United States
Environmental Protection
Agency
Office of
Underground Storage Tanks
Washington, D.C. 20460
June 1989
     Estimating Air Emissions from
     Petroleum LIST Cleanups
           Printed on Recycled Paper

-------

-------
ESTIMATING AIR EMISSIONS FROM
    PETROLEUM  UST CLEANUPS
     U.S. Environmental Protection Agency
      Office of Underground Storage Tanks

                June  1989

-------

-------
                           TABLE OF CONTENTS
 Section
Page
TABLE OF CONTENTS	i

LIST OF FIGURES	ii

1.0 INTRODUCTION	1-1

    1.1 Background	1-1
    1.2 Purposes of This Manual	1-2
    1.3 Approach and Organization	1-3
    1.4 Limitations	1-4

2.0 SOIL EXCAVATION	2-1

    2.1 Overview	2-1
    2.2 Factors that Affect Air Emission Rates from
        Excavated Soil	2-2
    2.3 Air Emission Estimation Procedures	2-4
    2.4 Variation of Emission Rates With Time	2-9

3.0 VACUUM EXTRACTION	3-1

    3.1 Overview	3-1
    3.2 Factors that Affect Air Emissions from
        Vacuum Extraction Systems	3-2
    3.3 Air Emission Estimation Procedures	3-3
    3.4 Variation of Emission Rates With Time	3-7

4.0 AIR STRIPPING	4-1

    4.1 Overview	4-1
    4.2 Factors that Affect Air Emissions from
        Air Strippers	4-2
    4.3 Air Emission Estimation Procedures	4-3
    4.4 Variation of Emission Rates With Time	4-10

5.0 REFERENCES	5-1

APPENDIX A: Physiochemical Properties of COM Synthetic
            Gasoline Blend	A-l

APPENDIX B: Air Emission Estimation Equations	B-l

APPENDIX C: Soil Gas Concentration Measurements	C-l

APPENDIX D: Units Conversion Table	D-l

-------
                            LIST OF FIGURES
Figure
 2-1    Estimated Soil Excavation Air Emission Rates for
        Gasoline VOCs	2-7
 2-2    Estimated Soil Excavation Air Emission Rates for
        Benzene	
                                                               .2-8
 2-3    Change in Emission Rate with Time for an Excavated
        Soil Pile	2-10

 2-4    Change in Emission Rates with Time for Disturbed
        Soil Piles	2-11

 3-1    Estimated Vacuum Extraction Air Emission Rates
        for Gasoline VOCs	3.5

 3-2    Estimated Vacuum Extraction Air Emission Rates
        for Benzene	„	4	3_g

 3-3    Change in Emission Rates  with Time for Vacuum
        Extraction Systems	....	3-8

 3-4    Change in Emission Rates  with Time for Pulsed
        Vacuum Extraction Systems	3_9

 4-1    Estimated Air Stripper Emission Rates  for
        Gasoline  VOCs (Removal Efficiency:  99.99%)	4-4

 4-2    Estimated Air Stripper Emission Rates  for
        Benzene (Removal  Efficiency:  99.99%)	„	4-5

 4-3    Estimated Air Stripper Emission Rates  for
        Gasoline VOCs (Removal Efficiency: 95.00%)	,	4-6

 4-4     Estimated Air Stripper Emission Rates  for
        Benzene (Removal Efficiency: 95.00%)	4-7

 4-5     Estimated Air Stripper Emission Rates  for
       Gasoline VOCs (Removal Efficiency: 85.00%)	4-8

4-6    Estimated Air Stripper Emission Rates for
       Benzene (Removal Efficiency: 85.00%)	4-9

4-7    Change of Emission Rates with Time for Air
       Strippers.	4-11
                                 11

-------
                                 FOREWORD
In our work to clean up contamination from leaking underground storage
tanks (USTs), it is easy to become caught up in the immediate danger to
our ground water supplies.  While this hazard to ground water is real and
must be addressed, it cannot be addressed with complete disregard to other
parts of the environment.  This manual is designed to address the issue of
air emissions that result from petroleum UST cleanups.  It will educate
both those who are conducting UST corrective actions, and those who are
regulating air emissions at the State and local levels.  This manual will
not answer the sometimes difficult policy questions that arise concerning
cross-media transfer of pollutants, but it will provide the means to make
more informed and responsible decisions.  We hope that this manual will
serve as one tool to employ in our common mission at the Environmental
Protection Agency to protect human health and the environment.

