A STUDY IN THE OXIDATION
OF KRAFT BLACK LIQUOR
PETER MAURO RICCA
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Abstract of Dissertation Presented to the Graduate Council
in Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy
A STUDY IN THE OXIDATION OF KRAFT BLACK LIQUOR
Peter Mauro Ricca
This dissertation was undertaken as a fundamental research project
at the University of Florida Air Pollution Laboratory. Black liquor
oxidation,a process which reduces odors and other air-borne chemical
losses, has seen limited use in the Kraft paper industry. Technical
difficulties caused by unique characteristics of southern black liquor
have impeded the employment of this process in the southeastern United
States. In this research an attempt was made to provide some general
process design criteria which are applicable regardless of location or
liquor characteristics. The relative importance of temperature, oxidant
pressure, method of contact, and type of gaseous oxidizing agent was
determined. The chemical reactions that occurred during oxidation were
A southern Kraft black liquor containing 6.1 grams per liter
of sodium sulfide was oxidized using pure oxygen. This alkaline
sodium base black liquor also contained polysulfide, thiosulfate,
sulfate, carbonate, hydroxide, and soluble organic matter. A non-
foaming bench scale contactor was used and certain physical and chemical
parameters varied. The predominant inorganic reactions that occurred
were the oxidation of sodium hydrosulfide to sodium polysulfide and
sodium thiosulfate. The thiosulfate reaction was slightly reversible
although the polysulfide reaction was not. The mole ratio of poly-
sulfide to thiosulfate formed by complete oxidation of the hydrosulfide
was relatively independent of temperature. Approximately 1.0 moles
of sodium polysulfide was formed for every 1.6 moles of sodium thio-
sulfate produced. Oxidation below 600C precipitated small quantities
of amorphous sulfur droplets from the liquor. These droplets upon
aging crystallized to rhombic sulfur which in turn decomposed and
re-dissolved in the alkaline black liquor.
The method of gas-liquid contact affected the rate of reaction
but not the quality nor quantity of the major inorganic reaction
products. Oxidation at 60C required the least volume of oxygen, but
oxidation proceeded most rapidly at 750C. Therefore the optimum
oxidation temperature lay between 600C and 750C. When adequate gas-
liquid contact was made available, the reaction rate was limited by
the oxidation rate of hydrosulfide. This oxidation very closely
approximated a first-order reaction.
Evidence collected suggested that the organic material present
in the black liquor catalyzed the oxidation of the hydrosulfide.
Oxidation took place in two distinct steps: first, absorption of
oxygen and, second, the utilization of this oxygen to oxidize the
hydrosulfide. Undesirable side reactions that also utilize some of
the absorbed oxygen were increasingly prevalent above 600C.
Pure oxygen at atmospheric pressures had only limited effective-
ness in oxidizing the odorous constituents of black liquor. Oxi-
dation fixed the inorganic hydrosulfide but at least a portion of the
organic still remained volatile. These materials were subject to loss
when the black liquor was recycled to reclaim the unused chemicals.
Thus it was necessary to further treat the oxidation exhaust gases to
completely eliminate odors. These residual odors were identified and
effective means to destroy them with ozone were investigated. Ap-
proximate costs and suggestions for mill-scale application of pure
oxygen and ozone were discussed.
I wish to express my appreciation to the Graduate Committee
chaired by Dr. E. R. Hendrickson, under whose direction and guidance
this study was conducted. The Committee, which included Professors
W. O. Ash, A. P. Black, F. W. Gilcreas, T. deS. Furman, and J. E.
Kiker, aided me greatly with continued encouragement and assistance.
To Professor Gilcreas, I owe special thanks for his help in obtaining
necessary equipment and materials for this study.
Appreciation is expressed to Mrs. Marjorie DuMez and Miss
Janice Brockett who aided in the preparation of the manuscript.
This investigation was supported in part by the National
Council for Stream Improvement, the United States Public Health
Service, and the University of Florida.
TABLE OF CONTENTS
ACKNOWLEDGMENTS . . . . .. . . . . . . ii
LIST OF TABLES . . . . . . .... . . . . .. iv
LIST OF FIGURES. .. . . . . . . . . . . vi
LIST OF PLATES . . . . . . . . . .. . viii
I. INTRODUCTION . . . . . . . . .. .... 1
II. PURPOSE AND SCOPE . . . . . . . . . 12
III. SURVEY OF PREVIOUS WORK . . . . . . ... .15
IV. THEORY . . . . . . . . .. . . . 31
V. EXPERIMENTAL . . . . . . . . ... . 39
VI. DISCUSSION OF RESULTS. .. . . . . . . 53
VII. SUMMARY AND CONCLUSIONS. . . . . . ... 98
VIII. SUGGESTIONS FOR FURTHER STUDY . . . . .... 104
I. METHODS OF CHEMICAL ANALYSIS . . . . . .. 107
II. DETERMINATION OF REACTION CONSTANTS . . . .. 129
III. COMPLETE BLACK LIQUOR ANALYSIS . . . . ... .133
BIBLIOGRAPHY . . . . . . . . .. . . . 136
BIOGRAPHICAL SKETCH . . . . . . . . .. .. . 140
LIST OF TABLES
1. SOURCES OF AIR-BORNE GASEOUS AND PARTICULATE EMISSIONS
IN THE KRAFT PROCESS . . . . . . . . . 7
2. EFFECT OF OXIDATION ON THE SULFUR COMPOUNDS IN A NORTHERN
BLACK LIQUOR . . . . . . . .... ...... 25
3. MOLAR SULFUR BALANCE OF DATA IN FIGURE 3 . . . .. 29
4. EFFECT OF OXIDATION ON HYDROGEN ION, HYDROXIDE, AND CARBONATE
CONCENTRATIONS IN BLACK LIQUOR . . . . . ... 55
5. MATERIAL BALANCE OF KNOWN AND UNKNOWN SULFUR COMPOUNDS . 58
6. SULFUR BALANCE BEFORE AND AFTER COMPLETE OXIDATION . .. 62
7. SULFUR BALANCE DURING COMPLETE OXIDATION . . . ... 63
8. CONCENTRATION OF AMORPHOUS SULFUR FORMED BY BLACK LIQUOR
OXIDATION . . . . . . . . ... ...... 66
9. RATIO OF THIOSULFATE TO POLYSULFIDE PRODUCED BY COMPLETE
OXIDATION . . . . . . .. . .. . . ... 72
10. EFFECT OF PURE-OXYGEN OXIDATION ON THE SULFUR COMPOUNDS IN
BLACK LIQUOR . . . . . . . . ... . . 73
11. EFFECT OF AIR OXIDATION ON THE SULFUR COMPOUNDS IN BLACK
LIQUOR . .. . . . . . . . . . . . 74
12. REVERSION OF POLYSULFIDE AND THIOSULFATE UPON ANAEROBIC
STORAGE . . . . . . . . ... . . . . 76
13. REGENERATION OF SULFIDE UPON ANAEROBIC STORAGE . . .. 78
14. PERCENTAGE REGENERATION OF SULFIDE UPON ANAEROBIC STORAGE 80
15. KINETICS OF BLACK LIQUOR OXIDATION . . . . ... 84
16. VOLUME OF OXYGEN ABSORBED BY BLACK LIQUOR DURING OXIDATION 87
17. UTILIZATION OF DISSOLVED OXYGEN . . . . . ... 89
18. ODOR THESHOLDS OF ORGANIC SULFUR COMPOUNDS FOUND IN BLACK
LIQUOR . . . . . . . . ... . . . . . 93
19. GAS CHROMATOGRAPHY INSTRUMENT CALIBRATION . . . . 128
LIST OF FIGURES
1. THE KRAFT (SULFATE) PROCESS FOR PULP PRODUCTION . .. 5
2. UTILIZATION OF OXYGEN BY BLACK LIQUOR . . . .. 18
3. FATE OF SULFUR COMPOUNDS DURING OXIDATION AT 100 120 PSI
PRESSURE USING PURE OXYGEN . . . . . ..... 27
4. HIGH VOLTAGE SILENT DISCHARGE OZONE GENERATOR . . . 41
5. FLOW DIAGRAM OF STATIC SYSTEM OXIDATION EQUIPMENT . . 43
6. FLOW DIAGRAM OF DYNAMIC SYSTEM OXIDATION EQUIPMENT . . 44
7. SAMPLING EQUIPMENT FOR CONCENTRATING AND COLLECTING
ODOROUS GASES . . . . . . . . ... . .. 52
8. EFFECT OF OXIDATION ON THE SULFUR COMPOUNDS IN BLACK
LIQUOR. STATIC OXIDATION WITH PURE OXYGEN AT 75C . . 56
9. RATE OF FORMATION OF KNOWN AND UNKNOWN SULFUR COMPOUNDS.
STATIC BLACK LIQUOR OXIDATION WITH PURE OXYGEN AT 750C . 59
10. THE OXIDATION OF SULFIDE TO POLYSULFIDE AND THIOSULFATE.
STATIC BLACK LIQUOR OXIDATION WITH PURE OXYGEN AT 750C . 61
11. SULFIDE OXIDATION AND REGENERATION IN BLACK LIQUOR. STATIC
OXIDATION WITH PURE OXYGEN AT 750C AND 12-HOUR ANAEROBIC
REGENERATION AT 75C . . . . . . . . .. 79
12. THE KINETICS OF SULFIDE OXIDATION IN KRAFT BLACK LIQUOR.
STATIC OXIDATION WITH PURE OXYGEN AT 750C . . . . 82
13. THE KINETICS OF SULFIDE OXIDATION OF KRAFT BLACK LIQUOR.
DYNAMIC OXIDATION AT 75C WITH PURE OXYGEN AND PURE OXYGEN
PLUS OZONE . . . . . . . . ... . . . 85
14. OXIDATION OF SULFIDE IN BLACK LIQUOR AND SYNTHETIC LIQUOR.
STATIC OXIDATION WITH PURE OXYGEN AT 750C . . ... 90
15. OXIDATION OF BLACK LIQUOR IN A STORAGE TANK. SUGGESTED
EQUIPMENT REQUIRING LITTLE CAPITAL INVESTMENT . . .. 95
16. OXIDATION OF BLACK LIQUOR IN A PRESSURE-DIFFUSER
CONTACTOR. SUGGESTED EQUIPMENT REQUIRING LITTLE CAPITAL
INVESTMENT . . . . . . . . . . . 96
17. OXIDATION OF BLACK LIQUOR IN A RECIRCULATING PACKED
TOWER . . . . . . . . .. . . . . . 97
18. SAMPLE STANDARD CURVE FOR SODIUM SULFIDE ANALYSIS . .. 109
19. SAMPLE STANDARD CURVE FOR COLORIMETRIC OZONE ANALYSIS 117
20. CURVE FITTING PROCEDURES FOR DETERMINING THE OXIDATION
KINETICS OF SODIUM SULFIDE IN BLACK LIQUOR. SYSTEM USING
STATIC OXYGEN FEED AND TEMPERATURES OF 600C TO 890C . 131
21. CURVE FITTING PROCEDURES FOR DETERMINING THE OXIDATION
KINETICS OF SODIUM SULFIDE IN BLACK LIQUOR. SYSTEM USING
DYNAMIC OXYGEN FEED AT A TEMPERATURE OF 750C ...... 132
LIST OF PLATES
1. PHOTOGRAPH OF EXPERIMENTAL EQUIPMENT . . . . ... 42
2. PHOTOMICROGRAPH OF FRESHLY PRECIPITATED AMORPHOUS SULFUR
DROPLETS . . . . . . . ... . . . 68
3. PHOTOMICROGRAPH OF PRECIPITATED SULFUR AFTER AGING FOUR
HOURS AT 250C. . . . . .. . . . .... 68
4. PHOTOMICROGRAPH OF PRECIPITATED SULFUR AFTER AGING THREE
DAYS AT 250C . . . . . . . . ... . . 69
5. PHOTOMICROGRAPH OF COMMERCIAL GRADE ELEMENTAL SULFUR . 69
Atmospheric Pollution and the Pulp and Paper Industry
Only in recent years have the expansion of industrial activity
and concentration of population created severe and widespread problems
known as atmospheric pollution. The Manufacturing Chemists Association
defines atmospheric pollution as "the presence in the air of substances
put there by the acts of man, in concentrations sufficient to interfere
with the comfort, safety, or health of man, or with the full use and
enjoyment of his property." Therefore, in order to have atmospheric
pollution problems, three conditions must exist simultaneously: there
must be a source of pollutants; a susceptible population or other
receptor; and a mechanism for transport between the two.
