EPA-650/2-74-005
January 1974
Environmental Protection Technology Series
.
•••••lii:
i:^
-------
EPA-650/2-74-005
INDI6ESTER BLACK
LIQUOR OXIDATION
FOR ODOR CONTROL
IN KRAFT PULPING
by
W . T . McKcan , Jr . and J . S . Gratzl
North Carolina State University
Department of Wood and Paper Science
Raleigh, North Carolina 27607
Grant No. AP-01269-02
ROAP No. 21ADC
Program Element No. 1AB015
EPA Project Officer: R. V. Hendnks
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
January 1974
-------
This report has been reviewed by the Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency, nor does
mention of trade names or commercial products constitute endorsement
or recommendation for use.
11
-------
ii
ABSTRACT
Laboratory studies have described the major reaction routes, key
inhibition reactions and kinetics of methyl roercaptan and dimethyl
disulfide during oxidation of black liquor. This provided the basis
for helping to explain low oxidation efficiencies with respect to
hydrogen sulfide, methyl mercaptan and dimethyl disulfide during black
liquor oxidation at temperatures from 60 to 90°C and suggests that high
temperature oxidation should be more afficient.
In small scale laboratory flow equipment softwood and hardwood
black liquor was oxidized at temperatures between 80 and 170°C. Oxidation
at temperatures above 100 to 120°C resulted in efficient oxidation of all
three malodorous compounds with no liquor reversion during subsequent
storage and distillation. Oxygen consumption was about 125% of theoretical
below 140°C but increased to about 200% at 170°C. Application of this
approach to batch and continuous digester systems is discussed.
Preliminary experiments show that injection of small amounts of oxygen
into the liquor circulation line during the early stage of pulping could
increase pulp yield by 1 to 3% depending on the final pulp kappa number. If
this yield increase could be obtained in combination with indigester black
liquor oxidation, the net costs for odor control in mills using batch digesters
would be very attractive.
Report submitted in fullfillment of Grant No. RO 1 AP-01269-01 & 02
-------
iii
CONTENTS
Page
Abstract ii
List of Figures v
List of Tables . vii
Acknowledgements viii
Sections
Conclusions 1
Recommendations 3
Introduction 5
A. Background 5
B. Rationale for Indigester Oxidation 16
Objectives 17
Discussion 18
A. Reactions of Methyl Mercaptan and Dimethyl
Disulfide in Sodium Hydroxide Solutions 18
1. Kinetics of oxidation of methyl mercaptan in
aqueous alkaline solution with molecular oxygen 18
2. Hydrolysis of dimethyl disulfide 21
3. Kinetics of hydrolysis of dimethyl disulfide 23
4. Summary of methyl mercaptan and dimethyl disulfide
reactions in aqueous, alkaline solutions 24
B. Reactions of Methyl Mercaptan and Dimethyl Disulfide in
Simulated Black Liquor 28
1. Methyl mercaptan oxidation by models compounds
featuring certain structures in kraft lignin 28
2. The effect of kraft lignin on the consumption of
methyl mercaptan 33
-------
iv
3. Effect of sodium sulfide on the oxidation of
methyl mercaptan 40
4. Methyl mercaptan reactions during black liquor
oxidation at temperatures below 100°C. 43
5. Methyl mercaptan reactions during black liquor
oxidation at temperatures above 100°C 47
C. Summary of Methyl Mercaptan and Dimethyl Bisulfide
Oxidation Kinetics in Simulated Black Liquor 51
D. Odor Reduction by In-digester Oxidation of Kraft Black
Liquor with Oxygen at Temperatures above 80°C 56
1. The influence of sodium sulfide throughout kraft
pulping 56
2. In-digester oxidation: Influence on inorganic
compounds 61
3. In-digester oxidation: Influence on methyl mercaptan,
dimethyl sulfide and dimethyl disulfide 66
4. In-digester oxidation: Influence on pulp yield and
pulp properties 70
5. In-digester BLO at variable temperatures 75
E. In-digester Oxidation of Hardwood Black Liquors 84
1. Influence of oxidation on methyl mercaptan, dimethyl
sulfide and dimenthyl disulfide 84
2. Influence of Oxidation on Pulp Yield 84
F. Odor Control in Batch and Continuous Digesters 88
G. Possible Pulp Yield Increase in Combination with Odor
Control by In-digester Oxidation 91
H. Cost Estimates and Effectiveness of Odor Control by
Indigester BLO Compared to Other Control Methods 95
I. Experimental
1. Equipment and Materials 106
2. Procedures 108
-------
V
FIGURES
Page
1 Projection of Production of Chemical Pulps in the U. S. 6
2 Major Atmospheric Pollutants Lbs/T 7
3 Pulp Yield as a Function of Lignin Removed 8
4 Modifications of Pulp Mill for Odor Control 12
5 Oxidation of Methyl Mercaptan at Varying pH 20
6 Oxidation-hydrolysis Cycle for Conversion of Methyl
Mercaptan to Nonvolatile Product 26
7 Methyl Mercaptan Consumption by Equlraolar Amounts of
Hydrogen Peroxide, Anthraquinone Beta Sulfonate,
o-quinone and 10 psig Oxygen Respectively 31
8 Consumption of Methyl Mercaptan in Aqueous Alkaline
Solution with Kraft Lignin Added 35
9 Formation of Dimethyl Bisulfide in Alkaline Solution
with Kraft Lignin Added 36
10 The Effect of Temperature on the Formation of Dimethyl
Disulfide in the Presence of Kraft Lignin 39
11 Oxidation of Methyl Mercaptan with 5 g/£ Sodium
Sulfide Added 41
12 Consumption of Methyl Mercaptan in Simulated Black Liquor 45'
13 Consumption of Methyl Mercaptan in Simulated Black Liquor 48
14 Schematic Outline of Reactions of Methyl Mercaptan during
Black Liquor Oxidation 52
15 Composition of Cooking Liquor as a Function of Cooking
Time, and Temperature in Regular Kraft Cooking and
Liquor Exchange Cooks of Loblolly Fine 58
16 Total Pulp Yield as a Function of Pulp Lignin Content 60
17 The Content of Sodium Sulfide in Black Liquor as a Function
of the Amount of Oxygen Injected at a Rate of Injection
of 5.3 g of 02/min. 63
-------
vi
Page
18 Effective Alkali in Black Liquor in Regular Kraft Cooks
and Cooks with Injection of Various Amounts of Oxygen
at a Rate of 5.3 g of 0 /min. 65
19 Methyl Mercaptan Content in Black Liquor at Various
Amounts of Injected Oxygen 68
20 Dimethyl Sulfide Content in Black Liquor at Various
amounts of Injected Oxygen 71
21 Tear Factor Versus Breaking Length for Regular Kraft
Pulps and Pulps Cooked with the Injection of Oxygen 74
22 Tensile Strength as a Function of Number of Revolutions in
the PFI-mill for Regular Kraft Pulps and Pulps Cooked
with the Injection of Oxygen 76
23 Representative Sodium Sulfide Oxidation Rates 81
24 Oxygen Consumption for Complete Oxidation of Sodium
Sulfide in Black Liquor Oxidation 82
25 Methyl Mercaptan and Dimethyl Disulfide Concentration as
a Function of Oxidation Time for Several Oxidation
Temperatures 83
26 Methyl Mercaptan Content in Black Liquor from Pulping
Red Gum 85
27 Total Pulp Yield as a Function of Kappa Number for
Pulping Red Gum 87
28 Pulp Yield as a Function of Lignin Content for Reference
Kraft Pulps and Kraft Pulps with Oxygen Pretreatment 94 •
-------
vli
TABLES
Page
I Air Pollutants from Kraft Pulp Mill with No Odor Control 10
II Ambient Air Odor Thresholds (3) 15.
Ill Rate Constants for the Oxidation of Methyl Mercaptan 22
IV Rate Constants for Dimethyl Disulfide Hydrolysis 25
V Effect of Quantity of Oxygen Injected on the Methyl
Mercaptan Content of Black Liquor 69
VI Total Pulp Yields in Cooks with and without the Injection
of Oxygen into the Digester at the End of the Cook 72
VII Physical Properties of Regular Kraft Pulps and Pulps
Produced with the Injection of Oxygen at the End
of the Cook 77
VIII Effect of Injection of Oxygen on Pulp Brightness 78
IX Total Pulp Yield for Kraft Cooks of Red Gum with and
without Oxygen Injection at the End of the Cook
Liquor to Wood Ratio 3.6:1, 90 Minutes from 25°C
to 170°C 86
X Effectiveness and Costs for Odor Control in a 500 T/day
Kraft Mill 97
-------
viii
ACKNOWLEDGEMENTS
The authors wish to thank Professor R. G. Hitchings of North Carolina
State University for valuable suggestions and continued interest in this
work. Mr. R. K. Stevens and members of the Field Methods Development
Section of the Environmental Protection Agency, Research Triangle, N. C.
provided invaluable assistance in design of analytical systems for analysis
of sulfur compounds. Dr. K. P. Kringstad and Dr. P. J. Kleppe initiated
the studies in this program and provided much guidance in the early phases
of the work.
The financial assistance of the Environmental Protection Agency is
gratefully acknowledged.
The financial assistance for J. Libert from the American-Scandinavian
Foundation (The Gunnar W. E. Nicholson Fellowship) and 1959 ars Fond for
Teknisk och Skoglig Forskning samt Utbildning is also gratefully acknowledged.
A gift of mill wood chips from the Riegel Paper Corporation, Riegelwood, N. C.
is gratefully acknowledged.
The authors appreciate the technical assistance of Robert Allison,
R. D. Shirley, Ms. Adrianna Kirkman and Ms. Elizabeth Wilson.
Graduate students, Dr. J. S. Bentvelzen and Mr. Kent Maurer contributed
immeasurably to the success of this project.
-------
1
CONCLUSIONS
During black liquor oxidation (BLO) there are complex relationships
between lignin, sodium sulfide, methyl mercaptan and dimethyl disulfide
which can limit oxidation efficiencies. It is well known that at con-
ventional oxidation temperatures (60-90°C) sodium sulfide in black liquor
can be oxidized, but is regenerated during storage and evaporation of the
oxidized liquor. The regeneration has been attributed to formation of ele-
mental sulfur during oxidation which later is partially converted back to
sulfide in alkaline liquors. The present studies suggest that another type
of regeneration occurs. Sulfide reacts with lignin during oxidation and
is slowly displaced from the lignin by hydroxide during storage and evapora-
tion resulting in an apparent reversion.
At normal oxidation temperatures mercaptide anion reacts with lignin in
an analogous manner and is later also displaced by hydroxide during evaporation.
Furthermore, oxidation of mercaptan to dimethyl disulfide suffers an apparent
inhibition in the presence of lignin or when the sulfide content of black
liquor is greater than about 0.2 g/&. Consequently, black liquor which has
been oxidized at normal temperatures contains significant quantities of
unoxidized methyl mercaptan and both mercaptan and sulfide are further regenerated
during storage and evaporation of the oxidized liquors. Thus, the utility of
BLO as an odor control measure at normal oxidation temperatures is limited by
these relationships.
However, at temperatures greater than 100 to 120°C black liquor oxidation
could be more effective as an odor control measure. Above these temperatures
the rates of sulfide and mercaptan oxidation are greatly enhanced relative to
rates during normal BLO. Furthermore, the inhibition of mercaptan oxidation by
lignin and sulfide becomes negligible. Finally, no reformation of sulfide or
mercaptan occurs during storage and evaporation of black liquors which have been
-------
2
oxidized at temperatures above 120°C. Consequently, the potential for
release of odorous substances would be greatly reduced by oxidation at
elevated temperatures.
The configuration of a high temperature BLO system would depend on the
type of digester. The results of the present work show that at oxidation
temperatures of 130-140°C reacrion rates are very rapid with oxygen con-
sumption of about 125% of theoretical. Consequently, for continuous digesters
the oxidation could be accomplished by direct injection of oxygen into the
liquor extraction line before the first flash tank.
For circulation batch digesters redesign of the circulation system would
be required to permit liquor circulation during the later stages of pulping.
Injection of oxygen into the circulation line near the end of the cook would
provide good odor control. In this case oxygen consumption would be somewhat
greater because of side reactions xfhich occur at digester temperatures (170°C).
Results show that pulp yield and pulp quality are retained if the black liquor
is oxidized during the last 30 to 45 minutes of a normal kraft cook producing
bleachable grades.
Digester relief and blow gases would need to be combusted since dimethyl
sulfide is not oxidized at temperatures below 170°C. Costs for collection and
combustion of these low volume gas streams in the lime kiln are reasonable
relative to other odor control costs.
Preliminary studies indicate that introduction of small amounts of oxygen
into the circulation line of batch digesters early in the pulping cycle could
result in some pulp yield increase. If such yield increase could be coupled
with the efficient odor control obtained by high temperature BLO in the digester
the economic incentive is very attractive. Similar yield increases may be
obtained in continuous digesters, though further studies are required to define
the operating limits of that system.
-------
3
RECOMMENDATIONS
The present studies have shown the limitations of conventional black
liquor oxidation (BLO) and the possible advantages of high temperature
black liquor oxidation in the digester as a method of reducing odor from
kraft pulping. These laboratory scale results suggest that high temperature
BLO should be more effective than conventional systems at equal or lower
cost. Confirmation of these results on a larger scale is required to provide
the basis for firm cost estimates. The following work is recommended:
1. Continuous digesters
High temperature oxidation could be accomplished by injection of oxygen
into black liquor extraction lines before the flash tank. A field test
should be made using a full scale continuous digester to confirm the
extent of odor control and oxygen consumption. If oxidation is done at
about 1AO°C the extraction system should be modified to have a residence
time of 5 to 6 minutes between oxygen injection and the time that black
liquor enters the first flash tank.
The modification of extraction system should require a reasonable capital
outlay and the major cost for a testing program would be for purchased
oxygen. For testing period delivered oxygen could be used so no onsite
generation and storage would be required. The total cost for a one month
demonstration on a 500 T./day digester would probably be between $20,000
and $40,000 including digester modification, oxygen costs, gas controls,
and testing for total reduced sulfur emissions from various sources in
the pulp mill.
2. Batch digesters
Similar field tests should be made using circulation batch digesters.
The liquor strainers will probably require extensive modification to
-------
avoid collection of fiber fines and pressure build-up which normally
occurs to the later stages of pulping. Some type of wiper arrangement
similar to continuous digesters may be required to maintain open strainers
and permit adequate circulation rates to oxidize all the black liquor in
the digester. Testing would provide confirmations of odor control,
oxygen consumption and influence of indigester BLO on the pulp properties.
If odor control can be coupled with an increase in pulp yield, high temperature
black liquor oxidation could be very economically attractive. Preliminary
studies reported herein have shown that some yield increase can be obtained
when small amounts of oxygen have been injected into the liquor circulation
line during digester heatup. This yield increase may be the result of
carbohydrate stabilization by oxidation of reducing end groups forming alkali
stable aldonic acids. The oxidation may be by oxidation with oxygen or poly-
sulfide formed from sodium sulfide in the liquor.
Stabilization will be influenced by the temperature, rate of liquor
circulation, rate of oxygen injection, sulfidity and effective alkali of liquor
and the rate of diffusion of oxygen or polysulfide into the wood chips.
Laboratory studies should be made to determine the influence of these parameters
on yield increases. This work would suggest the optimum conditions to obtain
maximum yield increase.
-------
INTRODUCTION
A. Background
The growing concern about odor emissions from kraft pulp mills is
reflected in numerous studies published in this area over the last decade.
This is a natural area for inquiry since the pulp and paper industry
comprises a significant portion of total industrial output. Furthermore,
of the nearly (250) pulp mills in the United States (126) are kraft mills.
