bar).

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—577

IV. BAKING EXPERIMENTS

Before doing the acid baking experiments, twoexperiments were performed to see if any phase trans-formations occur to enargite at 473 K (200 �C) in theabsence of sulfuric acid. The results indicated that nophase transformation took place, and enargite was

found to be stable at 473 K (200 �C). The bakingexperiments were performed under neutral (tube furnaceexperiment under a flow of nitrogen) and oxidative(muffle furnace experiment) conditions at a constanttemperature of 473 K (200 �C). In the acid bakingexperiments in the muffle furnace, the temperature wasincreased to 481 K (208 �C) within 3 minutes from the

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

O2 (g), kmol

Eq

uili

bri

um

Co

mp

osi

tio

n (

%)

As2O3 (A)

CuS

As4O6

Cu2S

CuSO4

H2O (g)

CuSO4.H2O

H2O

SO2 (g)

SO3 (g)

H2SO4

H2SO4 (g)

As2O5

Fig. 10—The equilibrium composition as a function of oxygen potential for enargite [input: 1 kmol enargite, 8 kmol H2SO4, total pressure:1 bar, and temperature 473 K (200 �C)].

0

2000

4000

6000

8000

10000

12000

5 10 15 20 25 30 35 40 45 50 55 60 65

2 °

Inte

nsi

ty

(a)

(b)

(c)

(d)

(e) A+B C

B

DD

B

B+D

D

BA

D

A+B

A+B

A: As 2O3

B: CuO.CuSO4

C: S8

D: CuSO4

C C C

C

C C

DA

CA+B

θ

Fig. 11—XRD spectra of (a) high-quality enargite, (b) baked sample in muffle furnace, (c) final leach residue of the baked sample in mufflefurnace, (d) baked sample in tube furnace, and (e) final leach residue of the baked sample in tube furnace (scan rate 2 deg/min). Baking condi-tions-muffle furnace: initial weight of enargite (�150+106 lm) = 1.97 g, weight of sulfuric acid = 3.35 g, baking temperature: 473 K (200 �C),baking time = 7 h. Tube furnace: initial weight of enargite (�150+106 lm) = 2.00 g, weight of sulfuric acid = 3.41 g, baking temperature:473 K (200 �C), baking time = 7 h, nitrogen flow rate: ~340 mL/min. First step leaching conditions in DI water: 45 min at 70 �C with a stirringspeed of 400 rpm. Second step leaching conditions in 0.5 M sulfuric acid: same as the first step. The results indicated correspond to the leachresidues from the second step leaching experiments. Powder diffraction file (PDF) numbers used for the identification of the phases are as fol-lows: As2O3 00-036-1490, CuSO4 00-015-0775, S8 00-008-0247, and CuOÆCuSO4 01-074-1590.

578—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

start of the experiment and remained at 481 K (208 �C)for 2 minutes, and cooled back down to 473 K (200 �C)in 11 minutes. In the tube furnace experiment, thetemperature increased to 505 K (232 �C) within 6 min-utes and remained at this temperature for 2 minutes, andcooled back down to 473 K (200 �C) within 12 minutes.The difference in temperature increase is most probablyassociated with the distance of the thermocouple fromthe material bed. At the end of 7 hours of heat treatment,the sample coming out of the tube furnace was com-pletely dry, corresponding to 15.5 pct weight loss at theend of experiment, while the sample coming out of themuffle furnace which had some moisture content due toresidual sulfuric acid, showed 9.4 pct weight loss uponbaking. Approximately 0.6 g elemental sulfur wasformed inside the tube; this was not measured for the

sample baked in the muffle furnace even though itsformation was observed on the furnace door. No H2Swas detected based upon the fact that the color of thelead acetate paper remained unchanged.The results of XRD for both the enargite (pattern (a))

and the baked samples (patterns (b) and (d)) are shown inFigure 11. While no significant difference is observed inthe spectra of the baked enargite samples, the samplebaked in the tube furnace (pattern (d)) shows strongercopper sulfate and elemental sulfur peaks, perhapsbecause it was dry. Enargite peaks are not labeled becausethe high-quality enargite pattern is inserted (pattern (a)),which was already identified. Considering the reactionproducts and the molar ratio of sulfuric acid/enargitewhich was used in the experiments (~7), the results can betentatively compared to Figures 7 through 10.