"The source may be any activity of man or nature which releases
to the atmosphere any dust, fume, gas, mist, odor, smoke, or vapor. The
population or receptor can be people, vegetation, domestic animals,
homes or manufacturing plants. Susceptibility of the population is
measured in terms of health, economics, aesthetics, and toxicity. For
a problem to exist, source and susceptible population must be brought
together by the common bond of transport which is controlled by topography
Numerous sources of pollution may be active in any given locality.
Industrial operations, steam-electric generating plants, food processing
operations, incinerators, and automobiles are some of the major contrib-
utors of air-borne material to the environment. Since atmospheric
pollution cannot be legislated out of existence, control activities must
be governed by sound economic, nuisance, and health factors.
Pulp and paper manufacturing is the fifth largest manufacturing
industry in the United States. Except in areas where low cost water and
rail transportation are available, pulp mills are frequently situated
near their sources of raw materials. The vast forests of the southern
Appalachian mountains and the gulf coast states have prompted industrial
growth in the wood and wood-products field. Today the southeastern
United States is one of the greatest pulp and paper producing areas of
Associated with pulp and paper making are various problems of
waste treatment and environmental change. Over the past few years the
Air Pollution Research Laboratory at the University of Florida has
studied the pollutional aspects of the air-borne emissions from the
sulfate paper industry. During this time methods have been developed
to identify and monitor these effluents.
The most prevalent complaint concerning the sulfate pulp and
paper industry is the emission of odorous gases into the atmosphere.
Under stable meteorological conditions the typical "rotten cabbage"
Kraft mill odor can be detected tens of miles from a plant site. The
odorous gases commonly emitted from the Kraft process result mainly
from the recovery of black liquor. These gases contain sulfur in
the form of hydrogen sulfide, sulfur dioxide, alkyl mercaptans, and
their oxidation products.
Commercial devices are available for controlling gaseous
emissions. However, the emission of odor-producing materials,because
of the many points of release and the low concentrations involved ,is
not amenable to reduction by conventional control devices. The heart
of the problem may be found at the point where these materials are
initially formed rather than where they are emitted into the atmosphere.
It is often easier to improve a process by preventing the evolution of
an elusive material than to try to collect the material after it is
formed. This study concerns the fundamental chemistry of black liquor
oxidation -- a process which may be used to reduce the quantities of
odorous gases lost from the chemical recovery system of the sulfate
The Kraft pulping process was introduced in Germany in 1879 as
an improvement over the soda method. It was discovered that if wood
chips are cooked in a liquor containing a mixture of sodium hydroxide
and sodium sulfide instead of plain sodium hydroxide (caustic soda),
the resulting pulp could be made into a high strength paper sheet.
Hence, the German word Kraft, meaning strength, was the name given to
the new process. This process is interchangeably known as the sulfate
process because the chemical make-up is added to the process in the
form of sodium sulfate or salt cake.
In Kraft pulping the wood is barked, chipped, and placed into
large pressure digesters along with the solution of cooking chemicals.
The chips are cooked at about 115 pounds pressure(3440F) for one to
six hours. Following the digestion period, the material is dumped into
a blowpit or blowtank where the liquor containing the noncellulosic
portions of the wood is drained from the pulp. During digestion the
sulfide in the cooking liquor combines with organic and inorganic
materials to form noxious gases which are released into the air when the
digester is opened.
The pulp is washed with hot water to remove the remaining chemicals.
The drainage from the blowpit and pulp washers is known as black liquor
and is stored hot in insulated tanks. The pulp is screened, refined,
thickened, sometimes bleached, and then converted into paper.
In the chemical recovery cycle the black liquor is concentrated
in multiple-effect and direct-contact evaporators, then burned, and the
heat used to produce steam. During evaporation and handling volatile
sulfur compounds continue to be lost. Evaporator exhaust contains gases
similar to those released in the digester blow tank. Effluent gases
from the recovery furnace contain both sulfurous gases and particulate
The ash or smelt from the recovery boiler contains the inorganic
chemicals, but in a chemically reduced form, that were present in the
black liquor. This smelt is dissolved in water to produce a green liquor
containing sodium sulfide and sodium carbonate. During solution some
gases are produced and these are vented into the atmosphere. The green
liquor is causticized by the addition of slaked lime and clarified to
produce white liquor which is returned to the digesters to process more
wood. A flow diagram of the Kraft process is shown in Figure 1.
w < xI
0 \ u
cr C I
0 U X<
(i-r 0U CIDV-
-J C Q-' 0 a-
Q O F- w
k\0 VJ -
.0 -0 V
0 0 s d
II LiI 4-
OD F-.) :01
0 jv 4 1
5Z I- 4
0r < 0r 0
LU~io 0~0 IL(
I z c
LU Z I 0
w II u, 0 u 9
CL V) a
Co u 0 44L
v, I W-
O- Z- :J
Z 0 < 0 w
L-- z 73 uo
o L4: C v
ui 0 wvl
ui 0 c L J
OL OL V )
0 w 0 D
m 1 ww ; c
Li wl :) 0 -
,on w n D _j
< CL L)Ld F- U < LL
-1 < IJ LL
V)- - CD (D n -
Kraft paper making has expanded greatly in the past 20 years.
This growth has been prompted by the economy of the process and the ever
increasing uses for its high-strength pulp. Presently more than 75 per
cent of the paper-grade pulp produced in the United States is made by
the Kraft process.
Pollution Control and Black Liquor Oxidation
Building and operating a completely odorless Kraft mill is
possible but expensive because a small amount of material must be
eliminated from a large amount of hot gas. For a Kraft mill to be
completely odorless all sources of emission must be controlled, mainte-
nance and operation must be excellent, and equipment must not be
overloaded. However, when sulfate mills are not located in densely
populated areas, 100 per cent control is neither necessary nor economical.
The sources of air-borne gaseous and particulate emissions from
a Kraft mill are tabulated in Table 1.
Chemical losses from the recovery boiler can be reduced by
installation of cyclones and electrostatic precipitators or wet-type
scrubbers. In some instances all three devices are used. Evaporator
chemical losses can be reduced by black liquor oxidation or gas scrubbers,
while digester losses may be reduced by by-product recovery, solution of
gases in liquor in a black liquor oxidation tower, or oxidation. Lime
kiln control devices include wet scrubbers and bag filters while on
power boilers electrostatic precipitators, cyclones, and wet scrubbers
may be used. Procedures intended to reduce chemical losses also reduce
SOURCES OF AIR-BORNE GASEOUS AND PARTICULATE
EMISSIONS IN THE KRAFT PROCESS6
Importance Source and Compounds Importance Source and Compounds
Recovery furnace -
H2S, SO2, and some
CH3SH and (CH3)2S2.
and some CH3SH and
Digesters H2S, CH3SH,
(CH3)2S, and (CH3)2S2.
Lime kiln some H2S.
Recovery furnace -
Na2SO4 and Na2CO3
Lime kiln CaO dust.
Minor Smelt dissolving tank. Minor Salt cake and lime
Lime mud filter.
Odor reduction may be achieved through black liquor oxidation,
absorption of gases in water or caustic spray, chlorination, or burning.
Black liquor oxidation and oxidation with chlorine are, to date, the
most practical systems for odor reduction. One progressive west coast
Kraft mill stores its dilute digester gases in a vaporsphere until a
sufficient amount is collected to be passed through a chlorination
system. Minor losses from dissolving tank, lime slaker, lime filter,
or pulp washers are seldom treated. The ideal piece of control equipment
is one which will recoup initial investment quickly by chemical savings
or by-product recovery.
Black liquor oxidation prior to concentration is effective in
reducing volatile sulfur loss in multiple-effect and direct-contact
evaporation. The idea was first patented by the Swedish engineers
Bergstroem and Trobeck. It is believed that black liquor oxidation
transforms the inherent malodors into products which are less volatile
and do not give rise to foul-smelling decomposition products. Because
of the conservation of sulfur in the recovery system, the amount of
sodium sulfate or sulfur make-up is decreased. This in turn decreases
the percentage of sodium carbonate present in the green liquor which,
therefore, reduces the quantity of lime needed for causticizing. By
conserving sulfur in the system, black liquor oxidation produces a
white liquor with high sulfidity (i.e., percentage of sodium sulfide in
the active chemicals) which means a better, more buffered cooking cycle.
Black liquor oxidation also improves the suitability of precipitated
lignin for by-product use and enables the evaporators to be operated
more uniformly and efficiently.
To date, the main drawbacks of black liquor oxidation have been
the expensive equipment cost and excessive foaming attributed to the
soap-like materials present in the liquor. Because of these drawbacks
at least part of the cost of such a system is frequently justified on
the basis of reduced odor and corrosion. At the present time some
full-scale black liquor oxidation units are operating in the north and
northwest United States, Canada, Mexico, and Sweden, although the
mechanisms of the process are not thoroughly understood.
Oxidation using pure oxygen rather than air has been suggested
by previous researchers.7' 8, 9 A closed cycle is necessary economically
to ensure that all of the oxygen introduced eventually is assimilated by
the black liquor. In such a system the foaming problem is minimized if
contact time is sufficient to enable any entrained bubbles to be
completely absorbed. Defoaming agents become unnecessary and equipment
size can be minimized since high oxygen tensions will shorten contact
time. Attempts have been made to add oxygen directly to a digester
towards the end of a cook.9 This study was abandoned at the bench scale
primarily because the oxygen costs exceeded the salt cake and lime savings.
Under present economic conditions black liquor oxidation in
general is a marginal operation. Future adjustments in chemical prices,
increased expansion, and social pressures caused by environmental
pollution could change this situation radically. The principal advantage
of the oxygen system over the air system is that the gas can be
introduced directly into existing equipment, such as pipe lines or
storage tanks, and achieve a degree of oxidation without expensive
At present black liquor oxidation has not met with success
in the southeastern United States partly because of the pronounced
foaming tendencies of southern Kraft black liquor. Although many
equipment patents have been registered, voids still exist in the basic
understanding of the mechanisms of the process. The technology has
advanced to a point where further refinements are difficult without a
better understanding of the fundamental aspects of the problem.
Although it generally is accepted that black liquor oxidation reduces
volatile sulfur losses and, hence, atmospheric pollution, the true
reasons underlying this stabilization are unknown.
Technical information applicable to the problem can be divided
into two general categories: the application and technology of black
liquor oxidation, and the chemistry of pure sulfur compounds. Neither
the chemical engineering nor the pure chemistry approach adequately
answers certain basic questions concerning black liquor oxidation.
These questions are: what changes occur during oxidation; in what
quantities do these changes occur; and why do the changes occur?
The previous research reported in the literature has been
conducted for very specific process applications, hence much of this work
cannot be applied comprehensively to the general case of black liquor
oxidation without verification and modification.
Dealing with black liquor presents some unique technical problems.
The black liquor contains approximately 5 per cent inorganic
and 10 per cent organic matter, and the interaction between the mineral
and the complex organic constituents is largely unknown. There are
some controversial factors and more knowledge about them is necessary
before the mechanisms of black liquor oxidation can be truly understood.