The dominant position is. also illustrated in terms of production output in
Figure 1 (1). Available projections indicate that this position will be
maintained at least into the mid 1980's. Consequently, environmental
difficulties with the process will be of continued concern and substantial
efforts to develop methods for minimizing emissions are well justified.
As a basis for discussion a flow sheet for a "typical" southern kraft
mill with ninimal odor control is shown in Fipure 2. Wood chips are treated
with an aqueous solution of sodium hydroxide and sodium sulfide at 170°C
for 2 to 4 hours. This treatment degrades and dissolves lignin and some
carbohydrates so the recovered fiber amounts to about 45-55% of the original
wood. While the pulp yield varies depending on species and pulping conditions,
Figure 3 (2) indicates the extent of removal of the classes of wood components
as a function of pulp lignin content. At the end of the cook residual pulp-
ing liquor is separated, and the pulp is washed.
-------
PROJECTION OF PRODUCTION
OF CHEMICAL PULPS IN THE U. S.
80
70
to
P 60
o
10
CO
o
50
40
S 30
§
i
o.
20
10
TOTAL
KRAFT
NSSC
SULFITE
DISSOLVING
SODA
1985
YEAR
Figure 1
-------
MAJOR ATMOSPHERIC POLLUTANTS
LBS/T
CH3SH =2:37 CH3SSCH3 » 1.86
1
ic:
8
to
i
x_<
DIRECT
CONTACT
EVAP.
- !>
RECOVERY
" FURNACE y
I
JLI'ME tt
'
ICAUSTICIZERJ \
1 'ISUAKERJ
MULTIPLE
EFFECT
EVAPS
600'
PULP MILL EFFLUENT
5000 GAL /T
BOD
Lbs/T
S.S. = l7Lbs/T
Figure 2
-------
8
Q
O
O
Ok
Q
-J
LU
j --EXTRACTIVES
TOTAL PULP YIELD
LIGNIN
CONTENT
HEMI CELLULOSE CONTENT
CELLULOSE CONTENT
20 40 60 80
LIGNIN REMOVED, %
Figure 3. Pulp Yield as a Function of Lignin Removed
-------
The remainder of unit operations shown in Figure 2 are intended to
recover and reuse the Inorganic pulping chemicals (NaOH and Na.S) and to
produce energy by burning organic material dissolved from the wood during
pulping. The combined residual pulping liquor and pulp wash water called
black liquor,1 normally has a solids content of 14-16% of which about 1/3
Is lignin degradation products, 1/2 carbohydrate degradation products,
1/10 sodium sulfide and sodium hydroxide and lass than 1% organic sulfur
compounds• After concentration to about 60% solids in multiple effect
evaporators, the resulting strong Black liquor is further concentrated to
65 to 70% solids in a direct contact evaporator and fired to the recovery
furnace. Combustion of the organic material providea hot water and steam
for the pulp mill. The molten smelt issuing from the bottom of the furnace
contains sodium carbonate and sodium sulfide. The smelt is dissolved in
water and treated with lime in the causticizer to convert carbonate to
sodium hydroxide• The resulting white liquor is recycled to pulp digesters.
Lime is regenerated by heating the precipitated calcium carbonate in a
lime kiln.
The major emission sources are identified in Figure 2 and typical
quantities of sulfur containing malodorous substances emitted from these
sources are shown in Table I. The volatile odorous substances of major concern
to the kraft pulping' Industry are hydrogen sulfide, methyl mercaptan, dimethyl
sulfide and dimethyl' disulfide. The total quantity of these four substances i*
termed Total Reduced Sulfur (TRS). Factors influencing the formation and release
of these substances have been thoroughly reviewed (3),." Hydrogen aulfide ie
-------
AIR POLLUTANTS
Source
KRAFT PULP MILL WITH HO
1 Lbc/Ton A.D..Pulp
CH3SH CH3SCK3 CH3SSCH3
HS
1. Turpentine
Decanter
2. 'Digester Blow
Tank
3. Washer Hood
Vent
4. Washer Seal
Tank Vent
5. Evaporator
Seal Tank
6. Direct Contact
Evaporator. 8
Recovery Furnace
-7 Dissolver
$ Lime Kiln
(with control)
0.02
0.45
O.I5
O.IO
O.5O
I.IO
0*El
O.O5
0.4O
I.4O
0.05
O.O5
O.IO
0.20
O.OI
005.
O.O3
I.5O
O.O5
0.03
O.O5
0.2O
»O
0.05 .
-0
O.05
O.05
O.O2
0.50
20.O
O.O3
QIO
TOTALS
2.37
2.20
1.86
2O.70
-------
11
derived from sodium sulfide which in alkaline liquors participates in
the pH dependent hydrolyses shown in Equation 1. Sulfide and hydro-
sulfide anions are formed in kraft mill liquors at concentrations
which are dependent on the ionization constants and on the amount of
added sodium hydroxide. The dissociation constant K has a value of
— _l_ 1 _l_ 1
S~ + 2 H ^ H S~ + H WH^S (1)
about 2 x 10 (4)> so hydrogen sulfide will be released from alkaline
liquors in a pulp mill only when the pH is reduced below a level of
about 9 to 10. This occurs mainly in the direct contact evaporator
where carbon dioxide in the flue gas neutralizes sodium hydroxide in
the strong black liquor.
Since hydrogen sulfide comprises a substantial part of the total
TRS, a major effort has been made by the pulp and paper industry to
control its emission, particularly from the recovery furnace area. One
approach has been to oxidize weak black liquor or strong black liquor
(location of oxidation units is shown by dashed lines in Figure 4)with
air or pure oxygen. The objective of black liquor oxidation (BLO) is to
convert sodium sulfide to non volatile sodium thiosulfate as shown in
Equation 2.
+ 20H~ (2)
If this conversion can be successfully accomplished x*ith high efficiency,
little or no hvdropen sulfide is released from the recovery furnace stack.
-------
MODIFICATIONS OF PULP MILL FOR ODOR CONTROL
12
g- • •
* -\
7 s
i
X\
*"
[
1 URP.
DF.CAN.
*
\
\
TO LIME
KILN
V
^W \L
t
. I .
x-S
^.^ ^^
oz
ml—
Y
t
1
o
s
(/)
J 1
•
**
f
X "\ . 1,
ll! ^AIR 0
r j. a<_ j_ ?
l^^^-.-l i^< ^ — LIMEKILN |
T i A.C.E. v^it^r^-* 8| j
1 . -^ "• ICAUSTICIZER
1 /
mRFCT i /
/»rtMTA/»T 1
UUl\ 1 AL. 1 | " AWr-0 1
FVAP, | i^ FT 1 ,S.L7^E°J
T^- — ' ! t IUISSOLVERI •
^ 1 •;-•—-—
; x-1-^ IWHITE LIQUOR h
'CT ,
-------
13
Obtaining a high, permanent conversion of sulfide to thiosulfate
can be a difficult problem, however. Unless the oxidation temperature
is maintained above about 70 to 80°C, elemental sulfur may form. This
will slowly dissolve and later reform sulfide during evaporation and -
storage of the oxidized black liquor. The details of this liquor
"reversion" have been discussed by Sarkanen (3) and Christie (5). In
a second type of reversion methyl mercaptan present during black liquor
evaporation may slowly disproportionate to reform sulfide ion (6).
A more recent approach to odor control has been to isolate the strong
black liquor from direct contact with recovery furnace flue gas. Either
strong black liquor (at 65-70% solids) is fired directly to the recovery
furnace or black liquor (60% solids) is concentrated in an indirect heat
exchanger (to 65 to 70% solids) by hot flue gas and then fed to the recovery
furnace. This option is also identified in Figure 4 as the A.C.E. system.
A variety of other control methods are discussed in reference 3.
Control of the malodorous organic sulfur compounds causes some
difficulty since they are emitted from many sources in the mill as shown
in Table I. Methyl mercaptan and dimethyl sulfide are formed only in the
digester and released throughout the mill. Dimethyl sulfide is a neutral
substance so is readily steam stripped from the alkaline liquors during
JUester relief and blow. Methyl mercaptan is a weak acid so it dissociates
In aqueous solution according to Equation 3. The dissociation constant has
•i value of 4.3 x lo"n(7) so free methyl mercaptan.may be stripped from
-------
black liquor at pH levels below 12 to 13. From Table I this clearly
happens in several parts of the kraft mill.
CH3SH K3 ^ CH3S~ + H+ (3)
Dimethyl disulfide is formed by air oxidation of methyl mercaptan
after the digester has been relieved to atmospheric pressure. This
neutral compound is also readily stripped out of black liquor and
emitted to the atmosphere. Dimethyl disulfide may also be formed by
oxidation of methyl mercaptan during black liquor oxidation. However,
oxidation in black liquor and release of methyl mercaptan and dimethyl
disulfide from black liquor during oxidation is much more complex than
expected and is highly dependent on oxidation conditions. A detailed dis-
cussion of these effects is given in a later section of this report.
Because of the diversity in composition and flow rate, any single
process modification can limit emissions of these odorous substances
from only a part of the sources. Consequently, development of control
measures has generally involved "add on1' technology to limit emissions at
each source. Significant progress has been made over the last two decades
and the approaches are described in detail by Sarkanen, et al. (3). Installa-
tion of these types of collection and treatment systems involves a considerable
capital outlay. Furthermore, the equipment must be operated at very high
annual efficiencies to maintain the ambient air concentrations in the
vicinity of the mills near or below the odor thresholds shown in Table II.
Consequently, it may be useful to minimize emissions by destroying the odorous
substances in the digester before blow down to atmospheric pressure.
-------
15
Table II. Ambient Air Odor Thresholds (3)
Compound Odor Threshold, ppb
H S 0.4-5
CH SH 2-3
CH SCH xv 1.0
1-5
-------
16
B. Rationale for In-digester Oxidation
It is well known that essentially all methyl mercaptan (dimethyl
disulfide) and dimethyl sulflde occuring during kraft pulping are
formed in the digester and released thereafter from various places as
described above. Consequently, a possible approach to odor control
could be oxidation of all malodorous substances in the digester before
the blowdown to atmospheric pressure. In-digester oxidation with
oxygen is attractive for several reasons. This process may require limited
changes in already existing mill equipment. Foaming problems are not
expected since all gas injected will be consumed. The formation of ele-
mental sulfur will be negligible at the high temperature in the digester.
Consequently, reversion of the oxidized liquor may be less than in conven-
tional oxidation systems. Furthermore, such a procedure may also lead to
an in-digester oxidation of methyl mercaptan, dimethyl sulfide, and
dimethyl disulfide. In addition to the beneficial effects normally ascribed
to conventional black liquor oxidation (BLO), the process should also reduce
emissions of odorous compounds from digester relief, blow, and pulp washing
operations which are presently not treated in many mills.
-------
17
OBJECTIVES
The overall objective of this research project was to determine
the feasibility of kraft mill odor control by oxidation of all malodorous
sulfur compounds in the digester before blow down to atmospheric pressure.
This method of black liquor oxidation (BLO) would permit oxidation of
odorous substances at digester temperatures and pressures and eliminate
emissions of hydrogen sulfide, methyl mercaptan and dimethyl disulfide
from the digester, pulp washing and recovery furnace areas. If successful
such a system would permit nearly odor free operation and eliminate the
need for installation of equipment for collection, ducting and combustion of
off gases from the numerous emission sources in a kraft mill.
The feasibility of this approach depends first on obtaining a high
efficiency with respect to oxidation of the malodorous sulfur containing
substances without excessive consumption of oxygen. In a first phase of
work laboratory studies were made to determine the interaction of sulfur
compounds, other black components and liquor during BLO and to study the
influence of reaction variables on oxidation kinetics and oxidation efficiency.
Second, for indigester BLO to be acceptable, oxidation must be conducted
in a way that pulp yield and physical properties are not adversely affected.
Therefore, studies were required to determine:
a) At what stage of pulping oxygen could be injected without reducing
pulp yield or affecting pulp physical and optical properties
b) The oxygen consumption and oxidation efficiency at conditions that
did not result in unacceptable changes in yield or pulp properties.
These studies were made for pulping of a softwood (Loblolly Pine) and a hardwood
(Red Gum) species.
Even if odor control could be successfully accomplished by this approach,
some expense would be incurred for modification of digesters and for oxygen
costs. However, if the modified digester could be operated in a way to
obtain a pulp yield increase, the resulting increase revenue could be taken
as a credit to offset the additional costs for odor control. Consequently,
preliminary studies were made using a modified oxygen injection system on
laboratory batch circulation digesters to determine if yield increases could
be realized in combination with odor control.
-------
18
DISCUSSION
A. Reactions of Methyl Nercaptan and Dinethyl Disulfide
in Sodium Hydroxide Solutions
As a basis for studies on in-digester BLO, this section describes
studies on the reactions of methyl mercaptan and dimethyl disulfide
in sodium hydroxide solutions. The reactions of methyl mercaptan are
highly influenced by other organic components present in the black liquor,
so interpretation of reaction routes can be obscured by side reactions.
Therefore, the oxidation of methyl mercaptan in water and sodium hydroxide
solutions was studied. To this system components of black liquor were
added in separate experiments, until a composition approaching commercial
black liquor was reached.
The background information obtained from these experiments provides
easier interpretation of results in actual BLO experiments.
1. Kinetics of the Oxidation of Methyl Mercaptan in Aqueous Alkaline
Solution with Molecular Oxygen
Mercaptans are readily oxidized in an aqueous alkaline solution
as reported by Oswald and Wallace (8). These detailed studies suggest
that the active species in the oxidation of mercaptans involves mercaptide
anions formed bv the reaction of a base as shown in reaction 4.
+ HO ^ RS + H0 (4)
-------
19
Thus, the general kinetic expression for the oxidation of methyl mercaptan
vjith molecular oxygen can be formulated as shown in Equation 5.
k[HM]M[0]N[Kf ]P (5)
2
At a pH where all the methyl mercaptan is ionized, the reaction would
become independent of base. For example, as illustrated in Figure 5,
the rate of methyl mercaptan oxidation increases with increasing initial
amounts of sodium hydroxide, until a pH between 12 and 13 is reached.
At higher pH values the rate of oxidation ts not further increased.
Methyl mercaptan has a pKa value of approximately 10, so it is practically
completely ionized at a pH between 12 and 13 and further addition of base
apparently has no effect on its rate of oxidation.
For all practical conditions during black liquor oxidation the pH
will be greater than about 12, thus Equation 5 simplifies to Equation 6,
since P = 0 and [H0~]° - 1.
dfMM] = k [MM]M[0,]H (6)
~ dt 2
The values of M and N were determined by the method of variation in
ratio of reactants (9). The overall empirical expression for the oxidation
of methyl mercaptan in sodium hydroxide solution is given by Equation 7.
d[HM] = k [MM]0'75 Po2°'66 (?)
dt
This form applies only at pH levels above 12 where all riethyl mercaptan is
ionized.
-------
20
Terap 20 C
0)
a pH 12
X PH 13
0
Figure 5 Oxidation of methyl mercaptan at varying pH.
-------
21
The rate of oxidation is sufficiently fast that reliable kinetic
studies could be made only at temperatures below about 25°C. The
temperature dependence was determined by conducting reactions at 10,
15, 20 and 25°C. Oxygen pressure was maintained constant so Equation 7
reduced to Equation 8. The reaction followed the Arrhenius relationship
k< P°2' - where k« - k Po °'66 <»
dt
so integration of Equation 8 and appropriate graphing resulted in the
rate constants and activation energy shown in Table III (9) .
