Industrial uses of high sulfur petroleum coke
by Lance Harold Ulrich
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Chemical Engineering
Montana State University
© Copyright by Lance Harold Ulrich (1991)
Abstract:
The purpose of this thesis is to find a use or uses for high sulfur petroleum coke. Research was
conducted to determine if high sulfur petroleum coke can be used as a reagent to convert molybdenum
ore, MoS2, to MoO2. The second experiment was to determine if the petroleum coke can be used as a
reagent to convert Cu2S ore to copper at a lower temperature than is currently used in industry. The
third area of research was to determine if high sulfur petroleum coke could be used in place of regular
coke to reduce iron ore, Fe3O4, to pig-iron (Fe). The last experiment was to determine if petroleum
coke could be substituted for regular coke in the initial lead ore roasting process.
Each of these experiments used a similar procedure. The petroleum coke was mixed with the metal ore,
and the mixture was heated in either an oxidizing or reducing atmosphere until the ore had been
converted to the final product. Then the product was chemically analyzed to determine purity and
percent conversion. For molybdenum, the ore/coke mix was cooked in air to facilitate the conversion of
MoS2 to MoO2. Air was blown through an experimental blast furnace to help convert Cu2S ore to
elemental copper. A reducing atmosphere was required to convert iron ore, Fe3O4, to iron, so the
ore/coke mix was insulated from the atmosphere with a top layer of coke. Lastly, air was blown
through the lead ore/coke mix to help convert PbS to Pb.
An 83% conversion of MoS2 to MoO2 was obtained by roasting an. 8:1 coke:ore mixture in air at
500°C for 38 minutes. The iron ore didn't reduce to iron at 1350°C. The best result with copper ore was
achieved by roasting a 0.5:1 coke:ore ratio at 700°C with a small amount of air blowing through it.
After two hours a 6 0% conversion of Cu2S to Cu was achieved. A 43% conversion of PbS to Pb was
obtained by roasting a 0.167:1 coke:ore ratio in air for 8 minutes at 800°C.
It was concluded that it is highly probably that high sulfur petroleum coke is effective in converting
MoS2 to MoO2 and it is highly probable that the petroleum coke can be used to convert lead ore to
lead. The coke is effective in converting copper ore to copper. Lastly, high sulfur petroleum coke
cannot convert Fe3O4 to iron at 1350°C; a higher temperature is probably needed. 
INDUSTRIAL USES OF HIGH SULFUR 
PETROLEUM COKE ,
by
Lance Harold Ulrich
A thesis submitted in partial fulfillment 
of the requirements for the degree
of
Master of Science 
in
Chemical Engineering
MONTANA STATE UNIVERSITY 
Bozeman, Montana
June, 1991
^  7 ii
APPROVAL
of a thesis submitted by 
Lance Ulrich
This thesis has been read by each member of the thesis 
committee and has been found to be satisfactory regarding 
content, English usage, format, citations, bibliographic 
style, and consistency, and is ready for submission to the 
College of Graduate Studies.
C 1 i 9 ^  I
Date
■----1----------
Chairperson^ G f rQduate Committee
Approved for the Major Department
-7 / 9 9 / J » -LxO
D^te H/eeid, Major Department
Approved for the College of Graduate Studies
2 ~ 4 p.i-,/ 99/
4Date” z™ Graduate tDean
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the 
requirements for a master's degree at Montana State 
University, I agree that the Library shall make it available 
to borrowers under rules of the Library. Brief quotations from 
this thesis are allowable without special permission, provided 
that accurate acknowledgement of source is made.
Permission for extensive quotation from or reproduction 
of this thesis may be granted by my major professor, or in his 
absence, by the Dean of Libraries when, in the opinion of 
either, the proposed use of the material is for scholarly 
purposes. Any copying or use of the material in this thesis 
for financial gain shall not be allowed without my written 
permission.
S ignaturq^ ^ ^ l ^ y
Date h € / 4. , I ^  ^ )— j
iv
TABLE OF CONTENTS
Page
BACKGROUND...................................................... 1
Petroleum Coke 
Molybdenum....
Copper........
Iron. . . . .....
Lead..........
THEORETICAL ASPECTS........................................... ..
Thermodynamics of Molybdenum Oxidation.......... ■____ n
Thermodynamics of Copper Reduction...................  19
Thermodynamics of Iron Reduction................... * [ 23
Thermodynamics of Lead Reduction............   25
RESEARCH OBJECTIVES..............'....... ........... . . ....... 27
MOLYBDENUM RESEARCH......................................   29
Molybdenum Experiments.................................  29
Molybdenum Results and Discussion................. .* .*! 31
COPPER RESEARCH.................................................
Copper Experiments...................................... 33
Copper Results and Discussion.........................  40
IRON RESEARCH...................................................
Iron Experiments........................................ 44
Iron Results and Discussion..........................  44
LEAD RESEARCH.........................    46
Lead Experiments........................................ 46
Lead Results and Discussion..................    43
SUMMARY AND CONCLUSIONS.........................................
RECOMMENDATIONS FOR FUTURE RESEARCH.......................... 54
H
 "d* LO f
' CO
TABLE OF CONTENTS— Continued
Page
REFERENCES CITED...........................................  55
APPENDICES..................................................  58
Appendix A— Molybdenum Experimental Data..............  59
Appendix B— Copper Experimental Data.................  63
Appendix C— Lead Experimental Data...........    65
V
vi
LIST OF TABLES
Table Page
1. Properties of EXXON Fluid Petroleum Coke..........  3
2. Composition of Lead Green Ore...................... 10
3. Thermodynamic Properties of Some
Molybdenum Compounds.............................. 12
4. Thermodynamic Properties of Some
Copper Compounds..................................  19
5. Thermodynamic Properties of Some
Iron Compounds....................................  23
6. Thermodynamic Properties of Some
Lead Compounds..................   25
7. Properties of Some Molybdenum Compounds........ . 35
8. Properties of Some Copper Compounds... ........... 40
9. 4:1 Molybdenum:Coke Samples..........................60
10. 8:1 Molybdenum:Coke Samples........................ 61
11. 16:1 Molybdenum:Coke Samples....................... 62
12. Copper Experimental Data............................ 64
13. Lead Experimental Data.............................. 66
LIST OF FIGURES
Page
1. Iron Blast Furnace........................................  8
2. Delta G vs Temperature for Molybdenum Oxidation....... 17
3. Delta G vs Temperature for Oxidation of MoS2 to MoO3 . . 18
4. Delta G vs Temperature for Reducing Cu2S to C u ........ 21
5. Delta G vs Temperature for Reducing Cu2S to Cu2O ...... 2 2
6. Delta G vs Temperature for Iron Reduction.............. 24
7. Delta G vs Temperature for Lead Reduction.............. 2 6
8. Molybdenum Experimental Procedure....................... 3 0
9. Molybdenum Results........................................  3 2
10. Molybdenum Average Yields...............................  3 3
11. Molybdenum Data Consistency............................  3 4
12. Experimental Copper Blast Furnace...................... 38
13. Copper Percent Conversion vs Time in Blast Furnace... 41
14. Lead Experimental Procedure............................  4 7
15. Lead Conversion vs Weight Ratio of Ore to Coke....... 49
vii
viii
TABLE OF NOMENCLATURE
Symbol Definition Units
Gibbs standard energy change of kcal/mol
reaction at constant temperature T
Standard enthalpy change of reaction kcal/mol 
at constant temperature T
T Temperature K
Standard entropy change of reaction cal/(K*mol) 
at constant temperature T
ACP° Standard heat capacity of reaction cal/(K*mol)at constant pressure
v. Stoichiometric coefficient of each none
component in the reaction, i 
(+) for products, (-) for reactants
cV Standard heat capacity of each component in the reaction, i
R Gcis constant
K Equilibrium constant
H  prod, 298 Sum
the
of the standard enthalpies 
products of reaction
of
"LJ0
n  react, 298 Sum
the
of the standard enthalpies 
reactants in the reaction
of
c°
° prod, 298 Sum
the
of the standard entropies 
products of reaction
of
g°
react, 298 Sum
the
of the standard entropies 
reactants in the reaction
of
cal/(K*mol)
cal/(mol*K) 
none 
kcal/mol
kcal/mol
cal/(K*mol)
cal/(K*mol)
ix
ABSTRACT
The purpose of this thesis is to find a use or uses for 
high sulfur petroleum coke. Research was conducted to 
determine if high sulfur petroleum coke can be used as a 
reagent to convert molybdenum ore, MoS2, to MoO2. The second 
experiment was to determine if the petroleum coke can be used 
as a reagent to convert Cu2S ore to copper at a lower 
temperature than is currently used in industry. The third area 
of research was to determine if high sulfur petroleum coke 
could be used in place of regular coke to reduce iron ore, 
Fe3O4, to pig-iron (Fe). The last experiment was to determine 
if petroleum coke could be substituted for regular coke in the 
initial lead ore roasting process.