Fig. 12—Backscattered SEM micrographs from the polished section of the high-quality enargite sample baked in (a)–(c) tube furnace and (d)muffle furnace. Baking conditions-muffle furnace: initial weight of enargite (�150+106 lm) = 1.97 g, weight of sulfuric acid = 3.35 g, bakingtemperature: 473 K (200 �C), baking time = 7 h. Tube furnace: initial weight of enargite (�150+106 lm) = 2.00 g, weight of sulfuricacid = 3.41 g, baking temperature: 473 K (200 �C), baking time = 7 h, nitrogen flow rate: ~340 mL/min).

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—579

The results from the experiments agree reasonablywell with the thermodynamic calculations where themajor reaction products were predicted to be CuSO4

and As2O3. Other phases predicted by thermodynamicsmay have formed, but were not detected due to their lowpercentage in the baked enargite. The amount ofenargite reacted can be easily measured by the sub-sequent leaching experiments, which are reported in thenext section. Characterization of the leach residues mayshow some of the phases which were not detected in thebaked enargite as they are concentrated in the leachresidue.

Polished sections of the baked samples were subjectedto further scanning electron microscopy (SEM) studies.The results show a very similar microstructure for bothbaked samples, with a mixed arsenolite-copper sulfateassociation. Figure 12 shows the SEM images of bothenargite samples baked in the tube and muffle furnace.The EDAX analysis of the mixed white-gray regionshowed that it is composed of Cu, As, S, and O. On theother hand, theEDAXspectrumof the areas free from thewhite phase indicates that the gray phase ismost probablycopper sulfate. Spot analysis of the white phases revealsthat they are most probably arsenic trioxide.

V. LEACHING EXPERIMENTS

As a result of first step leaching in DI water, ~81 pctof the sample baked in the tube furnace was leached inDI water. Approximately 9 pct of the leach residue fromthe first leach step was leached in the second acidleaching experiment. Also, about 80 pct of the samplebaked in the muffle furnace was leached in the first stepleach followed by ~7 pct weight loss in the second leachstep. The results for the second step leach in 0.5 Msulfuric acid may not be accurate because of the verysmall amount of material (<0.5 g) that was used in bothexperiments.

The results from leaching are very similar; indicatingthat ~80 pct of both baked enargite samples wereleached in the simple water leaching experiments. TheXRD results for the final leach residues are indicated inFigure 11. While the final residues are composed mainlyof enargite and elemental sulfur, it is noted that theleach residue of the sample baked in the muffle furnace(pattern (c)) shows stronger elemental sulfur peaks ascompared with the leach residue of the sample baked inthe tube furnace (pattern (e)). It is also inferred fromthe spectra of the leach residues that neither covellitenor chalcocite was formed in appreciable amounts as

expected from the thermodynamic analysis. In terms ofCu and As extraction, the results are summarized inTable V. The pH and Eh of the pregnant leach solution(PLS) for both solutions were ~2.2 and ~625 mV (vsSHE), respectively. According to the results reported inTable V, Cu and As extractions were almost same forboth the samples.

VI. CONCLUSIONS

A thermodynamic analysis was performed to furtherour understanding of the sulfuric acid baking of enargitein neutral and oxidative atmospheres. It was found thatthe major reaction products include CuSO4, As2O3, SO2

(g), and H2O (g). Sulfuric acid was found to be astronger sulfating agent than oxygen at low tempera-tures. No significant amounts of H2S (g), SO3 (g), andAs4O6 (g) were found to form in the baking reaction, aspredicted by thermodynamic calculations.Sulfuric acid baking experiments of a high-quality

enargite sample under oxidative and neutral atmo-spheres resulted in the formation of CuSO4, As2O3,CuOÆCuSO4, and S8, all of which were predicted basedon equilibrium composition calculations. No H2S wasdetected in the tube furnace experiment, adding credit tothe calculations listed in Table III and also equilibriumcomposition calculations.Two-step leaching of the baked enargite samples, first