2. Hydrolysis of Dimethyl Disulfide
In the presence of base and molecular oxygen, methyl mercaptan is
converted to dimethyl disulfide. Once formed, dimethyl disulfide does
not oxidize at conditions employed during black liquor oxidation, but
can undergo hydrolytic disproportionation (3) . This hydrolysis of
dimethyl disulfide has been studied in the present work for two
reasons: first, dimethyl disulfide, is very volatile and has a very
low odor threshold, so it contributes to the overall odor of kraft mills;
second, when dimethyl disulfide hydrolyzes, significant amounts of methyl
mercaptan are formed as one of the products.
In the study of this hydrolysis reaction, two approaches have been
taken'. First, the reaction products were determined by Nuclear Magnetic
-------
Tatile - Rate constants for the oxidation of methyl mercaptan.
Temp, C
10
15
20
25
k
1
1
2
3
,,, sec.
rcole1'75
oOl
*U7
,06
.25
ti
2
min.
12
8.5
6
1^
12.9 Kcal/mol
-------
23
Resonance (NMR) spectroscopy, and from the results the total hydrolysis
stoichiotnetry was formulated. Studies on the hydrolysis of disulfides
by sodium hydroxide have been reported (8, 10). It is generally suggested
that the mechanism involves an initial nucleophilic displacement at the
disulfide bond. From these studies it can be predicted that the major
products of dimethyl disulfide hydrolysis may be methyl mercaptan, methane
sulfinic acid and/or methane sulfonic acid. This was confirmed in the
present work using the ?C1P technique. The reader is referred to reference
(9) page 25 for a detailed discussion of the work.
The overall material balances for dimethyl disulfide hydrolysis by
hydroxide and hydrosulfide anions are given in reactions 9 and 10
respectively.
OH" ^ CH3S°2~ + 3 CH3S
CH3SSCH3 HS ^ 2 CH3S02 + 4 CI^S (10)
3. Kinetics of the Hydrolysis of Dimethyl Disulfide
As described in the previous section, dimethyl disulfide is hydrolysed
slowly by hydroxide and much more rapidly by hydrosulfide. Both components
are present in the black liquor, so it is appropriate to study their effects
on the hydrolysis of dimethyl disulfide in detail.
-------
24
Since the rate limiting step involves nucleophilic attack of the
anion at the disulfide bond, the simplest general kinetic expression
for the disappearance of dimethyl disulfide by hydrolysis with sodium
hydroxide and by sodium sulfide is shown in Equations 11 and 12
respectively. The powers M, N, P and 0, were determined to have the
d[DiffiS] = k [DMDS]M[HD
dt i
C12)
value of 1.0 (9), supporting the suggestion that dimethyl disulfide
hydrolysis is initiated by a nucleophilic, S 2 substitution. The rate
constants are summarized in Table IV.
Activation energies of 27.5 Kcal/mole were determined for hydrolysis
by hydroxide anion (Equation 11) and 16.5 Kcal/mole for hydrolysis by
hydrosulfide anion (Equation 12). This difference in temperature
dependence has a significant effect on the optimum conditions for black
liquor oxidation as will be discussed in a later section.
4. Summary of Methyl Mercaptan and Dimethyl Disulfide Reactions
in Aqueous Alkaline Solution
The scheme shown in Figure 6 summarizes the reactions discussed in
earlier sections. With a base and oxygen methyl roercaptan is readily
oxidized to dimethvl disulfide which in turn hydrolyzes rapidly with
-------
25
Table IV Rate constants for dimethyl disulfide hydrolysis.
Temperature
10
20
30
to
80
128
1U6
160
AE
HO"
k 1/mole-min.
(0.0015)*
0.02
0.1
3.6
27.3 Real/mole
HS"
k_ I/mole-rain.
0.2
0.7
1.7
U.3
(78)*
16.5 Kcal/mole
Calculated from Arrhenius expression.
-------
26
oxidation
hydrolysis
CH SSCH-
\j
K) slow
HS" fast
CHgSH
CH SOH
fast
CH SO H
3 2
slow
V
CH SO H
3 3
Figure 6 -
Oxidation-hydrolysis cycle for conversion of
methyl mercaptan to nonvolatile product.
-------
27
hydrosulfide or more slowly with hydroxide forming methyl mercaptan
and the non volatile methane sulfinic acid. From the overall equation
for the hydrolysis of dimethyl disulfide by sodium hydroxide (reaction 9)
it can be seen that 2 moles of dimethyl disulfide form 3 moles of methyl
mercaptan. Thus, from the oxidation of 4 moles of methyl mercaptanj 3
moles are regenerated. Similarly, the overall equation for the hydrolysis
of dimethyl disulfide with sodium sulfide (reaction 10) shows that 4
moles of methyl mercaptan are formed from 2 moles of dimethyl disulfide.
While this oxidation-hydrolysis cycle permanently destroys some methyl
mercaptan, the regeneration of substantial quantities makes inefficient
use of oxygen and will be troublesome in black liquor oxidation.
The hydrolysis of dimethyl disulfide in this cycle could be the
rate determining step depending on the conditions. At 50-90°C with
sodium sulfide present the rate is quite high, but is very slow with only
sodium hydroxide present. Thus, to complete the conversion of methyl
mercaptan to its non-volatile product under practical conditions, sodium
sulfide is required. However, under oxidizing conditions sodium sulfide
will also rapidly oxidize to sodium thiosulfate by reaction 2 repeated here.
While sodium sulfide oxidation is desired in black liquor oxidation, its con-
sumption causes significant reduction in the rate of conversion of methyl
mercaptan and dimethyl disulfide to non volatile products.
5/2 QZ ^ Na2S2°3 + 2 Na°H
-------
28
The activation enerpy for the hydroIvsis reaction by hydroxide
is quite high, consequently the rate is greatly effected by tempera-
ture chanpes. For instance by increasing the temperature to 125°C
and above, the rate of hydrolysis by hydroxide will become quite high.
In summary, with oxygen and base present, the conversion of
methyl mercaptan to the non volatile methane sulfinic acid, proceeds
through a very inefficient oxidation - hydrolysis cycle. The hydrolysis
step in the sequence is possibly rate determining depending on the
conditions. At lower temperatures, 50-90°C, the hydrolysis depends on
the presence of sodium sulfide which, however, is also rapidly oxidized.
At temperatures greater than about 125°C, the rate of hydrolysis by
hydroxide becomes verv rapid so the oxidation-hydrolysis cycle can be
effectively completed even when the hydrosulfide concentration is very
low.
Interpretations must be further modified to account for the influence
of other compounds in black liquor. In the following sections the effects
of other components in black liquor on the conversion of methyl mercaptan
to its non-volatile product will be discussed.
B. Reactions of Methyl Mercantan and Dimethyl Disulfide in
Synthetic Black Liquor
1. Methyl Mercaptan Oxidation by Model Compounds Featuring Certain
Structures in Kraft Lignin
The reactions of methyl nercaptan are also influenced to a high
degree by lignin present in the black liquor. Earlier investigators
-------
29
(11, 12, 13) noticed that the consumption of methyl raercaptan was
accelerated by organic material dissolved in black liquor. It is
generally suggested that phenolic and catecholic structures in the
lignin contribute to the increased rate of consumption of methyl raercaptan.
Although no detailed studies have been made to describe the reactions
under conditions employed during black liquor oxidation, results
obtained by workers in the petrochemical field suggest possible
mechanisms by which polyphenolic structures in lignin may interact
with ipethyl mercaptan under oxidizing conditions.
The chemical composition of black liquor is not accurately defined;
for this reason, a discussion on the reactions of methyl mercaptan with
model compounds featuring certain structures in lignin will be given
first. Results derived from these studies will help to interprete
observations made during oxidation of mixtures of methyl mercaptan
and Kraft lignin in alkaline solutions.
As the result of deiaethoxylation reactions which occur during pulping,
Kraft lignins contain about 0.3 moles of catechol structures per one
thousand grams (14). In alkaline solution, these structures are readily
oxidized by oxygen forming the corresponding quinones and hydrogen
peroxide as shown in reaction 13. Furthermore, during oxidation of phenolic
°2
H° ' - (13)
-------
30
structures some side chain elendnation may take place with formation
of p-quinonoid structures (15) as illustrated for moieties with a
carbonyl sidechains.
HO
I
+ OO
°CH3 OH
(14)
Ifeouerian (16) and Oswald (8) studied the catalytic effects of
polyphenolic structures on the oxidation of a variety of mercaptans
and suggested two possible reaction routes:
1. Oxidation of the nercaptan, by quinonoid structures, to
the sulfide
2. Nucleophilic 1,4-addition of the mercaptide ion to the
quinone.
To test these concepts, methyl mercapten was oxidized by several
oxidants under nitrogen atmosphere and the results are shown in
Figure 7. As a reference one experiment at 10 psig oxygen was also
included. The first oxidant chosen was o-quinone, since it most
closely represents the quinonoid structure in the Kraft lignin. The
following reactions may take place:
Oxidation
CILjSfl
1,^-addition
(15)
-------
31
2.8
2.0
1.6
1.2
0.8
O hydrogen peroxide
D anthraquiijone beta
sulfonate
Figure 7 Methyl mercaptan consumption by equimolar amounts
of hydrogen peroxide, anthraquinone beta sulfonate,
o-quinone and 10 psig oxygen respectively.
-------
32
Figure 7 shows the results of the experiment where equimolar
amounts (2.5 imnole/1) of methyl nercaptan and o-quinone were mixed
together-at 10°C. Two general observations xvere made: first, methyl
mercaptan was rapidly consumed; secondly,, no dimethyl disulfide was
detected. Tentatively, the following conclusion is drawn: methyl
mercaptan reacts very rapidly x^ith o-quinone in a nitrogen atmosphere
apparently by nucleophilic 1,4-addition only.
To examine more closely the relative effects of the 1,4-addition
and direct oxidation of methyl mercaptan by quinoness anthraquinone
beta sulfonate was selected as the oxidant. Because of its condensed
ring structures 1,4-addition cannot occur. Again equimolar amounts
of methyl mercaptan and anthraquinone beta sulfonate (2.5 mmole/1)
were mixed together at 10°C (Figure 7)» The disappearance of
methyl mercaptan was markedly slower compared to the experiment with
o-quinone; but all methyl mercaptan consumed was converted to dimethyl
disulfide. Thus, methyl mercaptan is readily oxidized by anthraquinone
beta sulfonate in a nitrogen atmosphere, to dimethyl disulfide only.
Furthermore, the nucleophilic 1,4-addition of the mercaptide ion to
the quinone has a faster rate than the oxidation of methyl mercaptan
to dimethyl disulfide by the quinone. The oxidation of methyl mercaptan
by the anthraquinone derivative is shown in reaction (16).
+ CH SSCH
(16)
-------
33
When organic material present in black liquor is oxidized with
molecular oxygen a variety of peroxide intermediates and hydrogen
peroxide are formed (17) which could be responsible for the increased
rate of consumption of methyl mercaptan. To test this hypothesis, an
experiment with hydrogen peroxide under nitrogen gas was designed.
The results presented in Figure 7 show that at 10°C the rate of
methyl mercaptan disappearance is quits slow when compared to
experiments with the quinor.aSj and the methyl mercaptan consumed was
all converted to dimethyl disulfide according to reaction 17. Further
experiments with hydrogen peroxide at higher temperatures showed similar
rates when compared to oxygen at these temperatures s and a stoichiometry
as written for reaction 17.
2 CE SH -}- HO !> CH SSCH + 2H 0 (17)
From the results of these experiments the following order of
the rate of disappearance of methyl mercaptaii is proposed: o-quinone>
anthraquinone beta sulfonate > oxygen^-' hydrogen peroxide, and
nucleophilic Is4-addition » oxidation by quinone to the disulfide.
2. The Effect of Kraft Lignin on the Consumption of Methyl Mercaptan
In the. previous sections orienting studies with model compounds
featuring certain lignin structures show the importance in reactions with
methyl mercaptan. They are:
-------
34
1. Nucleophilic 1,4-addition
2. Direct oxidation to the disulfide
Since the model compounds used feature only certain structures, it is
necessary tc work with. Kraft lignin to verify the aboye observations.
Figures 8 and 9 show the results of experiments in which: Kraft
lignin was added to an alkaline solution containing methyl raercaptan
at 10°C and 10 psig oxvgan. It is shown that tha sedition of Kraft
lignin results in a significant iricrss.se' :'.n the rate of methyl mer-
capcan disappearance eecoiapanied by a simultaneous increase in the
rate of disisthyl aisul::ic3 fcrjiation,, FurthermoreB the yield of
dimethyl disulfide decreases with increasing amounts of lignin.
Apparently in';the presence of lignins methyl marcaptan undergoes a
reaction which leads tc other products in addition to dimethyl
disulfide,,
These observations may be interpreted by assuming that initially
the nucleophilic 1,4-addition takes place with the quinonoid structures
present in the lignin. When all available addition sites are occupied,
the remaining methyl mercaptan reacts with oxygen and/or quinones and
peroxides to form dimethyl disulfide„ This would explain the rapid
disappearance of methyl mercaptan, and the formation of less dimethyl
disulfide at a higher rate. The greater number of quinonoid structures
available; with greater quantities of lignin result in more methyl
mercaptan addition and less dimethyl disulfide formation, Thuss_ with lignin
and oxygen present the following reactions may occur. First, methyl mercaptan
may be oxidized to dimethyl disulfide according to reaction 18:
-------
35
2.8
QJ
H
ra
0
Temp 10°C
D 0 g/l lignin
g/l lignin
O 7 g/l lignin
10
15
Time, min.
20 ' 25
Figure 8 .Consumption of methyl mercaptan in aqueous alkaline
solution with Kraft lignin added.
-------
36
0)
H
CO
U
l.U
1.2
1.0
n O
0.8
0.6
0.2
0
0
Temp 10 C
D 0 g/1 lignin
A b g/1 lignin
O 7 g/1 lignin
X25 g/1 lignin
10 15 20
Time, min.
25
Figure 9 Formation of dimethyl disulfide in alkaline
solution with Kraft lignin added.
-------
37
2CILSH
HO
CILSSCIL +
J —'
OCH
(13)
With oxygen present and at temperatures below approximately 100°C
the catechols formed may be reoxidized and are available for further
mercaptan oxidation. Thus, small amounts of catechols present in
lignin may oxidize large quantities of mercaptans by this redox cycle.
The second alternative is the nucleophilic addition of mercaptide
anions to quinones to either mono-, di- and tri-thiopolyphenols by a
series of successive steps of oxidation and nucleophilic addition
as shown for the o-quinonoid structures in lignin as illustrated in
reaction 19:
CH_S
-SCH HO'
HO
R
CH_S
(19)
-------
38
In addition, it has already been shown that reactions 20 and 21 occur:
•an
2 CH3$H + 1/2 02 —— > CH SSCH + HO (20)
+ 2 H0 (21)
The rates of these reactions may be listed in the order 19 >18
21. Thus, in alkaline solution, methyl mercaptan should be oxidized to
the disulfide; but with Kraft lignin present, the major reaction is the
1,4-addition as shown in reaction 19, and less dimethyl disulfide is
formed.
It was mentioned earlier that during the oxidation of black liquor
side chain elimination of certain phenolic lignin structures could result
in the formation of new (para) quinonoid moieties. Presumably, this
formation increases with increasing temperatures. For instance, when
methyl mercaptan in the presence of Kraft lignin was oxidized at 10, 50
and 75°C respectively, lesser amounts of dimethyl disulfide are formed
at the higher temperatures (Figure 10). This could be due to the increase
in quinonoid structures formed by side chain elimination. Furthermore, 5 g
Kraft lignin/1 which is approximately 10% of the amount actually present in
black liquor, results in the formation of about 90% less disulfide from
5 mmole/1 methyl mercaptan at 75°C. The concentration of methyl mercaptan
and the temperature of 75°C are comparable to practical conditions of black
linuor oxidation with approximately 50 g/fc of Kraft lignin, probably all
mfithvl mercaptan is consumed by the nucleophilic 1-4 addition and very
little is oxidized to the dimethyl disulfide at between 50 and 90°C.