Each of these experiments used a similar procedure. The 
petroleum coke was mixed with the metal ore, and the mixture 
was heated in either an oxidizing or reducing atmosphere until 
the ore had been converted to the final product. Then the 
product was chemically analyzed to determine purity and 
percent conversion. For molybdenum, the ore/coke mix was 
cooked in air to facilitate the conversion of MoS2 to MoO2. Air 
was blown through an experimental blast furnace to help 
convert Cu2S ore to elemental copper. A reducing atmosphere 
was required to convert iron ore, Fe3O4, to iron, so the 
ore/coke mix was insulated from the atmosphere with a top 
layer of coke. Lastly, air was blown through the lead ore/coke 
mix to help convert PbS to Pb.
An 83% conversion of MoS2 to MoO2 was obtained by roasting 
an. 8:1 coke:ore mixture in air at SOO0C for 38 minutes. The 
iron ore didn't reduce to iron at 13500C. The best result with 
copper ore was achieved by roasting a 0.5:1 coke:ore ratio at 
700°C with a small amount of air blowing through it. After two 
hours a 6 0% conversion of Cu2S to Cu was achieved. A 43% 
conversion of PbS to Pb was obtained by roasting a 0.1.67:1 
coke:ore ratio in air for 8 minutes at 800°C.
It was concluded that it is highly probably that high 
sulfur petroleum coke is effective in converting MoS2 to MoO2 
and it is highly probable that the petroleum coke can be used 
to convert lead ore to lead. The coke is effective in 
converting copper ore to copper. Lastly, high sulfur petroleum 
coke cannot convert Fe3O4 to iron at 1350°C; a higher 
temperature is probably needed.
IBACKGROUND
Petroleum Coke
Heavy oils that are left over after vacuum distillation 
in a refinery are worth very little since they can only be 
used as fuel. However, light oils can be further processed to 
make saleable products. Lighter oils can be produced .from very 
heavy ones solely by thermal cracking. This process is called 
"coking" because the by-product is petroleum coke.
The feed to a coker is usually the heavy gas oil that is 
left over from vacuum distillation of the lubricating residue. 
,The lubricating residue comes from crude oil distillation. The 
coker produces either delayed coke or fluid coke. Delayed coke 
is available in chunks while fluid coke is in the form of 
small spherical balls. In either case, the chemical 
composition of'the coke is 93 - 99% carbon and the balance is 
sulfur. [1]
Both delayed and fluid coke are valuable as electrolytic 
reducing agents if their sulfur content is below 3%. In the 
refining of sulfur containing crude oil, the pitch, asphalt 
and coke fractions contain the most sulfur. If the crude is a 
high sulfur crude, the coke fraction can become 5 - 7 %  sulfur; 
After coking, most of the sulfur in this fraction ends up in 
the petroleum coke. Currently, high sulfur petroleum coke is
2almost worthless. Even as a fuel, the current restriction is 
one pound of■sulfur per million BTU1s . [2]
The EXXON refinery in Billings, Montana uses a high 
sulfur crude as its feedstock. The petroleum coke produced by 
this refinery is about 94% carbon and 6% sulfur so it cannot 
be used as an electrolytic reducing agent or as a fuel. The 
research in this thesis is based upon trying to find a use for 
the high sulfur petroleum coke produced at the EXXON Refinery 
in Billings. [3] The properties of the EXXON fluid coke are 
presented in Table I. Properties of EXXON Fluid Petroleum
Coke.
3Table I. Properties of EXXON Fluid Petroleum Coke [4]
Sieve Analysis
Cumulative, % Retained Non-cumulative, % Retained
on; IOm --- 2 on: 12m --- 4.2
20 --- 5 20 --- 3.9
48 --- 11 4 0 --- 6.4
6 0 --- 20 5 0 --- 37.6
8 0 --- 44 7 0 --- 29.2
100 --- 65 10 0 --- 10.0
150 --- 88 200 --- 6.4
200 --- 95 3 2 5 --- 1.2
Through 200 --- 5 Through 325 --- 1.1
Particle Density, a/cm3 
Bulk Density. IbZft3 
Calorific Value, BTU/lb (ASTM D-
1.3
55.9
271) 14,100
Proximate Analysis, wt. % (ASTM D- 271)
Moisture 
Volatile Matter 
Fixed Carbon 
Ash
Ultimate Analysis, wt. % (ASTM D-
0.3 
6.0 
93.4 
0.3
271)
Carbon
Hydrogen
Sulfur
Metals, (wt. % on coke)
Nickel
Vanadium
Iron less
90.0
2.0 
6.0
0.013 
0.034 
than 0.01
Calcium less than 0.01
Silicon less than 0.005
Titanium less than 0.001
Sodium less than 0.02
4Molybdenum
Molybdenum was discovered in 1778 by C. W. Scheele when 
he produced a new oxide from molybdenite (MoS2) , thus 
distinguishing the mineral from graphite, with which it had 
been thought to be identical. Today, molybdenum is obtained 
from molybdenite and is a byproduct from the production of 
copper. In both cases, MoS2 is separated by floatation and 
then roasted in air to produce MoO3. About 85% of the MoO3 is 
used in the manufacture of stainless steel and high-speed 
tools. It is used directly or after conversion to 
ferromolybdenum by the aluminotherimc process.
Molybdenum chemicals are synthesized from ammonium 
molybdate which is obtained by dissolving MoO3 in aqueous 
ammonia, then crystallizing the ammonium molybdate. Pure 
molybdenum, which is used in catalysts for a variety of 
petrochemical processes and as an electrode material, is 
obtained from hydrogen reduction of ammonium molybdate. [5]
The intent of the molybdenum research is to determine if 
molybdenum ore, MoS2, can be converted to MoO2 using high 
sulfur petroleum coke as a reagent, thus demonstrating a use 
for the coke. MoO2 is of interest because it contains 11% more 
molybdenum per pound than MoO3.
Several investigators have reported methods of converting
5molybdenite to HoO2:
"V. H . Zazhigalov (1975) employed hydrogen and 
elemental sulfur. N. Satani (1975) used hydrogen- 
thiophene mixtures to reduce the MoO^ hydrate to 
MoO2. J . 0. Besenhard (1976) carried out the 
reduction of MoS2 in dimethylsulf oxide. A. Wold 
(1964) prepared pure MoO2 crystals by the 
electrolytic reduction of MoO3— NaMoO4 solution. W. 
Kunnman (19 61) used a mixture of CO and CO2 to 
reduce MoO3 to MoO2. S . I . Sobol (19 61) used hydrogen 
and CO under pressure at 200°C. to reduce NaMoO4 to 
MoO2. V,. Angelova (1972) reacted molybdenite with 
TiO2 at 50 0°— 900 ° C to produce MoO2, SO2 and 
elemental Ti. F. Zabo (1963) reduced MoO3 with 
ammonia." [6] .
Additional work on the subject includes Conversion of
Molybdenite to Molybdenum Dioxide Using Petroleum or Coal Tar
Pitches. This patent states that "Molybdenite, MoS2, is
completely converted into molybdenum dioxide, MoO2 by mixing
MoS2 with petroleum or coal tar pitches and heating in air at
400 °-600 ° F ." [7]
Conner
Copper is one of the most important metals in the history 
of man. It was in use as far back as 5000 BC, was obtained 
from charcoal reduction in 3500 BC, and was combined with tin 
to establish the "Bronze Age" in about 3000 BC. Today, copper 
is still one of man's most important metals. .
The majority of copper comes from huge open pit mines. 
The ore only contains about 1/2% Cu so it is concentrated to 
15-20% Cu by froth floatation before further processing.
6Silica is added to the concentrate, and the mixture is melted 
in a reverberatory furnace at about 1400 °C. The iron in the 
melt (FeS) is more readily converted to the oxide than the 
Cu2S ore, so it forms an upper layer of iron silicate slag. 
This leaves a lower layer of copper matte which is mostly Cu2S 
and FeS. The liquid matte is poured into a converter, more 
silica is added, and a blast of air is forced through it. The 
air and silica transforms the remaining FeS to FeO and then to 
slag, while the Cu2S is converted to Cu2O and then to copper:
2FeS + 3O2  > 2Fe0 + 2S02
2Cu2S + 3O2  > 2Cu2O + 2S02
2Cu2O + Cu2S ---> 6Cu + SO2
This "blister" copper is usually further purified 
electrolytically for. use in the electronics industry. [8]
The research in this thesis involves using petroleum 
coke as a reagent to convert Cu2S ore to Cu at a lower 
temperature. If this is proven, it will demonstrate a use for 
the high sulfur petroleum coke. The probable reaction is as 
follows:
Cu2S + C + 2O2 ---> 2 Cu + SO2 + CO2
This reaction represents the overall conversion of Cu2S to Cu. 
There is probably an intermediate step where the copper is in 
an oxidized state, but this is not important because the above 
equation is only used to obtain a rough approximation of 
temperatures at which Cu2S can be converted to Cu.
IIron
As with copper, iron has had a tremendous impact on man's 
history. Iron beads dating from about 4000 BC were probably of 
meteoric origin. Iron was first made by low temperature 
reduction methods, but this produces a spongy material that 
could only be shaped by prolonged hammering. The high 
temperature smelting process did not evolve until about 1200 
BC— the start of the "Iron Age". More recently, the use of 
coke as the reducing agent had far-reaching effects, such as 
the start of the Industrial Revolution.