Tle main purpose of black liquor oxidation is to stabilize the
volatile sulfur compounds in the liquor. The effectiveness of this
oxidation in reducing gaseous losses depends on the method and degree
of oxidation. Oxidation stabilizes the liquor for evaporation but some
gases may be lost during oxidation.
Advantages and Disadvantages of Black Liquor Oxidation
Summarizing the advantages and disadvantages of black liquor
oxidation, the advantages include:
1. Reduction of salt cake make-up and lime required for
2. Reduction of corrosion in multiple-effect and direct-contact
evaporators and scrubbers as well as at the plant site.
3. Better and more buffered cooking liquor.
4. Improved by-product recovery.
5. Reduction of odors released from evaporators.
The disadvantages are:
1. Capital investment in equipment which will not directly
2. Additional operational costs to maintain oxidation equipment.
3. Increased chemical costs if pure oxygen is used.
PURPOSE AND SCOPE
This work was conducted as a fundamental research project in
atmospheric pollution control at the Air Pollution Research Laboratory,
University of Florida. The purpose of the study was to investigate
completely the oxidation of Kraft black liquor using pure oxygen and
a mixture of oxygen plus ozone. The fate of the principal sulfur
compounds during oxidation was determined and their technological
significance evaluated. The technical and economic feasibility of
mill scale oxidation using pure oxygen was discussed. The possibilities
of eliminating pulp mill odors with ozone was also investigated.
Through this paper the relative importance of the parameters
-controlling black liquor oxidation has been established. Where
conflicting reports existed in the literature, experimentation was
conducted to define the process accurately. It is hoped that in
putting black liquor oxidation on a more quantitative basis, this study
aids the development of a commercial system applicable to Kraft mills
in the southeastern United States.
Black liquor oxidation was investigated under a variety of
conditions to determine the effect of different parameters on the process.
The principal physical and chemical parameters examined were:
2. Oxygen partial pressure.
3. Method of contact.
4. Oxidizing agent.
The practical temperature range considered was from 60 C to
90 C. Reaction pressures were not raised above one atmosphere since
previous studies9' 10 indicated that elevated pressures were conducive
to excessive oxygen consumption. Several different methods of contact
were used and the oxidation efficiency of oxygen, and oxygen enriched
with ozone were contrasted.
By varying the pertinent parameters, the following areas were
1. The nature of the predominant inorganic reactions that occur
during oxidation and the effect of temperature, oxygen pressure, and
method of contact on the products of reaction.
2. The precipitation of elemental sulfur from 1e black liquor.
3. The completeness of these predominant reactions and a sulfur
balance for the entire system.
4. The degree of reversibility of these reactions.
5. The kinetics of oxidation and the effect of temperature,
oxygen pressure, method of contact, and oxidizing agent on the rate of
6. The oxygen requirements of the black liquor and the
7. The catalysis of sodium sulfide in black liquor.
8. The emission of oderiferous gases during oxidation.
9. The elimination of these gases by oxidation with ozone and
the application and approximate cost of similar oxidation to other
10. The application and the approximate costs of a mill scale,
pure oxygen oxidation system.
SURVEY OF PREVIOUS WORK
The idea of stabilization of some of the sulfur compounds in
Kraft black liquor was first suggested by M. G. Schmitt in 1938.
Schmitt noted that in the recovery operation sulfur losses exceeded
sodium losses by a ratio of four to one. The high sulfur loss was
thought to be due to thermal instability of residual sodium sulfide
in the black liquor. To remedy this situation Schmitt proposed
aeration of the black liquor as well as other methods of treatment.
Since its inception, most of the work on black liquor oxidation
has been conducted by a few groups of investigators in Sweden, Canada,
and the United States. To date, studies of black liquor oxidation have
been aimed primarily at reducing chemical make-up rather than odor or
atmospheric pollution control. From this aspect the economics of the
system have been marginal.
In the period between 1939 and 1942, Bergstroem and Trobeck11
studied the changes that occur in black liquor after storage. They
found that at elevated temperatures in the presence of air the sulfide
content decreased. Based on this work, Bergstroem pioneered the first
installation of a black liquor oxidation system at the Norrsundet,
Sweden, mill in 1952. Bergstroem and Trobeck11 reported reduction in
sulfur losses from 26 to 10 per cent at this installation. They also
noted decreased evaporator corrosion, increased white liquor sulfidity,
increased viscosity, and no decrease in heat value of black liquor.
They found, however, that oxidation efficiency dropped off when the
solids content went above 30 per cent.
The Bergstroem-Trobeck system was developed in conjunction with
high-vacuum evaporation of black liquor to dryness (i.e., 10 per cent
moisture). The weak black liquor was run into a tower containing
perforated shelves. Air blown into the bottom of the tower converted
much of the liquor into foam which was carried up through the unit and
out into a cyclone to separate foam and droplets. The liquid from the
cyclone was evaporated to dryness and burned. A number of Swedish,
Canadian, and American patents have been obtained covering the Bergstroem-
A Bergstroem-Trobeck oxidation system has been installed in the
Kraft mill of Loreto and Pena Pobre near Mexico City and its operation
investigated by Trobeck, Lenz, and Tirado2 in 1959. In 1961, Tirado
et al.,10 conducted bench scale oxidation studies using air under
pressure. The reactor consisted of a vertical pipe with an air diffuser
at the bottom. Liquor and air in variable quantities were fed through
the bottom and the products of reaction escaped through a pressure relief
valve at the top. At 50 pounds per square inch and 1490F the liquor was
well oxidized if air was supplied at a rate of 2.3 times the theoretical
requirement. The oxidation of black liquor by air at 25 to 75 pounds per
square inch showed promise, especially regarding the stabilization of
mercaptans which are difficult to oxidize in conventional processes.
Trobeck, 13 stated that complete oxidation would eliminate
sulfur lost as hydrogen sulfide and mercaptans in the evaporators. He
suggested that the sulfide present as sodium sulfide may be oxidized
according to the following formula:
(1) Na2S + 202 + H20 --- Na2S203 + 2NaOH.
The author mentioned, however, that it is likely that some of the
hydrogen sulfide was tied up in the liquor in the form of organic
compounds. Trobeck said that corrosion was caused by gaseous hydrogen
sulfide which oxidized to form droplets of sulfuric acid in the evaporator
(2) H2S + 202 -- H2S4.
Reducing the amount of hydrogen sulfide in the evaporators increased the
tube life almost ten-fold.
Trobeck also presented data on the oxygen absorbed during
oxidation of black liquor (see Figure 2).
The sharp increase in oxygen uptake after 30 minutes was
accompanied by a decrease in liquor heating value which indicated
oxidation of organic matter. Since this loss in heat value was not
found in less than 30 minutes of oxidation, it was concluded that oxida-
tion of sulfide is possible without wasteful destruction of organic
Tomlinson4 and co-workers in Canada developed a nonfoaming
oxidation tower as part of their recovery process. Tomlinson placed
great emphasis on the fact that hydrogen sulfide was lost from the
concentrated black liquor in direct contact (disc, cyclone, or cascade)
evaporators due to the reaction of sodium sulfide with acidic gases
ZI O1 8 9 V7 z 0
aonbl-[ pwoq 5o im 3ad Im 'uojdumnsuoo u9sXxo
such as sulfur dioxide, sulfur trioxide, and carbon dioxide. They
found that when a completely oxidized black liquor was evaporated by
hot flue gases in a venturi scrubber, the effluent gas contained only
180 parts per million of hydrogen sulfide.
Tomlinson et al., suggested that stabilization was achieved
because dissolved hydrogen sulfide was oxidized to sulfur, sulfites, or
sulfates which are no longer subject to destructive hydrolysis.
Laboratory pilot plant work employed a counter-current tower packed with
coke or Raschig rings and the optimum reaction temperature was found to
be about 70 C. Designs based on this study supplied five times the
theoretical quantity of air necessary to react with the sodium sulfide.
Full-scale units have been erected in Windsor and Marathon, Canada. At
Marathon, the gaseous effluent from the counter-current oxidation tower
is fed into the recovery furnace and the residual odorous material
burned. Several United States and Canadian patents have been issued
covering these systems.
A chemical recovery system of the Tomlinson type, including
black liquor oxidation, was installed at the South African Pulp and
Paper, Ltd., mill in Springs, S. A.15 Eucalyptus is one of the species
being pulped and it has been noted that isopropyl mercaptan and
isopropyl disulfide are detectable in digester relief gases. Hisey
reported that oxidation tests at Springs showed 85 per cent sulfide
reduction and that 48 hours additional contact did not completely fix
the remaining sulfur.
A combination oxidation and absorption tower used to strip
sulfurous gases from the digester exhaust has been investigated.
Collins passed noncondensible evaporator and digester gases along with
fresh air into an experimental Tomlinson oxidation tower. He noted that
the odor of both the gas and the black liquor was reduced while the
biochemical oxygen demand of the liquor was not markedly decreased.
In addition to the Bergstroem-Trobeck and Tomlinson systems
for oxidizing black liquor, it has been recognized that oxidation
occurred in the Industrikemiska Aktiebolaget 1(INKA) system used
for evaporating sulfate liquor. This system was independently patented
by Oman and Goth, and Naucler and Ledin in 1927 and 1928, respectively.
In the INKA evaporator the black liquor was heated indirectly by the
recovery furnace stack gases; the hot liquor then flowed over a
perforated plate through which dry air was blown. The dry air evaporated
the black liquor and oxidized it to a certain extent. Actually, many of
the INKA patents concerned the design of an aerating and evaporating
device which brought the liquor in contact with air without causing
excessive foam. The sodium sulfide in the liquor was reduced
approximately 80 per cent in its 20 passes through the INKA evaporator.
A detailed study of black liquor oxidation was carried out by
Bialkowsky and DeHaas7 18 of the Weyerhauser Timber Company. These
workers noted a very definite decrease in sulfide content when black
liquor was passed through the pulp washers. From their observation that
additions of sodium sulfide to oxidized black liquor proceeded to oxidize
rapidly and from unpublished polarographic data, the authors concluded
that organic compounds in the black liquor catalyzed the oxidation
reaction. They also found that more oxygen reacted with the black
liquor than was necessary to oxidize the sulfide to thiosulfate according
to reaction 1. No increase in sodium hydroxide was detected,
however, and it was thought that this increase may have been masked
by the buffering action of the organic compounds. Bialkowsky and
DeHaas also investigated the addition of air to a pilot plant digester.
They found that the sulfide concentration of the black liquor was
lowered and the bleachability of the pulp was increased.
The basic oxidation studies of Bialkowsky and DeHaas were
carried out in a concurrent-flow glass tower. Using "wetted wall"
surfaces with careful control of the temperature, humidity, air, and
liquor flow rates, they determined the optimum transfer coefficients for
oxidizing the sulfide. They found that foaming decreased and oxygen
consumption increased as the temperature rose. Studies in a cascade-
type evaporator showed that unoxidized liquor lost all of its hydrogen
sulfide while oxidized liquor lost little hydrogen sulfide when
evaporated to dryness. Next, the possibility for absorbing and oxidizing
noncondensable gases from digester relief and blow gases with black
liquor was studied in a laboratory oxidation tower. Hydrogen sulfide,
methyl mercaptan, and dimethyl sulfide were mixed with air and passed
through a packed tower. Fresh black liquor was passed concurrently
through the packed column. Hydrogen sulfide was completely absorbed
and oxidized; 5 per cent of the methyl mercaptan passed unchanged,
40 per cent left the tower as methyl di-sulfide,and the rest was absorbed
in the liquor. Over 90 per cent of the methyl di-sulfide passed through
the tower unchanged.
Other tests showed that when temperature was increased to around
250 C,some hydrogen sulfide was evolved from the black liquor in spite
of the fact it did not contain any residual sodium sulfide. These
results indicated that a reduction of hydrogen sulfide losses could be
expected as far as the direct contact evaporators were concerned, but
any localized overheating in the evaporators and high temperature
combustion in the furnace would release hydrogen sulfide gas.