-------
39
ra
w
20
Figure 10 The effect of temperature on the formation of
dimethyl disulfide in the presence of Kraft
lignin.
-------
40
3. Effect of Sodium Sulfide on the Oxidation of Methyl Mercaptan
Sodium sulfide has two effects on the oxidation of methyl mercaptan
which highly influence the efficiency of reactions during black liquor
oxidation. The first is an apparent inhibition which is dependent on
temperature and sodium sulfide concentration.
For example, when 5 g/1 sodium sulfide is added to an aqueous
alkaline solution containing 2.5 mmole/1 methyl mercaptan the methyl
mercaptan consumption depends on the reaction temperature as shown in
Figure 11. Throughout this experiment the pH of the reaction mixture
remained between 12 to 13 and the oxygen pressure at 2.5 psig. The
results in Figure 11 suggest an apparent inhibition of the oxidation of
methyl mercaptan of at least 60 minutes at 60°C and approximately 15
minutes at 100°C. Analysis showed that sodium sulfide was oxidized to
sodium thiosulfate, and when the methyl mercaptan finally started to
react only very small amounts of sodium sulfide remained in solution
(approx. 0.2 g/1). Thus, it seems apparent that the presence of sodium
sulfide inhibits the oxidation of methyl mercaptan. The results of the
experiment at 150°C are different and will be discussed later in this
section.
To explain the above observations the following is suggested. The
oxidation of both methyl mercaptan and sodium sulfide seems to involve
radical mechanisms (18, 8, 19). Results of a study on the photodecom-
position of acetaldehyde catalysed bv thiols (20) provides a possible
explanation for the sulfur inhibition in systems involving radical
-------
£5 O— o O-D -O D
Figure 11 Oxidation of methyl mercaptan with 5 g/1 sodium
sulfide added.
-------
42
mechanisms. From this study, the apparent order of stability of
thivl radicals was determined: HS* > CH,S'. > C.H_S' . Thus, it
i J • 2 5
appears that the following competing reactions may occur;
II CH0S v
HS
CH_S
HS
(22)
Where R* could be any radical present, e.g. hydroxyl-»perhydroxy-
radical or the biradical oxygen. Since HS* is more stable than CH S",
route I should be preferred. If route II occurs the rate of CH_S*
formation should be slower than HS" formation. Furthermore, transfer
of an electron would form the more stable CH.S ion and KS" radical.
Thus, if hydrosulfide is abundant, the concentration of CH S* is
too small or its life time too short to undergo further reactions
which would lead to the disulfide.
However, the experiment conducted at 150°C (Figure 11) does
not show any inhibition for methyl mercaptan oxidation. An analysis
at the point where approximately 25% of methyl mercaptan had been
converted showed the presence of significant amounts of sodium sulfide.
Thus, apparently under these conditions methyl mercaptan oxidation
was not inhibited by the presence of sodium sulfide. Birrell (20)
-------
43
has reported a change in relative stability of HS" and CH,S* radicals
with temperature. For instance it was suggested that at temperatures
between 125 and 150°C the HS* and CH-S' radicals have about equal
stability. Thus, at 150°C route I and II are equally possible and
reaction (23) should not be important.
CH S' + HS~ * CH3S~ + HS' (23)
It has been shown that the presence of sodium sulfide is required
for the hvdrolysis of dimethyl disulfide, and thus for permanent
conversion of methyl mercaptan to nonvolatile methane sulfinic acid.
At the same time, at temperatures below 100°C the concentration of
sodium sulfide must be below about 0.2 g/1 to permit oxidation of
methyl mercaptan to dimethyl disulfide. Thus, efficient oxidation of
methyl mercaptan will be very difficult to achieve under practical
black liquor oxidation conditions. On the other hand, at higher
temperatures nethyl mercaptan oxidation is not retarded by sodium sulfide,
and dimethyl disulfide hydrolysis by sodium hydroxide is rapid. As a
consequence high temperature black liquor oxidation should be more efficient
for odor control.
The second effect of sodium sulfide is discussed in the following section.
A. Methyl Mercaptan Reactions During Black liquor Oxidation at
Temperatures Below 100°C
In the following experiment sodium sulfide, Kraft lignin and methyl
mercaptan were added to an alkaline solution of pH of 13 and 75°C to simulate
-------
44
black liquor and the results are shown in Figure 12. As expected
from earlier discussions methyl mercaptan rapidly disappeared. Apparently
all methyl mercaptan was consumed by rapid 1,4-addition to the quinonoid
structures in lignin since no dimethyl disulfide could be detected.
Surprisingly, methyl mercaptan was slowly regenerated following its
initial rapid consumption. This rapid consumption followed by slow
regeneration may be explained in terms of the nucleophilicity of ionic
species in black liquor and their relative concentration at different
stages of the oxidation.
Turunen (21) and Goheen (22) have shown that the nucleophilic
strength of anions is in the following order; CH S~ > HS > H0~ .
Thus, the methyl mercaptide anion should compete most effectively
for quinonoid structures and be rapidly consumed by nucleophilic 1-4
addition. In early stages of oxidation when methyl mercaptan
concentration is relatively large the weaker hydrosulfide nucleophile
competes ineffectively with mercaptide for the quinone structures.
However, when the mercaptide concentration becomes depleted (for
example after about 2 minutes in Figure 12 the ratio of hydrosulfide
to mercaptide is much greater than at zero time. Thus, although
hydrosulfide is a weaker nucleophile, as a result of its much higher
concentration a slow displacement of the mercaptide may occur according
to reaction 24.
-------
45
2.5
2.0
3"
2.5 psig 0,
Temp 75°C '
O 5 g/1 Na2S
10 g/1 Na2S
g/1 NaOH
50
Figure 12 Consumption of methyl raercaptan in simulated
black liquor.
-------
+ HS
CILS
+ CHJ
(24)
When the concentration of sodium sulfide was increased to 10 g/1
significantly more methyl mercaptan was reformed as is shown in Figure 12.
Large amounts of the weaker hydroxide nucleophile should also displace
the mercaptide ion as shown in reaction 25.
HO-i
HO
CH_S
(25)
Indeed, this was found and is shown in Figure 12. If the conditions
are appropriate, reaction (26) could possibly also take place.
HS -i
HO
+HS
(26)
Although no detailed experimental studies were made, a slow displacement
of hydrosulfide from lignin by tlie hydroxide ion during storage and
evaporation of oxidized black liquor could partially account for sulfide
'"reversion" reported by numerous Kraft mills.
-------
47
Methyl mercaptan in an aqueous alkaline solution will be oxidized
by molecular oxygen, but in these experiments the regenerated methyl
mercaptan did not oxidize. It has been shown that even small amounts
of sodium sulfide could effectively inhibit the oxidation of methyl
mercaptan. Analysis after 50 minutes (see Figure 12) revealed that
there was still sufficient sodium sulfide left to inhibit methyl
marcaptan oxidation. This suggests that the stability of methyl mer-
captan in this simulated black liquor oxidation results from inhibiting
effects similar to those earlier discussed.
In summary, methyl mercaptan oxidized under simulated practical
conditions with molecular oxygen, is rapidly consumed without
formation of dimethyl disulfide. It has been proposed that almost
all methyl mercaptan undergoes an initial rapid 1,4-addition with
the quinonoid structures present in lignin. However, methyl
mercaptan can be slowly displaced from the 1,4-addition product by
hydrosulfide or hydroxide at rates dependent on relative concentrations
of the three anions. At higher temperatures -and longer times the
hydrosulfide should be slowly displaced by the hydroxide ion which is
present in black liquor at concentrations of about 0.1 to 0.5 mole/1.
5. Methvl 'fercaptan Reactions During Black Liquor Oxidation at
Temperatures Above 100°C
Figure 13 illustrates the fate of methyl mercaptan when 2.5 mmole/1
was oxidized with 10 psig oxvgen in a solution containing 5 g/1 Kraft
-------
48
2.8
2.0
03
1.2
0.8
D
A
10 psig
Temp A Ito C
a i3o°c
Time, min.
8 10
Figure 13
Consumption of methyl me reap tan. in simulated
black liquor.
-------
49
lipnin and 5 p,/l sodium sulfide at 130 and 140°C respectively.
'fpthyl rcercmtan was rapidly consumed with no detectable amounts
of dimethyl disulfide at any time during the oxidation. Further-
more, no methyl mercaptan was regenerated. The following two events
may occur:
1. The fact that methyl niercaptan is not regenerated suggests
the absence of the 1,4-addition products.
It is well known that lignin related quinonoid structures
are rapidly oxidized to a variety of aliphatic acids at temperatures
above 100°C (23). Thus, quinonoid structures generated at these high
temperatures could react by two general routes as shown in equation 27.
R
oxidation
R
C-OH
Ring opening
on further
oxidation.
OH
(27)
Further oxidation of the quinonoid structures with or without thiomethyl
group (s) in the ring will result in opening of the ring (16). It is
suggested that the thiomethyl group cannot be displaced from these further
oxidized (muconic acid) structures.
-------
50
2. Aoparentlv, at these high temperatures sodiun sulfide does
not inhibit the oxidation of methyl mercaptan.
While studying the inhibition effect of sodium sulfide in
a pure system, the oxidation of methyl mercaptan appeared to be no longer
affected by sodium sulfide at 150°C. Since the inhibition was observed
in the presence and absence of lignin, it is suggested that results
obtained in a pure system may be used to interprete results obtained
when litmin was present. Thus, it is possible that at high temperatures
methyl mercaptan is oxidized very rapidly to dimethyl disulfide even in
presence of sodium sulfide.
The activation energy for the hydrolysis of dimethyl disulfide
by hydroxide was found to be quite high (27.3 Real/mole), indicating
a significant temperature effect. In fact, at 130 to 140°C the rate
of hydrolysis of dimethyl disulfide was found high enough to keep
the concentration of the disulfide at an undetectable level.
It can be concluded that methyl mercaptan when oxidized in a
system resembling black liquor at temperatures of 130-140°C is con-
sumed rapidly and no regeneration takes place. Thus .oxidation at
high temperatures resultsin efficient destruction of methyl mercaptan
and dimethyl disulfide.
-------
51
C. Summary of Methyl Mercaptan and Dimethyl Bisulfide Oxidation
Kinetics in Simulated Black Liquor
The reactions of methyl mercaptan during black liquor oxidation
have been described. From the available information a general scheme
outlined in Figure 14 describes the reactions for the conversion
of methyl mercaptan to its non-volatile products. The temperature at
which black liquor oxidation is carried out is the single most
important variable. Above about 70°C sodium sulfide may be oxidized
completely to sodium thiosulfate (3,5). But a below 90-100°C, a
complete and permanent conversion of methyl mercaptan to non-volatile
products does not seem possible. However, at above about 125°C the
reactions taking place are significantly different, resulting in a
complete and permanent conversion of methyl mercaptan.
When black liquor containing methyl mercaptan is oxidized at
60-90°C with oxygen the first reaction to take place is the oxidation
of the polyphenolic structures in the lignin to quinonoid structures
followed by a very rapid 1,4-addition with methyl mercaptide anion.
This results in the formation of mono, di- and tri-thiopolyphenols by a
series of successive steps of oxidation and addition. However, under
these low temperature conditions, methyl mercaptan is regenerated under
the influence of hydrosulfide and hydroxide. With an excess of sodium
sulfide practically all methyl mercaptan initially present could be
regenerated, suggesting that possibly all methyl mercaptan undergoes the
-------
52
rapid l,U-add.
CH.SH : -—
3 quinonoid
A
mono, di and
tri-thiopolyphenol
T <
[Na S] > 2.5 in mole/e
T < 125°C
Hvdrolvsis
T < 100°C
[Na2S"] > 2-5 m mole/e .f
-*• i
> 110 oxidation
quinonoid quinonoid
CH SSCH
quinonoid
» rapid
fast
CH-SOgH
Isle
a3?03H
HO", slow
displacement HS , rapid
[Na S] > 2.5 n mole/e
T <
slow Figure 14 Schematic outline of reations of methyl
CH SCLH mercaptan during black liquor oxidation.
-> CH^SH
-------
53
1,4-addition with the quinonoid and the direct oxidation to diciethyl
disulfide either by oxygen or the quinonoid is negligible. In the
presence of sodium sulfide, the oxidation of methyl mercaptan is
inhibited, until the sodium sulfide has diminished by approximately
95-98%. This is usually the condition at which black liquor oxidation
is stopped in practice. Thus, when the oxidation is terminated the
regenerated methyl mercaptan is left in the black liquor while the
remainder is in the form of mono, dl-or tri-thiopolyphenol. If
one desires total oxidation of methyl mercaptan,the oxidation
should be continued. With prolonged time methyl mercaptan oxidation
follows a continuous oxidation-hydrolysis cycle with dimethyl disulfide
Intermediate in which one cycle converts four moles of mercaptan to .
one mole methane sulfinic acid and with regeneration of three moles methyl
mercaptan. A sufficient number of cycles will lead ultimately to
complete conversion of methyl mercaptan and dimethyl disulfide to the non-
volatile methane sulfinic acid. The completion of this oxidation-
hydrolysis cycle, however, will be very difficult to obtain. Because of
the very low concentration of hydrosulfide the 'intermediate dimethyl
disulfide must be hydrolysed by hydroxyl. Unfortunately, the rate of
this reaction is very slow at these low temperature conditions. Moreover,
methyl mercaptan Is continuously, slowly displaced from its addition product
by' hydroxyl. These results suggest how substantial quantities of methyl
mercaptan and dimethyl disulfide could survive a conventional black liquor
oxidation.
-------
54
^fids grora the
The. etaisflioaa of saotbyl aateaptan
x'^,e;?-*/p.jfy area ©f a a5.ll e@uid be frd
vxijisoy.' msidaeioa ey§£QBO_@Epleyia8 gue© ossygaa do aot . sraqiil;?© yeatiag.
?so wwtbyl ffiGseapfefta aad disofehyl^ diswlf ids will see bo .'
•f,ut-! ©sdJiaago Hedtseod ' "hold up91 tima BoteecR fell® b,lae,
ri:^eGt; fioatGefe O¥a,gersf;or and raepvesy.fum'QeQ -would. la
9.tm 6! law 5?oi©aoQ ©£ sofchyl asaseaptaa f sosi. its addleia.^:
V)nBt', nyse©m 'caployiag eQnventiomal eendifeisaa ^m?ld . a^jsaar fe© bo ofereag
. ©ssidafeion ifltli diroet food tot© feha &$>•£&$& essfcaet
oovosal .
irst „ a
o It to aofe
Ufa eiae @£ fefeo
oddifelea ^feh '"fcfec
talm glaeo fefeo
a£ tMn feisa
i
-------
55
Finally, the activation energy of hydrolysis by hydroxide is
about 28 Kcal/mole while that by hydrosulfide is 16 Kcal/mole.
Therefore, above 125°C the rate of hydrolysis by hydroxide becomes
quite significant because of the high specific rate constant at
those temperatures and the constant high concentration of hydroxide.
The following sections describe laboratory studies of an in-
digester black liquor oxidation to confirm the results predicted by
the earlier sections of this report.
-------
56
D. Odor Reduction by In-Digester Oxidation of Kraft
Black Liquor with Oxygen at Temperatures above 80"C
This section reports on investigations to study the feasibility of
in-digester BLO with oxygen as the oxidant. The following specific
questions are addressed: a) At what stage of pulping should oxygen
be injected into the digester? b) What is the oxygen consumption and
utilization efficiency with respect to the oxidation of sodium sulfide,
methyl mercaptan, dimethyl sulfide, and-dimethyl disulfide? and c)
What is the influence of the oxygen injection on the pulp yield and the
physical and optical properties of the pulp?