We would like to thank Rebecca Zarba and Bernadette Kolb of Camp Dresser &
McKee Inc. for their excellent work on this document, and the many
individuals from the Federal, State and local Air and UST offices who
contributed their time and thoughts to make this into a better, more
useful document for everyone.

-------

-------
                             1.0  INTRODUCTION
1.1  BACKGROUND

A variety of corrective strategies may be used to remediate sites where
underground storage tanks (USTs) have released gasoline into the ground.
Frequently, gasoline must be removed from either the soil or ground water,
and sometimes both.  The corrective strategies that remove contaminants
from soils include vacuum extraction, soil washing, and soil excavation.
Corrective strategies that remove contaminants from water include air
stripping and granular activated carbon.   Some of these technologies
result in the transfer of contaminants from the soil and ground water into
the air.  For example, air stripping towers, which are commonly used to
remove volatile organic compounds (VOCs) from water, transfer a high
percentage of the contaminants  from the liquid phase to the vapor phase,
where they may be released to the atmosphere.  Likewise, two widely used
technologies of soil remediation, soil excavation and vacuum extraction,
involve bringing soil and/or vapors to the surface of the ground where
vapors may move freely into the atmosphere.

The release of contaminants into the atmosphere may be undesirable for two
reasons.  First, exposure to VOCs could pose public health risks if they
are released in sufficient quantities for sufficiently long periods of time
(Fancy, 1987).  Benzene, which  commonly makes up to 2 to 4 percent of
gasoline, is of particular concern because it is a known carcinogen  (COM,
1988b).  Second, VOCs are ozone precursors, thus making them even more
undesirable in regions that are unable to meet current ozone regulations.
For these reasons,  some State and local air offices have adopted standards
that  restrict the emissions of  total VOCs and benzene at UST cleanup sites.

The air emission estimation procedures presented in this manual were
developed by examining relevant literature, and through discussions with
industry and regulatory contacts.  Gasoline is a mixture of many
hydrocarbons, the exact blend of which can vary widely due in part to
                                     1-1

-------
 differences in the original  crude  oil,  and differences  in  the  refining
 process (COM,  1987b).  A specific  blend of gasoline had to be  assumed in
 order  to perform some  of the emission  rate calculations in this manual.
 This blend was used in other projects  performed  for the Environmental
 Protection Agency's (EPA) Office of Underground  Storage Tanks  (OUST), such
 as  the LOCI Research Report  (PEI,  1988)  and the  vapor phase modeling report
 (COM,  1988b),  and is included in Appendix A.  Characteristics  of other
 fuels  such as  diesel,  jet fuel, and home heating oil, and  their overall
 effect on emission rates, are also discussed in  Appendix A,,

Whenever possible,  data  from existing UST cleanup sites were used to verify
 that the procedures presented in this manual provide realistic estimates of
 emission rates.   The data used was provided by regulatory  and  industry
 sources,  and from published  literature.  In general, comparisons between
 the estimated  and actual emission  rates  indicate that the  procedures are
adequate in providing  order  of magnitude estimates of emission rates.
1.2  PURPOSES OF THIS MfiNUAL

The purposes of this manual are:

    (1) To provide State UST offices with a means of estimating air
        emission rates of VOCs and benzene at individual UST cleanup sites
    (2) To provide general information on air emission rates that may be
        used to develop policies for air emissions at UST cleanup sites in
        a given State or locality