Wright and associates8 at the British Columbia Research Council
began work on odor control in 1948. Methods of analysis for hydrogen
sulfide and methyl mercaptan were developed and the kinetics of the
oxidation process were studied using a respirometer. The effect of
interfacial mass transfer was minimized by making the gas-liquid
interface very large. This was achieved by absorbing black liquor on
filter paper, thus presenting maximum surface area. They found that
sodium sulfide and mercaptans in black liquor were oxidized readily and
that the oxygen absorption continued indefinitely thereafter but at a
steadily diminishing rate. Tests also showed that oxygen consumption
increased with temperature in the range of 400C to 900C. Wright also
indicated that 1 to 3 volumes of oxygen were needed to oxidize completely
1 volume of black liquor. Since the black liquor contained approximately
1.7 grams per liter of sodium sulfide, this meant each mole of this
sulfide consumed 2 to 5 moles of oxygen during oxidation.
It was found that with infinite area the oxidation of sulfur
compounds still required from 2 to 3 minutes and hence nothing was
gained by increasing the superficial area beyond a certain point.
To test laboratory data a pilot plant oxidation unit was built
at the Port Alberni,1 20 British Columbia, mill of Bloedel, Stewart,
and Welch. Concurrent air and liquor flow was used in the tower and
noncondensable digester gases were added to the air supply. These gases
were practically completely absorbed and the exit air carried little, if
any, Kraft odor. This statement contradicted the findings of Bialkowsky
In 1959, Murray at the British Columbia Research Council
published a comprehensive study of the kinetics of oxidation of Kraft
black liquor using both air and pure oxygen, Murray reported that the
maximum rate of oxidation occurred at 72 C using pure oxygen and that
sulfur formation was favored by temperatures below 80 C, low sulfide
concentration, and high oxygen pressures. Murray found that oxidation
rates were higher when pure oxygen rather than air was used, but a
five-fold increase in oxygen pressure did not bring about a proportional
increase in oxidation rate. It was noted that storing oxidized black
liquor at elevated temperatures in the absence of air caused some of
the sulfide to be regenerated. This was especially true when sulfur was
an oxidation product. It was also reported that oxidation rates were
not greatly affected by changes in solids content. Experiments showed
that an increase in thiosulfate concentration reduced the black liquor
oxidation rate while an increase in sodium hydroxide increased the
oxidation rate. The addition of elemental sulfur to a digester produced
sulfide according to equation 3; the over-all effect was to lower the
sodium hydroxide concentration and increase the thiosulfate and sulfide
(3) 4S + 6NaOH----- Na2S + Na2S203 + 3H20.
On the basis of the kinetic data Murray concluded that the oxidation
takes place in two definite steps: the first a rapid oxidation and then
a slower reaction. Therefore, he suggested that a two-stage oxidation
system should be more efficient than a one-stage system.
Collins 11, 16, 22 published a series of papers on black liquor
oxidation. On the basis of preliminary studies he presented data on
the effect of oxidation on the sulfur compounds in black liquor (see
Table 2). Collins found that the sulfide, thiosulfate, and sulfite
oxidized to thiosulfate and sulfate. Collins also showed that chemical
oxidation of black liquor with sodium peroxide was impractical. Based
on the results of pilot plant studies at Thilmany, Wisconsin, Collins
developed a commercial system for black liquor oxidation. In this
system air was blown into the black liquor to form a stable foam. The
oxidized foam was then collected in a cyclone and converted back into
liquor. A material balance on the Thilmany evaporators indicated that
the use of oxidized black liquor greatly reduced sulfur losses.
A comprehensive literature review published by Collins1 in 1953,
included most of the significant basic and related papers on sulfate
black liquor oxidation. He mentioned an early patent in which alkaline
waste liquors were purified by concentrating and heating with hot air
in the presence of caustic until the organic substances carbonized.
Addition of an oxidizing agent such as saltpeter was also suggested.
Collins noted another source which pointed out that sodium sulfide in
weak wash water discharged to the sewers was partially oxidized by the
dissolved oxygen in the water. He also mentioned Basberg who discovered
that at room temperature in the absence of air, black liquor sulfide
content did not change after 5 weeks of storage.
EFFECT OF OXIDATION ON THE SULFUR COMPOUNDS
IN A NORTHERN BLACK LIQUOR22
in Fresh Liquor in Oxidized Liquor
Grams per Liter Grams per Liter
Na2S 9.01 1.02
Na2S203 4.07 11.75
Na2SO3 0.18 0.00
Na2SO4 3.97 5.48
Total Sulfur (by analysis) 6.56 6.45
Total Sulfur (by addition) 6.29 6.42
According to Collins, Venemark stated that some of the sulfur
compounds in black liquor were subject to air oxidation at room temper-
ature. Venemark also reported that when evaporation was carried out in
direct contact with flue gases, sulfur losses occurred due to the action
of acidic gases on the alkaline black liquor.
In his literature survey Collins11 discussed the laboratory
findings of Bozza and Colombo. Their kinetic studies showed that the
oxidation of sulfide was a first order reaction, but the oxidation of
mercaptan was intermediate between a first and a second order reaction.
To determine the retention of sulfur in black liquor after oxidation
of the liquor, they distilled various samples in a current of nitrogen
and analyzed the gases evolved. Oxidation resulted in the reduction of
the amount of hydrogen sulfide lost during distillation but mercaptans
were not affected.
Heath, Bray, and Curran22 referred to the early work of Klason
and Segerfelt in which the oxidation of black liquor when stored was
first noted. This observation was never followed up and no attempt was
made to apply black liquor oxidation.
Fones and Sapp oxidized Kraft black liquor at digester temper-
atures using pure oxygen. Their experiment was divided into three phases:
the oxidation of pulp alone; the oxidation of pulp and black liquor at
the end of a normal Kraft cook; and the oxidation of black liquor alone.
The first two phases produced undesirable reduction in pulp quality.
On oxidizing the black liquor alone at 100 to 120 pounds per square inch,
the sodium sulfide was oxidized to sodium thiosulfate, which in turn was
oxidized to sodium sulfate (see Figure 3). Unfortunately, a sulfur
dy .y / o o
balance on the data in Figure 3 does not account for all the sulfur
compounds (see Table 3). After 150 minutes only 76 per cent of the
sulfur initially present as sulfide was accounted for by the thiosulfate
and sulfate formed. After 217 minutes, however, all of the sulfur was
accounted for. It would appear that some other compound or compounds,
either organic or inorganic, were formed during oxidation.
Conclusions Based on Previous Literature
On the basis of the literature survey of the information
available on black liquor oxidation, certain conclusions can be drawn:
1. Although the possibility of oxidizing black liquor has been
known for at least 20 years, little theoretical information is available
on the basic mechanisms of the process. The majority of the literature
covered the application of a few basic ideas to specific mill problems.
2. Full scale units have been successfully installed and
operated in areas other than the southeastern United States.
3. The principal reason for installation of a black liquor
oxidation system is to reduce chemical losses and corrosion rather
than to improve cooking liquor or reduce atmospheric pollution.
4. Knowledge of the chemistry of the stabilization of the
sulfur compounds in black liquor is indefinite. It is suspected that
the dissolved organic material catalyzes the oxidation of the sulfide
but little has been reported about this mechanism.
5. The relative importance of the physical parameters of the
system, such as temperature, contact method, and choice and quantity of
oxidizing agent have not been reported.
MOLAR SULFUR BALANCE OF DATA IN FIGURE 3
Time Sulfide Thiosulfate Sulfate Sum of the
Expressed as Expressed as Expressed as Sulfur
Moles per Moles per Moles per Accounted For
Liter Sulfur Liter Sulfur Liter Sulfur
0 0.155 0.077 0.058 0.290 100.0
27 0.146 0.080 0.063 0.289 99.7
38 0.098 0.076 0.068 0.242 83.4
45 0.107 0.095 0.070 0.272 93.8
55 0.069 0.128 0.077 0.274 94.5
62 0.055 0.134 0.070 0.259 89.3
78 0.026 0.150 0.074 0.250 86.2
95 0.023 0.136 0.084 0.243 83.8
114 0.017 0.099 0.118 0.234 80.7
150 0.014 0.039 0.168 0.221 76.2
191 0.010 0.019 0.250 0.279 96.2
217 0.006 0.006 0.278 0.290 100.0
6. There is some doubt as to whether or not black liquor can be
oxidized without the emission of odors.
7. There is also some doubt as to whether or not black liquor
oxidation using pure oxygen is technologically or economically feasible.
The work on black liquor oxidation over the past 20 to 30 years
has added knowledge as well as conflicting information to the technology
of the pulp and paper industry. Black liquor oxidation as a means of
reducing chemical losses and atmospheric pollution from the recovery
cycle of the Kraft process has definite potential. In some instances
oxidation has been made to work, while in others, technical and economic
limitations have hindered its application.
Oxygen, if properly utilized, may prove feasible on a mill scale
although it is a more expensive oxidizing agent than air. The potential
advantages of an oxygen system over an air system are:
1. The elimination of the foam problem.
2. Higher oxygen pressures over the liquor, which will decrease
the contact time and hence the size of the equipment.
3. Improved by-product recovery from black liquor.
Oxidation States of Sulfur
Sulfur in various degrees of oxidation is found in black liquor.
The more common oxidation states with their corresponding oxidation
potentials24 in basic solution are shown below:
--0.61 --i 0.76 1
(4) S= 0.48 S 0.74 S203 0.58 S03= 0.93 S04
Polysulfides are also stable in alkaline solution but have not been
included in the above diagram because their oxidation potentials are
indefinite. Polysulfides are intermediates between sulfide and elemental
sulfur and include the disulfide, S2", and the tri-, S3=, tetra-, S4",
and penta-, S5", sulfides. The couples shown in equation 4 are not
subject to rapid reversibility and hence they cannot be used to calculate
Other oxidation states of sulfur include hyposulfite, S204 ,
dithionate, S206", and the persulfates or thionates such as S306",
S406", and S506=. The problem of assigning polar oxidation states
to these compounds is similar to that encountered with the polysulfides,
since polar numbers are rather ambiguous because the sulfur atoms
are linked by covalent bonds.
Evaporation of Sulfate Black Liquor
The active chemicals in Kraft white liquor are sodium sulfide
and sodium hydroxide in concentrations of 15-30 and 40-65 grams per
liter, respectively. During cooking the lignin and other noncellulosic
portions of the wood are hydrolyzed into soluble compounds, freeing the
fiber bundles. The products of hydrolysis include carbohydrates,
alcohols, acids, thiolignin, mercaptides, and other soluble organic;
the black liquor contains these constituents as well as unutilized
sodium sulfide, sodium hydroxide, and other soluble mineral matter.
The sulfide in the liquor serves a dual purpose5 by buffering
the system during cooking and by supplying sodium hydrosulfide (see re-
action 5) which combines with lignin to form alkaline soluble thiolignin.
(5) Na2S + H20 --- NaHS + NaOH.
At the end of the cooking period the sodium sulfide is at
least partially hydrolyzed and the concentrations of active chemicals
(i.e., Na2S and NaOH) are substantially reduced.
When the black liquor is concentrated in vacuum multiple-effect
evaporators, some sodium sulfide is lost as hydrogen sulfide gas.1
(6) Na2S + 2H20 --- 2NaOH + H2S.
Another odorous evaporator loss, but a relative minor one in
terms of volume, is methyl mercaptan. Methyl mercaptan and its sodium
aalt, sodium mercaptide, are formed by the reaction of hydrosulfide
on free methyl groups during the cooking process. During evaporation
some of the sodium mercaptide hydrolyzes and liberates methyl mercaptan.
(7) CH3SNa + H20---- NaOH + CH3SH.
Methyl sulfide is another compound which is formed in the digesters
possibly by the liberation of hydrogen sulfide from two molecules of
(8) 2CH3SH--- CH3SCH3 + H2S.
