United States
                                 Environmental Protection
                                 Agency
                                 Industrial Environmental Research
                                 Laboratory
                                 Cincinnati OH 45268
v°xEPA
                                 Research and Development
                                 EPA-600/S2-80-186  Dec 1980
Project Summary
                                 Evaluation  and  Application  of
                                 SO,  Measurement  Procedures
                                 for  Kraft  Recovery  Furnaces

                                 A. K. Jam, R. 0. Blosser, D. B. Newport, and H. S. Oglesby
                                  A research program was initiated to
                                 determine sulfuric acid (SOa/hbSOi)
                                 and sulfur  dioxide (SOz) emissions
                                 from kraft recovery furnaces using an
                                 extractive   sampling  system.  The
                                 Goksoyr and Ross controlled conden-
                                 sation  technique,  which  uses  a
                                 modified  Grahm condenser,  was
                                 chosen  because of  its  reported
                                 accuracy. Equipment was designed
                                 and fabricated to evaluate the effect of
                                 coil length, frit porosity, temperature,
                                 flow rate, and concentration on the
                                 efficiency of H 2864 capture.
                                  Laboratory tests were conducted to
                                 determine the optimum design and
                                 operating  conditions  to  minimize
                                 SO3/H2SO4 losses in filter holders
                                 used  in  the sampling train  for
                                 particulate separation. The effects of
                                 recovery furnace particulate on
                                 SO3/H2SO4 losses were investigated.
                                 The alkaline  particulate from the
                                 recovery furnace reacted with a part of
                                 SO3/H2SO4 in the sample. Although
                                 the losses could not be correlated with
                                 the alkalinity of the particulate, they
                                 were not considered significant.
                                  Field studies were conducted on five
                                 representative kraft recovery furnaces
                                 to  determine  SOa/hhSO and S02
                                 concentrations  in flue  gases. The
                                 SO3/H2SO4  concentrations  varied
                                 from 0 to 2.98 ppm, and the range of
                                 SO2 concentrations varied from 1 4 to
                                 416 ppm.  A  comparison  of  these
                                 emissions  with the  SO3/H2SO4
                                 and SO2 emissions from oil- and coal-
                                 fired utility boilers shows the kraft
                                 recovery furnace emissions  to be
                                 much lower.
                                 Introduction
                                   Because of their adverse effect on
                                 respiratory  functions,  acid  sulfate
                                 aerosols in the ambient air are of special
                                 concern in  air quality management
                                 programs One of the  precursors  of
                                 ambient acid sulfate aerosols is sulfur
                                 trioxide (SOa), which is formed during
                                 the  combustion of sulfur containing
                                 fossil fuels
                                   In the kraft pulping industry, the con-
                                 centrated spent cooking liquor (black
                                 liquor), which consists of wood compon-
                                 ents, inorganic chemicals, and sulfur
                                 compounds, is sprayed into a  furnace
                                 and combusted  Although the  kraft
                                 recovery furnace burns black liquor m a
                                 manner to minimize sulfur dioxide (SOa)
                                 generation,  the  presence of  high
                                 concentrations of sulfur in the black
                                 liquor  has raised the question of the
                                 possible presence of sulfur  trioxide
                                 (SO3)/sulfunc  acid (H2S04)  in  kraft
                                 recovery furnace flue gas
                                   The objective of this investigation was
                                 to determine representative SO2 and
                                 SO3/H2SO
-------
investigation  was divided into four
areas'

 1.  Development  and  laboratory
    testing of  a  suitable method to
    measure  SOs/HaSCU concentra-
    tion   m  a  particulate-free  gas
    stream,

2.  Determination of the effect of the
    presence of kraft recovery furnace
    paniculate on the measurement of
    S03/H2S04,

3   Selection and testing of a suitable
    paniculate  separation device to
    minimize  paniculate  interaction
    with SO3/H2S04 in the sampling
    tram if paniculate is found to inter-
    fere  in SO3/H2SO4 measurement,
    and

4   Measurement  of  S02  and
    SO3/H2SC>4 levels in kraft recovery
    furnaces from five  representative
    sources in the southeast