1. The Influence of Sodium Sulfide Throughout Kraft Pulping
The mechanisms involved in the formation of methyl mercaptan and di-
methyl sulfide during a kraft cook has been well established (3). In
a first step, hydrosulfide ions react with lignin methoxyl groups to form
methyl mercaptan. The latter dissolves in the cooking liquor forming
methyl mercaptide ions. These react further with lignin methoxyl groups
to yield dimethyl sulfide. In the kraft pulping of softwood, the major
portions of methyl mercaptan and dimethyl sulfide are formed after a
temperature of 1708C has been reached (3). If the sulfide ion is partly
or fully removed by oxidation with oxygen at this stage, the formation of
methyl mercaptan and dimethyl sulfide may also be reduced or prevented.
-------
57
A preliminary series of cooks were studied to determine if reduction
or removal of sodium sulfide after 170°C has been reached will influence
the rate of delipnification and the obtainable pulp yield. This may not
necessarily be so, since Gierer and Smedinan have shown that considerable
amounts of thiol groups are already introduced in the lignin molecule at
150°C (24).
Loblolly pine (Finus taeda L.) chips were pulped in 2.8-liter
autoclaves as described in the experimental section. Autoclaves were
removed at different time intervals, and the pulping was arrested by
cooling the autoclave in cold water. Figure 15 shows the content of
effective alkali and sodium sulfide in the liquor, as a function of cooking
time and temperature, in regular kraft pulping of this wood species.
From an oxidation point of view, it is important to note that the content
of sodium sulfide decreases by about 50% throughout the pulping and
that about 80% of this decrease apparently occurs before the maximum
temperature has been reached. Since only a slight increase in the content
of sodium thiosulfate and no significant changes in the contents of sodium
sulfite and polysulfide were measured, the decrease in the content of
sodium sulfide may be largely the result of early introduction of sulfur
into the lignin in accordance with Gierer and Smedman's results.
Figure 15 illustrates the approach used to study the influence of a
reduction or removal of sodium sulfide during the latter part of the
pulping on the rate of delignification and the pulp yield. Regular kraft
cooks were interrupted as soon as the maximum temperature of 170°C was
-------
TEMPERATURE
EFFECTIVE ALKAU
175 200 225
Figure 15.
COOKING TIME, rnin
Composition of cooking liquor as a function of cooking time, and temperature in regular
kraft cooking and liquor exchange cooko of loblolly pine. Cooking conditions: effective
alkali 18.52 (KaOH), aulfidity 25% liquor to wood ration A:l £7 Regular kraft cooks.
A Liquor exchange cooks with OZ au'lfldity in second stage liquor, o Q liquor exch'angc
cooks with 50Z liquor replacement in second stage.
-------
59
reached (100 min.) and rapidly cooled to room temperature. In some
cooks, 50?! of the total volume of black liquor was replaced by sodium
hydroxide solution with a content of effective alkali equivalent to the
effective alkali content of the withdrawn black liquor. Subsequently,
the autoclaves were rapidly reheated to 170°C to complete the pulping.
The content of sodium sulfide in the second stage was approximately
50% of the sodium sulfide content of the regular cook (Figure 15). In
other cooks,the total volume of black liquor was removed from the
partlv cooked chips by washing with water. Again, a sodium hydroxide
solution was added yielding a content of effective alkali equivalent to
the effective alkali of the removed black liquor. In a third series,
regular kraft cooks were interrupted after 150 minutes, cooking time,
50% of the black liquor was replaced by sodium hydroxide solution, and
the pulping was then completed. Regular kraft cooks which had been
cooled and reheated at the same stages as the two-stage cooks served
as control.
The relationship between total pulp yield and lignin content of
regular kraft pulps and two-stage pulps is shown in Figure 16. All of
the two-stage cooks with 50% or 0% of original sodium sulfide in the
second-stage liquor have lower pulp yields compared at equal lignin
contents. The yield difference seems to be of the same magnitude for
both 50% and 0% sodium sulfide and may vary from about 0.5% (on wood.)
at intermediate yield levels to about 2% at low levels.
-------
60
-
Vx u-
V..-' •"* - •_
O W l !
""• — .*> i
-" GO —
5 AQ -
O !• V>
O
143
Q. '-
_
<40
b39
1
REGULAR KRAFT
COOKS
«7 LIQUOR EXCHANGE
•^ COOKS
O
I I I I
I I I
. .
I 2 3 4 5 6 7 8 9 i'O II !2 13
LIGNiN COiSJTENT^/os.on pulp)
Figure 16
Total pulp yield as a function of pulp lignin Content.
-------
61
Although, surprisingly, no significant differences in the rate of
delignification were found between regular kraft cooks and liquor
exchange cooks, the early sulfidation of lignin did not prevent the
loss in carbohydrate yield that was observed under two-stage cooking
conditions. Hartler, Andersson, and Bergstrom studied two stage cooks
with liquor exchange at 155°C and reduced sulfidity in the second stage
(25). These authors also found a loss in total pulp yield in such
cooks.
These results, therefore, suggest that injection of oxygen into
the digester to oxidize sodium sulfide should be carried out as late in
the cook as possible.
2. In-Digester Oxidation: Influence on Inorganic Components
Various amounts of oxygen were injected into the circulation line
of a laboratory batch digester from 2 to 30 minutes before blowing the
digester. The black liquor was analyzed for sodium sulfide, thiosulfate,
and effective alkali before and after oxygen injection. The oxidation
of sodium sulfide should result in thiosulfate formation following
the overall reaction 1 repeated here.
2 S + 202 + H20 *> S203 + 2 OH (1)
-------
62
The sodium sulfite and polysulfide contents were determined on selected
cooks, but only negligible quantities were detected. Thus, sulfur
material balances showed that disappearance of sulfide corresponded with
almost stoichiometric formation of thiosulfate as predicted by reaction
(2). Since polysulfide and free sulfur were not formed in detectable
quantities, the resulting oxidized black liquor should not be subject
to reversion with subsequent formation of sulfide anion. This is
presently being studied in more detail.
Oxygen requirements at various sodium sulfide oxidation efficiencies
for one series of cooks are shown in Figure 17. For example, 90% oxidation
of liquor containing approximately 8 g/liter sodium sulfide requires about
70 g when using an oxygen injection rate of 5.3 g/min. under the present
conditions. Proportionally, higher amounts of oxygen were required to
achieve a higher degree of sodium sulfide oxidation.
The consumption of 70 g of oxygen is equivalent to two moles of oxygen
per mole of oxidized sodium sulfide or 2.3% (calc.) based on wood or 4.5-5%
based on- pulp. The stoichiometric amount of oxygen required is one mole
per mole of sodium sulfide according to reaction (2). Therefore, other
oxygen consuming reactions must also proceed under the present conditions
of oxidation. Since sulfur balances show only sodium sulfide and thio-
sulfate as maior components in the liquor, reactions which cause oxygen
consumption in excess of requirements for sodium sulfide oxidation very
likely involve the oxidation of lignin and/or polysaccharide degradation
products dissolved in the black liquor. Grangaard has shown that in such
reactions, various carboxylic acids are formed in high, yields (26).
-------
63
LL
O
f-
•—y
*£.
O
f V
3
2
1
\
— ^ V. «^>^
\ .
— \ —
^o
^"•.^ O
X) •
AMOUNT OF OXYGEN INJECTED, g
^
Figure 17
The content of sodium sulfide in black liquor as.a function of the amount of
oxygen injected at a rate of injection of 5.3 g of O./min. Total amount of
liquor - 12 liter.
-------
The formation of acidic groups is also reflected by a decrease in
effective alkali. The consumption of effective alkali is significantly
increased in oxygen injection cooks as compared to the composite average
for several reference cooks (Figure 18). Removal of sodium sulfide
should cause an apparent loss in effective alkali since one mole of
sodium sulfide contributes 1/2 mole to effective alkali of the liquor.
However, this is compensated for by a new formation of alkali in the
oxidation reaction 2 described above. Thus, the reduction in effective
alkali shown in Figure 18 is not explained by sodium sulfide oxidation.
The present studies employed injection rates of 5-100 g/min., the
lower rates tending to require higher oxygen consumption to reach the
same sulfide oxidation efficiency. In their studies Fones and Sapp (27)
recorded oxygen requirements of about 4.5 moles of oxygen per mole of
oxidized sodium sulfide which are significantly higher than the present
work. This low oxygen utilization efficiency may have been caused by
the very low injection rates and poor mixing used in that study.
In most cases black liquor samples were withdrawn from the digester
approximately 2 minutes after the injection of oxygen. In. some separate
cooks, the pulping was continued for periods of up to 60 minutes after the
oxygen injection was completed. Analysis of the black liquor showed no
increase in the content of sodium sulfide during the continuation of the
pulping indicating that analyses represented the total amount of black
liquor present in the dipester, including the portion penetrating the. chips.
-------
65
o
z
14
13
12
10-
t < 3
O
o
6
-i
i i
i i i i i i
REGULAR COOK
k
\>60g 02
1
1
1 1
50 60 70 80 SO 100 110 120 130
COOKING TIME AT MAX. TEMR min.
Figure 18
Effective alkali In black liquor in regular kraft .cooks and cooks with
injection of various amounts of oxygen at a rate of 5.3 g of O./mln. Total
amount of liquor = 12 liter.
-------
66
In preliminary studies, the temperature in the digester increased
about 4°C with the injection of 80 g of oxygen. In most cooks this
was compensated for by reducing the supply of heating steam, and no
increase in the digester pressure was found during the injection of
oxygen. This indicates that the injected oxygen is consumed rapidly
by the black liquor at digester conditions.
3. In-Digester Oxidation: Influence on Methyl Mercaptan,
Dimethyl Sulfide, and Dimethyl Disulfide
Black liquor samples taken before and after oxygen injection were
analyzed to determine the content of the malodorous sulfur compounds,
raethvl mercaptan, dimethyl sulfide, and dimethyl disulfide. Samples
were withdrawn from the circulation system through the liquor sampling
valve.
In these cooks the amounts of methyl mercaptan and dimethyl sulfide
generated during the heating up and constant temperature period prior to
oxygen injection were consistent with other pulping studies (3). In all
cooks the digester was evacuated before liquor injection to remove air
and to facilitate impregnation of the wood chips by the liquor. Consequently,
oxygen was absent from the digester during the heating up and constant
temperature period of the cook so that methyl mercaptan oxidation was
negligible and only trace amounts of dimethyl disulfide were found prior
to oxygen injection.
-------
67
At the end of the normal cooking time, various amounts of oxygen
were injected to study the efficiency of oxidation of sulfur compounds.
The results for methyl mercaptan oxidation are shown in Table V. The
efficiency of oxidation is increased profoundly with increase of
injected oxygen. Injection of 60 g of oxygen will destroy over 99.8%
of the methyl raercaptan present at the end of a kraft cook.
After oxygen injection, pulping was continued for 45 minutes with
black liquor samples analyzed at 15-minute intervals. Figure 19 shows
that injection of 80 and 100 g of oxygen will cause almost total
permanent methyl mercaptan removal. Injection of 20, 40, and 60 g of
oxygen result in an initial drop in methyl mercaptan content of black
liquor which is then followed by renewed formation. Injection of these
lower amounts of oxygen does not totally destroy the sodium sulfide
content of the black liquor. Consequently, methyl mercaptan formation
can continue during the prolonged cooks by further demethylation of
lignin.
The oxidation of methyl mercaptan involves conversion to dimethyl
disulfide, and at conditions of conventional black liquor oxidation,
substantial amounts of dimethyl disulfide are formed as follows:
4 CK SNa + 2H 0 + 0 ^ 2 CH.SSCH + 4 NaOH (28)
J £ f. J J
At the higher pulping temperatures, only trace amounts of dimethyl
disulfide were measured following injection of oxygen. Regardless of the
-------
68
I_J I ! \ I
20 40 60 80 100 120 140
COOKING TIME AT MAX. TEMP., min.
Figure 19
Methyl mercaptan content in black liquor at various amounts, of injected oxygen.
Total amount-of liquor - 12 liter.
-------
69
Quantity of
Oxygen Injected,
g
20
40
60
80
100
Methyl Mercaptan
Concoucration at
End of Normal
Cooking, g/£
_1
2.14 x 10
_1
2.16 x 10
_1
2.24 x 10
_1
2.15 x 10
_1
2.31 x 10
Methyl Her cap tan
Concentration at
End of Oxygen
Injections g/fc
-1
1.01 x 10 -1
„!
0.85 x 10 L
—i
0.01 x 10 -1
_i
0.001 x 10 -1
trace
Final Concentration
Initial
Concentration
0.47
0.27
0.005
O.OOOi
0
Table V; Effect of quantity of oxygen injected on the methyl mercaptan
content of black liquor.
-------
70
amount of oxygen injected, no accumulation of dimethyl disulfide occurred.
Consequently, in-digester oxidation by injection of oxygen near the end
of technical kraft pulping provides a possible method for odor reduction
by destruction of both methyl mercaptan and dimethyl disulfide.
Conversely, in-digester oxidation has much less influence on the
content of dimethyl sulfide as shown in Figure 20. At low levels of
oxygen injection, the formation of dimethyl sulfide is temporarily
interrupted by the partial destruction of methyl mercaptan shown in
Figure 19. Injection of 80 and 100 g of oxygen totally oxidizes the
sodium sulfide and methyl mercaptan content of the black liquor, effectively
halting further demethylation to form malodorous sulfur species. Under
these conditions dimethyl sulfide content of black liquor decreased only
slightly, indicating a mild oxidation of this substance.
4. In-Digester Oxidation: Influence on Pulp Yield
and Pulp Properties
Table VI and Figure 16 show total pulp yields versus the lignin content
of pulps obtained in regular kraft cooks and in cooks where 80 or more grams
of oxygen were injected at the end of the cook with injection rates varying
from 2.7 to 20 g/min. Under the present conditions the injection of oxygen
does not noticeably influence the pulp yield when compared at equal lignin
contents (Figure 16). Similar results were obtained in cooks with lower
amounts of oxygen injected. Also, no significant differences were found
between regular and oxygen injection cooks with regard to screened yields
and the rate of delignification.
-------
r- 6.0
.40 gr 02
60gr 02
80gr 02
I
§
o
JJ
s
&
•a
o
u
u
o
o
in
O
*4
U
CO
-------
72
Cook
time
at max.
temp.
min.
80
80
90
90
100
130
150
60
80
90
100
100
Eff.
alkali
in white
liguor
g/fc
47.5
45.8
48.2
46.5
47.2
46.8
43.7
49.2
49.8
48.2
45.5
45.0
Na2S
inwhlte
ligvor
g/£
13,1
12.4
13.3
12.6
13.0
12.5
12.8
12.3
10.9
12.6
13.1
12.8
o2inj.
g
0
0
0
0
0
0'
0
80
80
96
80
80
Inj . rate
g/rain
—
_—
—
_—
—
—
5.3
5.3
9,6
20.0
2.7
Total
pulp
yield
%
48.7
50.3
48.3
49.1
48.6
46.0
45.6
51.1
48.6
51.4
47.9
47.4
Lignin
cont.
%
7.8
8.1
7.4
7.6
7.0
4.8
4.9
9.1
7.3
7.7
6.5
6.5
Kappa
No
-,„-„-
60
53
52
48
36
32
--
—
53
46
47
Table VI: Total pulp yields in cooks with and without the injection of
oxygen into the digester at the end of the cook. Liquor to
wood ratio 4:1, 90 minutes from 25° - 170°C.