The methyl sulfide and any other low boiling gases such as
hydrogen sulfide or methyl mercaptan are stripped readily from the
black liquor during vacuum evaporation.
When unoxidized black liquor is passed through direct-contact
evaporators and heated with flue gases containing carbon dioxide,
hydrogen sulfide is lost.
(9) Na2S + CO2 + H20 --- Na2CO3 + H2S.
Thus hydrolysis, desorption, and reaction with acidic gases as
shown in reactions 6 through 9 are the generally accepted causes of
volatile sulfur losses during evaporation.
Black Liquor Oxidation
Black liquor oxidation reduces sulfur losses since the products
formed are apparently nonvolatile and stable at evaporator conditions.
Reactions i and 109, II, 13, 22
Reactions 1 and 10 11, 13, 22 have been suggested to explain the
formation of stable inorganic sulfur compounds.
(1) 2Na2S + 202 + H20-----Na2S203 + 2NaOH.
(10) Na2S203 + 2NaOH + 202 2Na2SO4 + H20.
Neither sodium thiosulfate nor sodium sulfate is subject to destructive
Prior to the experimental work, the following hypotheses were
drawn from studies of the previous literature.
It seems likely that when black liquor is oxidized, several
different oxidation states of sulfur will be formed. Certain compounds
are more likely than others to be present in the strongly alkaline
black liquor. Elemental sulfur, for example, is unstable in alkaline
solution and slowly decomposes into sulfide and thiosulfate.24' 25
(11) 4S + 6NaOH---- Na2S203 + 2Na2S + 3H20.
It is unlikely, therefore, that elemental sulfur is a stable oxidation
product of sodium sulfide. Further circumstantial evidence that elemental
sulfur is not to be expected in black liquor is the fact that in an
alkaline solution containing sodium sulfide and sulfur, the two compounds
combine to form polysulfides according to reactions 12 through 15.25
(12) Na2S + S -- Na2S2'
(13) Na2S + 2S --- Na2S3.
(14) Na2S + 3S --- Na2S4.
(15) Na2S + 4S --- Na2S5.
Sulfur also reacts with sodium sulfite to form sodium thiosulfate.25
(16) S + Na2SO3 -Na2S203.
Under certain conditions, however, sulfur is formed and is
apparently stable in the black liquor. The reasons for this sulfur
formation will be investigated and discussed in this paper.
Yeoman found that pure soluble metal sulfides in dilute
solution slowly hydrolyze and in the presence of air produce both
polysulfides and thiosulfates.
(17) Na2S + H20 NaHS + HaOH.
(18) 4NaHS + 02 -- 2Na2S2 + 2H20.
(19) 2NaHS + 202 ---- Na2S203 + H20.
It may be possible that the sulfide in the black liquor reacts similarly
when subjected to oxidation.
Sodium sulfite with an oxidation number of +4 is another compound
which may be formed when black liquor is oxidized. Near the boiling
point, however, sodium sulfite and disodium disulfide react quickly to
form sodium sulfide and sodium thiosulfate.2
(20) Na2SO3 + Na2S2 -- Na2S + Na2S203.
Hence, it is unlikely that both disodium disulfide and sodium sulfite
exist simultaneously in the hot black liquor.
It is possible that only thiosulfate and sulfate, as suggested
in reactions 1 and 10, are formed during oxidation. It is more
likely, however, that a mixture of the many oxides of sulfur are
formed during black liquor oxidation.
The free alkyl mercaptans, their derivatives,and salts are all
subject to partial or complete oxidation. Under mild oxidation mercaptans
form a stable disulfide.2
(21) 2RSH + 0----- RSSH + H20.
A more vigorous oxidation produces the corresponding sulfonic acid.2
(22) RSH + 30---- RS3H.
Pure oxygen normally oxidizes methyl mercaptan to methyl disulfide while
a peroxide oxidizes methyl mercaptan completely to methyl sulfonic acid.
In a similar manner aliphatic sulfides are oxidized to sulfoxides
and sulfones by mild and strong oxidation respectively.26
(23) RSR + 0 ----RSOR.
(24) RSR + 20 ---- RSO2R.
There is some doubt as to just how efficient black liquor
oxidation is in attacking the mercaptan fraction. Tirado10 et al.,
reported that oxidation using air at a pressure of 70 pounds per square
inch increased the oxidation of mercaptans from 45 to 85 per cent.
There is some question as to whether or not some of the oxidation
products of methyl mercaptan are stripped from the black liquor during
oxidation. It appears that methyl disulfide, CH3SSCH3, and methyl
sulfonic acid, CH3SO3H, with boiling points of 117C and 1600C will
remain condensed. Methyl sulfide, CH3SCH3, methyl sulfoxide, CH3SOCH3,
and residual methyl mercaptan, CH3SH, with boiling points of 380C, 1000C,
and 7.6C, respectively, possibly will be volatilized from the liquor.
during oxidation. This indicates that black liquor oxidation may emit
some odoriferous sulfur gases.
Black Liquor Oxidation with Oxygen Plus Ozone
Stabilization of the soluble sulfur compounds may be accelerated
by use of a stronger oxidizing agent than pure oxygen. A mixture of
ozone and oxygen provides additional oxidizing power. Ozone is a strong
electron acceptor and its oxidation potential is second only to fluorine.2
(25) 02 + 20H" 03 + H20 + 2e" EB = -1.24V.
Ozone has a very characteristic pungent odor and is more soluble
than oxygen in a pure alkaline solution. Upon absorption in basic
solution the ozone hydrolyzes and releases the powerful perhydroxyl ion.24
(26) 03 + 220H 2 + H20 AF = 5.1 Kcal.
Ozone is produced by a high voltage corona discharge according
to reaction 27.2
(27) 302 -~ 203 AF -69 Kcal.
It is thermally unstable and cannot exist at temperatures above
2000C. High concentrations decompose violently into oxygen with the
liberation of heat.
Ladenberg27 produced 86 per cent ozone in liquid oxygen, but
Saunders and Silverman2 found the equilibrium concentration of ozone
at room temperature to be 16.5 per cent. The latter produced this
concentration using a silent electrical discharge of 5.5 Kv, across a
2 mm gap, in an atmosphere of pure oxygen. Equilibrium was attained after
a corona discharge of 70 to 90 minutes. Saunders and Silverman found
that, using a 2 mm gap, 1.8 to 2.0 mg of ozone could be produced per
coulomb of electrical energy--a value which was relatively independent
of the potential used.
Ozone may also be useful in oxidizing odoriferous gases from a
pulp mill. Ozone will oxidize methyl mercaptan directly to methyl
sulfonic acid.25, 26
(28) CH3SH + 03 CH3SO3H.
Ozone will probably deodorize by oxidation other similar sulfurous
The oxidation of weak black liquor cannot be described by any
simple chemical or absorption rate theory. The complexities of the
oxidation occur because the temperature region of practical interest
is a transition region in which the products of reaction, sulfide
concentration, and oxygen partial pressure act together to influence
the oxidation rate.
Using the semi-empirical equation 29, the effects of chemical
reaction and oxygen absorption can be separated.8
(29) dC K (C + bP).
where: C = concentration of Na2S in grams per liter-at any time t, in
K = velocity constant for oxidation reaction.
P partial pressure of oxygen in atmospheres.
b constant representing the rate of oxygen absorption.
By varying experimental conditions the effect of temperature,
oxygen pressure, method of contact,and oxidizing agent can be determined.
If the gas-liquid interface offers no resistance to oxygen absorption,
constant b will be zero and the oxidation is a first-order reaction.
Similarly, if oxidation is independent of sulfide concentration,the
reaction is of zero order, depending solely on the factor K (bP). These
are the two limiting cases of equation 29 and in reality the actual
situation probably lies somewhere between these two extremes.
Equipment and Procedures
Experimentation on black liquor oxidation requires a ready
supply of liquor, an oxidizing agent, and a method of bringing the two
reactants into intimate contact.
The black liquor used in this study was obtained from the Hudson
Pulp and Paper Company Kraft mill at Palatka, Florida. Black liquor was
drawn off from a point between the pulp washers and the multiple-effect
evaporators. It was collected in 5-gallon glass carboys which were
sealed and allowed to cool. The space above the black liquoY was
filled with inert nitrogen or helium and the liquor was displaced as
needed into a 4-liter flask. Analyses showed that liquor could be
stored in these carboys up to five weeks with no apparent change in the
initial sulfide concentration.
Compressed oxygen with a rated purity of 95 per cent was obtained
in 200 cubic foot cylinders from the Matheson Company. When an oxidizing
agent stronger than pure oxygen was required, a mixture of ozone and
oxygen was produced and supplied to the system.
The ozone was produced by passing a stream of pure dry oxygen
through a group of high voltage aluminum plates. The plates were
insulated from each other with mica and a high voltage transformer was
used to create a silent corona discharge between alternate plates.
A portion of the oxygen was ionized by the corona discharge and recombined
as ozone. The ozone concentration was controlled by varying the voltage
across the plates. The oxygen flow-rate was held constant at 450 cubic
centimeters per minute. Concentrations from zero to 1450 parts per
million of ozone could be produced in the generator shown in Figure 4.
The actual concentration of ozone being fed into the chamber was
determined by passing a known volume of the ozone-oxygen mixture through
an alkaline solution of potassium iodide. The quantity of ozone collected
in the alkaline scrubbing solution was analyzed photometrically (see
The oxidation chamber consisted of a 12-liter cylindrical stain-
less steel chamber (see Plate 1 and Figures 5 and 6). Eight bolts around
the perifery of the unit fasten the plexiglass lid to the chamber. A
gasket made from 1/16 inch Cordet sheet packing provided a gas-tight
seal between the plexiglass and the stainless steel. A fan which blows
the oxygen down onto the exposed liquor surface and a stirring disc
which agitates the liquor were both mounted on a stainless steel shaft.
This pulley-driven shaft was mounted in a gas-tight bearing in the center
of the plexiglass top. The shaft was turned at 98 revolutions per
minute by a 1/30 horsepower, 110 volt, AC motor. The oxidizing agent
was fed into the center of the chamber through an 11-millimeter glass
tube and circulated by the fan. The liquor was introduced into the
chamber through a funnel mounted on a length of 15-millimeter glass
tubing. A three-way valve mounted on an auxilliary piece of 11-millimeter
glass tubing served as a relief and exhaust valve.
Sca r w
,.cL o -0
o o .j
0c -4 r
U, ** C
= u M-
3 o 0
0 0 X
o. E ao
0 C I 0
(. C 0
U -, U
JJ (U o
0> o cr
: -4 -4 -4
w ui r ?/ \ i n cE;5
>1C II C
C) U 3 l .C 0 0
u z3 u
S4\ t3 (3 0
Ir V / 6
-4 1 I
Black liquor samples of 0.15 to 20.0 milliliters were withdrawn
from the chamber through a sampling port with a hypodermic syringe.
The sampling port consisted of a length of 15-millimeter glass tubing
capped with a serum-bottle stopper. Liquor samples taken at different
time intervals during oxidation were analyzed for sulfide, polysulfide,
thiosulfate, sulfite, and sulfate.
The oxidation chamber was kept at a constant temperature by
immersion in a thermostatically controlled water bath. A 2000-watt
immersion heater kept the 15 gallons of bath water at the desired
temperature, and the water was circulated by an electrically-driven
Three methods, one static and two dynamic, were used to supply
gas to the oxidizing chamber. In the static system the vessel was
purged with oxygen, liquor poured in, and the chamber sealed. Oxygen
was made available at a constant pressure of 50 millimeters of water
and, as the liquor utilized the oxygen above it, more gas entered the
chamber through a wet-test meter. The fluid in the meter previously
had been saturated with oxygen. With this static system the volume of
the consumed oxygen was determined. The static oxygen system is shown
diagramatically in Figure 5.