Conclusions
  The kraft recovery furnace SOX study
results can be  divided  into two parts
laboratory investigation and field mea-
surements of SOx emissions
Laboratory Investigation
  A  study  of  the Goksoyr and Ross
method for S03/H2SO4 measurement
showed that the efficiency of SO3/H2SO4
capture in  the modified Grahm con-
denser  was dependent upon  the  frit
porosity and coil length (Table 1) A con-
denser with Type C (Ace Glassjfrit and a
200-cm coil maintained at 75°  to 85°C
(1 67-1 95°F) was the most suitable for
use in the sampling train because of its
ability to provide high  H2SC>4  capture
efficiency and acceptable sampling rate
in the range of 6-8 L/m
  Tests to find a suitablefilter holderfor
removing  paniculate  from  the gas
stream  prior to SOa/I^SCU  measure-
ment showed that the conventional EPA
Method-5 type of filter holders could not
withstand the  desired  temperature of
260°C(500°F) A quartz filter assembly,
developed for another EPA project, was
fabricated and  tested in the laboratory
The study showed no significant loss of
H2SC>4 in the  filter support  assembly
maintained above 260°C (500°F)
  The effect of kraft recovery-furnace
paniculate   upon  the  passage   of
S03/H2S04  through the filter holder
Table 1.    Effect of Frit Porosity Upon SOa Condenser Efficiency.
Max. Pore Length of
Frit D/'a. Flange Condensing Coil
No. Type* (fjM) (cm)
1 B 70-100 100
2 D 10-20 200
3 C 25-50 200
Flow Rate % H2SO4
(Lpm) Capture
2 20-40
2 99
2-8 90-99
*Ace Glass
Table 2.   H2S04 Losses in Tests With Recovery Paniculate
H2SOs Cone , H2S04 Loss,
ppm
200
230
288
323
8.8
11 1
222
194
177
ppm
3.4
60
4.3
0.3
1 4
4 1
24
2 7
25
Percent
17
26
15
1
16
37
1 1
14
14
H2SO4 Loss*
Possible
ppm
1 4
1 6
1 4
1 0
3 1
32
0 7
06
0.5
Percent
7
7
5
3
35
29
3
3
3
Difference
ppm
2.0
4.4
2.9
-07
1 7
09
1 7
2.1
2.0
Percent
10
19
10
- 2
-19
8
8
11
11
 t all alkalinity present was neutralized
was determined by continuously load-
ing  the  quartz  filter with  recovery-
furnace paniculate while passing a gas
stream containing H2SO4 through the
filter The results (Table 2) show sub-
stantially higher losses of H2SC>4 than
were expected based upon the alkalinity
of the paniculate and may be due to the
adsorption of H2SC>4 on the paniculate
surface. The absolute values of H2SO4
losses were, however, low and, for pur-
poses  of  estimating  emission  levels
from kraft recovery furnaces,  of little
significance
  The alkaline paniculate present in the
kraft recovery furnace stack gas offered
a potential for SOs/H2SO4 loss from the
sample   To  minimize   the  contact
between the gas and the collected parti-
culate and to reduce  the potential for
SO3/H2SC>4 loss, tests were conducted
with an  electrostatic  precipitator
obtained from the  EPA Environmental
Sciences  Research Laboratory  Tests
showed that  the precipitator  removed
99 percent of the paniculate present in
a stack gas following a precipitator, and
there was no conversion of SO2 to SOs
in the precipitator


Field Measurement of SO«
Emissions
  The  SO, emissions from five kraft
recovery  furnaces were  measured
Included were two  furnaces with direct
contact  evaporators (DCE) and  three
without any contact evaporators One of
the non-contact furnaces was a cross-
recovery furnace in which 30 percent of
the liquor was from the sodium-based.
semichemical cooking process       U
  The sampling tram is shown in Figure^
1  Table 3 summarizes test results by