This manual discusses the type of information that is required and the
procedure that should be followed in order to obtain an air emission rate
estimate at a particular site.  The type of information needed includes the
cleanup technology being used, and some site specific conditions such as
soil type and contaminant concentrations.
                                    1-2

-------
1.3  APPROACH AND OBC»NIZATION

The intended users of this manual are State and local UST officials who
must comply with air emission regulations at cleanup sites.  A simplified
approach for estimating emission rates at UST cleanup sites is presented
herein.  The cleanup technologies considered in this manual are soil
excavation, vacuum extraction, and air stripping.  Although passive venting
is sometimes used to vent gasoline vapors to the atmosphere, industry
officials indicate that the expected emission rates would be negligible due
to low air flow rates emanating from the vents (Johnson, 1988).  Soil
washing is not used as frequently as excavation or vacuum extraction to
remove gasoline from soils because it tends to be a more complicated and
expensive technology.  Despite the fact that its use is increasing,
incineration also tends to be a more costly and less widely used cleanup
technology.  Incinerators frequently have emission control devices in
place, and hence air emissions are generally not an issue.  For these
reasons, passive venting, soil washing, and incineration will not be
discussed in this report.  The contaminants considered are:  (1) total VOCs
from gasoline and (2) benzene.

The  format of this manual has been arranged so that each corrective action
technology is presented separately.  Soil excavation, vacuum extraction and
air  stripping will be covered in Sections 2.0, 3.0, and 4.0  respectively.
Each section  is organized as follows:

     o A brief overview of the technology and expected emission  rates
     o A discussion of factors that affect air  emission rates  and how to
         reduce air emission  rates
     o Air emission estimation procedures and graphs
     o  The manner  in which  the emission rates are expected to  vary with
         time
A review of the  equations  used  to  calculate  the  emission  rates  is included.
 in Appendix B.
                                     1-3

-------
1.4  LIMITATIONS
Limitations of this manual include the following:


    (1) The emission rates calculated using the procedures jprovided in this
        manual are "ballpark" estimates.  These estimates should be
        adequate for determining whether air emission controls or permits
        may be required at an UST site.

    (2) The information pertaining to the change in emission rates with
        time is general in nature, as the actual change with time will be
        highly dependent upon many site specific conditions.

    (3) This manual does not address the issues of emissions dispersion or
        the health risks that may be associated with emissions at UST
        cleanup sites.
                                   1-4

-------
                            2.0 SOIL EXCAVATION
2.1  OVERVIEW

Soil excavation is a widely applied.corrective action technology.  It
involves the transfer of contaminated soil from the subsurface to the
surface of the ground, where it is commonly dumped in a pile to be treated
and/or disposed of at a later date.

Gasoline that has leaked from an UST frequently becomes trapped in the soil
pore spaces as it moves through the unsaturated zone.  For this reason,
excavated soils frequently contain some liquid gasoline, the amount of
which depends upon the soil type and soil moisture conditions.  When liquid
gasoline and air are present within the soil pores, the concentration of
gasoline vapors in the pore spaces will reach an equilibrium concentration
based upon the vapor pressures of the constitutive chemicals that make up
the gasoline.  The equilibrium concentration also represents the maximum
concentration of the gasoline vapors that can exist in the pore spaces.

Before volatilization begins, the air in the pore spaces of the entire soil
pile will have essentially the same equilibrium vapor concentration.  The
vapor concentration in the air surrounding the soil pile, however, will be
negligible.  This change in vapor concentration across the soil pile
surface results in vapor molecules within the pile moving into the
atmosphere in an effort to attain equal vapor concentrations within the
pile and in the surrounding air.  Thus, volatilization begins when vapor
molecules move from a region having a high vapor concentration to a region
having a low vapor concentration.  As the gasoline vapors escape from the
surface of the pile, more liquid gasoline will volatilize to the vapor
phase in an effort to maintain the equilibrium concentration described
above.

