United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-86/013 July 1986
Project Summary
Environmental Assessment of an
Enhanced  Oil  Recovery Steam
Generator Equipped  with  an
EPA  Heavy  Oil  Low-N0x  Burner
C. Castaldini, L R. Waterland, and R. De Rosier
  The report discusses emission results
obtained from sampling the flue gas from
a 16-MW (55 million Btu/hr) enhanced oil
recovery steam generator equipped with
an EPA heavy oil low-NOx burner firing
high-nitrogen  (about  1  percent) Kern
County crude oil. The tests consisted of
a comprehensive flue gas monitoring pro-
gram to characterize emissions from the
steamer with the burner operation set to
achieve NOX levels below 85 ppm (at 3
percent O2) with low CO and smoke emis-
sions. In addition, an extended 30-day flue
gas monitoring program was conducted to
measure low-NOx burner performance
under typical unattended steamer opera-
tion.  Emission measurements for the
comprehensive tests included:

  • Continuous monitoring of flue gas
    emissions.
  • Source assessment sampling system
    (SASS) testing with subsequent lab-
    oratory analysis of samples to give
    total flue gas organics in two boiling
    point ranges, specific quantitation of
    the semivolatile organic priority pollu-
    tant species, and flue gas concentra-
    tions of 73 trace elements.
  • Volatile organic sampling train (VOST)
    testing with subsequent laboratory
    analysis to give volatile organic prio-
    rity pollutant species emissions.
  • EPA methods 5 and 8 sampling for
    paniculate and SOX emissions.
  • Controlled condensation system (CCS)
    sampling, also for SOX emissions.
  • Emitted particle size distribution mea-
    surements using Andersen impactors.
  • N2O emission sampling.
  During the comprehensive tests, NOX
emissions averaged 70 ppm at 3 percent
O2 with  CO  levels generally below 30
ppm. SO2 emissions measured with  a
continuous monitor ranged between 500
and 750 ppm at 3 percent O2. Measure-
ments of SO2 by EPA Method 8 (and
subsequently by controlled condensation)
indicated generally good agreement with
the continuous emission analyzer. The
average of two paniculate emission mea-
surements was 27ng/J (96mg/dscm). The
average condensible paniculate emission
level was 14 ng/J (50 mg/dscm). Two par-
ticle size distribution measurements were
performed. Results indicate that 90 per-
cent of the paniculate matter had a mean
particle diameter less than 1.4 and 11 /urn,
respectively, for the two tests.
  Total organic emissions from the stea-
mer were relatively low, at 300 jig/dscm
(85 pg/J), and relatively evenly distributed
between the semivolatile  (boiling point
about 100 to 300 °C) and nonvolatile (boil-
ing point greater than about  300 °C)
categories. Of the volatile organic priority
pollutants, emissions of benzene, toluene,
and ethylbenzene were quantitated in the
2 to 60 fjg/dscm (0.4 to 20 ppb) range. Of
the semivolatile organic priority pollutants,
emissions of  naphthalene and phenol at
about 1 Mg/dscm (0.3 ppb) were measured.
  Extended 30-day continuous monitoring
of flue gas emissions confirmed the bur-

-------
ner's ability to maintain NOX emissions
below 80 ppm. The average for this dura-
tion was calculated at about 70 ppm at 3
percent O2. CO emissions were also low,
generally less than 30 ppm at 3 percent
O2, throughout the test program.
  This Project Summary was developed
by EPA's Air and Energy Engineering Re-
search Laboratory, Research Triangle Park,
NC,  to  announce key findings  of the
research project that is fully documented
in two separate volumes of the same
title (see Project Report ordering infor-
mation at back).

Introduction
  Remaining petroleum reserves in older
fields often require enhanced  oil recovery
(EOR) techniques for continued produc-
tion. One EOR technique involves steam
flooding of oil fields to lower the viscosity
of heavy crude so it can be recovered. The
steam for injection is most often raised by
crude-oil-fired steam generators, termed
steamers. In Kern County, California, the
aggregate NOX  emissions from  these
steamers have received recent regulatory
attention since that area  is currently in
borderline attainment of the NO2 ambient
air quality standard.
  Over the  past several years, EPA has
sponsored the development of a low-NOx
                 burner capable of achieving  NOX levels
                 below 85 ppm with low CO and smoke
                 generation while burning high nitrogen fuel
                 oils such as the Kern County crude. These
                 programs have led to the recent full-scale
                 demonstration of a burner prototype in an
                 operating steamer. This steamer, equipped
                 with a full-scale 16-MW (55 million Btu/hr)
                 burner prototype, was tested under the
                 Combustion Modification Environmental
                 Assessment (CMEA) program. Results of
                 these tests are presented in this report.