-------
furnace type and sample conditioning
techniques The data for all  the tests
show  a  SC>3/H2SC>4   concentration
range of 0 to 2 98 ppm, with an average
value of 0 81  ppm.  When  separated
according to the  type of furnaces, the
data show that DCE furnaces have an
average  SOa/HjSO*  concentration of
0 24  ppm  in  stack  gases;  non-DCE
furnaces have a slightly higher average
concentration  (1 07  ppm)   Non-DCE
furnaces had an average concentration
of 356 ppm of flue gas SC>2, substanti-
ally higher than the average concentra-
tion in furnaces with  DCE (80 ppm)
  The data in Table 3 indicate that non-
DCE  furnaces  have  an  average
S03/H2SCu level of 0.90 ppm when a
quartz filter was  used in the sampling
tram,  a  lower   average  than  the
SC>3/H2S04 concentration of  1 62 ppm
obtained with a miniature electrostatic
precipitator (ESP) in the sampling tram
for paniculate separation These differ-
ences,   too   small   to  be   of  any
importance, may be due to source vari-
ability,  oxidation of SO2 to S03 in the
ESP during sampling, or lower losses of
SO3/H2SO4 in the ESP.
                                         The lower SOzconcentrations in DCE
                                       furnaces were anticipated and are gen-
                                       erally due to SO2 scrubbing in the DCE
                                         To determine  any  linear correlation
                                       between the SO3/H2S04 and the SO2
                                       concentrations, the values of coefficient
Table 3.
                      Emission Data Summary
SOz/HzSOt Cone, ppm
Samples
AH Samples
No.
26
Max.
298
Mm
0
Avg
0.81
SO2 Cone.,
ppm
260
Types of Furnaces
-DCE Furnaces              9    1  10      0    0 24        80
-non-DCE Furnaces        17    298   017    107       356

Non-DCE Furnaces
-Quartz Filter Only         13    263   0.17    090       430
-ESP Only                 4    298   0.91    162       305
                                           Insulated
                                          'Box
Heated
Probe
w/Quartz
Tube
                                                               High Temperature
                                                            r~ Heating Elements
                                                Thermocouple
                                                           Greenburg
                                                           Smith
                                                           Impmgers
                                                                                                      Silica Gel
                                                                                                      Drying Tube
                                                        Thermometer
                                                     \-Water
                                               3% Hydrogen
                                               Peroxide
                                                                                                      Needle  Valve
                                                                                                  Pump
                                                                Rate Meter
Figure 1.    SO, sampling tram

-------
   of correlation (r) between
   and  862  were  determined for each
   recovery furnace and are tabulated in
   Table  4  These  results,  except  for
   recovery furnace C, show a very poor
   correlation between  SO3/H2SC>4 and
   862. If a high level of SOz is noted in a
   particular furnace's stack gas, it should
   not be assumed that the same furnace
   also emits  a  high  concentration  of
   S03/H2S04.
   Table 4.     Coefficient of Correlation
               SO3/H2SO4 and S02.
Recovery
Furnace
A
B
C
D
E
Coefficient of
Correlation
-0.33
-0.50
099
-042
-046
     To  understand the  relative signifi-
   cance of the SOx levels in kraft recovery
   furnace stack gases, these levels must
   be compared to  emissions from oil- or
   coal-fired  utility or  industrial boilers.
   Some of the recently published SOa and
   h^SCU emission data from oil- and coal-
   fired boilers show that both theSOaand
   the H2S04 concentrations in flue gases
   from oil- and coal-fired boilers are much
   higher than the  levels m kraft recovery
   furnace stack gases
           A. K. Jam, R. O. Blosser, D. B. Newport, and H. S. Oglesby are with the National
              Council of the Paper Industry for Air and Stream Improvement, Inc., New York,
              NY 10016.
           Michael D.  Strutz is the EPA Project Officer (see below).
           The complete report, entitled "Evaluation and Application of SO* Measurement
              Procedures for Kraft Recovery  Furnaces," (Order  No.  PB 81-109092;
              Cost: $7.00, subject to change} will be available from:
                    National Technical Information Service
                    5285 Port Royal Road
                    Springfield, VA 22161
                    Telephone: 703-487-4650
           The EPA Project Officer can be contacted at:
                    Industrial Environmental Research Laboratory
                    U.S. Environmental Protection Agency
                    Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency       I  U-SJ
EPA 335
Official Business
Penalty for Private Use $300
    L
                       J

-------