in DI water and then followed by leaching in 0.5 Msulfuric acid solution, resulted in more than 80 pct of thematerial leached for both samples, leaving behindenargite and elemental sulfur in the leach residues.Neither chalcocite nor covellite was detected in the leachresidues. No significant amounts of the water-leachedsamples were leached in 0.5 M sulfuric acid solution,indicating that most reaction products (CuSO4, As2O3,CuOÆCuSO4) are completely leached in the water leachstep.A preferred operating window can be selected based

on the current study to maximize the formation ofwater-soluble copper and arsenic compounds. Not onlyare the results of the current study useful for the analysisof the baking reaction, but they can also be useful forpredicting enargite behavior during the roasting process.

ACKNOWLEDGMENTS

The authors would like to thank Newmont MiningCorporation for providing the funds for this enargiteproject. The authors also wish to thank Ms. DorrieSpurlock for proofreading of the manuscript.

REFERENCES1. M.B. Babanly, Z.T. Gasanova, L.F. Mashadieva, V.P. Zlomanov,

and Y.A. Yushibov: Inorg. Mater., 2012, vol. 48, pp. 225–28.2. M.S. Safarzadeh, M.S. Moats, and J.D. Miller: Miner. Process.

Extr. Metall. Rev., 2012, DOI:10.1080/08827508.2012.723651.3. M.S. Safarzadeh, M.S. Moats, and J.D. Miller: Miner. Process.

Extr. Metall. Rev., 2012, DOI:10.1080/08827508.2012.725683.

Table V. The Results of Cu and As Extraction Following

Water Leaching

Sample

Metal Extraction (Pct)

Cu As

Tube furnace 83.9 77.5Muffle furnace 86.0 76.9

580—VOLUME 45B, APRIL 2014 METALLURGICAL AND MATERIALS TRANSACTIONS B

4. M.S. Safarzadeh, M.S. Moats, and J.D. Miller: Hydrometallurgy,2012, vols. 119–120, pp. 30–39.

5. M.S. Safarzadeh, M.S. Moats, and J.D. Miller: Miner. Metall.Process., 2012, vol. 29, pp. 97–102.

6. M.S. Safarzadeh, M.S. Moats, and J.D. Miller: SME AnnualMeeting, Seattle, WA, USA, Preprint Number 12-068, 2012.

7. J.R. Craig and P.B. Barton: Econ. Geol., 1973, vol. 68, pp. 493–506.8. J.H. Wernik and K.E. Benson: J. Phys. Chem. Solids, 1957, vol. 3,

pp. 157–59.9. S. Maske and B.J. Skinner: Econ. Geol., 1971, vol. 66, pp. 901–918.10. R.R. Seal, II, R.A. Robie, B.S. Hemingway, and H.T. Evans, Jr.:

J. Chem. Thermodyn., 1996, vol. 28, pp. 405–412.11. D.C. Lynch: Arsenic Metallurgy: Fundamentals and Applications,

The Metallurgical Society, Montreal, 1988, pp. 3–32.

12. HSC Chemistry ver. 5.1: Outokumpu Research Oy, Espoo, 1994.13. H.H. Huang: STABCAL, Montana Tech, Butte, 2012.14. J.D. Prater, P.B. Queneau, and T.J. Hudson: JOM, 1970, vol. 22,

pp. 23–27.15. R.S. Boorman, R.S. Salter, and D.W. Davis: Mining Mag., 1984,

September issue, pp. 225–35.16. C.J. Ferron and J. De Cuyper: Int. J. Miner. Process., 1992,

vol. 35, pp. 225–38.17. A.B. Whitehead and R.W. Urie: Proceedings of the Aus. I.M.M.,

1961, vol. 199, pp. 51–85.18. Y.K. Rao: Stoichiometry and Thermodynamics of Metallurgical

Processes, Cambridge University Press, Cambridge, 1985.19. A. Roine: HSC Chemistry 5.1, Outokumpu Research Oy, Espoo,

1994.

METALLURGICAL AND MATERIALS TRANSACTIONS B VOLUME 45B, APRIL 2014—581