-------
73
Burst, fold, tensile, and tear strength of 3 regular kraft pulps
and 4 pulps which had been pulped with the injection of 77-96 g of
oxygen were compared to study the influence of the oxygen injection on
the physical properties of the pulp. All pulps had a lignin content
between 6.7 and 8% (Kappa numbers from 48 to 58). In Figure 21 the
injection of oxygen has no significant effect on the tear-tensile
strength relationship"of the pulps. The data in Table VII show that
tear-burst relationship and folding endurance are compared to values
obtained from normal kraft cooking. (Altogether more than 100 cooks
were made in this study and data points are averages from replicated
cooks.
Retention of strength characteristics do not agree with the con-
clusion made by Fones and Sapp who showed that the addition of oxygen
at the end of a kraft cook significantly reduces pulp strength (27).
However, these authors based their conclusion primarily on the signifi-
cant strength losses observed in experiments in which unbleached,
regular kraft pulps, buffered in a sodium carbonate solution, were
pulped in an oxygen atmosphere. It is a moot question if such conditions
may be-considered as fully representative for the conditions existing
at the end of a Icraft cook. In the latter case sodium sulfide, lignin,
and carbohvdrate degradation products dissolved in the black liquor
appear to favorably compete with the pulp in consuming the injected oxy-
gen, .thereby preventing strength losses.
-------
74
200
o:
o
i~"
o
<
LL.
i 150
UJ
1-
100
^
— r - r • i I I I i I
~" o
~~ a QX
— \.
D ^^
X
vv a
— x
"N
_ X
Sox a°
a x^ o
O NS
_ a -REGULAR KRAFT PULP ^ o
f\ X
_. 0 -PULPS COOKED WITH 02- INJECT. ° N
—
- i 1 1 1 1 1 1 1 1
1 4 5 6 7 8 9 .10 II
1
—
—
—
—
—
.
*
—
—
i
12
TENSILE STRENGTH, 1000m
Figure 21. Tear factor versus breaking length for regular kraft pulps and pulps cooked with'
the injection of oxygen.
-------
75
In accordance with Femes and Sapp's results, the present investigations
indicated that the pulps with oxygen injection were somewhat easier to
beat to a given tensile strength (Table VII and Figure 22). Table VII
additionally indicates that, at equal tensile strength, the oxygen.
injection pulps have a higher freeness.
Also, in agreement with the latter authors, a decrease in the pulp
brightness varying from 0.5-2 units was found as a consequence of the
injection of oxygen (Table VIII). This decrease is, however, much lower
than previously reported.
5. In-Digester BLO at Variable Temperatures
While high temperature BLO should be more efficient at odor control,
it is important to know the influence of oxidation temperature upon the
oxygen consumption. Experience shows that in conventional BLO systems
(60-80°C) oxygen consumption often amounts to a 5 to 10% excess over the
stoichiometric requirements for reaction (2). At high temperatures (170°C)
about 100% excess oxygen was required as discussed in the last section.
While an increase in oxidation temperature may cause Increased oxygen
consumption by reaction with lignin, the critical temperatures have not
been established. This section describes experiments made between 80
and 150°C to determine oxygen consumption and oxidation efficiency with
respect to sodium sulfide, methyl mercaptan and dimethyl disulfide.
-------
76
E "
O
o 10
H 9
uj 8
o:
h-
CO 7
LJ
i*
^»
1 1 1 ' 1
.''"J'--^-
- /'''
. 1!
^ D- REGULAR KRAFT PULP
- / 0- PULPS COOKED WITH 02- -
^ D INJECT.
1 1 ! 1 1
0246
REVOLUTIONS IN PFI MILL, 1000
Figure 22. Tensile strength, as a function of number of revolutions in the
PFI-mill for regular kraft pulps and pulps cooked with, the
injection of oxygen.
-------
77
Total
Yield
%
48.3
48.6
49.1
47.6
47.9
48.9
51.4
Lignin
Cont.
%
7.4
7.0
7.6
7.1
6.5
8.0
7.7
02 inj.
g
0
0
0
80
80
77
96
Beating
PFI-
mill
Revolutic
0
2000
4000
6000
0
2000
4000
6000
0
2000
4000
6000
0
2000
4000
6000
0
2000
4000
6000
0
2000
4000
6000
0
2000
4000
6000
Freeness
m£
ns
f
—
274
730
630
352
240
700
631
380
252
722
642
457
223
710
628
331
—
710
665
503
345
—
650
—
220
Tensile
Strength
km
4.1
8.3
9.4
10.3
5.6
8.7
9.4
9.6
4.9
8.3
8.7
9.4
5.5
8.6
9.7
10.4
5.6
8.8
9.4
10.2
5.4
9.5
10.4
10.7
4.1
9.3
10.2
11.2
Burst
Factor
26
57
63
71
37'
65
70
70
38
63
67
72
37
60
69
74
38
65
68
73
35
62
68
73
20
62
65
75
Tear
Factor
197'
172
150
128
202
144
—
129
182
130
—
132
206
148
131
115
196
135
120
112
217
155
143
139
198
142
157
123
Fold
30
760
1050
2850
340
1280
1860
220.0
240
1240
1750
2750
600
860
1500
2600
330
1400
1900
2000
280
900
1100
1350
50
900
1020
1700
Table VII Physical properties of regular kraft pulps and pulps produced with
the injection of oxygen at the end of the cook. Liquor to wood
ratio 4:1.
-------
78
Table VIII. Effect of Injection of Oxygen on Pulp Brightness
Inj. Rate, Brightness,
0 inj., g g/min % Elrepho
0 23.4
96 9.6 22.0
77 25.9 22.5
0 21.7
80 5.3 19.8
-------
79
Black liquor was generated bv a conventional kraft cook in a
one cubic foot laboratory digester. At the end of the cook the
digester and its contained pulp and black liquor were cooled to the
desired oxidation temperature. Oxygen was injected at the outlet
side of the circulation pump, so the circulation line acted as a flow
reactor similar to the equipment employed by Cooper (28) and Galeano
(29). The system behaved similar to a back mix flow reactor since the
liquor charge was 12& and the circulation rate was about 361 per minute.
Black liquor samples were taken at about two minute intervals and
analyzed for sodium sulfide, methyl mercaptan and dimethyl disulfide.
Analysis of selected samples showed nearly quantitative conversion of
sodium sulfide to sodium thiosulfate with no detectable quantities of
polysulfide or elemental sulfur and only small amounts of sodium sulfate.
Sodium sulfide content of the black liquor was between 7 and 8 g/fc. In
this work the circulation rate was not controlled, so the Reynolds number
varied from about 150,000 at 1708C to 45,000 at 80°C and was always above
the critical values reported by Cooper C28) for efficient oxygen
absorption.
The rate of sodium sulfide consumption is shown in Figure 23 for
oxidation at 100°C and 140°C with oxygen supply rates of 10 and 8.2 g/min.
respectively. The amount of oxygen required to totally oxidize sodium
sulfide may be calculated and is shown in Figure 24 for various temperatures.
™ith 12!, of liquor at 7.8 g/fc sodium sulfide the theoretical oxygen
required is shown as 1.2 moles. Thus, at temperatures below about 140CC
-------
80
the sodium sulfide could be totally oxidized with about 20% excess
oxygen consumption by side reactions. Above that temperature large
quantities of oxygen are consumed presumably by lignin oxidation
reactions.
Figure 25 shows the efficiency of methyl mercaptan and dimethyl
disulfide oxidation at various oxidation temperatures. When oxidizing
at 130 to 140°C the methyl mercaptan was about 95% consumed by the
time sodium sulfide had been reduced to undetectable levels. No dimethyl
disulfide could be detected at any time during oxidation at these
temperatures. At 80 and 100°C less than 50% of the mercaptan was con-
sumed when sodium sulfide was at undetectable level. Dimethyl disulfide
was present throughout the oxidation period. Prolonged oxidation could
reduce the mercaptan and disulfide to lower levels, but at the expense
of greater oxygen consumption.
Oxidation at higher temperatures results in greater destruction, of
the odorous organic sulfur compounds than conventional BLO at the expense
of slightly higher oxygen consumption. It's practical application would
involve substantial changes from existing black liquor oxidation systems.
At higher temperatures a closed oxidation system would be required and
the system would probably be best used in a flow reactor configuration with
molecular oxygen. The oxidation efficiency is high for all sulfur compounds
and no "reversion" of hydrosulfide or methyl mercaptan occurs. Thus,
oxidation of black liquor before pressure releasfi from the digester system
should result in less emission of TRS substances from all sources between
the digester and the direct contact evaporator.
-------
81
8
7H- o
<_
-------
82
2.8
to
z
O
Theoretical Oxygen Required
for Solution Containing
'O.1 mole/liter of
UJ
O
Oi
I
1
1
I
I
SO 100 120 140 16O 180
TEMPEATURE, °C
Fig.24 Orygen consumption for complete oxidation of sodium
sulfide in black liquor oxidation
-------
I ' I
O 80°C
A 10O°C
O13O°C
v140°C
X 170°C
468
TIME(min
Fig. 25 Methyl mercaptan and dimethyl disulfide concentration as
a function of oxidation time for several oxidation temperatures
-------
84
E. Indigester Oxidation of Hardwood (Red Gum) Black Liquor
A hardwood species (Red Gum, Liquadatnbar styraciflua) was pulped and
oxygen was injected into the digester in the latter stages of pulping
to determine the effectiveness for odor control and the influence on
pulp yield. In these studies the rates of oxidation of sodium sulfide
were nearly identical to those discussed earlier for softwood species.
1. Influence of oxidation on methyl mercaptan, dimethyl sulfide and
dimethyl disulfide
The formation of methyl mercaptan and its rate of oxidation by
oxygen injection is shown in Figure 26. The amount of mercaptan formed
during kraft pulping of the hardwood species is somewhat greater than
softwood species (3). Injection of 80 g of oxygen was adequate to
essentially destroy all methyl mercaptan. No dimethyl disulfide was
detected in the oxidized black liquor. Essentially no dimethyl sulfide
oxidation was observed in agreement with the soft-wood results.
2. Influence of oxidation on pulp yield
Table IX and Figure 27 summarize the results from reference kraft
cooks and oxygen Injections cooks of Red Gum. The results confirm the
conclusion given earlier for softwoods that oxygen injection in the
latter stages of a cook does not change the pulp yield, kappa number
relationship which applies to normal kraft cooks.
-------
85
1.0
tO
O
•-I
x
O
o-
19
4J
81
u
0.1
0.01 _
§
u
I
20
40
60
80
100
120
140
Figure 26. Methyl Mercaptan content in Black Liquor from
Pulping Red Gum
-------
86
White
Liquor
Eff. Alk.
g/i
33.1
43.5
35.0
30.0
30.5
36.7
40.3
41.2
35.3
40.2
37.0
39.6
39.9
Na2S in
White
Liq^.
g/1
. 12.3
12.6
12.5
10.7
21
11.0
11.2
11.9
11.8
10.4
10.6
11.2
10.9
°2
Inj.
Rate
0
0
0
0
0
0
0
0
5.3
5.3
5.3
5.3
5.3
Total
°2
g
0
0
0
0
0
0
0
0
.-80
80
80
80
80
Total
Pulp
Yield %
62
42
44.2
46.0
45.2
45.7
47.6
43.3
56.5
56.2
48.0
45.8
47.4
Kappa
No.
50.7
40.2
16.9
19.0
13.3
28.4
20.9
20.9
46.7
43.0
22.0
22.4
23.7
Table IX. Total Pulp Yield for Kraft Cooks of Red Gum with and
without Oxygen Injection at the end of the Cook Liquor
to Wood Ratio 3.6:1, 90 minutes from 25°C to 170°C.
-------
65
87
60
55
50
45
40
35
30
Kraft Cooks
Kraft Cook with Oxygen
Injection
10
20
30
40
50
60
Figure 27. Total Pulp Yield as a Function of Kappa Number for
Pulping Red Gum
-------
88
F. Odor Control in Batch and Continuous Digesters
1. Odor Control in Batch Digesters
The results given in this report were obtained using 1 cu. ft. batch
circulation digesters which were operated somewhat differently than
normal industrial digesters. Black liquor was circulated throughout
the heat up and cooking cycle. This was necessary to avoid temperature
drop in the uninsulated digester during cooking. The liquor charge
was 12X. (4:1 liquor wood ratio) and liquor was pumped at lOfc/min.
. so the "turn over" time was less than two minutes. The digester and
circulation line could then be treated as a backmix reactor. By slow
injection of oxygen (total time of 2 to 40 minutes) into the circulation
line the oxygen was uniformily distributed in the black liquor.
Commercial sized batch reactors are classed as- direct or indirect steam
heated. In the first case steam enters the bottom of the closed digester
through a. special nozzle. The major amount of steam is injected during
heat up with only small amounts being required during cooking since heat
losses from the insulated digesters is small. Uniform heating of the
wood chips results from convective circulation of heated liquor.
In indirect heated digesters liquor removal is through strainers in the
digester walls, pumped through a steam-liquor heat exchanger and returned
to the digester top and bottom. Liquor circulation rates are high during
heat up, but are reduced during the cooking cycle since little heat loss
occurs from the insulated digester. Circulation rates are limited in the
later stages of cooking by accumulation of fiber fines on the strainers
with resultant pressure build-up across the strainer.
-------
89
Black liquor oxidation in the direct heated type of digester would
be accomplished by either direct injection of oxygen into the digester
shell or injection of oxygen into the blow line during the blow. Neither
digester shell would probably not result in uniform oxygen distribution.
That portion of pulp mass exposed to oxygen would likely suffer oxidative
damage and the resulting pulp would be weaker as reported by Fones and
Sapp (27). Furthermore, with poor oxygen distribution portions of the
black liquor would remain unoxidized with resultant low oxidation
efficiencies with respect to the sulfur compounds.
Injection of oxygen into the blow line would provide adequate mixing.
However, resident times of the liquor and oxygen in the blow line
would be from 5 to 10 seconds which is too short for oxidation of
the sulfur substances. This alternative was not studied in the pre-
sent work because of inadequate facilities to simulate conditions in
the blow line.
Indirect heated batch digesters offer more promise for odor control,
oxygen may be injected into the circulation line where it is normally
consumed before black liquor enters the digester. Distribution can be
controlled by black liquor circulation rate and oxygen injection rate.
The strainer system would have to be redesigned to permit high liquor
pumping rates near the end of the cook so all liquor in the digester
could be circulated at least once within the last 30 to 45 minutes of
the'cook. Odor control by this type of indigester BLO would probably
be accomplished at the conventional digester temperatures OSfl70°C).
At these temperatures, oxygen consumption would be approximately 200%
of theoretical (cf. page-75).
-------
90
2. Odor Control in Continuous Digesters
High temperature black liquor oxidation could be employed as an odor
control measure in mills using continuous digesters. Liquor extraction
lines could be modified between the extraction point and the first flash
tank for injection of oxygen. This type of arrangement would permit
more flexibility than in the case of batch digesters. Normally liquor
temperatures are in the range 120 to 120°C at which oxidation rates
are more rapid for all malodorous sulfur compounds. At the same time
excessive oxygen consumption would not occur (Figure 24).
-------
91
G. Yield Increase in Combination with. Odor Control
If the concept of indigester BLO is accepted as a feasible approach
for odor control, some modification of existing batch digesters would
be required to permit introduction of oxygen into the liquor recirculation
system. With this modification it may be possible to obtain increased
pulp yields by introduction of small amounts of oxygen into the liquor
during the heating up of the digester. Preliminary studies were made
to determine the magnitude of yield increase that could be obtained.
When oxygen is introduced into the circulating liquor during heat
up, at least two categories of reactions will take place. First the
sulfide content of the liquor may be partially converted to poly-
sulfide which can react to stabilize wood carbohydrates against alkaline
induced peeling. This should result in a yield increase.
1. Sodium sulfide in the circulating liquor will be oxidized in
several reactions. Sulfide may be converted to polysulfide
as shown in reaction (29) .