Today, most iron is used as steel of one form or another. 
The first step to convert iron ore to steel is the blast 
f u m a d e . The Fe2O3 ore is reduced to Fe using coke as the 
reducing agent while limestone (CaCO3) removes sand or clay as 
slag (see Figure I, Iron Blast Furnace). The molten iron is 
cast into molds or ingots for further processing. This iron is 
in an impure form containing about 4% carbon and is called 
"cast-iron" or "pig-iron". [9]
8Charge (o re , lim estone, coke)
Waste
gases
-  -  200“ C
SFe2O j +  CO 
CaCO3 —  
Fe3O4 +  CO -
1— —  700° C
C +  CO2
> Fe(S) + CO2
I \ — 1200° C --------------------------------------------------------------------
I \ Impure iron melts
I I Molten slag (largely CaSiO3) forms
J J— 1500° C -------------------------------------------------------------------
/  Z  Phosphates and silicates reduced
Z P and S pass into molten iron
Z_ — ■ —' 2000°C — — — —— — 2C +  O2 ' ► 2CO —  —A ir  b last \ A
(~ 9 0 0 °C )
Hearth
Blast FurnaceFigure Iron
The purpose for this research on iron is to find out if 
petroleum coke could be used in place of regular coke to 
reduce the iron ore to pig-iron.
Lead
Lead is another of man's oldest metals. It was used in 
ancient Egypt for glazing pottery in 7000-5000 BC. The Romans
9used lead for water pipes and plumbing, extracting 6-8 million 
tons in four hundred years. Today, over half the lead produced 
is used in batteries. The remaining amount is used in cable 
sheathing, sheet, pipe, foil, tubes, Pb (CH2CH2)4, solders, 
pigments, and chemicals. Pb (CH2CH2)4 is used as an antiknock 
additive in gasoline, b u t ■ is being phased out by the 
Environmental Protection Agency.
Most lead is obtained from PbS ore. This is concentrated 
from low-grade ores by froth flotation, then roasting in air 
to produce PbO. The PbO is then mixed with coke and limestone, 
and the mixture is reduced in a blast fufnace. The following 
equations illustrate the process.:
PbS + I.SO2 ---> PbO + SO2
PbO + C ---> Pb(Iiq) + CO
' PbO + CO — > Pb(Iiq) + CO2
Another possible reduction method is to replace the reduction 
of the roasted ore with fresh ore (PbS):
PbS + 2PbO -— > SPb(Iiq) + SO2 (g) [10]
The "Green Ore" used at the ASARCO lead refinery in East 
Helena has the following composition:
10
Table 2. Composition of Lead Green Ore [11]
Element W t . %
Lead (PbS) 3 4 — 36
Copper (Cu2S) 3.9-4.I
Arsenic (As2S2, As2S3) 0.8-1.0
Antimony (Sb2S3) I.1-1.3
Sulfur (as the metal sulfides) 57.6-60.2
The ASARCO lead refinery in East Helena uses the 
roasting/blast furnace process. However, they combine the lead 
ore with regular coke in the initial roasting process to speed 
up the production of SO2. The SO2 is then used to produce 
sulfuric acid. The purpose of the lead research in this thesis 
is to determine if petroleum coke could be substituted for 
regular coke in the initial roasting process.
11
THEORETICAL ASPECTS
All of the reactions for molybdenum, copper, iron and 
lead involve rather high temperatures. If petroleum coke is 
going to be used in the production of the metals in question, 
it would be useful to know the thermodynamics of the systems. 
Since the reactions for each experiment are fairly well known, 
as are the thermodynamic, properties for the compounds in 
question, one Can calculate the temperature of neutral 
equilibrium, the Gibbs standard energy change of reaction, and 
the heat of reaction. Since the Gibbs standard energy change 
of reaction must be equal to or less than zero for a feasible 
reaction, and AG = 0 at neutral equilibrium, one can find the 
feasible temperature range for the reaction (see 
Thermodynamics of. Molybdenum Oxidation). Once a feasible 
temperature range is known, one can use it to design 
appropriate experiments and to better evaluate the results of 
these experiments.
Thermodynamics of Molybdenum Oxidation
Molybdenite is to be converted to molybdenum dioxide by 
the following reaction:
MoS2cs0 + Ccs0 + 402(g) -— > MoO2cs0 + 2S02Cg) + C02(g)
12
This reaction is probably not the only reaction taking place 
when MoS2 is oxidized to MoO2, but it will allow a good 
estimate of the temperature of neutral equilibrium and the 
heat of reaction. The following information is available from 
the Handbook of Chemistry and Physics: [12]
Table 3. Thermodynamic Properties of Some Molybdenum Compounds
Standard
Entropy
Compound cal/deg*mol
Gibbs Standard
Energy
kcal/mol
Standard
Enthalpy
kcal/mol
MoS? 14.96 — 54 -56.2
C 1.372 0 0
O2 49.003 0 0
MoO2 11.06 -127.40 -140.76
MoO3 18.58 -159.66 -178.08
SO2 59.30 -71.748 -70.944
CO2 51.06 -94.254 -93.963
The heat of reaction is calculated by subtracting the sum of 
the enthalpies of the reactants from the sum of the enthalpies 
of the products. A negative quantity indicates an exothermic 
reaction, a positive quantity means the reaction is 
endothermic. [13]
Heat of Reaction = ,2 (Product Enthalpies)
- 2 (Reactant Enthalpies)
Heat of Reaction = [-140.76 + .2 (-70.944) + -93.963]
-[-56.2 + 0 + 4 ( 0 ) ]
13
Heat of Reaction = -320.411 kcal/mol 
So the reaction is very exothermic.
An approximation of the temperature range at which a 
reaction is feasible can be determined by calculating the 
Gibbs standard energy change of reaction and the neutral 
equilibrium temperature. A proof of this begins with:
Ag" = Ah" - TAs" (i)
The equation for Ag at one temperature, using AH and AS at 
another temperature is:
T
A — A 1^9 a f A  C p d T
2 9 8
''2 9 8 I
2 98
A C n
(2)
Since
ACp = S v i Cpi  ( 3 )
If you assume that CP(products) = Cp(reactants) then Equation 2 
becomes:
A C t ~ AffgBS — fACggg (4)
It is also know that
14
A G t = -RTlnK (5)
If you arbitrarily choose K - I  for a feasible reaction,
(K. = I gives about a 50% conversion) then InK = 0, and
Equations 4 and 5 can be combined as:
A Gt ~ A 1^298 — lAS^gg = 0 (6)
Equation 6 can be rearranged as:
_ ^ 2 9 8  _ H prod, 298 - H r.eact, 298
1 N-E. -  -------—  “  — ---------------------------:------------ ( ' I
298 S prod, 298 ~ S react, 298
where Tne is the "Temperature of Neutral Equilibrium".
[14]
15
For the molybdenum reaction, the temperature of neutral 
equilibrium is:
1000 cal
kcal
-140.76 jcca^  + 2 (-70.944) kca} + -9 3.96 3-^ccaj
mol mol mol
LN.E.
1000 cal 
kcal
\-56.2kcal + 0 kCal + 4 (0) kcal] 
mol mol mol
I1 1 - 0 6 ^ o l  + 2 < 5 9 -30) + 5 1 - 0 6 ^ o l
14.96 cal 
' K-mol
+ 1.372- Cal + 4 (49.003) -6 Cal
' Kwol ' Kwol
Tn. E. = 10,131 °K
By observation of Equation 5, one can determine that Ag t° 
must be less than or equal to zero for a feasible reaction. 
The Gibbs standard energy change of reaction is calculated by 
subtracting the sum of the AG298°'s of the reactants from the 
sum of the AG298"'s of the products. [15] A negative 
quantity indicates a feasible reaction at 298 K, a positive 
quantity means that the reaction is not feasible.
Gibbs Standard Energy Change of Reaction = S (AG298" Products)
-E(AG298" Reactants)
Gibbs S.E.C.R = [-127.40 + 2(-71.748) + -94.254]
-[-54 + 0 + 4(0)]
Gibbs S.E.C.R = -311.15 kcal/mol
16
Since the Gibbs S.E.C.R is negative, the reaction is 
thermodynamically feasible from 298 K to 10,131 K, the 
temperature of neutral equilibrium. Note that the compounds in 
question probably do not exist at 10,131 K, but the 
thermodynamics state that if they did exist at that 
temperature, the reaction would be feasible.
In addition, the following thermodynamic properties are 
calculated for
m o s Z(S) +  C  +  9/ 202< g ) ----> M o 03(s) +  2s 0 Z(S) +  C 0 2(g)
using the methods outlined above:
Heat of Reaction (AHf298)............. -357.64 kcal/mol
Gibbs Energy of Reaction (AGf298).... -3 4 3.41 kcal/mol
Temp, of Neutral Equilibrium (Tn e )... 7.36 K
17
Figure 2, Delta G vise Temperature for Oxidation of MoS2 to 
MoO2, and Figure 3, Delta G vise Temperature for Oxidation of 
MoS2 to MoO3 provide a comparison of the thermodynamic 
feasibility for the reactions in question.