A dynamic or circulating system was also employed to oxidize
the black liquor. An oxygen flow of 450 cubic centimeters per minute
was used so that the gas above the liquor was changed approximately
every ten to fifteen minutes. The oxygen flow rate was kept low in
order to simulate static conditions as much as possible and not to
upset the thermal equilibrium in the chamber. An oxygen-ozone mixture
was also supplied using this dynamic or circulating system (see Figure 6).
When plain oxygen was utilized dynamically, the ozone generator was not
Detailed descriptions of the operating procedures and testing
sequences for the static and dynamic flow systems are given below.
Static Flow Procedures
1. Fix the oxidation chamber in place and bring the bath up to
the desired working temperature.
2. Close the chamber, flush thoroughly with oxygen, and test
3. Withdraw samples from the 4-liter black liquor flask and
analyze for sulfide, polysulfide, thiosulfate, sulfite, and sulfate
(see Appendix I).
4. Open the relief valve and quickly pour 4 liters of black
liquor at room temperature into the oxidation chamber.
5. Reseal the chamber, start the stirrer, open the gas valve,
and make oxygen available to the system at a constant head of 50
millimeters of water. The liquor is heated in the reaction chamber
and temperature equilibrium is achieved in less than 10 minutes.
6. Read the wet-test gas meter at 5-minute intervals to determine
the oxygen utilized by the black liquor. Withdraw liquor samples at 10-
to 30-minute intervals during oxidation and analyze for sulfide,
polysulfide, thiosulfate, sulfite, and sulfate.
Dynamic Flow Procedures
1. Fix the oxidation chamber in place and bring the bath up to
the desired working temperature.
2. Close the chamber, flush thoroughly with oxygen, and test
3. Open the relief valve and adjust the flow rate of oxygen or
ozone-oxygen mixture to 450 cubic centimeters per minute.
4. Divert the ozone-oxygen stream from the chamber after
equilibrium has been established in the ozone generator (after 2 to 5
minutes). Bubble the gas through two midget impingers containing alkaline
iodide scrubbing solution. Analyze the contents of the impingers (Appen-
dix I) and calculate the ozone concentration in the oxygen stream.
5. After sampling, re-direct the gas through the chamber and
vent the exhaust to the hood. (If ozone is not used omit steps 4 and 5.)
6. Connect "freeze-out" traps to exhaust line to condense and
concentrate the odorous sulfurous compounds in the off-gases. (If
exhaust gases are not monitored for odors omit step 6.)
7. Withdraw samples from the 4-liter black liquor flask and
analyze for sulfide, polysulfide, thiosulfate, sulfite, and sulfate
(see Appendix I).
8. Quickly pour the 4 liters of black liquor at room temperature
into the oxidation chamber.
9. Reseal the chamber except for the relief valve and start the
stirrer. The liquor is heated in the reaction chamber and temperature
equilibrium is achieved in less than 10 minutes.
10. Adjust the oxygen flow to 450 cubic centimeters per minute.
Check this flow periodically throughout oxidation. Withdraw liquor
samples at 10-to 30-minute intervals during oxidation and analyze for
sulfide, polysulfide, thiosulfate, sulfite, and sulfate (see Appendix I).
Other oxidation studies were conducted using an impingement-
type glass bubbler. This unit was mounted in the constant temperature
bath and to minimize foaming the oxidizing gas impinged upon, rather
than under, the surface of the black liquor. Placing the impinger tube
under the surface of the liquor was found to be undesirable since the
bubbles formed forced large quantities of foam into the exhaust line.
This rapid "excess-air" technique was used to verify some tests conducted
by a previous research worker in the field. The impinger was also used
to supply small quantities of well-oxidized black liquor for miscellaneous
The methods of chemical analysis used were obtained from a
variety of sources and in some cases had to be modified for use with
black liquor. Methods for sulfate, thiosulfate, and sulfite in black
liquor were obtained from TAPPI standards29 T625m-48. I
The determination of soluble sulfates in black liquor is a
gravimetric analysis. The organic matter in a diluted sample of black
liquor is precipitated by the addition of hydrochloric acid. These
interference are removed by double filtration. A dilute solution of
barium chloride is added to the sample to precipitate barium sulfate.
The precipitation is made in acidified solution near the boiling point.
The precipitate is filtered off, washed with water, ignited to redness,
and weighed as barium sulfate.
The thiosulfate and sulfite analyses are volumetric iodine-starch
titrations. The sulfide and organic matter in the liquor is precipitated
with zinc carbonate solution and the precipitated interference removed
by filtration. The filtered sample is acidified with acetic acid and
titrated with standard iodine solution to the blue-starch end point.
This titration determines the total quantity of thiosulfate and sulfite
present. A similar filtered sample is acidified and titrated with
standard iodine solution after the sulfite has been destroyed by the
addition of formaldehyde. This titration determines the total amount
of thiosulfate present in the black liquor. The sulfite concentration
is the difference between the two titrations. In general, the values
for these titrations are somewhat high due to organic reducing agents
invariably present. This is, however, the method recommended by TAPPI
and apparently the best available at this time.
Sulfide is determined colorimetrically by the method of Strickland
and Risk.30 The dissolved sulfide reacts quantitatively with p-Phenylene-
diamine dihydrochloride in the presence of ferric chloride forming a deep
purple color. The color is formed in a fixed volume of aqueous solution
containing dye, oxidizing agent (ferric chloride), and wetting agent
alkyll dimethyl benzyl ammonium chloride). The sample is appropriately
diluted and the absorption measured at 600 millimicrons against a
similarly prepared blank. A reference absorption curve is prepared using
a standard solution of sodium sulfide. This method was selected over the
TAPPI standard method because it required less time and produced more
Polysulfides are determined gravimetrically by a method modified
from Scott's Standard Methods of Chemical Analysis.31 Soluble polysulfides
quantitatively precipitate sulfur in a strongly acidified solution.
Interfering sulfides are precipitated with zinc chloride and interfering
organic materials with hydrochloric acid; then both are removed by
filtration. The pH is then lowered to 1.0 with hydrochloric acid to
precipitate flowers of sulfur. The precipitate is filtered off, washed
with water, dried,and weighed.
Ozone is analyzed colorimetrically using a variation of the
method of Smith and Diamond. The oxidants are absorbed in an alkaline
iodide solution and form hypoiodite ion. The hypoiodite quantitatively
liberates iodine when the solution is acidified with acetic acid. The
color is formed in a fixed volume of alkaline iodide solution and the
absorption is measured at 352 millimicrons against a distilled water
blank. A reference absorption curve is prepared using a standard
solution of sodium iodate. Possible interference in the oxygen stream
include nitrogen dioxide, sulfur dioxide, and peroxides. The nitrogen
dioxide interference is eliminated by adjusting the sample to the proper
pH immediately before measuring the absorbence. The possibility of
sulfur dioxide or any peroxides being present is unlikely since pure dry
oxygen is used to produce the ozone.
Methods for total solids, specific gravity, percentage inorganic
matter (sulfated ash), percentage organic matter, pH, hydroxide, and
carbonate were also obtained from TAPPI standards29 T625m-48.
The specific gravity and total solids are determined by weighing
a known volume of black liquor before and after drying. The specific
gravity is the ratio of the weight of one volume of black liquor to an
equal volume of distilled water. The total solids are the oven-dry
residue after all the water is completely driven off. This residue is
ashed, treated with sulfuric acid, ignited to redness, cooled,and weighed.
The material remaining is the inorganic matter (sulfated ash), while the
loss in weight is the volatile or organic matter.
The hydrogen ion concentration is determined on a pH meter
equipped with a glass and a calomel electrode. The hydroxide concentration
is determined from a potentiometric titration of a diluted liquor sample
with standard hydrochloric acid to a phenolphthalein end point. The
carbonate concentration is determined from a potentiometric titration
of a diluted liquor sample with standard hydrochloric acid to a methyl
orange end point.
Odorous organic gases were analyzed by gas chromatography
according to procedures developed by Adams33' 34, 35, 36 for the National
Council for Stream Improvement. The gases are concentrated by freezing
them out on silica gel at -78.50C. Moisture is removed prior to freeze-
out by passing the gas through an ice water condenser and a Drierite
adsorption trap (see Figure 7). The sample is desorbed from the silica
gel at 1000C and analyzed by gas chromatography (see Appendix I).
Detailed particulars covering the analytical procedures previously
described are presented in Appendix I beginning on page 107.
(0 ( Com- ~ a
- 41 0 00
a,= = "4 -0 ~
Co w r.
DISCUSSION OF RESULTS
Black liquor oxidation was investigated at different conditions
to determine the relative importance of certain variables on the system.
In a typical mill black liquor is kept near boiling; hence the temper-
ature range of practical interest is 600C to 900C. Oxidation was
conducted at atmospheric pressure using pure oxygen. As the temperature
increased, thus increasing the water vapor pressure, the oxygen partial
pressure was in turn decreased. Therefore, temperature and oxygen
pressure are not independent but are dependent variables which vary
The contact surface, quantity of agitation, and other contact
variables have an influence on the oxidation rate. When the temperature
and oxygen pressure were studied, the contact variables were held fixed.
When the method of contact was investigated, the temperature and oxygen
pressure were held constant. In this manner the effect of temperature,
oxygen pressure, and method of contact on the underlying chemistry of
the system was determined.
The use of ozone to eliminate odor has been suggested by persons
experienced in the field of sanitary engineering.37 Ozonating pulp mill
odors and accelerating black liquor oxidation with ozone have been
Chemistry of Black Liquor Oxidation
The initial phase of this investigation dealt with the inorganic
reactions that transpire during oxidation and the effect of temperature
and oxygen pressure on the products formed. Analysis conducted according
to standard procedures (TAPPI T625m-48) showed considerable concentrations
of sulfide, thiosulfate, and sulfate but not sulfite present. Other
analyses conducted on the unoxidized black liquor included total solids,
specific gravity, percentage inorganic and organic matter, hydrogen ion
concentration (pH), hydroxide, and carbonate. Once a clear picture of
the major mineral constituents of black liquor was obtained, oxidation
tests were conducted and the changes in the inorganic compounds observed.
Initial tests were conducted at 750C which was the median
condition in the temperature range (600 to 900C) under consideration.
Samples were withdrawn at 10- to 15-minute intervals during static
oxidation and analyzed for sulfide, thiosulfate, sulfite, and sulfate.
The hydrogen ion, hydroxide, and carbonate concentrations were determined
before and after oxidation. Chemical changes produced by oxidation are
shown in Table 4 and Figure 8.
No change in pH, hydroxide, or carbonate was found to occur with
oxidation. The sulfur compounds, however, were affected by oxidation
(Figure 8). Thiosulfate was formed by the oxidation of sulfide but the
sulfate content remained unchanged. Apparently the conditions studied
were not severe enough to drive sulfide all the way to its sulfate form.
No sulfite was detected before, during, or after oxidation and it was
suspected that thiosulfate was the highest common oxidation state of
sulfur formed under normal conditions.
EFFECT OF OXIDATION ON HYDROGEN ION, HYDROXIDE, AND
CARBONATE CONCENTRATIONS IN BLACK LIQUOR
Static Oxidation for 180 Minutes With Pure Oxygen at 750C
Material ___ Difference
Before Oxidation After Oxidation
Hydrogen Ion (pH) 11.8 11.8 0
Hydroxide as Grams per
Liter NaOH 11.0 10.9 -0.1
Carbonate as Grams per
Liter Na2CO3 11.7 11.7 0
x 0 *N 5
x 0 *C t
\ / -2 ^
o \i /
ia3T[ -lad 9urIBJ *U'p3BHiauo3U0
A material balance was calculated to determine whether or not
all the sulfur initially present was accounted for in oxidized form
after stabilization. Table 5 shows that 25.6 per cent of the sulfide
sulfur was not accounted for by the thiosulfate produced. The rate of
formation of the unknown compound or compounds was calculated by
difference and plotted in Figure 9.