The air emission estimation procedure presented in this section assumes
that the extent of contamination in the soil pile is such that both liquid
                                    2-1

-------
gasoline and air are present in the soil pores.  The equilibrium
concentration of the gasoline vapors can then be used to obtain the maximum
emission rate that can be expected from a given soil pile.  If there is no
liquid gasoline in the pore spaces, the emission rate estimated using this
procedure will be greater than the actual emission rate.

Total VOC and benzene emission rates from excavated soil are significantly
higher than those of other types of corrective actions, but these emissions
are also of the shortest duration.  The maximum emission rate from soil
piles occurs immediately after excavation, and can be thought of as an
initial burst which rapidly declines over time.  For total VOCs, the
maximum emission rate for an average soil pile having an exposed surface
area of 2000 ft2 is between 50 and 200 Ibs/hr, depending upon the
temperature.  Benzene emissions for the same soil pile would range from
0.5-2 Ibs/hr, again depending upon the temperature.  The air emission rate
calculated using the charts in Section 2.3 represents the maximum emission
rate, which occurs immediately after the soil is excavated.
2.2  FACTORS THAT AFFECT AIR EMISSION RATES FROM EXCAVATED SOIL

a)  Temperature

Vapor pressure is highly dependent upon temperature, and consequently
temperature greatly affects the volatilization rate (PEI, 1988).  The
emission rates presented in Section 2.3 have been calculated assuming
temperatures of 86°F, 68°F, 50°F, and 32°F (30°C, 20°C, 10°C, and 0°C).  If
the temperature is warmer at a site in question, emission rettes increase.
Correspondingly, if the temperature at a site in question is cooler, the
emission rates decrease.

b)  Soil Type and Soil Moisture

Other parameters that affect the emission rate of contaminants from soil
include soil type and soil moisture (PEI, 1988).  The ability of
contaminant vapors to move through the pore spaces depends upon how large
and continuous the pore spaces are.  Contaminants will take longer to move
                                    2-2

-------
through soils having small pore spaces (such as clays) than they will in
soils where the pore spaces are large and interconnected (such as gravels).
Water can occupy the pore spaces of soils, and can coat the individual soil
grains, thus reducing the size of the voids.  Consequently, the presence of
soil moisture will further inhibit the movement of contaminant vapors.
Because fine-grained soils can retain more moisture than coarse soils,
vapors will take longer to escape from clays than they will from soils such
as sands and gravels.  The emission rates in Section 2.3 have been
calculated assuming a soil having a medium grain size (such as a sand) with
a porosity of 0.35 (or 35 percent), and a small amount of moisture
(moisture content = 0.08).

c)  Wind

Wind plays an important role in enhancing volatilization from soil piles by
mixing the contaminated air at the soil surface with the uncontaminated air
above.  The wind reduces the amount of the contaminant in the air
surrounding the soil pile to negligible levels.  As a result, the
difference between the vapor concentration in the soil pore spaces and in
the air is maximized.  This produces a larger flux of contaminant vapors
away from the soil pile (PEI, 1988).  The procedure presented for
estimating emission rates assumes that the wind is sufficient to mix  the
air above the soil pile, thus keeping the vapor concentration in the  air
negligible and the flux at a maximum.

d)  Surface Area

The exposed surface area of a soil pile directly affects the rate at  which
contaminants are emitted.  Piles with greater surface area tend to emit
larger amounts of VOCs and benzene than piles with lesser surface area.
Piles can be managed, therefore, to decrease the rate at which contaminants
are volatilized.  For example, to decrease emissions, a given volume  of
soil can be spread in a thick layer, as opposed to a  thin layer; thus the
surface  area and the emission rates are reduced.
                                     2-3

-------
2.3  AIR EMISSION ESTIMATION PROCEDURES


The following procedure should be followed in order to estimate an emission

rate from an excavated soil pile:


(1)  Determine the approximate shape of the soil pile:


     a.  Horizontal layer:
                                                    HEIGHT
     b.  Cone:
                     -DIAMETER
(2)   Determine the surface area of the  soil pile:
     a.   Horizontal layer:

         Knowing the total volume  of  the  soil pile  (in ft3) and the
         thickness of the soil  layer  (in  ft), use the following formula to
         compute the surface  area:
                                   2-4

-------
    Surface area = (total volume)/thickness
              Vol. = 100yd
3MC
                                             Surface Area =
 Volume
 Height

100yd3   27ft3
 3 ft     1 yd3
                                                       = 900ft
    or
    Estimate the length and width of the soil layer  (in ft), and use
    the following formula to compute the surface area:

    Surface area =  (length) x  (width)
              -45 ft-
                                             Surface Area = Length X Width

                                                      = 45x20 ft2

                                                      = 900 ft2
 b.   Cone:
    Approximate  or measure  the diameter at the base of the pile,  and
    the height of the pile  (in ft).   Then use the following formula to
    calculate the surface area:
     Surface area = 3.142 x r x /r2  + h2

     where r = 0.5 x diameter
                                       Surface Area _ 3 142 r /
                                                = 3.142(10)/100+225
                                                = 566.4 ftf
                                2-5

-------
          or
          If the radius of the pile is approximately equal to the height of
          the pile,  the following formula may be used to estimate the
          surface area:
          Surface area = 1.11 x (diameter)2
                                              Surf ace Area = 1.11 x (diameter)2
                                                       = l.llx(20)2
                                                       = 444. ft.2
                  20ft
Appendix D has been  included  to  assist  in converting other units of length
and area to feet and ft2.

(3)  Use Figures 2-1 and 2-2  to  determine the  expected maximum emission
rate from excavated  soil of gasoline VOCs and  benzene, respectively.
Locate the surface area of the soil pile  on  the x  axis,  and draw a vertical
line up to the line  on the graph.  Find the  line corresponding to the
temperature at the site.  At  the intersection  of these lines,  draw a
horizontal line over to the y axis.  The  emission  rate can be  read from the
y axis at this point.

It should be noted that the emission rates presented in  Figures 2-1 and 2-2
are for a soil having a medium grain size, such as a sand.   Coarse soils
such as gravels have emission rates which are  approximately six percent
greater, while fine-grained soils such  as clays have emission  rates that
are approximately 4  percent less than the emission rates depicted in
Figures 2-1 and 2-2.
                                    2-6

-------
ro

—i
                s

                I
                o
                t/3
                      400
                      300
200
100
                                           1000
                                         2000
                                   SURFACE AREA OF SOIL PILE OR SOIL LAYER (ft2 )
                                                                                    86°F (30°C)
  68°F (20°C)




  50°F(10°C)


  32°F(0°C)
3000
                                                               FIGURE2-1
                                   ESTIMATED SOIL EXCAVATION AIR EMISSION RATES FOR GASOLINE VOCs

-------
ro
i
00
                  L.


                  S
                  01

                  Z

                  O
                  i
                                               1000
2000
                                                                                             86°F (30°C)
                                                                                             68°F (20°C)
                         50°F(10°C)




                         32°F (0°C)
3000
                                                Ann* x-ve c/-yii on c ^\D c/^li I AVCD /ft
                                                nivEA vrr ov-rii. r ILL •o-i\ crvrii. i^t i tn. vi«.
                                                                        FIGURE2-2

                                              ESTIMATED SOIL EXCAVATION AIR EMISSION RATES FOR BENZENE

-------
2.4  VARIATION OF EMISSION SATES WITH TIME

In general, emission rates for excavated soil piles can be characterized by
a sharp burst of emissions followed by a rapid decline to negligible
levels.  The amount of time required for contaminants to volatilize from
soil piles is highly variable, but tends to be on the order of a few days
to weeks.  The emission rate calculated using the charts in Section 2.3
represents the initial burst of emissions illustrated in Figure 2-3.