-150
g -200
-250
-300
2000 4000 6000 8000 10000 12000 14000
Temperature, K
Figure 2, Delta G vs Temperature for Molybdenum Oxidation
18
ble Region
-100
o -150
o -200
-250
-300
150
Temperature, K
OxidationFigure Delta Temperature
MoO3
Note that AG is more negative at lower temperatures in the
MoS2 ---> MoO3 case than in the MoS2 ---> MoO2 case. This shows
that it is more thermodynamically feasible for molybdenum ore 
to oxidize to MoO3 than to MoO2- However, limiting the amount 
of oxygen would modify reaction conditions to favor oxidizing 
the ore to MoO2, and make MoS2 ---> MoO2 possible.
19
Thermodynamics of Copper Reduction
Copper ore is converted to blister copper by the 
following reaction:
C U 2S (s )  +  C (s )  +  2 0 2(g )  ^  2 C U (s )  +  S 0 2(g )  +  C 0 2(g)
The thermodynamic properties of copper are listed in the 
Handbook of Chemistry and Physics: [16]
Table 4. Thermodynamic Properties of Some Copper Compounds
Compound
Entropy
cal/deg*mol
Gibbs Standard
Energy
kcal/mol
Standard
Enthalpy
kcal/mol
Cu2S 28.9 -20.6 -19.0
C 1.372 0 0
O2 49.003 0 0
Cu 7.923 0 0
SO2 59.30 -71.748 -70.944
CO2 51.06 -94.254 -93.963
Cu2O 22.26 -34.9 -40.3
The following thermodynamic properties are calculated for
CUgS(S) + C;,,) + 2 0 ^ ^ ---> 2CU(s) + S(^g) + COg^)
using the methods outlined above:
Heat of Reaction (AHf298)..............-145.907 kcal/mol
Gibbs Energy of Reaction (AGf298).... -145.402 kcal/mol
Temp, of Neutral Equilibrium (Tn e )...70,418 K
In addition, the following thermodynamic properties are
calculated for
2Cu 2S(s) + 302(g) > 2Cu20(g) + 2S02(g)
using the methods outlined above:
Heat of Reaction (AHf298)............. -184.48 kcal/mol
Gibbs Energy of Reaction (AGf298).... -172.096 kcal/mol
Temp, of Neutral Equilibrium (Tn e )... 4.12 K 
Both of these reactions are very exothermic. In addition, the 
Gibbs energy of reaction is negative for both reactions. This
means that the reaction Cu2S ---> Cu is feasible for all real
temperatures and the reaction Cu2S ---> Cu2O is feasible from
4.12 K on up.
21
The feasibility of making Cu is shown in Figure 4, Delta G vs 
Temperature for Reducing Cu2S to Cu:
-100
-150
15000 30000 45000 60000 75000 90000
Temperature, K
Figure 4, Delta G vs Temperature for Reducing Cu2S to Cu
22
The feasibility of making Cu2O from Cu2S is shown in Figure 5, 
Delta G vs Temperature for Reducing Cu2S to Cu2O :
No i -  Feasible
O' -100
5 -120
-140
-160
-180
150
Temperature, K
Figure ReducingDelta Temperature
A comparison of Figures 4 and 5 indicates that the reduction 
to Cu2O is more thermodynamically feasible than the reduction 
to Cu because the AG is more negative for all temperatures in
the Cu2S ---> Cu2O case. However, if the reaction conditions
favor the Cu2S ---> Cu reaction, it could happen. Notice that
the Cu2O reduction requires three oxygens but the Cu reduction
23
only requires two oxygens. If oxygen is limited in the, 
reaction atmosphere, the reduction to Cu would probably be 
favored.
Thermodynamics of Iron Reduction
The iron ore under consideration for this project is 
primarily Fe3O4, not the traditional Fe2O3 because this is what 
is available in Montana mines which are close to Billings. For 
this particular ore, the reaction in question is:
F e 3 ° 4 ( s )  +  2 C (s )  — >  3 F e (s )  +  2 C 0 2(g )
The thermodynamic properties of these iron compounds are
listed in the Handbook of Chemistry and Physics: [17]
Table 5. Thermodynamic Properties of Some Iron Compounds
Compound
Entropy
cal/deg*mol
Gibbs Standard
Energy
kcal/mol
Standard
Enthalpy
kcal/mol
Fe3O4 35.0 -242.7 -267.3
C 1.372 0 0
Fe 6.52 0 0
CO2 51.06 -94.254 -93.963
The following thermodynamic properties are calculated for
F e 3 ° 4 ( s )  +  2 C (s )  — >  3 F e (s )  +  2 C 0 2(g )  
using the methods outlined above:
24
Heat of Reaction (AHf298)..............+79.374 kcal/mol
Gibbs Energy of Reaction (AGf298).... +54.192 kcal/mol
Temp, of Neutral Equilibrium (Tne ) ... 946 K 
The reaction is endothermic. In addition, since AGf298 is 
positive, this particular reaction is only feasible above 946 
K, the temperature of neutral equilibrium, as shown in Figure 
6, Delta G vs Temperature for Iron Reduction:
O
E
X
□
U
o'
©
Cl
70 
60 
50 
40 
30 
20 
10 
0
—  10
0 200 400 600 800 1000 1200
i t -
I eglon Fensihle
cN
Region
X
\
\
V
\
X
Temperature, K
Figure 6, Delta G vs Temperature for Iron Reduction
25
Thermodynamics of Lead Reduction
The lead reaction in question is a combination of the
reactions mentioned in the BACKGROUND:
PbS + C + 202 ---> Pb + SO2 + CO2
The thermodynamic properties of these lead compounds are
listed in the Handbook of Chemistry and Physics: [18]
Table 6. 1Thermodynamic Properties of Some Lead Compounds
Gibbs Standard Standard
Entropy Energy Enthalpy
Compound cal/deg*mol kcal/mol 'kcal/mol
PbS 21.8 -23.6 -24.0
C 1.372 0 0
O2 49.003 0 0
Pb 15.49 0 0
SO2 59.30 -71.748 , -70.944
CO2 51.06 -94.254 -93.963
The following thermodynamic properties are calculated for 
PbS + C + 202 ---> Pb + SO2 + CO2
using the methods outlined above:
Heat of Reaction (AHf298)..............^14 0.9 07 kcal/mol
Gibbs Energy of Reaction (AGf298).... -142.4 0 2 kcal/mol
Temp, of Neutral Eguilibrium (Tn E )..-30,159 K 
Once again, the reaction is very exothermic. The temperature 
of neutral equilibrium is below 0 K, absolute zero, so it is
26
only hypothetical. However, the AG vs T line slopes downward 
from AG = 0 on, so the reaction is feasible for all real
temperatures. The AG vs T line is shown in Figure 7, Delta G 
vs Temperature for Lead Reduction:
-139
Feasible-139.5
-140
-140.5
-141
2 -141.5
-142
-142.5
150
Temperature, K
ReductionFigure Delta Temperature Lead
27
RESEARCH OBJECTIVES
The primary objective of this research is to find a use 
or uses for high sulfur petroleum coke. To this end, research 
was conducted on molybdenum, copper, iron, and lead ores.
The objective of the molybdenum research is to find a use 
for high sulfur petroleum coke in the molybdenum processing 
industry. This is done by determining if molybdenite, MoS2  ^
can be converted to MoO2 using the high sulfur petroleum coke 
as a reagent. This is an attractive area of research because 
most molybdenum used in steel making is currently obtained by 
the conversion of molybdenite to MoO3 by roasting the MoS2 in 
a kiln at about 1100 °F. If the ore could be converted to MoO2 
at 500 °F by using coke as a reagent, the molybdenum 
processing companies would probably save energy. In addition 
to the energy savings, the processing companies could save 
shipping charges because MoO2 contains 11% more molybdenum per 
pound than MoO3.
The objective of the copper research is to find a use for 
high sulfur petroleum coke in the copper industry. The 
experimental approach is to determine if copper ore 
concentrate, Cu2S , can be converted to elemental copper at 
about 700 °C using high sulfur petroleum coke as a reagent. 
This objective was chosen because the current industrial
28
process requires a furnace temperature of about 1400°C and a 
silica reagent. Producing copper at a VOO0C lower temperature 
would probably result in energy savings and therefore a market 
for the high sulfur petroleum coke.
The objective of the iron research is to find a use for 
the petroleum coke in the iron smelting industry. The possible 
use is substituting high sulfur petroleum coke for coke 
obtained from coal in the iron ore smelting process. Regular 
coke costs about $70 a ton plus shipping; high sulfur 
petroleum coke would probably only cost the freight to ship 
it. The reason for this is that high sulfur petroleum coke has 
to much sulfur to be used in traditional petroleum coke 
applications. With this in mind, an iron smelter would save a 
lot if the substitution is successful and the cost of 
petroleum coke remains less than regular coke.