The question of identifying these unknown compounds and
establishing routine methods of analysis was next approached. The
situation was complicated by the fact that no fewer than five stable
oxidation states are known to exist between sulfide sulfur with an
oxidation number of -2 and thiosulfate sulfur with an oxidation number
of +2. If the more oxidized forms than thiosulfate are considered and,
excluding the sulfate and sulfite which have been accounted for, the
matter is further complicated by the persulfates, hyposulfate, and
Since no sulfite or sulfate was produced, it seemed likely that
at least some of the unaccounted for sulfur was present in oxidation
states between the sulfide and thiosulfate form. The stable oxidation
states in this region fall into two categories, elemental sulfur and the
polysulfides. Elemental sulfur is insoluble in aqueous solution although it
decomposes slowly under alkaline conditions. Some visual evidence of
sulfur, if present, should be detectable in the oxidized black liquor.
At 750C no sulfur formation was observed. Therefore it seemed unlikely
that sulfur was a major stable product of oxidation. The polysulfides are
stable in alkaline solution and are a possible decomposition product of
20 60 0
O ca *
"0 0 u
01 *H 1 3
*4 0 (0
o4 r a)
1 0 d
cO M 4a
U C U) 0.
OL 09 0S 0+7 Oc
Injwns go a8u9uaoIad aloR
F- 4 0
ri 1 *
Of the polysulfides four stable oxidation forms -- the di, tri,
tetra, and penta-sulfide -- are known. Separation of the four forms is
difficult and since the disulfide is believed to be the most predominant
form, quantitative analysis usually determines total polysulfides and
expresses the results as disodium disulfide, Na2S2. Scott's Standard
Methods of Chemical Analysis suggests a quantitative method for sodium
polysulfide. Since the analysis is based on the fact that the soluble
polysulfides will precipitate sulfur quantitatively in a strongly acid
solution, it is not directly applicable to black liquor without elimina-
tion of certain interference. The method was adapted by precipitating
the sulfides and lignin in the black liquor with zinc chloride and
hydrochloric acid, respectively. Filtering out these precipitates left
a fairly pure solution from which polysulfide sulfur could be precipitated
by acidification and digestion. The sulfur was collected, dried, and
weighed in sintered glass crucibles. To check the purity of the
precipitate, several of the crucibles were extracted with carbon disulfide
and reweighed. The precipitate was completely soluble in carbon disulfide
indicating that the sulfur was relatively free from impurities.
This analysis showed polysulfides were present in both the fresh
and the oxidized liquor. Once a second oxidation product was established,
the oxidation at 750C was repeated to determine the changes that occurred
in sulfide, thiosulfate, and polysulfide concentrations. At 600C and
890C the above mentioned sulfur compounds as well as pH, hydroxide, and
carbonate concentrations were determined before and after oxidation.
The oxidation of sulfide to polysulfide and thiosulfate is shown in
Figure 10, and the sulfur balances over the entire temperature range is
presented in Tables 6 and 7.
01 6 8 L 9
0 4 *4
00 E4 0 34
o or -i
z 1 0
') 0 r- r'-
0 0 0
04.J \0 \0 \L
44 0 r, I'D r
1 U 4- 0 0 0
[3 0 0 0
to pq a C; Ca
co 0-1 0 0
I So0 0 0
En to 3 cc f f
1-4 LIn \O a% ~ 0
0 4 4n 9Lr\
0 r0 00 n 00 0
O O O O O O O
(O \D YCO CN in O r-
\DO 0' -1 (N (NJ
0 0 1- i-4 -4 .-4
C; C C;C; ; a
-4 n C. 0<
0 0 0 0 0 0 0
O O O O O
0'% 0 14 n --1 0 0
- .n 1-4 0 0 0 0
0 0 0 0 0 0 0
O O o oo O
0 0 0 0 0 0 0
M \.0 C CN Lin 00
r-4 -4 ,-4
) a 4-4
..4 ) 4
& 4 CJ 4/
Q) a) L
Under the conditions studied, the sulfur balance (see Table 6)
accounted for all of the sulfide oxidation products in the system with
an average error of 3.9 per cent. The sulfur balance calculated at
30-minute intervals throughout oxidation (Table 7) confirmed the fact
that no appreciable amounts of unknown inorganic sulfur compounds were
formed during any stage of the oxidation. Therefore, it was thought
that most of the experimental error was caused by inaccuracies in the
thiosulfate analysis. Since the thiosulfate determination tended to
produce high results (see Appendix I) and the error of closure was in
the same direction, a considerable portion of the error may be attributed
to the determination. The remaining differences were attributed to
various other forms of experimental error.
It is entirely possible, of course, that compounds of sulfur
other than polysulfide and thiosulfate were formed by oxidation. These
compounds, however, were not formed in significant concentrations from
the inorganic hydrous sulfide. Thus, these trace compounds were of
little practical importance. Because all the sulfur was accounted for
within reasonable limits of experimental error, the presence of trace
compounds other than elemental sulfur was not determined.
The matter of trace quantities of sulfur present in the black
liquor was investigated for two distinct reasons. The first was the
theoretical consideration that elemental sulfur should not be stable in
a strongly alkaline solution. The second reason was that the precipi-
tation of sulfur during oxidation was reported by a previous worker in
To determine the conditions conducive to sulfur formation,
samples of black liquor were oxidized in the reaction chamber (Figure 5)
at 890C, 750C, 60C, 450C, and 300C. No sulfur was formed at 89C and
750C. Traces of sulfur were noticed at 600C while at the two lower
temperatures a definite scum formed. The sulfur was very finely divided
and continued stirring led to agglomeration. To determine the order of
magnitude of the precipitated sulfur, 100-milliliter samples were placed
in a gas impinger and oxidized with oxygen. When analysis indicated that
the sodium sulfide content had dropped to 0.05 grams per liter or less,
the samples were filtered through sintered glass crucibles and thoroughly
washed. The precipitate was dried at 75 C and the weight of sulfur
expressed as grams per liter (see Table 8). The concentration of sulfur
precipitated during oxidation was in the range of 0.004 to 0.005 grams
per liter. This meant that the amount of sulfur formed was about one
thousand times smaller than the polysulfide and the thiosulfate formed,
Since this sulfur formation was negligible compared to other reactions,
it is easy to see why a major discrepancy did not appear in the sulfur
balance in Table 6.
To investigate the stability of the precipitated sulfur, a
portion was collected and washed free of black liquor. Examination under
a microscope showed the precipitate to be comprised of irregular bunches
of tiny spherical sulfur droplets. These droplets, which were insoluble
in carbon disulfide, exhibited no crystalline structure under 1455X
magnification. When the precipitated sulfur was aged about four hours
at room temperature, the drops began to grow into irregular crystals.
At room temperature (250C) about 3-days storage in aqueous solution was
CONCENTRATION OF AMORPHOUS SULFUR FORMED
BY BLACK LIQUOR OXIDATION
Oxidation in Gas Impinger With Oxygen
Sulfur Concentrations Expressed as Grams per Liter Sulfur
30 45 60
0.004 0.001 0.007
0.006 0.006 0.004
0.003 0.005 0.005
Average 0.004 Average 0.004 Average 0.005
required to complete this crystallization. The resulting crystals were
completely soluble in carbon disulfide. Photomicrographs of this
metamorphosis from amorphous droplets to rhombic crystals are shown in
Plates 2 through 4. The comparison between the crystal structure of
the aged droplets and the commercial-grade elemental sulfur can be seen
in Plates 4 and 5. From the 0.5- to 1.0-micron particle size it appears
that the sulfur was precipitated very slowly, possibly as a decomposition
product of an organic complex.
The trace of amorphous sulfur will eventually disappear if aged
long enough in the oxidized black liquor. Experimentation showed it was
apparently immaterial whether the liquor was stored aerobically or
anaerobically. The controlling factor was storage temperature. Above
60 C the re-solution of sulfur was definitely accelerated. The probable
reason for this disappearance was that the amorphous or p sulfur crystal-
lized into rhombic or o< sulfur which is unstable in alkaline solution
(see reaction 30).
(30) REDUCED SULFUR COMPOUNDS ---- S (amorphous) Te- So (rhombic).
P Time oc
The crystalline rhombic sulfur will decompose probably according to
(11) 4S (rhombic) + 6NaOH----- 2Na2S + Na2S203 + 3H20.
Under aerobic conditions the sodium sulfide formed will in turn be
oxidized to thiosulfate and polysulfide.
On the basis of the previous discussion and the data presented
in Tables 4 through 8, and Figures 8 through 10, certain generalizations
can be made regarding the reactions that take place during oxidation.
PL. 2.-Photomicrograph of freshly precipitated amorphous sulfur
PI. 3.-Photomicrograph of precipitated sulfur after aging four
hours at 250C.
PI. 4.-Photomicrograph of precipitated sulfur after aging three
days at 250C.
PI. 5.-Photomicrograph of commercial grade elemental sulfur.
The sulfide in black liquor was completely hydrolyzed and present
in the form of hydrosulfide ion, HS If this was not true and the
sulfide ion, S was present, an increase in hydroxide concentration
according to reaction 1 would occur during oxidation. Since no such
increase was found (Table 4), the conclusion was drawn that all the
sodium sulfide was hydrolized to sodium hydrosulfide and sodium
hydroxide during the digestion period (see reaction 4).
On the basis of the curves in Figure 8 and the sulfur balance in
Table 6, the premise that sulfate ion was a product of black liquor
oxidation cannot be verified. These findings are in direct contradiction
to the work of Collins (Table 2). Collins' work was conducted in an air
bubbler at 82 C, and it is difficult to explain how he could have oxidized
some of the soluble sulfide to sulfate. These results may be accounted
for by differences in the black liquor or possibly by different methods
of analysis. It is believed, therefore, that equations 31 and 32 describe
the primary inorganic reactions that take place during black liquor
(31) 4NaHS + 02 2Na2S2 + 2H20.
(32) 2NaHS + 202 P Na2S203 + H20.
It is also believed that reactions 31 and 32 more precisely describe the
fate of the dissolved sulfide than the unconfirmed reactions 1 and 10.
A relatively minor reaction that also occurs during oxidation is the
formation of amorphous elemental sulfur from the liquor.
(33) REDUCED SULFUR COMPOUNDS + 02 S (amorphous).
The oxidation states of sulfur as disodium disulfide and sodium
thiosulfate are -1 and +2, respectively. It was initially suspected that
in the absence of any data to the contrary, thiosulfate might be an
oxidation product of the more reduced polysulfide form. The shapes of
the thiosulfate and polysulfide curves in Figure 10 indicate, however,
that these two products were formed simultaneously rather than
consecutively. The flat slope of the polysulfide curve and the lack of
a maximum point showed that the thiosulfate was not an oxidation product
of polysulfide. Therefore, oxidation reactions 31 and 32 were parallel
rather than consecutive as might be suspected. The latter reaction
proceeded at a faster velocity than the former and produced a greater
concentration of oxidation products from the black liquor hydrosulfide.
Optimum conditions of temperature and oxygen pressure are
important considerations in black liquor oxidation. In a static oxygen
contact system (Figure 5) at a pressure of 1.0 atmosphere, a rise in
temperature from 60 C to 890C lowered the oxygen tension above the liquor
from 0.803 atmospheres to 0.334 atmospheres. These two dependent parame-
ters appeared to have little or no over-all effect on the products of
reaction. Table 9 shows the ratio in mole fraction of the thiosulfate
to polysulfide produced. The variation in this ratio which averaged
1.60 was small and it was concluded that the oxidation products formed
were relatively independent of temperature.