The highly volatile constituents of gasoline such as isobutane, n-butane,
isopentane, and 1-pentene volatilize first, leaving behind the less
volatile constituents.  Because of their lower vapor pressures, the less
volatile constituents will have lower concentrations in the soil pores
(Johnson et al., 1988).  Also, as the contaminants near the surface of the
soil pile escape to the atmosphere, the vapors that are deeper in the soil
pile will have to travel farther in order to reach the surface.  Both of
these effects contribute to the rapid decrease in emission rates with time,
as shown in Figure 2-3.

Frequently, the surface of the soil pile is intentionally disturbed,
perhaps through land farming or rototilling to increase volatilization.
Land farming consists of removing the top several inches of soil from a
horizontally distributed pile, thus exposing the more contaminated soil
lower in the pile to the atmosphere.  Rototilling greatly increases the
size of the void spaces in the soil by physically breaking it up and
loosening it.  This can also expose the more contaminated portions of the
soil pile to the air, thus increasing the emission rate temporarily.
Figure 2-4 illustrates the effect of disturbing the soil pile upon the
emission rate.
                                    2-9

-------
en
CO

HI
                     TIME (DAYS -^-WEEKS)
                                    FIGURE 2-3
                        CHANGE IN EMISSION RATES WITH TIME
                            FOR AN EXCAVATED SOIL PILE

                            2-10

-------
                          TIME OF DISTURBANCE
LU
5
DC
                      TIME (DAYS -^-WEEKS)
                                    FIGURE 2-4
                        CHANGE IN EMISSION RATES WITH TIME
                             FOR DISTURBED SOIL PILES

                             2-11

-------

-------
                           3.0  VACUUM EXTRACTION
3.1  OVERVIEW

Vacuum extraction can remove as much as 99.99 percent of contaminants from
soils through the use of one or more suction wells, or a series of air
injection and suction wells (COM, 1987a).  Because the soil is treated in
place, vacuum extraction can be less expensive and less disruptive than
excavation, and as a result, it is becoming a popular choice as a
corrective action technology (CDM, 1987a).

The pressure  in the ground at a suction well will be lower than the
pressure some distance away from the well.  This pressure difference
induces the air in the soil pore space to move toward the suction well.
The contaminants, which are in the vapor phase, move along with the air to
the suction well.  As with excavated soil piles, as the contaminant vapors
are removed from the soil pores, more volatilize from the liquid phase to
the vapor phase.  Naturally, vacuum extraction systems will be more
effective in soils which allow air to flow freely through its pore spaces.

In order to estimate air emission rates, the procedure and charts presented
in Section 3.3 rely on actual soil gas contaminant concentrations (which
have been measured as part of the site investigation), and the pumping rate
of the vacuum extraction system.  By using measured contaminant
concentrations, site specific conditions such as soil type, soil moisture,
and temperature need not be considered in the emission rate determination.
These factors do affect the overall performance of a vacuum extraction
system however, and they are discussed in greater detail in Section 3.2.
It should be noted that many soil gas contaminant concentrations are
measured as part of a typical site investigation.  In order to estimate the
maximum emission rate using the graphs presented in Section 3.3,  the
maximum soil gas concentration should be used.  If an average soil gas
concentration is used instead,  the estimated emission rate will most likely
represent the short-term average (perhaps over the first month of
                                    3-1

-------
 operation)  of the actual emission rate.   It should be noted that the  use  of
 an average  soil gas concentration will tend to result in an air  emission
 rate which  may be significantly higher than the long term average emission
 rate (i.e.,  the emission rate  over several  months).