Similarly, the objective of the lead research is to 
determine if high sulfur petroleum coke can be substituted for 
regular coke in the initial ore roasting process. The economic
incentive is the same as in the iron case.
29
MOLYBDENUM RESEARCH
Molvbdenu-m Experiments
Recall that the purpose of this experiment is to find a 
use for high sulfur petroleum coke in the molybdenum industry 
and that using the coke a reagent to convert MoS2 to MoO2 is 
the specific area of interest. The. experiment is graphically 
depicted in Figure 8, Molybdenum Experimental Procedure. The 
coke was ground to a fine powder in a ball mill before use. 
After grinding, a known ratio of coke to molybdenum ore is 
cooked in a crucible for about 3 8 minutes. As the mixture 
cooked, a white gas was evolved, probably SO3. The mixture 
turned from black to a lead gray, and yellow crystals formed 
a thin crust on the top of the mixture and on the sides of the 
crucible. The yellow crystals turned white after cooling.
The ratio calculations considered coke as 94% carbon 6% 
sulfur. The experimental ratios were 4:1 moles carbon to moles 
molybdenum, 8:1 moles carbon to moles molybdenum, and 16:1 
moles carbon to moles molybdenum. It was assumed that burning 
the coke and ore resulted in a mixture of unconverted ore 
(MoS2) , MoO2, and MoO3. This is verified in Molybdenum Results 
and Discussion. The white crystals on the sides of the 
crucible were assumed to be MoO3 since it is the
30
W e i g h  O re  & C o ke C o o k  O re  & C o ke
500 c
38 ml ns
S ca  I e
A d d  S u l f u r i c  A c i d  T o  M i x
P o u r  M i x t u r e  I n t o  B e a k e r
F I I t e r  S o l u t i o n  I n t o  A 
P r e - W e l g h e d  B e a k e rD i s s o l v e  R e m a i n i n g  M oS2
n A c  I d
Coke, Mo02, MoOS
H o t  P I a t e
D IssoIvedm I n s
W e i g h  B e a k e r  W i t h  M oS2 
In  I t  T o  D e t e r m in e  %  Y i e l d E v a p o r a t e
H 2 S 0 4 -M o S 2MoS2
So I u t i o n
O v e r n i g h t
H o t  P I a t e
Lo Med
S c a  I e
O f f
Figure 8, Molybdenum Experimental Procedure
31
most probable solid molybdenum compound which can be both 
yellow or white in crystalline form. [19]
After burning, the mixture was cooled and then leached 
with sulfuric acid to remove any MoO3, MoS2, MoS4, and Mo that 
may have been in the mixture while leaving MoO2. ,[20] 
[21] [22] [23] The leaching involved heating the 
mixture in a beaker with various concentrations of sulfuric 
acid for about 30 minutes. Different concentrations of H2SO4 
were tried to determine if acid concentration had an effect on 
the leaching process.
The compound(s) that dissolved in the sulfuric acid were 
assumed to be primarily unconverted MoS2. It is unlikely that 
MoS4 or Mo were created by roasting MoS2 in air, but this 
assumption is noted. To determine yield, the sulfuric acid was 
evaporated to dryness in a pre-weighed beaker, see Figure 8. 
The pre-weighed beaker was weighed after the sulfuric acid had 
evaporated to determine the amount of MoS2 that did not 
convert to MoO2, and the yield was calculated.
Molvbderium Results and Discussion
'
A problem with the procedure was determining if all the 
unconverted MoS2 had been dissolved in that particular acid 
concentration, or if ,a more concentrated sulfuric acid 
solution was necessary. To determine this, three samples at
32
20, 40, 60, and 70 weight% H2SO4 were tested. This was done for 
the 4:1, 8:1, and 16:1 Carbon:Mo molar ratios, for a total of 
36 tests. The averages of each three-run phase are graphically 
depicted below in Figure 9, Molybdenum Results:
20% 40% 60% 70%
Percent Acid
§§5] 4:1 CiMe Samples WM 8:1 CiMo Samples KSXI 16:1 CiMo Samples
Figure 9, Molybdenum Results
33
As can be see, the acid concentration had no significant 
effect on yield. However, the molar ratio of carbon:Mo seems 
to have a significant effect. This is better demonstrated in 
Figure 10, Molybdenum Average Yields which shows the average 
yield for each carbon:Mo molar ratio.
I UU
90 _
O A _ IOU 
70 -
|
o  / v 
£0 -
:
H - U U
— c n _ Ia> u U
X
-t— AO-
|
C 4-V 4)U
7 0 - I j j i ia) OU CL Ij
20 -
I u
10~ I m
o ■
4:1
Molar Ra
8:1
Ho of Carbon:Mo in
16:1
Samples
Figure 10, Molybdenum Average Yields
Excellent yields of MoO2 were obtained, from 79%-85%. The 
yield of MoO2 seems to get better as the ratio of Carbon to
34
Molybdenum decreased, however, a Population Standard Deviation 
was calculated on the Percent Yield in each category to 
determine data consistency. The results are shown on Figure 
11, Molybdenum Data Consistency. The standard deviations 
ranged from 1.5% to 5.7%, so the data is quite consistent. 
However, the range of Percent Yields in Figure 10 is only 5%. 
This means that experimental error could be the cause of 
Percent Yield getting better as carbonzMo ratio decreased, so 
the trend illustrated in Figure 10 should not be taken as 
absolute fact.
Figure 11, Molybdenum Data Consistency
The data is. very consistent, but is it correct? Recall 
that the mixture in the crucible was assumed to be primarily 
MoO2. Analytical tests were performed on the mixture to 
eliminate other possible molybdenum compounds. Table 7, 
Properties of Some Molybdenum Compounds is an exhaustive list 
of all known compounds that have any combination of 
molybdenum, carbon, sulfur, and oxygen.
3 5
Table 7. Properties of Some Molybdenum Compounds [24]
Compound Color/form Density, g/cm3 Solubility
MoS2 Black
MoO2 Lead Grey
MoO3 White
MoC Grey
Mo2C White
Mo2S3 Steel Grey
MoS4 Brown Powder
MoS3 Black Plates
4.80 H2SO4
6.47 -------
4.692 H2SO4
8.20  ' --------------------
8 .9 --------
5.91 d h HNO3
---- h H2SO4
MoS3 can be eliminated from the list of possible compounds 
since no black plates were observed. MoS2, MoO3, and MoS4 can 
be eliminated because they dissolve in sulfuric acid and would 
have been removed in the leaching process. To confirm that all 
the soluble compounds werte completely removed, two leaches 
were performed. The weight of the solids was reduced by only 
0.7% after the second leach. This was probably the weight of 
the fines that were entrained in the filter paper, so only one
36
leach was necessary. Some of the mixture was place in 100+°C 
nitric acid to see if xit would decompose, indicating the 
presence of Mo2S3. Nothing happened after 20 minutes so there 
was no Mo2S3 present. Only MoC and Mo2C are left. Since both of 
these compounds are only slightly soluble in concentrated 
H2SO4, the MoO2 was cooked until it turned to white crystals, 
probably MoO3. The crystals were then dissolved in 50% H2SO4. 
If all the crystals dissolved, then no MoC or Mo2C was 
present. All but 0.8% of the crystals dissolved easily in hot 
H2SO4. The compound that didn't dissolve was probably unreacted 
coke, so very little MoC and Mo2C were present. These tests 
confirm that the mixture is, with high probability, MoS2, MoO2, 
and MoO3 as was assumed in the experimental procedure.
Finally, a density test showed that the density of the 
mix was 6.0 g/cm3. Since the mixture was a powder, it was 
firmly packed into a graduated cylinder. However, there 
probably was a small void fraction which would give a slightly 
lower density. In any case, the experimentally determined 
density of 6.0 g/cm3 is consistent with a mixture of 80% MoO2 
and 20% MoS2 which would have a calculated density of 6.136 
g/cm3. This provides more supporting evidence that the mixture 
was MoS2, MoO2, and MoO3.
Experimental error could produce misleading data. There 
are several possible sources of error in this experiment. When 
filtering the solution, the filter paper would occasionally
37
float, and some of the solid would get into the filtrate. On 
these occasions, the faulty test was disregarded.
Another source of error occurred When the solids were 
washed out of the beaker during filtration. The literature 
stated that MoS2 was only soluble in hot H2SO4 [25] . Room 
temperature distilled water was used to wash the solids into 
the filter, thus lowering- the temperature and diluting the 
solution. This did not have a major effect since very little 
wash water was used, so the temperature Was only lowered 2 - 
3 0 C .
The analytical tests show that no other known compounds 
that contain molybdenum, sulfur, oxygen, or carbon were 
present. However, without evidence confirming the presence of 
MoO2, one can only say that there is a high probability that 
the data is both correct and consistent. Therefore, it is 
highly probably that high sulfur petroleum coke can be used to 
convert molybdenum ore (MoS2) to MoO2.