4 Na.S + 0 + 2 H 0 2 Na S + 4 NaOH (29)
Simultaneously polysulfide may be converted to sodium thio-
sulfate according to reaction 30 and sodium sulfide may
2 Na2S2 + 302 _ ^2 Na^C^ (30)
be partially oxidized to sodium sulfite according to reaction 31.
NaS -f 0 _ NaS0 (31)
-------
92
Once formed polysulfide is subject to two types of reactions.
At low temperatures polysulfide solutions are relatively
stable (in the absence of sulfite, see below discussion),
but at elevated temperatures and alkalinity decompose to
thiosulf ate as shown in reaction C32) .
S2= + OH" + 3/4 H20 ____^l/2 HS~ + 1/4 S^ (32)
In the presence of sulfite polysulfide is decomposed forming
sulfide and thiosulf ate as shown in reaction (331
S03 + S2 * S203~ + S C33)
The rates of these reactions, of course, depend on reaction
conditions, but in general at elevated temperatures and
»
alkalinity and after extended periods substantial amounts of
the generated polysulfide are lost by decomposition C30, 31).
Consequently, a system where the polysulfide is generated in the
digester where it may diffuse quickly to the reaction sites-and
stabilize reducing end groups (reaction 34) before decomposition
occurs has some attraction.
2 RCHO + Na2S2 + 4 NaOH ^
2 RCOONa + 2 Na2S + 4 H20 (34)
-------
93
2. If some oxygen Is carried unreacted with, the liquor into
the digester, a second type of carbohydrate stabilization
may occur. It is known that even at low temperature C32)_
carbohydrate reducing end groups will be oxidized to
aldonic acids.
Formation of aldonic acid end groups by oxidation of carbonyl
reducing end groups by either polysulfide or oxygen should result in
less peeling and greater pulp yields. A series of experiments was
made to test if introduction of oxygen into the liquor circulation line
would indeed result in improved yield. The results shown in -Figure 28
illustrate that a yield increase of from 1 to 3% may be obtained by
introduction into the digester. In these experiments oxygen was
injected into the circulating liquor at 35°C and a rate of 3 g/min
for 7 minutes. This amount of oxygen corresponds to about 1.25% on
pulp. The 12£ of unoxidized white liquor contained 13 g/£ of sodium
sulfide and 48 g/£ effective alkali, so the amount of oxygen introduced
corresponded to approximately a 20% consumption of the available sodium
sulfide.
Recently a U. S. patent has been issued to R. G. Barker C33) in
which these results have been confirmed. No further work, has been
completed on this project to date, but the approach, is- very encouraging
and should receive further study.
In principle this approach may also be used on continuous digesters.
The application would require some modification of digester internals to
provide for adequate control of the oxygen with white liquor Into the
impregnation zone. Of course, odor control as described in the previous
section would be an independent oxygen injection system. No studies were done.
-------
94
•a
o
o
§
at
•H
52
51
C f\
50
49
48
47
46
45
44
43
42
41
40
39
i i | i l | I I 1 1 i I 1
A J^,
/ SQ
/ ° /
/ / o
A /
/ °
: (/
/ 0,0
A/ y/O
/y^°
cro
x^
o x
y^o
/^ O Reference Kraft Cooks
- Cooks with Oxygen
Pretreatment
-
_
| i I i i | 1 1 1 I i i i i
-
-
-
-
-
„
_
-
-
_
-
—
:
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Lignin Content, % Con pulp)
Figure 28. Pulp Yield as a Function of Lignin Content for
Reference Kraft Pulps and Kraft Pulps with Oxygen
Pretreatment. 4:1 liquor to wood ratio
-------
95
H. Estimates for Cost and Effectiveness of Odor Control by Indigester
BLO Compared to other Control Methods
In this section preliminary estimates of cost and effectiveness
are given for odor control. The effectiveness of various odor control
systems are based on well run systems in non overload. The basis
for calculations draws heavily on the data provided by Hendrickson,
et. al. (1). The calculations are made based on a 500 T/day kraft
mill with total TRS emissions of 27 #/T with no odor control (see
Table I). Oxygen costs assume that other uses such as oxygen bleaching
or waste water treatment will bring the total usage to about 50 T/day.
Values for recovery furnace emissions are based upon optimum operating
parameters. The NAFCA report (1) indicates that the value of this emission
can vary between 0.16 and 2.55 Ibs. TRS/ton of pulp, with a mean value of
0.50 Ibs./ton. Because studies have shown that 0.30 Ib./ton should be
achievable with moderate controls, that figure will be used in this paper.
Values for lime kiln operation are based on the use of fresh water for
causticizing and installation of a 99% efficient particulate scrubber on
the discharge of the stack gases. Use of condensate instead of fresh
water could add an additional 0.25 Ibs. TRS/ton pulp, while use of an 80%
efficient particulate scrubber will add another 0.30 Ibs- TRS/ton pulp.
If no particulate scrubber of any kind is installed, and condensate is used
-------
96
for causticizing, the lime kiln emissions are used because it is desirable
to see how low the emissions can be reduced using practical and economical
equipment.
Calculations are made for the cases listed in Table X. Annual
operating cost and effectiveness are given with each case. A more detailed
development of the figures is given on pages 99 to 105 of this section.
The cost and effectiveness of these alternatives for odor control are
summarized in Table X. The following comments can be made at this pre-
liminary stage:
1. Assuming no yield increase the cost of indigester oxidation
is comparable to other systems at equal or better effectiveness.
2. Assuming a 1% yield increase by low temperature pretreatment
with oxygen the resulting credits make the annual cost of
indigester oxidation in batch digesters about zero.
3. The high effectiveness in control of odors by in-digester
oxidation results from elimination of all emissions except
dimethyl sulfide in the digester blow and this low volume stream
' can be treated in the lime kiln. Furthermore, liquor which has
been oxidized at high temperature in the digester does not undergo
reversion during evaporation and storage.
-------
Table X. Effectiveness and Cost of Odor Control in a 500 T/D Kraft Mill
Case
1. Indigester Oxidation
2. Indigester Oxidation
(assuming 1% yield
increase)
3. Indigester Oxidation with
lime kiln combustion of
digester relief and blow
4. Weak BLO with Oxygen
5. Case 4 with lime kiln
Combustion of digester
Blow and Relief Gases
6. Weak BLO with air
Annual Cost
$142,000-Batch Digesters
96,000-Continuous
$0 - Batch
96,000-Continuous
$172,000-Batch
110,000-Continuous
$120,000-Batch
$150,000-Batch
$ 90,000-Batch
Unit Cost
0.78
0.52
0
0.52
0.94
0.60
0.49
Remaining Emission Lbs./T
0.3 Rec.Fur. & DCE
0.3 lime kiln
0.5 Dig. Relief
1.1
Same as Above
0.3 Rec. Fur. & DCE
0.3 lime kiln
(TIT
0.3 Rec. Fur. & DCE
0.3 Washers
0.2 Multiple Effect. Evap.
3.5 Digester Blow
0.5 Digester Relief
0.3 Lime Kiln
5.1
0.3 Rec.Furn & DCE
0.3 Washers
0.2 MEE
0.3 Lime Kiln
1.1
0.3 Rec. Fur. & DCE
0.3 Washers
0.2 Multiple Effect Evap.
3.5 Digester Blow
0.5 Digester Relief
0.3 Lime Kiln
5,1
\o
-------
Table X - Continued -
Case
Annual Cost
Unit Cost
Remaining E:m.bsion Lbs./T.
7. Case 6 with lime kiln
combustion of digester
relief and blow gases.
MEE and WK BLO off gas
combustion in Rec. Furnace
8. Strong BLO with air
$136,000
0.74
$ 78,000
9. Case 8 with lime kiln
combustion of digester
relief and blow gases. MEE
and 5BLO off gas combustion
in Rec. Furn.
10. Air Contact Evaporator
with same incineration as
Case 7
$126,000-Batch
115,000-Continuous
400,000*
0.42
0.69
0.63
2.20
0.3 Rec. Furnace & DCE
0.3 Washers
0.3 Lime Kiln
0.9
0.3 Rec. Furn. & DCE
0.4 SBLO Oxidation tower
0.3 Washers
0.9 MEE
3.5 Blow gases
0.5 Relief gases
0.3 Lime Kiln
6.2
0.3 Rec. Furn.
0.3 Washers
0.3 Lime Kiln
0.9
1.0-1.5*
*Estimate
-------
99
4. In-digester oxidation should be applicable to:
a. Older mills x*ith no odor control systems.
b. Mills with black liquor oxidation systems already installed,
but operating at overload. The capital investment is very
low so incremental costs to help unload an existing BLO
unit will be low.
c. New mills which will probably have continuous digesters
can use in-digester oxidation. This may be the best application
since the liquor could be oxidized at the best conditions for'
high oxidation efficiency at the lowest possible oxygen con-
sumption. However, oxidation in this case would probably be
carried out in the liquor extraction line upstream from the
flask tank.Consequently no increase in yield could be realized.
The following section gives a more detailed discussion of the basis
for cost and emission estimates.
1. Digester Relief - .50 Ibs TRS/ton pulp - 35 cf gas/ton @ 120°F
During the cooking process, it is necessary to relieve the gases that
form in the digester. These gases are, in part, air and stream that
have been trapped in the woodchip during the pretreatment stage,
and also reduced sulfur gases that form during the cooking process.
And since the cook is carried out under pressure, it is sometimes
necessary to vary that pressure to control the cook. The relief
gases from this pressure control plus a partial pressure relief at
the end of the cook add to those relief gases mentioned above.
Because the cook is carried out at a high pH, most of the H S and
-------
100
RSH stay in solution. The primary TRS gases are, therefore,
RSR and RSSR. The gas stream is a low volume/high concentration
stream.
2. Digester Blow - 3.45 Ibs TRS/ton pulp - 300 cf gas/ton @ 120°F
At the end of the cook, the contents of the digester are blown
into a blow tank to relieve the remaining pressure and to break
up the chips. Here again, due to the high pH, the H S and RSH
stay in solution, and the major TRS compounds are RSR and RSSR.
This gas stream, like the digester relief, is a low volume/
high concentration stream.
3. Washers and Screens - .30 Ibs TRS/ton pulp - 95,000 cf gas/ton @ 120°F
The next step in the process is the washing of the pulp to remove the
cooking liquor. The typical washer is a vacuum droplet type, which
pulls air from the atmosphere as well as TRS gases in the process.
Here, however, the major TRS compound is RSH which leaves the
solution as the pH is lowered by the diluting effect of the water.
The emission stream from this source, however, is a high volume/low
concentration stream. It is, therefore, difficult to treat.
4. Multiple Effect Evaporators- .90 Ibs/ton pulp - 35 cf/ton @ 120°C
Emissions from operations following the washer stage are all concerned
with the recovery of the chemicals. The first step in the recovery of
the cooking chemicals is the evaporation of water from the diluted
wash liquors to a 50% solids concentration. The process is primarily
a boiling process, and in the process TRS gases are stripped from
solution. Most of the TRS gases are H S and RSH, since the RSR and
RSSR have been removed in the preceding operations. Some RSR does form,
however, probably due to a decomposition of RSH. As with the blow and
-------
101
relief gases, this is a low volume/high concentration stream.
5. Direct Contact Evaporator/Recovery Furnace- 12.5 Ibs TRS/ton
pulp - 470,000 cf @ 325°F.
The direct contact evaporator concentrates the liquor from.50%
solids to about 65% solids by direct contact of the hot flue
gases of the recovery furnace with the liquor from the multiple
effect evaporation stage. This operation has long been recognized
as the major source of reduced sulfur; of the total of 17.5 Ibs
TRS/ton pulp, better than 12 Ibs is emitted from the recovery
furnace and direct contact evaporator. Theoretically, if the
recovery furnace itself is operated at less than 130% of rated
capacity, the amount of reduced sulfur leaving the top of the
furnace should be insignificant. The major source of the reduced
sulfur emissions, therefore, is the direct contact evaporator.
It is here that the CO generated in combustion goes into
solution, lowering the pH and causing the evolution of I^S and RSH.
Some mass transferring also occurs due to a concentration graduant.
The generation of TRS gases can, however, be almost completely
•
eliminated by oxidizing the black liquor prior to entry into the
direct contact evaporator. The NAPCA report estimates that the
total evaporator emissions from a furnace operating at proper levels
with 99.9+% oxidation of the black liquor, should be about .5 Ibs/ton
pulp. Two independent tests run by a recovery furnace supplier and
by a pulp mill indicate that this figure can be as low as .3 Ibs/ton.
-------
102
Obviously, this emission source is an extremely high volume stream.
Treatment by any means other than, say, catalytic oxidation would
seem impractical.
6. Lime Kiln with 99% efficient scrubber - .30 Ibs. TRS/ton pulp
The source of this sulfur gas is the sulfur normally present in the
oil burned in the lime kiln, and reduced sulfur in the lime mud.
As stated in the summary, the reduced sulfur that would be added
from use of evaporator condensate in the causticizing operation can
be eliminated by using fresh water. This stream of gases is, again,
primarily H S and RSH, and it is a high volume/low concentration
stream.
Costs of Various Systems for 500 T/D Pulp Mill
1. Indigester Oxidation
A. Oxygen Costs -
1. The Batch digester present study as well as independent
tests run by paper companies indicate that the oxygen consumption would be
about 5% based on pulp production. A 500 T/D pulp plant would use, therefore,
500 T/D x 5% = 25 T/D 0 . At a minimum oxygen cost of $15/ton, annual
oxygen costs would be 25 tons x $15 x 365 days = 137,000/yr.
day ton year
2. Continuous Digester
If oxidation is done in the extraction line at 130°C, oxygen
consumption will be about 2.5% on pulp. 2.5 x 500 =12.5 T/day. At
$20/T (for the reduced capacity) annual oxygen costs will be 12.5 T x
day
$20 x 365 day = $91,000/yr.
T year
-------
103
B. Oxygen system costs
The cost of equipment will vary with the total number of
digesters in the plant. For the sake of discussion it will be assumed
that a total of about 10 digesters (continuous + batch) are in use.
If it is also assumed that a total of 3000 ft. of oxygen piping @ $2/ft.,
and individual controls and associated wiring of $1000/digester are
required, equipment costs could easily run as high as $2 (3000) + 10
($1000) = $16,000. Using the figures generated in the NAPCA report,
i.e. annual costs of about 35% based on capital costs, annual costs for
equipment would be 35% x $16,000) = $5000/yr. Total annual costs, therefore,
for indigester oxidation would be about $137,000/yr. + $5000/yr. = $142,000/yr.
2. Indigester Oxidation Assuming a 1% yield Increase Resulting from
Oxygen Pretreatment
i.
A. Assuming a 1% yield increase, at a value of $100/T the credits from
additional pulp would be 5 T x $100 x 365 days = $182,500.
T year
Against this must be charged the additional cost of oxygen for pre-
treatment amounting to about 1.25% on pulp.
1.25 x 500 T_ X $15 x 365 day = $34,200/year.
day T year
Thus, a total profit of about $142,000 + 34,200 - 182,500 = $6,300 may be
realized.
3. Strong Black Liquor Oxidation with Air
From the NAPCA report, the new annual costs for operation of a strong
black liquor oxidation system in a 500 T/D pulp plant would be about
$78,000/year. This figure is in agreement with the Owens Illinois figures
for breakeven costs between weak black liquor oxidation with oxygen and
-------
104
and strong black liquor oxidation with air. Although 0-1 stated
breakeven oxygen costs of 14/ton - $18/ton, a careful review indicated
that if the questionable improvement in tall oil recovery and evaporator
life are neglected, the breakeven oxygen cost would be about $10/ton.