                 Summary and Conclusions
                 Source Description
                   The tests were performed on a crude-
                 oil-fired steamer retrofitted with the low-
                 NOx burner illustrated in Figure 1. The
                 burner consists of a large, 13 m3 (460
                 ft3), refractory-lined chamber. Steam-ato-
                 mized oil and combustion air are well mix-
                 ed in this chamber at a stoichiometric ratio
                 of about 0.6 to 0.65. Primary air to the
                 chamber is preheated by a regenerative
                 chamber design in  which the air passes
                 through  an  inner shell before being in-
                 jected into the chamber. Secondary com-
                 bustion air is  injected both  radially and
                 axially in the interface connecting the
                 burner to the steamer. This staged com-
                 bustion  Iow-N0x  burner enhances the
decay of total fixed nitrogen species (NH3
and HCN primarily), formed early in the
mixing zone  from fuel  nitrogen, to  N2
prior to injection of secondary combustion
air. The large refractory-lined combustor
volume provides both the long residence
time (about 0.6 sec) and the high tem-
perature (about 1,400 °C) required for the
decay of the  fixed nitrogen species.

Test  Program and Burner
Operation
  The  test program called for  a com-
prehensive evaluation of steamer flue gas
emissions with the burner operation set to
achieve NOX  levels  below  85 ppm with
low CO and smoke emissions. In addition,
the test program called for a continuous
30-day emission monitoring program to
monitor  the longer-term  NOX  control
capability of the burner under  typical
steamer operation.
  Table 1 summarizes the steamer/burner
operating conditions during the compre-
hensive tests on February 1, 1984. The
burner fired Kern County crude with a ni-
trogen and sulfur content of 1.04 and 1.06
percent, respectively. Measurement of pri-
mary and secondary airflows indicated a
first-stage combustor stoichiometry in the
range of 0.61 to 0.65, with an estimated
   Ceramic Fiber
     (external
    insulation)
    Tile Support
      Ring
       Primary
       Register
                   Secondary
                    Air Inlet
          /              i— Inner,

^•*-«- ^-v--*-^ I •*.-».-*•». •*- -v -v-^-v -v^ -yl -v-v -v-v -w -v T
                    WV/WY/\//\//\/A
                                           25 4 mm Expansion Joints
                                                          L
                                                                                                           \    30%
                                                                                                             ^— Alumim
                                                                                                                Plastic
              Insulating
               Firebrick
                                                                                    90% Alumina
                                                                                   Brick (Greenal 90)
Figure 1.   EPA low-NO* burner

-------
bulk gas residence time of 0.65 sec. Stea-
mer operation was uninterrupted during
this test.

Emission Measurements and
Results—Comprehensive Tests
  The sampling and analysis procedures
used in this test conformed to an extended
EPA Level 1 protocol. Emission measure-
ments  included:
  •  Continuous monitoring for 02, C02,
     CO, NOX, and SO2.
  •  SASS sampling for trace element and
     semi- and non-volatile organic
     emissions.
  •  VOST for volatile organic emissions.
  •  EPA Method 5 sampling for solid and
     condensible participate emissions.
  •  EPA  Method  8 sampling for SOX
     emissions in conjunction with Met-
     hod 5.
  •  Particle  size  distribution measure-
     ments using Andersen impactors.
  •  CCS for SOX emissions.
  •  Grab sampling for laboratory gas chro-
     matographic analysis of N20.
All  sampling took place at the stack,
downstream of the steamer economizer.
The analysis protocol for the SASS train
samples included:
  • Analyzing methylene chloride extracts
     of particulate and XAD-2 resin for
     total organic content in two boiling
     point ranges:  semivolatile organics
     with boiling points  between  100 and
Table 1.    Steamer/Burner Operation
    300 °C (nominally C7 to C16 organ-
    ics) by total chromatographable or-
    ganic (TCO) analysis and nonvolatile
    organics with boiling points greater
    than 300 °C (nominally C16+ organ-
    ics) by gravimetry.
  • Obtaining infrared (IR) spectra of the
    gravimetric residues of the extract
    samples.
  • Analyzing the organic sorbent module
    extract samples for the semivolatile
    organic priority pollutants, including
    several polynuclear aromatic hydro-
    carbon (PAH) species, by gas chro-
    matography/mass spectrometry (GC/-
    MS) according to EPA Method 625.
  • Analyzing particulate, XAD-2 resin,
    and impinger solutions for 73 trace
    elements using primarily spark source
    mass spectrometry (SSMS) and ato-
    mic absorption spectroscopy.
VOST train samples were analyzed by
thermal desorption, purge and trap GC/MS
for the volatile organic priority pollutant
compounds as outlined in the EPA VOST
protocol.
  Table 2 summarizes the emissions mea-
sured during  the comprehensive tests.
Emissions of NOX,  CO,  and  S02 were
relatively constant throughout these tests.
NOX emissions averaged about 70 ppm
dry corrected to 3 percent 02, well below
the burner target of 85 ppm. CO emissions
averaged 24 ppm. SO2 emissions, as de-
termined by both the continuous monitor
                                           Range
                         Average3
Fuel flow, l/min (bbl/day)
Heat input, MW (JO6 Btu/hr)
Fuel temperature, °C (°F)
Feedwater flow, l/min (103 bbl/day)
Steam pressure, MPa tpsig)
Steam temperature, °C (°F)
Stack temperature, °C f°FJ
Burner primary airflow, m3/s fscfm)
Secondary airflow, m3/s (scfm)
First stage stoichiometry
First stage residence time,b sec
Fuel/Oil analysis, percent weight
Carbon
Hydrogen
Sulfur0
Nitrogen0
Ash
Moisture
Oxygen
Gross heating value, MJ/kg IBtu/lb)
Specific gravity (API) at 16°C <60°F>
23.4
16.4
129
380
7.03
279
232
2.71
2.38
0.61