38
COPPER RESEARCH
Copper Experiments
Recall that the purpose of the copper research is to find 
a use for high sulfur petroleum coke in the copper industry, 
specifically, converting Cu2S to copper at 700 °C using the 
coke as a reagent. The copper concentrate that was used in the 
experiments came from Montana Resources in Butte. This 
concentrate contained about 26 wt% copper. The experiment 
began by combining 66 wt% concentrate and 33 wt% Coke in an 
experimental blast furnace (see Figure 12, Experimental Copper 
Blast Furnace):
Fur nace Copper/Coke Mix
Bur ner
ExperimentalFigure BlastCopper
39
About 2 standard cubic feet per hour of air was then blown 
into the "furnace" to provide the necessary oxygen for the 
copper reduction reaction. Very little air was blown into the 
furnace to try to maintain a reducing atmosphere. This was 
necessary to minimize the conversion of Cu2S to CuO. Little, 
if any, CuO was produced as shown in the analytical test for 
pure copper in Conner Results and Discussion.
The furnace was heated to about 700°C. This temperature 
was measured on the outside of the furnace, so it only 
approximates the actual temperature of the reaction. Since the 
reaction is very exothermic, the actual temperature inside the 
furnace was probably much higher. Because of the design of the 
furnace, it was not possible to measure the temperature inside 
it. After heating for two hours, the powdered ore and coke had 
formed a lump of solid copper at the bottom of the furnace. 
The copper was separated from the ore and excess coke by water 
flotation. This procedure separates components on the basis of 
density— the lighter components float away, leaving the 
heavier ones. The table below shows the properties of the 
copper compounds in question.
40
Table 8. Properties of Some Copper Compounds [26]
Compound Color Density, g/cm3
Cu2S Black 5.6
CuO Black 6.3
Cu Reddish 8.92
Since the unconverted ore, Cu2S , and coke are much lighter 
than copper, the ore and coke would be washed away, leaving 
the copper.
The actual procedure is as follows: the lump of copper, 
along with the unconverted ore and excess coke was scraped 
out, and placed in a large beaker. A hose was placed with its 
end about 1/4" from the bottom of the beaker and water was 
forced into the beaker. The upward momentum of the water 
carried the lighter components (Cu2S and coke) away, while 
leaving the solid copper lump on the bottom of the beaker. 
After the lump was washed, its color was reddish, and it felt 
like metallic copper. The blister copper was then dried and 
weighed, and the percent conversion was calculated.
Conner Results and Discussion
The data that was obtained from using the method outlined 
in EXPERIMENTAL PROCEDURE is shown Figure 13, Copper Percent 
Conversion vs Time in Blast Furnace:
41
OAOV
7 0  H
c 60" O
E? R n -
>
O j A  _U  4  U 
E TA _O
Q>orv _ZU
I U '
h i : :_r_I L I 3 IU "I
I ' I ' 1 I I
I
2
ime
2
in
2
Bla
2
st F
2
urn
2
DCC
2
i, F
2
r
2 2 2
VZ1ZA I hr. Samples H  2  hr. Samples BTO Deviation From Avg.
Figure 13, Copper Percent Conversion vs Time in Blast 
Furnace
The average conversion of Cu2S to blister copper was 60%. If 
the reaction continued for one hour, the average conversion 
was reduced to about 30%. The data is very consistent, as 
shown by the Deviation from Avg. series.
Again, one must ask if the data is correct. The only 
assumption in the procedure was assuming the solid at the 
bottom of the experimental blast furnace was blister copper. 
The color of the lump was the same as copper, so it couldn't 
have been CuO which is black. However, to show that the lump 
was pure copper, it was analytically tested by the following
42
procedure:
1. Dissolve the questionable copper in concentrated 
nitric acid
2. The acid will turn from clear to green
3. Boil the Solution, evolving brown fumes
4. If the copper has 95% or better purity, the solution 
will turn from green to blue, indicating the 
presence of the Cu2++ anion.
5. If the copper does not have a 95% purity or better, 
the solution will remain green, the color of the Cu1+ 
anion. [27].
This procedure is represented by the following reaction:
Cu + .ZVO3" + 2H* a Cu2++ ■+ AZO2T + H2O
The solid on the bottom of the blast furnace turned green when 
it was dissolved in nitric acid. The solution was heated and 
brown fumes (NO2) were evolved as the solution turned from 
green to a bright blue. The bright blue color positively 
confirms the presence of the Cu2+ ion which will only form if 
the copper has a purity of 95% or better. If the copper does 
not have this purity, it will not form the Cu2"1" ion, instead, 
it will remain as the green Cu1"1" ion. Therefore, the lump in 
the bottom of the furnace was at least 95% pure copper, and 
the above assumption is valid.
43
The experimental error in this procedure is very minimal. 
While separating the unconverted ore from the copper, some 
copper could have been washed away. Since the copper came but 
of the furnace in large lumps, this error is negligible.
After evaluating the assumptions and experimental error,. 
it has been shown that the data is consistent and correct so 
high sulfur petroleum coke can be used to reduce copper ore to 
copper.
J
44
IRON RESEARCH
Iron Experiments
The experiments began by combining various ratios of ore 
to coke in a porcelain crucible. Both the coke and ore were in 
a powdered form. The crucible was then heated with a Fisher 
Blast Burner. This burner resembles a regular bunsen burner, 
but it uses compressed air. Propane was used as the fuel and 
pure oxygen was added to the forced air stream in an effort to 
increase the flame temperature. The coke/ore mix was cooked in 
this manner for about one hour at 1350°C. If elemental iron 
was created, the ore would have become liquid iron. However, 
during the cook the ore changed color, from a dark black to 
dark red to grey, but never actually became a liquid. After 
cooling, it was observed that the ore/coke did not form a lump 
of iron, instead, the ore had changed color but was still in 
a powder form. The experiment was terminated at this point.
Iron Results and Discussion
The iron experiments did not go as planned. The solid 
iron ore, Fe3O4, remained in the powdered form, never entering 
the liquid phase. Instead, it underwent two color changes from
45
black to dark red, then to lead grey. Since Fe3O4 is black, 
Fe2O3 is rust, and FeO is lead grey [28], it is assumed 
that the iron ore was reduced to FeO, but not all the way to 
elemental iron. Since iron melts at 15350C and the blast 
burner only achieved 1350 °C, it probably did not get hot 
enough to complete the reduction from Fe3O4 to Fe. Many 
different heat sources were used in an attempt to get the 
crucible hotter, but none were able to exceed the
oxygen/propane blast burner temperature of 13 500 C .
46
LEAD RESEARCH
• Lead Experiments
The information in this section is pictorially presented 
in Figure 14, Lead Experimental Procedure. To begin, the Green 
Ore used in ASARCO's East Helena plant and powdered coke were 
weighed in specific ratios. Then the mix was cooked in a 
crucible above a bunsen burner at about SOO0C for about 8 
minutes. A constant air flow was directed through the mixture 
while it was stirred continuously. During this time, the ore 
reacted vigorously, evolving much white gas (SO3) and forming 
little liquid spheres of metal. Recall that the green ore 
contains lead, copper, arsenic, and antimony sulfides. All of 
the sulfides except lead are in small concentrations, but they 
can still introduce experimental error. Because of this, the 
unreacted PbS as well as the copper sulfide impurity were 
dissolved in hot sulfuric acid. [29] These compounds had 
to be removed from the mixture because both PbS and the copper 
sulfide are soluble in nitric acid, and would have produced 
experimental error later in the procedure. [30] The hot 
sulfuric acid did not dissolve either the arsenic or antimony 
sulfides. However, these compounds don't have to be removed 
because they won't dissolve in any acid, so they won't cause
47
W e i g h  O re  & C o ke C o o k  O re  & C oke  
w / S t T r r I n g  & A T r
D i s s o l v e  PbS I n  
S u l f u r  I c  A c  I Cf
BO C
PbS
zI 0 m i ns
A  T r 8 0 0  C w/ St i r r i ng
8 m  I ns
Hot. P I a t e
S ca  I e Lo M e d
O ff
P u l v e r i z e  L e a d -  
I m p u r l t y  M i x t u r e
F I I t e r  L e a d ,  
I m p u r i t i e s  F ro m  
D i s s o  I v e d  PbS
Rem ove
G ra v e
Lead
<—  ImpurItIesF ro m
D I s s o I vedL e a d / I m p u r I t I e s
D i s s o l v e  L e a d  In  
H o t  N i t r i c  A c  I d
F I  I t e r  D i s s o l v e d  
L e a d  F ro m
I m p u r I t I e s
W e i g h  L e a d -  
I m p u r i t y  M i x t u r e
m  I ns
Impurities
H o t  P l a t e
D i sso IvedLo MedSca  I e
Lead
Ca  IcuI a t e  %
W e i g h  I m p u r i t i e sC o n v e r s  I o n
C O . 35 X Ore) ~
C D IssoIv e d  Pb)
C O . 35 * Ore)
S ca  I ec %  Conv O-IOO
ProcedureLead
48
experimental error later in the procedure. [31] The
dissolved PbS and Cu2S were then filtered from the
lead/impurities mixture and the mixture was dried. The green 
ore contained some 1/8" pieces of gravel. The gravel was 
easily identified at this point because the sulfuric acid 
bleached them from grey to white. The pieces of gravel were 
removed from the mixture by hand. The gravel was not weighed 
because this information is not necessary to determine the 
percent conversion of lead ore to lead. The mix was then 
pulverized. After pulverization, the mix was weighed. The lead 
in the mixture was then dissolved in nitric acid, and 
filtered. The filtrate was dried and weighed, then by 
subtraction the amount of pure lead created in the initial 
cook could be calculated. Since the Green Ore contains 35 wt.% 
lead, the percent conversion from Green Ore to lead could then 
be calculated.