At 4% oxygen utilization based on pulp weight, this gives an annual
breakeven cost for the oxygen system of $10/ton 0 x 4% x 500 TID x
365 days/year = $73,000/year. This is also, then the cost of a comparable
SBLO system and compares with the NAPCA figure of $78,000/year.
4. Weak Black Liquor Oxidation . with Oxygen
A. Oxygen Costs - As stated above, the oxygen consumption for WBLO
with oxygen was confirmed in the 0-1 test to be 4%. At $15/ton 0 , the
annual oxygen costs would be $15 x 20 ton 0 x 365 days = $110,000/yr.
B. System Costs - 0-1 estimated system costs at $30,000. Again
using the NAPCA report's average annual cost of 35% of capital cost,
the annual system costs would be 35% x $30,000 = $10,500/year.
Total cost for the WBLO system using oxygen would be $110,000/yr. +
$10,500/yr. = $120,500/yr.
5. Lime Kiln Incineration of Emissions from Digester Blow, Digester Relief,
and MEE.
The NAPCA study shows that the total annual cost for the incineration
of batch digester relief gases, blow gases, and MEE noncondensible from a
500 T/D plant is about $30,000/year. The cost for incineration of continuous
-------
105
digester TRS emissions is only $14,000/yr. the difference being the flexible
surge chamber required in batch operation to "even out" the gas flow to
the lime kiln. Since many companies have both continuous and batch
digesters, and would, therefore, require a surge chamber, the $30,000/yr.
figure will be used.
6. Incineration of SBLO Oxidation Gases in Power Boiler.
A. Incineration Cost
The gases from the oxidation tower are emitted at a temperature of
150°F. It is assumed that the BTU's provided by the incineration of the
TRS compounds is negligible, and that the energy required to incinerate
this gas stream is merely the BTU's required to rasie the temperature from
150°F to 350°F. This is certainly not absolutely accurate, since the
SBLO gases are about 12% oxygen and would provide some oxygen for combustion.
For the purposes of this calculation however, it is assumed that this
might be offset by a reduction in boiler efficiency. A rough annual cost
for incinerating the oxidation tower gases is, therefore,
.25 BTU x 900 Ibs gas x 500 tons pulp x 365 days x (350°F - 150°F) x
lb°F ton pulp day year
$.65
1,000,000 BTU = $5400/year
B. Piping Costs
The piping would be about 17" diameter piping. Assuming a length of
1000 feet, and a cost/ft of $30.00, the total capital cost would be
$30,000. At 35% annual charges based on capital costs, the annual cost
would be $10,500/year.
Total annual cost, therefore, for the incineration of the SBLO oxidation
gases in the lime kiln would be $5400/year + $10,500/year = $16,000/year.
-------
106
I. Experimental
1. Equipment and Materials
A. Laboratory Reactor
All reactions were carried out in a 1000 cc T316 stainless
steel, high pressure reactor. (Parr Instrument Company, 211 Fifty Third
Street, Moline, 111., Cat. No. 4521). The reactor was equipped with a
pressure gauge, gas inlet valve, liquid sampling valve and gas tight
stirring shaft. Modifications were made to enable rapid injection of
reactants by means of a syringe. The liquid sampling valve was fitted
with a syringe needle so rapid Capproximately 3 seconds) sampling with
exclusion of oxygen was possible. To maintain constant temperature
the reactor was placed in a constant temperature bath which could be
controlled within limits of + 0.1°C. By changing pulleys on the stirring
shaft between 200 and 1000 revolutions per minute could be obtained.
Gas Chromatograph
In laboratory kinetic studies discussed under sections A and B
analysis of methyl mercaptan and dimethyl disulfide was carried out
using a Perkin Elmer Model 990 gas chromatograph equipped with a flame
ionization detector. Analytical data were recorded using a Servo/riter
II Potentiometric recorder from Texas Instruments Incorp., Houston,
Texas 77066.
For quantitative work, the column packing was prepared by depositing
20% by weight of Carbowax 20M on acid washed chromosorb 80-100 mesh.
-------
107
This material was packed into stainless steel tubing 10 feet long
and 1/4 inch OD. Operating conditions were 100°C and 75cc per minute
nitrogen carrier gas.
A. In pulping studies the methyl mercaptan, dimethyl sulfide and
dimethyl disulfide content of black liquor samples was determined as
follows: The samples for gas chromatographic analyses were collected
using the technique described by Andersson (35). Carbon tetrachloride
extracts were analyzed with a Perkin Elmer 990 gas chromatograph equipped
with a flame photometer detector CMeloy Laboratories, Inc. Model 100 AT)
which provided specific response for compounds containing sulfur. The
28 ft by 1/8 in. o.d. teflon column was packed with 60-80 mesh Haloport
F coated with 10% diisodecylphthalate. The oven was operated at 100°C
with a 75 ml/min nitrogen carrier. Helium and oxygen flow to the detector
were 75 and 16 ml/min, respectively. Retention times of hydrogen sulfide,
methyl mercaptan, dimethyl sulfide, and dimethyl disulfide were 0.6, 1.2,
2.0, and 8.5 min, respectively. The sulfur specificity of this detector
avoided interference by other substances extracted from black liquor by
the carbon tetrachloride.
Calibration was done by iterative addition of measured amounts of
the sulfur compounds to sample vials' containing carbon tetrachloride and
the buffered black liquor sample. The carbon tetrachloride phase was
analyzed following each addition.
B. Materials
Methyl mercaptan was obtained from Air Products and Chemical Inc.,
dimethyl disulfide and methane sulfonic acid from Sigma Chemical Company.
-------
108
Methane sulfinic acid was synthesized according to the procedure used
by Claesson (36). Sodium hydroxide, sodium sulfide and organic solvents
used were reagent grade purchased from Fisher Scientific and were used
without further purification. The 1,4-dioxane was distilled over sodium
borohydride immediately before use. Kraft lignin (Indulin ATR RXL 3340;4)
was obtained from Westvaco.
2. Procedures
A. Preparation of Solutions
Freshly distilled water was stored under nitrogen and oxygen was
excluded during preparation of solutions to prevent undesired oxidation
of sulfur compounds in the reaction mixture.
Alkaline solutions of sodium methyl mercaptide were prepared daily
by absorbing the appropriate weight of methyl mercaptan gas into a IN
sodium hydroxide solution.
B. Sampling from Laboratory Reactor
The sampling valve of the 1A. laboratory reactor was equipped with a
hypodermic needle. By placing this needle through the rubber septum of
the sampling bottle and opening the sampling valve, very rapid sampling
was possible from the pressurized reactor.
For sampling, a 35 ml vial was prepared by filling it with 10 ml
carbon tetrachloride and 10 ml 0.75 N sulfiiric acid. The vials were
closed by an airtight rubber septum and aluminum cap and part of the
air was removed by vacuum. The purpose of removing the air from the
vial was to facilitate the accommodation of an approximately 5 gram
sample taken from the reactor. Accurate determination of samples
taken was obtained by measurement of weight differences.
-------
109
Sulfuric acid was added to quench the base catalysed reactions
by neutralizing the base.
C. Liquor Exchange Pulping
Freshly cut loblolly pine wood was chipped in an Appleton laboratory
chipper, air dried, and screened. The fraction passing a 1 1/4 inch mesh
screen and retained by a 1/2 inch screen was used for pulping experiments.
The cooks were carried out in 2.8 liter ..stainless steel autoclaves.
Autoclaves and chips (400 g o.d.) were evacuated for 30 min. at about
23 mm Hg to facilitate liquor impregnation (18% effective alkali, 25%
sulfidity at 4:1 liquor to wood ratio). Temperature was raised from
25°C to the maximum cooking temperature of 170°C in 100 min by a hot air
heating unit providing rotating motion to the autoclaves.
D. Oxygen Injection Pulping
Commercially produced chips of loblolly pine were used for the oxygen
injection cooks. Charges of 3000 g (calc. o.d.) were vacuum impregnated
and cooked in a 28-liter circulation digester equipped for liquid and gas
sampling. Oxygen (prepurified grade) was injected at 130 psi at the outlet
of the circulation pump. Reynolds numbers were estimated at 100,000 (34)
providing adequate oxygen mixing. The oxygen flow was regulated with a
needle valve and measured quantitatively by a calibrated flowneter (Flowrator
Meter, Model 10A356A, manufactured by Fisher and Porter, Warminster, Pa.)-
-------
110
E. Pulping Testing
A PFI-mill was used for pulp beating. Physical properties, lignin
contents, and kappa numbers of the pulps were determined according to
TAPPI standard methods. The brightness was measured according to SCAN-
Cll:62 using the Elrepho Reflectance Photometer.
F. Black Liquor Analysis
The contents of sodium sulfide, thiosulfate, polysulfide, and sulfide
in white and black liquors were determined by potentiometric titration
using the methods developed by Danielsen, Johnsen, and Landmark C37).
-------
Ill
Literature Cited
1. Hendrickson, E. R., Roberson, J. E. , Koogler, J. B. "Control of
Atmospheric Emissions in the Wood Pulping Industry", U. S.
Department of Health, Education and Welfare, March 1970.
2. Kleppe, P. J. Tappi .53 (1) , 35 (1970).
3. Sarkanen, K. V., Hrutfiord, B. F., Johansen, L. N., Gardner, H.S.,
Tappi, 53_ (5), 766 (1970).
4. Loy, H. L., Himmelblan, D. II., J. Phys. Chera., 65_, 264 (1961).
5. Christie, R. D., P. & P. Mag. Can., _73_ (10), 73 (1972).
6. Martin, J. Pulp and Paper, 43_ (6), 125 (1969).
7. Shih, T. C., Hrutfiord, B. F., Sarkanen, K. V., Johansen, L. N.,
Tappi, 50 (12), 634 (1967).
8. Oswald, A. A., Wallace, T. J., The Chemistry of Organic Sulfur
Compounds, Vol. 2, 2nd ed., Ed. N. Kharash, Pergamon Press,
London (1967).
9. Bentvelzen, J. M. , Doctoral Dissertation, Department of Wood and
Paper Science, North Carolina State University, Raleigh, N.C. 1973.
10. Barringer, C. It., Ind. Eng. Chem., 47, 1022 (1955).
11. Bilberg, E., Landmark, P., Norsk, Skogind, 5_, 221 (1961).
12. Lindberg, J. J., Nordstrom, C. G., Paperi ja Puu, 41^ (2), (1955).
13. Wright, R. H., Tappi, 35_ (6), (1952).
14. Falkehag, I., personal communication, 1973.
15. Musso, H., Maassen, D., Liebigs Ann. Chem., 689, 93 (1965).
16. Mequerian, G. H., J. Am. Chem. Soc., 77, 5019 (1955).
17. Schubert, M., J. Am. Chem. Soc., 6>9, 712 (1947).
18. Schmidt, U., Trans. Far. Soc., 62^, 379 (1966).
19. Cullis, C. F., Roselaar, L. C., Trans. Far. Soc., 55_, 1562 (1959).
20. Birrell, R. N., Smith, R. F., Trotman-Dickenson, A. F., Vilkie, H. J.,
Chem. Soc., 2807 (1957).
-------
112
21. Turunen, J. , Dissertation, Helsinki Univ., Soc. Scient. Fenn. Comment
Phys. Mathem., 2£, 9 (1963).
22. Goheen, D. W., Forest Prod. J. 1.2_ (1963).
23. Lay, K., Anpew, Chem., 70 (2), 79 (1958).
24. Gierer, J. F., Smedman, L. A. Acta. Chem. Scand., 19, 1103 (1965).
25. Hartler, N. , Andersson, K. , Bergstrom, J. G. J., Kemisk. Tidsskr, 2^,
?2 (1969).
26. Grangaard, D. H., 144th ACS meeting, April 4, 1963.
27. Fones, ?,. E. , Sapp, J. E. , Tappi 43_ (4), 369 (1960).
28. Cooper, H. B. H., Rossano, A. T., Taper presented at the Tappi
Alkaline Pulping Conference, Memphis, 1972.
29. Galeano, S. F. , Arasden, C. D., Paper presented at 73rd National
AIChE Meeting, Minneapolis, 1972.
30. Teder, A., SvenskPapp., 72/8), 245 (1969).
31. Bilberg, E., Landmark, P., Norsk, Skog., 5_, 221 (1961).
32. Bamford, C. H., Collins, J. R., Proc. Royal Chem. Soc. (London),
Ser. A., 62 (1951).
33. Barker, R. G., U. S. P. 3,723,242, March 1973.
34. Davis, D. S., The Paper Ind., Feb. 1097 (1955).
35. Andersson, K., Bergstrom, J. G. T., Svensk. Papperstid., T^dl) (1969),
36. Claesson von, P. J., Prakt. Chem. 2^ (15) 193 (1877).
37. Danielson, A. J., Johnsen, K., Landmark, P. A., Norsk, Skog., 23
C3), 77 (1969)^-2.3 (12), 378 (1969).
-------
BI3LIOGRAPHIC DATA ]- Report No. 2.
SHEET EPA-650/2-74-005
,4. Titra,ar.d auotitie
Iindigester Black Liquor Oxidation for Odor Control
in Kraft Pulping
7. Aut'uor(s)
W. T. McKean, Jr. and J.S. Gratzl
9. Performing Organization Name and Address
Department of Wood and Paper Science
North Carolina-State University
Raleigh, NC 27607
1 2^Sponsoring Organization Name and Address
EPA1, -Office of Research and. -Development
NERC-RTP , • 'Control Systems . Laboratory
Research Triangle Park, NC,. 27711
3.N^ecipient's Accession No. (
~3. Report Date
J anuary 1974
6.
&• Performing Organization Rept.
No.
10. Puject/Task/Work Unit No.
ROAP 21ADC
11. Contract/Grant No.
Grant AP-01269-02
13. Type of Repor&-&-Period
Covered
Final
14.
15. Supplementary Notes
16. Abstracts Tjje report.gives results of laboratory studies describing the major reaction
routes,. key inhibition reactions, and kinetics of methyl mercaptan (CH3SH) and di-
methyl disulfide (CH3SSCH3) during oxidation of black liquor. The studies help explain.
low oxidation efficiencies with respect to hydrogen sulfide, CH3SH, and CH3SSCH3
during black liquor-oxidation at 60-90°C and suggest that high-temperature oxidation
should be more efficient. Softwood and hardwood black liquor was oxidized at 80-170°C.
Oxidation at above 100-120 °C resulted in efficient oxidation of all three malodorous
.compounds with no liquor reversion during subsequent storage and distillation. Oxygen
consumption was about 125% of theoretical below 140°C, but increased to about 200% at
170°C.. Application of this approach to batch and continuous digester systems is .disc-
ussed. Preliminary work shows that injecting small amounts of oxygen into the liquor
circulation line during the early stage of pulping could increase pulp yield by 1-3%,
depending on the final pulp kappa num-
17. Key Words and Document Analysis. 17o. Descriptors
Air Pollution
Sulfate- Pulping
Black-Liquors
Odor .'Control
Oxidation
Kinetics
Cost-Estimates
17b. Ucr.:ii:e:s/Open-£nded Terns
Air Pollution- Control
Stationary Sources'
Ihdigester Black Liquor Oxidation
Oxygen Injection
jReduced Sulfur Compounds
ber. If this yield increase could be
obtained in combination, with.indigester
black liquor oxidation, the neb.costs for
odor control in mills using batch diges-
ters would be very attractive-.--
e
iroup
07A.'07D. 13B
! 13. Availability Statement
i
:
1
»-
Unlimited
19. Security Class (This -
Report)
UNCLASSIFIED
20. Security Class (This
Page
UNCLASSIFIED
21. No. of Pages
22. Price
l NTI5-35 (PEV. 3-721
THIS FORM MAY BE REPRODUCED
USCOMM-OC 14B52-P72
------- |