1.00
1.00





to
to
to
to
to
to
to
to
to
to




to
to





"Spans the entire 1-day period of comprehensive
24.1
16.8
132
400
9.10
>300
254
2.87
2.95
0.65




1.10
1.09





testing.
(212
(55.8
(264
to
218)
to 57.4)
to
(3.4 to
(1,020
(525
(450
to
to
(5, 740
(5,040

—

_
—


—
_
—
—
—














269)
3.6)
to 1,320)
>570
490)
to 6,080)
to 6,260)













23.8
16.6
130
390
8.41
—
242
2.80
2.65
0.64
0.65

86.2
11.3
1.06
1.04
0.03
0.57
0.4
42.78
12.6

(215)
(56.61
(267)
(3.5)
(1,220)
—
(468)
(5,940)
(5,630)










(18,430)


bAssumes a combustor temperature of 1,430°C (2,600°F).
c Analyses of sulfur and nitrogen were
performed
in triplicate.
and the average of two Method 8 deter-
minations, are shown in the table: the two
agree within 10 percent. Average S03
emissions by Method 8 were about 9 ppm
(3 percent 02), corresponding to about
1.4 percent of total SOX (SO2  + S03)
emissions: this is a significantly lower frac-
tion than is typical for oil-fired sources.
Subsequent measurements  using CCS
gave S02 emissions of 600 ppm (average
of two  measurements) with S03  emis-
sions of 49 ppm: in these subsequent test,
S03 accounted for about 7.5 percent of
the total SOX emitted, more typical for oil-
fired sources.
  Table 2 shows that particulate emissions
were 146 mg/dscm (average of two mea-
surements) with about 35 percent of the
particulate being condensible. Total organ-
ic emissions from the steamer were quite
low, at 300pig/dscm. These were roughly
equally  split among the  semi- and non-
volatile  boiling point ranges.
  Of the volatile organic priority  pollu-
tants, benzene and alkyl substituted ben-
zenes were detected  at concentrations
below about  20 ppb. No other volatile
organic  priority pollutants were detected.
Of the semivolatile organic priority pol-
lutants,  naphthalene and phenol were de-
tected at concentrations of 0.3  and  0.2
ppb, respectively. The dimethyl phthalate
noted  in  Table  2  is  most  likely a
contaminant.
  Trace element analyses of SASS sam-
ples indicate that sodium, chlorine (con-
densed  chlorides), copper, iron, nickel,  and
zinc are emitted at the highest levels with
emissions exceeding  100 ^g/dscm.  The
most likely source of the sodium and chlo-
rine is the  residual brine in the crude oil
fuel. Sodium levels in the fuel were quite
high, about 120 /^g/g.  Figure  2 illustrates
the particle  size  distributions resulting
from  two  separate Andersen impactor
measurements. The data indicate that 90
percent of the total particulate mass  had
a mean particle diameter of less than  1.4
pirn for run 1  and  11 ^irn for run 2.