Lead Results and Discussion
Various ratios of coke/ore were tested. The 6:1 grams ore 
to grams coke ratio was best because the reaction went to lead 
in the least amount of time. Figure 15, Lead Conversion vs 
Weight Ratio of Ore to Coke, graphically shows the 
experimental results.
49
cO
M 30
<B
>
CO
u 20
.75 :1  1:1 1.5:1 3:1 1:0 1:0 1:0 1:0 1:0 6:1 6:1 6:1 6:1 6:1
Ore Io Coke Weight Ratio
Figure 15, Lead Conversion vs Weight Ratio of Ore to Coke
The average conversion with the 6:1 weight ratio is 43%, with 
a 2.8% Standard Deviation.
It is possible to reduce lead ore (PbS) directly to lead 
without a reducing agent. N . N. Greenwood and A. Earnshaw 
state that "Alternatively, the carbon reduction can be 
replaced by reduction of the roasted ore with fresh galena:
[32]"
50
2 PbS + 3 O2 2 2 PbO + 2 S1O2 T 
PbS + 2PbO 2 3Pb (Iiq) + SO2T
The samples that were tried without coke for this research 
(1:0 grams ore to grams coke) appear to have a high 
conversion, see Figure 14. However, they took longer to react, 
15+ minutes compared to 8 minutes.
Figure 14 shows that the data is consistent. However, is 
it correct? The first assumption made in the procedure is that 
the sulfuric acid dissolves all of the unreacted PbS, but none 
of the lead. To begin, PbS dissolves completely in sulfuric 
acid. [33] Since the sulfuric acid was used at a 50% 
concentration, and the dissolving process took place at 60oC, 
only a very negligible amount of lead would have been 
dissolved in the H2SO4. [34] The second assumption is that 
all the lead is dissolved in nitric acid. The Handbook of 
Chemistry and Physics states that lead is soluble in HNO3. To 
insure that all the lead would dissolve, 150 grams of hot 
concentrated HNO3 was added to a maximum of 1.05 grams lead. [35]
There is only one source of experimental error in this 
procedure. When filtering the solutions, the filter paper 
would occasionally float, and some of the solid would get into 
the filtrate. On these occasions, I would disregard the faulty 
test and start over from the beginning of the procedure, so
51
the amount of experimental error is negligible.
Since there is little experimental error, and no 
obviously bad assumptions, the data appears to be correct and 
consistent. However, since the mix was never analytically 
tested to confirm the presence of lead, one can only say that 
it is highly probable that lead was created. This is 
acceptable because the intent of the lead research is to see 
if high sulfur petroleum coke can be used in any capacity in 
the lead refining industry.
52
SUMMARY AND CONCLUSIONS
Through carefully selected and conducted experiments, it 
was found that:
1. It is highly probable that high sulfur petroleum coke can 
be used to convert molybdenite to MoO2 because 83% of 
molybdenite, MoS2, was oxidized to MoO2 by combining 0.660 
grams high sulfur petroleum coke with one gram of MoS12 and 
roasting in air at 500°C for 38 minutes.
2. High sulfur petroleum coke can be used to reduce ■ copper 
concentrate to copper because 60% of the Cu2S in copper 
concentrate can be converted to blister copper. The procedure 
is to mix 66 wt% concentrate with 34 wt% high sulfur petroleum 
coke and roast the mixture in a blast furnace at 700°C for two 
hours.
3. Iron ore, Fe3O4, cannot be converted to elemental iron by 
combining it with high sulfur petroleum coke and roasting in 
air at 1350°C. Instead, it is reduced to FeO.
53
4. It is highly probable that high sulfur petroleum coke can 
be substituted for regular coke to reduce lead ore to lead 
because 43% of the PbS in lead ore was converted to elemental 
lead by combining 17 wt% high sulfur petroleum coke with 83 
wt% ore and roasting in air at 800°C for 8 minutes while 
blowing air through the mixture.
Three of the four areas of research show promise for uses for 
high sulfur petroleum coke, a substance that was previously 
thought of as an unwanted waste product. This research should 
be very useful to any refinery that produces fluid petroleum 
coke with a sulfur content over 3%, high sulfur petroleum
coke.
54
RECOMMENDATIONS FOR FUTURE RESEARCH
This research is only the first step in the process to 
use high sulfur petroleum coke in the metals refining 
industries. Future research should include finding and 
employing an analytical test to absolutely confirm the 
presence of MoO2 and elemental lead. Since the iron 
experiments at 13500C were not successful, future experiments 
should be conducted at 1600°C, above the melting point of 
iron. Other than that, there is little research that can be 
performed in a lab. The experiments, when run on a lab scale, 
were usually successful. However, the coke needs to be proven 
on an industrial scale before total success can be claimed.
More specifically, the scale up in the molybdenum 
industry should include mixing the MoS2 with powdered coke and 
then roasting at a lower temperature to obtain MoO2 instead of 
MoO3. The copper industry should try adding coke to the 
ore/silica mix in the blast furnace. The experiments in this 
thesis indicate that adding the coke would allow the 
conversion to proceed at a lower temperature. Finally, the 
lead industry could prove the utility of petroleum coke by 
combining it with the lead ore in the initial roast. If the 
petroleum coke works as well in the actual industry as it 
worked in the lab, it could be substituted for the regular 
coke breeze the lead industry is currently using.
REFERENCES CITED
56
1. Austin, G., Shreve's Chemical Process Industries, 5th ed., 
McGraw-Hill, New York, 1984, pp. 722-742.
2. Berg, L., and L. Ulrich, U.S . Patent Application 
00/000,000, filed on May I, 1991, p. 3.
3. Ibid., p. 4.
4. Suiter, R.C., Silicon Carbide Synthesis Using High-Sulfur 
Petroleum Fluid Coke and Montana Silicon. Ph.D. Thesis in 
Chemical Engineering, Montana State College, Bozeman, 
Montana, June 1965.
5. N. N. Greenwood and A. Earnshaw, Chemistry of the Elements, 
1st ed., Pergamon Press Inc., New York, 1984, p. 1168.
6. Berg, L., U.S. Patent Application 4,687,647, January I, 
1991.
7. Ibid., p. 2.
8. N . N. Greenwood and A. Earnshaw, op. cit., pp. 1365-1366.
9. Ibid., pp. 1243-1244.
10. Ibid., pp. 429-430.
11. McIntyre, T.F., Process Engineer, ASARCO Lead Refinery, 
Letter to L. Berg, December 7, 1990.
12. Weast, R.C., M-Jv Astle, and W.H. Beyer, CRC Handbook of 
Chemistry and Physics, 65th ed., CRC Press, Boca Raton, 
Florida, 1984, pp. B-116 to B-117.
13. Felder, R., and R.W. Rousseau, Elementary Principles of 
Chemical Processes, 2nd ed., John Wiley and Sons, New 
York, 1986, p. 422.
14. Dr. Frank P . McCandless, Ph.D., Chemical Engineer, private 
communication, May 8, 1991.
15. Smith, J .M ., and H.C. Van Ness, Introduction to Chemical 
Engineering Thermodynamics. 4th ed., McGraw-Hill, New 
York, 1987, p. 504.
16. Weast, R.C., M.J. Astle, and W.H. Beyer, op. cit., pp. 
B-92 to B-94.
17. Ibid., p. B-104.
18. Ibid., pp. B-105 to B-107.
19. Ibid., p. B-116.
20. Ibid., p p . B-116 to B-117.
57
21. Parker, G.A., Analytical Chemistry of Molybdenum. 
Springer-Verlag, Berlin, Germany, 1983, pp. 8-32.
22. Browning, P.E., Introduction to the Rarer Elements, 2nd 
ed., John Wiley and Sons, New York, 1908, pp. 125-131.
23. Samsonov, G., The Oxide Handbook, 2nd ed. , IFI/Plenum Data 
Company," New York, 1982, pp. 329-330.
■24. Weast, R.C., M.J. Astle, and W.H. Beyer, op. cit., pp. 
B-116 to B-117.
25. ■Ibid., p. B-117.
26. Ibid., p p . B-92 to B-94.
27. Dr. Kenneth Emerson, Ph.D ., Physical Chemist, private , 
communication, June 25, 1990.
28. Weast, R.C., M.J. Astle, and W.H. Beyer, op. cit., pp. 
B-105 to B—107.
29. Ibid., pp. B-94 to B-107.
30. Ibid., p. B-107.
31. Ibid., pp. B-74 to B-75.
32. N. N . Greenwood and A. Earnshaw, op. cit., p. 430.
33. Weast, R.C., M.J. Astle, and W.H. Beyer, op. cit., p. 
B-107.