To determine whether or not the method of contact influenced the
oxidation products, experiments were performed at 750C using a dynamic
oxygen flow (Figure 6). Tests were also conducted to verify the work of
Collins.16 This was done by oxidizing black liquor at 82 C in an air
bubbler. At these experimental conditions large excesses of oxygen and
air were provided. Tables 10 and 11 show the change of certain sulfur
O ein ON
0o r- 00
a aj -i 0
.-I (0 Q CM~
0 (D co 4 4
44 '10 O1 4N
to to o1
-4 W- CO)
S0W4M 00 C* 0%0
C; Ca CM
0 O 00 0 0o
3-w 4 4
* * *
)3 00 00 00
compounds upon oxidation. Even with great excesses of oxidizing agent
sulfate was not a product of oxidation. It was concluded that in the
temperature range considered the rate of oxidation may depend on the
method of contact but the ultimate oxidation products formed do not.
Miscellaneous other studies showed that application of 1450 parts per
million (0.1450 per cent) of ozone was insufficient to oxidize any of
the reduced sulfur compounds to sulfate.
Reversibility of the Oxidation Reactions
When in contact with oxygen at elevated temperatures, the sodium
hydrosulfide in black liquor spontaneously oxidizes to disodium
disulfide and sodium thiosulfate. The degree to which these reactions
are reversible when the oxygen is removed was investigated. This
reversibility may create odor problems since in practice the black liquor
may be stored anaerobically for a length of time after oxidation and
before evaporation. If volatile sulfur compounds are regenerated, some
of the effect of oxidation will be lost. Murray reported that sulfide
was regenerated when black liquor was stored anaerobically, but he did
not explore this observation or attempt to explain it. This regeneration
can be explained if reactions 31 and 32 are reversible.
To test this hypothesis, samples of partially oxidized liquor
were withdrawn periodicallyfrom the oxidizing chamber, analyzed for
sulfide, polysulfide, and thiosulfate, and then stored anaerobically at
75 C for 12 hours. At the end of that period the samples were again
analyzed for sulfide, polysulfide, and thiosulfate, and the results
reported in Table 12. From these data it was concluded that the
c. c -i
4u4T 0 otn 0 on o t n o0 00oo
r- Q) C4 C4 0 ON Ca 4 r' ln C 4O C O
3w a 1. eo a . . W..
S1 0 r- 0 CM
0 0 0 0 0
-4 W OI
-4 () 4 M e n Cn C4 It cn 00 a-t -4 o
CQ CO w C C m
U1 0 N. 0o
0 O o in L O
*d a) M 00 C4 t r '-d LIn Ln 0 n
4 U) Z -1( 0n 0 I 00 It 00 00C
1:$) C4 COC ; 1 1 4 C
regeneration of the sulfide is due mainly to the reduction of some of
the thiosulfate while the polysulfide formed is relatively stable. The
more completely oxidized the black liquor was, the more irreversible was
the sodium thiosulfate reaction. Therefore, the oxidation of hydro-
sulfide (reaction 32) to thiosulfate is partially reversible whereas
the oxidation to polysulfide is not.
A more detailed picture of the regeneration is shown in Table 13
and Figure 11. The solid line in Figure 11 shows the rate of oxidation
of black liquor at 750C using a static oxygen system. The dotted curve
shows the hydrosulfide concentration if oxidation were stopped at that
point and liquor stored anaerobically at 75 C for 12 hours. The
reactions involved were much more reversible in the early stages of
oxidation when the system was still actively absorbing oxygen. After the
oxygen transfer ceased, the hydrosulfide regenerability of the liquor
dropped sharply. This would seem to indicate that one mechanism of
reaction predominated early in the oxidation while a different mechanism
controlled during the final stages. The average degree of regeneration
to be expected with this particular liquor at various stages of oxidation
is shown in Table 14. To prevent less than 10 per cent regeneration after
12 hours of storage, 80 per cent or more of the black liquor hydrosulfide
had to be oxidized. This pointed up the importance of evaporating the
liquor immediately after aeration to derive the fullest benefits of
oxidation. Regeneration was not pronounced in samples stored anaerobically
at room temperature.
REGENERATION OF SULFIDE UPON ANAEROBIC STORAGE
Static Oxidation With Pure Oxygen at 75 C
Anaerobic Storage for 12 Hours at 750C
Initial Sulfide Final Sulfide Percentage Sulfide
Concentration Concentration Regeneration
Expressed as Expressed as Based on Initial
Grams per Liter Grams per Liter Concentration of 6.1
Na2S Na2S Grams per Liter Na2S
01 6 8 L 9 7 c 1 0
SZuN jaITT jad surejS 'uoj~eivuaauo3
PERCENTAGE REGENERATION OF SULFIDE UPON ANAEROBIC STORAGE
Static Oxidation With Pure Oxygen at 75 C
Anaerobic Storage for 12 Hours at 750C
Based on Curve in Figure 12
Percentage Sulfide Percentage Sulfide Total Percentage
Oxidized Regenerated o Sulfide Present After
Based on Initial After 12 Hour Storage at 75 C Regeneration
Concentration of Based on Initial Based on Initial
6.1 Grams per Liter Concentration of 6.1 Concentration of 6.1
Na2S Grams per Liter Na2S Grams per Liter Na2S
0 16 116
20 24 104
40 24 84
60 21 61
80 9 29
90 7 17
For effective oxidation sufficient quantities of oxygen or air
must be brought into intimate contact with the black liquor. Wright7
found that infinite contact area did not produce an instantaneous
reaction, and he proposed that nothing is gained by increasing the area
past a certain point. Assuming that adequate gas-liquid contact can be
provided on a mill scale, the reaction kinetics and stoichemetric oxygen
requirements become the paramount considerations. In the studies
conducted by the author on the kinetics of oxidation, the effect of gas-
liquid contact was not eliminated but merely held at a fixed value. This
was achieved by holding the stirring rate and gas feed constant. By
varying the temperature and hence the oxygen pressure within the system,
their effect on the reaction rate could be evaluated. Although it would
be difficult to extrapolate the reaction kinetics to actual plant design
criteria, comparison of these data at different physical conditions
showed the relative importance of the controlling parameters. Once these
parameters are evaluated, the gathering of design data for an actual
installation is greatly simplified.
The first step was to determine the effect of temperature on
oxidation of the dissolved sulfide. Experimental temperatures of 600C,
75 C, and 89 C were investigated to determine the optimum reaction
temperature. A comparison of oxidation at these three temperatures is
shown in Figure 12.
When the semi-empirical equation 29, dC/dt K (C + bP) was
fitted to the curves in Figure 12, some quantitative comparisons can be
made about the relationship between temperature, reaction rate, and
LC 4C 0
D CD D-
C- r-4 C4 D
000 ca .
000 r E-- "
C C 00 p -0
C 0 0 *0
t I i 0
00 0 o4o
0 0 o o -04
CD 0 4'
v a ad survi 'uoTlua u0. 0
S^N.I3T-i' sr-S aj3au~u4^
oxygen absorption. The details of determining the reaction constants K
and b in the Appendix II.
A tabulation of the velocity and absorption constants K and b
is presented in Table 15. If the oxidation is a true first-order
reaction, oxygen absorption will not affect the rate and constant b will
be zero. Similarly, if oxidation is independent of the sulfide concen-
tration, the reaction will be of zero order depending solely on the
factor K (bP). As can be seen from Table 15, the static system with
very low values of bP is very nearly a first-order reaction. This was
especially true at 890C.
A normal "rule of thumb" in chemical kinetics is that the rate
constant is doubled with every 100C rise in temperature. In the case of
black liquor oxidation the velocity constant, K, varied only slightly
with temperature. The value of K at 750C was only 23 per cent higher
than the value at 600C. When the temperature was raised to 89 C, the
velocity constant began to drop off. Thus, of the three temperatures
tested, the reaction velocity was fastest at 75 C.
In connection with the ozone experiments black liquor was
oxidized dynamically at 750C. The oxidation of soluble hydrosulfide
in this dynamic system, both with and without ozone, is shown in Figure
13. Apparently 1450 parts per million (0.1450 per cent) of ozone was
insufficient to affect the oxidation of the dissolved sulfide, but the
ozone did react with the odoriferous gases above the black liquor. A
more thorough treatment of the odor problems involved with black liquor
oxidation is discussed later. Equation 21 was fitted to this curve and
the reaction constants (Table 15) were compared to those of the static
KINETICS OF BLACK LIQUOR OXIDATION
Based on Curves in Figures 12 and 13
See Appendix II for Details of Calculation
Type of Kinetic Constants
Oxygen o Velocity Oxygen Absorption Oxygen Absorption
FlConstant, K Constant, b Product, bP
60 -0.0241 0.175 0,140
Static 75 -0.0312 0.177 0.110
89 -0.0284 0.029 0.010
System 75 -0.0100 5.480 3.400
0 o o
E o u
o o -
' I 4y
S 0 C
a -d )
L OT 0
/ o *r<
S BN la3^^ lad suro-i 'UOT3BJ3U30UOO
system. In dynamic oxidation the reaction velocity was only one-third
as great as that of the static system. The oxygen absorption factor,
however, was increased by a factor of 30. Thus, the over-all reaction
was less efficient dynamically than statically and more dependent on the
rate of oxygen absorption. Apparently escaping water vapor hindered the
oxygen transfer at the gas-liquid interface.
The indirect inferences which may be drawn from the comparison
between the two systems should not be overlooked. In both cases the
oxidation was conducted under the same conditions except for the method
of oxygen feed. The static system was very nearly a first-order reaction,
but the dynamic system was significantly influenced by oxygen absorption.
This confirms that it was relatively easy for the contact parameters to
be the limiting factor in oxidation. In properly designed large-scale
equipment, however, contact should be sufficient so that the chemical
reaction rate which is a relatively fixed quantity is the limiting factor.
Black Liquor Oxygen Requirements
Additional tests were conducted using a closed static system
which enabled the utilized oxygen to be metered. Table 16 shows the
variation of oxygen absorbed with changes in temperature. Oxygen
absorption was almost doubled as the temperature was raised from 600C to
89 C. Since the black liquor contained 6.1 grams per liter of sodium
sulfide, the theoretical amount required to convert the sulfide to
polysulfide and thiosulfate according to reactions 31 and 32 was
approximately 1.1 milliliters of oxygen per milliliter of black liquor.
At temperatures above 6000C considerably more oxygen was absorbed than
VOLUME OF OXYGEN ABSORBED BY BLACK LIQUOR DURING OXIDATION
Static Oxidation for 180 Minutes With Pure Oxygen
Temperature Volume of Oxygen Absorbed by Black Liquor
o Milliliters of Oxygen per Milliliter of
C Black Liquor
Volume of oxygen corrected to 25 C and 760 millimeters of mercury.
was required to oxidize the sodium sulfide. Thus, as the temperature
increased, more and more oxygen was used in side reactions. The
percentage of absorbed oxygen used in these side reactions is shown in
Table 17. Sixty degrees centigrade was the most desirable temperature
of the temperatures tried because most of the oxygen absorbed was
utilized in oxidizing the inorganic sulfides, This was not the case at
the higher temperatures and at 890C almost half of the absorbed oxygen
was used in extraneous reactions.
Catalysis in the Oxidation of Sodium Sulfide
It has been suggested that the organic matter dissolved in black
liquor catalyzes the oxidation of sodium sulfide. This author found
that in the range of 60 C to 89 C oxygen absorption practically ceased
after 40 to 50 minutes when 1.5 to 1.7 grams per liter of sulfide had
yet to be oxidized. This remaining sodium hydrosulfide continued to
oxidize with little apparent oxygen absorption. The time lag between
oxygen absorption and utilization could be due to chemisorption of oxygen
by the liquor.
To test the hypothesis that soluble materials in black liquor
catalyze oxidation, a synthetic liquor containing 6.1 grams per liter
of sodium sulfide, 5.4 grams per liter of sodium thiosulfate, and 11.0
grams per liter of sodium hydroxide in distilled water was prepared.
This liquor contained the same concentration of the principal inorganic
compounds found in the black liquor. In a static contact system at
750C the synthetic liquor oxidized much more slowly (see Figure 14)
than the black liquor. Thus it is evident that the black liquor contained