Emission Measurements and
Results—30-Day Monitoring
  The sampling and analysis protocol for
this portion of the test program consisted
of continuous monitoring of flue gas for
02, C02, CO, NOX, and SO2 with certifi-
cation of  NOX analyzer readings  using
EPA Method 7. Flue gas was monitored
from January 21 to February 24, 1984.
Figures 3 through 5 illustrate the emission
data collected. Apart  from two periods
when the steamer was shut down, one for
inspection and another due to electrical

-------
Table 2.    Summary of Flue Gas Emissions
           Component
                                                 Flue gas concentration3
ppmf3
mg/dscm
Total organics:
  Semivolatile organics (C7 to C,6i
    by TOO
  Nonvolatile organics C16+l by
    gravimetry
Total C7+

Volatile organic priority pollutants:
  Benzene
  Toluene
  Ethylbenzene

Semivolatile organic priority pollutants:
  Naphthalene
  Phenol
  Dimethyl phthalate
   0.018
   0.00083
   0.00039
   0.0003
   0.0002
   0.0004
                   0.17

                   O.13

                   0.30
    0.060
    0.0032
    0.0017
    0.0014
    0.0007
    0.0036
ng/J°
Criteria pollutant and other gas-phase
species:
NOX (as NO2)
CO
SO2
— Continuous monitor
- Method 3d
SO3
- Method 3d
Paniculate
- Solid"
— Condensibled
Total particulated
N20


69
24

556
606

8.7

—
—
—
7


140
29

1,560
1,640

29

96
50
146
12


39
8.1

435
456

8.1

27
14
41
3.5
                 0.050

                 0.035

                 0.085
  0.017
  0.0009
  0.0005
  0.0004
  0.0002
  0.0010
aAverage flue gas O2 and C02 were 2.7 and 13.1 percent, respectively.
bCorrected to 3 percent O2.
cHeat input basis.
dAverage of two separate measurements.
problems, operation throughout this test
period was relatively steady. 02 levels,
shown in Figure 3, were  maintained  at
about 3 percent. Only at the start and end
of the 30-day test period  did the 02 in-
crease to over 4 percent. It is not known
if this increase was caused by a change
in primary  or secondary combustion air-
flow. NOX emission levels, shown in Fig-
ure 4, were below 80 ppm  corrected to 3
percent O2 throughout. Average NOX for
the entire test period was about 70 ppm.
CO emissions were generally below 30
ppm. However, there were no CO readings
after day 23. SO2  emissions ranged be-
tween 500 and 750 ppm corrected to 3
percent 02. The notable increase in SO2
during the  last days of testing may have
been caused by an  increase in sulfur con-
tent of the crude.

Quality Assurance
   Results  of  several  quality assurance
(QA) activities performed  in conjunction
with this test program are discussed in the
report. Conclusions are that the quality of
data obtained is of an acceptable level in
terms of the stated project QA objectives.

Summary
  Emissions from a crude-oil-fired steamer
equipped with  the EPA Iow-N0x burner
were tested. Field testing of the steamer
over a 30-day monitoring period supports
the capability of the burner to maintain
NOX levels below 80 ppm with CO emis-
sions below 30 ppm. Paniculate emissions
were in the range typical of  other  EOR
steamers. Total organic emissions were
relatively low. Low concentrations of sev-
eral aromatics (benzene, toluene, ethylben-
zene,  phenol,  and   naphthalene)  were
quantitated.

-------

Q)

ID
Q
    10°-
    10'
                                                                                          n  Run 1
                                                                                          A  Run 2
        01
                   .05  1  .2   .5   1   2     5    10    20   30 40 50  60  70  80    90   95    98  99 99.5   99.9   99.99

                                          Cumulative Wt Percent Less Than Diameter of Particle (DP)

Figure 2.    Particle size distribution.

-------
   o
   o
   f)
   3
   to


   Uj
       20-,
       16
       
-------
   800-,
   700-
                                                                       O
   600-
I
<4
§
I
 «
o
to
   400-
   300
       21       26        31        5        10        15       20

           January 1984                            February 1984

Figure 5.    Flue gas SOs for the extended test period.
                                                                          25
   C.  Castaldini, L. Water/and, and R. De Rosier are with Acurex Corp., Mountain
     View, CA 94039.
   Joseph A. McSorley is the EPA Project Officer (see below).
   The complete report consists of two volumes,  entitled "Environmental Assess-
     ment of an Enhanced Oil Recovery Steam Generator Equipped with an EPA
     Heavy Oil Low-NO* Burner,"
     "Volume/. Technical Results," (Order No. PB 86-191  475/AS; Cost: $16.95)
     "Volume II. Data Supplement," (Order No.  PB 86-191 483/AS; Cost: $16.95)
   The above reports will be available only from: (cost subject to change)
           National Technical Information Service
           5285 Port Royal Road
           Springfield.  VA 22161
           Telephone: 703-487-4650
   The EPA Project Officer can be contacted at:
           Air and Energy Engineering Research  Laboratory
           U.S. Environmental Protection Agency
           Research Triangle Park, NC 27711
                                                                       *U.S.Government Printing Office: 1986 — 646-116/40619

                                                                               7

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S7-86/013

               0000329    PS

               U  S  ENVIR  PROTECTION  AGENCY
               REGION 5  LIBRARY
               230  &   DEARBORN  STREET
               CHICAGO                 IL    60604

-------