34. Hofmann, W., Lead and Lead Alloys Properties and 
Technology. 2nd ed. , Springer-Verlag, New York, 1970, pp. 
268-275.
Weast, R.C., M.J. Astle, and W.H. Beyer, op. cit., p. 
B-107.
35.
58
APPENDICES
59
APPENDIX A
MOLYBDENUM EXPERIMENTAL DATA
60
Table 9. 4:1 Molybdenum:Coke Samples
4:1 Samples 20% Acid
Starting
MoS2
Weight
(g)
2.85
2.85
2.85
Beaker
Before
Filtrate
(g)
70.34 
69 . OS 
111.10
Beaker Unreacted Population
After Dried MoS2 % Yield Average Standard
Filtrate Weight Mo02 Yield Deviation
(g) (g) (%) (%)
70.43 0.09 96.84 91.23 4.06
69.35 0.30 89.47
111.46 0.36 87.37
4:1 Samples 40% Acid
Starting
MoS2
Weight
(g)
2.85
2.85
2.85
Beaker
Before
Filtrate
(g)
69.01 
67.28 
69.58
Beaker Unreacted
After Dried MoS2 % Yield
Filtrate Weight Mo02
(g) (g)
69.53 0.52 81.75
67.45 0.17 94.04
69.85 0.27 90.53
Population 
Average Standard 
Yield Deviation 
(%) ' (%)
88.77 5.16
4:1 Samples 60% Acid
Starting 
MoS 2 
Weight
(g)
2.85
2.85
2.85
Beaker
Before
Filtrate
(g)
68.90
67.28
69.50
Beaker Unreacted
After Dried MoS2 
Filtrate Weight
(g) (g)
69.06 0.16
67.59 0.31
70.04 0.54
% Yield 
Mo02
94.39
89.12
81.05
Population 
Average Standard 
Yield Deviation 
(%) (%)
88.19 5.48
4:1 Samples 70% Acid
Starting
MoS2
Weight
(g)
2.85
2.85
2.85
Beaker
Before
Filtrate
(g)
69.00 
67.26 
69.48
Beaker Unreacted
After Dried MoS2 % Yield
Filtrate Weight Mo02
(g) (g)
69.78 0.78 72.63
67.88 0.62 78.25
70.02 0.54 81.05
Population 
Average Standard 
Yield Deviation 
(%) (%)
77.31 3.50
61
Table 10. 8:1 Molybdenum:Coke Samples
8:1 Samples 20% Acid
Starting
MoS2
Weight
(g)
1.43
1.43
1.43
Beaker
Before
Filtrate
(g)
69.01 
67.27 
68.34
Beaker Unreacted
After Dried MoS2
Filtrate 
(g)
69.24 
67.49 
68.39
Weight
(g)
0.23
0.22
0.05
% Yield 
Mo02
83.86 
84.56 
96.49
Average
Yield
(%)
88.30
8:1 Samples 40% Acid
Starting 
MoS 2 
Weight 
(g) '
1.43
1.43
1.43
Beaker
Before
Filtrate
(g)
68.32 
70.35 
6 8.9 0-
Beaker Unreacted
After Dried MoS2
Filtrate
(g)
68.60 
70.66 
69.27
Weight
(g)
0.28
0.31
0.37
% Yield 
Mo02
80.35
78.25
74.04
Average
Yield
(%)
77.54
8:1 Samples 60% Acid
Starting 
MoS 2 
Weight
(g)
1.43
1.43
1.43
Beaker
Before
Filtrate
(g)
69.66
70.33
68.90
Beaker Unreacted
After Dried MoS2
Filtrate
(g)
69.82 
70.51 
69.19 \
Weight
(g)
0.16 
0.18 
0.29
% Yield 
Mo02
88.77
87.37
79.65
Average
Yield
(%)
85.26
8:1 Samples 70% Acid
Starting 
MoS 2 
Weight 
(g)
1.43
1.43
1.43
Beaker
Before
Filtrate
(g)
67.15
67.46
69.60
Beaker Unreacted
After Dried MoS2
Filtrate
(g)
67.35 
67.71 
69.80
Weight
(g)
0.20
0.25
0.20
% Yield 
Mo02
85.96
82.46
85/96
Average
Yield
(%)
84.80
Population
Standard
Deviation
(%)
5.80
Population
Standard
Deviation
(%)
2.63
Population
Standard
Deviation
(%)
4.01
Population
Standard
Deviation
(%)
1.65
62
Table 11. 16:1 Molybdenum:Coke Samples
16:1 Samples 20% Acid
Starting 
MoS 2 
Weight 
(g)
0.72
0.72
0.72
Beaker
Before
Filtrate
(g)
68.30
70.30 
68.88
Beaker Unreacted
After Dried MoS2 
Filtrate Weight
(g) (g)
68.43 0.13
70.41 0.11
69.01 0.13
Population
% Yield Average Standard
Mo02 Yield Deviation
(%) (%)
81.82 82.75 1.32
84.62
81.82
16:1 Samples 40% Acid
Starting Beaker 
MoS2 Before
Weight Filtrate
(g) (g)
0.72 67.15
0.72 67.45
0.72 69.59
Beaker Unreacted
After Dried MoS2 % Yield
Filtrate Weight MoD2
(g) (g)
67.22 0.07 90.21
67.54 0.09 87.41
69.65 0.06 . 91.61
Population 
Average Standard 
Yield Deviation 
(%) (%)
89.74 1.74 -
16:1 Samples 60% Acid
Starting
MoS2
Weight
(g)
0.72
0.72
0.72
Beaker
Before
Filtrate
(g)
68.33
70.34 
68.90
Beaker Unreacted
After Dried MoS2 % Yield
Filtrate Weight MoQ2
(g) (g)
68.51 0.18 74.83
70.56 0.22 69.23
69.08 0.18 74.83
Population 
Average Standard 
Yield Deviation 
(%) (%)
72.96 2.64
16:1 Samples 70% Acid
Starting 
MoS 2 
Weight 
(g)
0.72
0.72
0.72
Beaker
Before
Filtrate
(g)
68.31
70.31 
67.43
Beaker Unreacted
After Dried MoS2 % Yield
Filtrate Weight Mo02
(g) (g)
68.47 0.16 77.62
70.53 0.22 69.23
67.60 0.17 76.22
Population 
Average Standard 
Yield Deviation 
(%) (%)
74.36 3.67
63
APPENDIX B
COPPER EXPERIMENTAL DATA
64
Table 12. Copper Experimental Data
Time in Starting Starting Starting After
Blast Copper Ore Copper Coke Cook Cu Coppe
Furnace Weight Weight Weight Weight
(min) (g) (g) (g) (g) (%)
120 50 13 25 8.6 65.8
120 50 13 25 6.6 50.8
120 50 13 25 7.7 59.5
120 50 13 25 7.5 57.3
120 50 13 25 8.2 62.7
120 50 13 25 9.9 76.2
120 50 13 25 5.9 45.1
12 0 50 13 25 8.1 62.5
120 50 13 25 6.2 51.6
120 50 13 25 8.4 64.3
120 50 13 25 8.6 66.2
120 50 13 25 4.6 35.4
60 50 13 25 4.7 36.0
60 50 13 25 2.8 21.5
60 50 13 25 5.1 39.4
60 50 13 25 3.5 26.9
60 50 13 25 3.1 23.5
Yield
APPENDIX C
LEAD EXPERIMENTAL DATA
66
Table 13. Lead Experimental Data
Green
Ore
(9)
Coke
(g)
Ore:coke 
Ratio
Lead in 
Green Ore 
(35% of Green 
Ore, g)
Wt. After 
Gravel Removed, 
Sulfuric Acid 
Treatment, (g)
Wt. After 
Nitric Acid 
Treatment 
(g)
Lead in
Sample
(g)
% Conversion 
Green Ore to 
Lead, (%)
3.00 4.00 .75:1 1.05 1.39 0.92 0.47 44.763.00 3.00 1:1 1.05' 0.90 0.56 0.34 32.383.00 2.00 1.5:1 1.05 1.15 0.82 0.33 31.433.00 1.00 3:1 1.05 0.84 0.56 0.28 26.673.00 0.00 1:0 1.05 1.50 1.06 0.44 41.903.00 0.00 1:0 1.05 1.90 1.36 0.54 51.43
3.00 0.00 1:0 1.05 1.90 1.54 0.36 34.29
3.00 0.00 1:0 1.05 1.85 1.36 0.49 46.67
3.00 0.00 1:0 1.05 1.70 1.26 0.44 41.90
3.00 0.50 6:1 1.05 1.35 0.91 0.44 41.90
3.00 0.50 6:1 1.05 1.28 0.86 0.42 40.00
3.00 0.50 6:1 1.05 1.26 0.76 0.50 47.62
3.00 0.50 6:1 1.05 1.35 0.91 , 0.44 41.90
3.00 0.50 6:1 1.05 1.35 0.87 0.48 45.71
Standard 
Deviation 
of 17%
Coke Samples: 
2.80
Average %Conv. 
of No Coke 
Samples:
43.24