 Emission rates from vacuum extraction systems  are  generally less than those
 for excavated soil,  and greater than those  of  air  strippers.  Maximum
 emission rates tend to  be under 50 Ib/hr  for total gasoline VOCs and  under
 2  Ib/hr  for benzene.  The duration of the emissions from vacuum  extraction
 systems  is  on the order of weeks to months;  hence,  they  tend to  be longer
 in duration than those  from excavated soil  piles,  and less than  those of
 air strippers.   The procedure  presented in  Section 3.3 is commonly used by
 industry officials to estimate air emission rates.
3.2  FACTORS THAT AFFECT AIR EMISSIONS FROM VACUUM EXTRACTION SYSTEMS

Factors that can  affect the  removal rate of VOCs from contaminated soil
include the soil  permeability, moisture content, applied suction pressures,
air flow rate, and temperature (Krishnayya, et al., 1988).  The most
significant of these factors are temperature and soil type and soil
moisture, and they are discussed in detail below.

(a)  Temperature

The contamination level within a soil can be affected by temperature due to
the large effect  of temperature on the vapor pressures of the contaminants.
Because the procedures presented in Section 3.3 use actual contaminant
concentrations to arrive at emission rates, the effect of temperature on
the emission estimate is negligible as long as the variation in the soil
temperature during the period of operation is not large.  If the soil gas
concentrations are measured during a cold season, and the operation of the
vacuum extraction system runs well into the summer, the emission rate may
tend to increase  over expected levels.
                                    3-2

-------
(b)  Soil Type and Soil Moisture

Typically, air tinder pressure can flow freely through soils having large
grain sizes (such as gravels and coarse sands), while air generally cannot
flow freely through fine grained soils such as clays.  This relationship
can be complicated by the moisture content of the soil.  In general, wet
soils restrict the movement of air through the pore spaces.  Hence, vacuum
extraction systems are more effective in dry soils.

These soil characteristics determine the ability of an extraction well to
form a flow field in the soil which can encompass the contamination.  The
radius of influence of a single vacuum extraction well having a specific
pumping rate can range from tens to hundreds of feet, depending on soil
type and the depth to ground water (Terra Vac, 1987).  Therefore, within a
given soil, the pumping rate can be lowered such that the radius of
influence is reduced.  Consequently, the well will encompass less
contamination, and the emission rate can be somewhat reduced.  This may be
desirable in regions where air emissions are a problem.  It should be
noted, however, that reducing the pumping rate can substantially reduce the
volume of soil being treated, and hence can prolong the cleanup time.
 3.3  AIR EMISSION ESTIMATION PROCEDURES

 It should  be  noted that vacuum extraction  systems  frequently consist of
 more than  one extraction well.   In these cases,  emission rates  must be
 determined for each well,  and then summed  to determine  the  emission rate  of
 the  entire extraction system.  The following procedure  should be followed
 in order to estimate  an emission rate for  one vacuum extraction well:

 (1)  Determine the pumping rate of the well in cubic feet per minute (cfm).
 Appendix D has been included to assist in  converting other  systems of units
 to cfm.                                                	     	
                                     3-3

-------
 (2)   a.  To  obtain the most conservative  estimate  of  the  air  emission  rate,
         determine the maximum concentration  in ppm-v (parts  per
         million-volume)  of the gasoline  and/or benzene in  the  soil gas.
         This data should be obtained  as  part of the  site investigation.
         The use  of the maximum soil gas  concentration will provide an
         estimate of the  maximum air emission rate.

      b.  To  obtain an air emission  rate which represents  more of a short
         term average, determine the average  soil  gas concentration in ppm-v
         of  the gasoline  and/or benzene.  This concentration  can be
         approximated by  averaging  the soil gas concentrations  that were
         obtained as part of the site  investigation.

Care  must be taken to ensure that  the concentration  used in  this procedure
is the soil gas  concentration as opposed to  the total soil concentration.
Appendix C  contains information to assist in converting  total  soil
concentrations to soil gas  concentrations.   A procedure  for  converting soil
gas concentrations in units of /ug/1 to ppm-v is also contained in
Appendix C.  It  should be noted that  the maximum  concentration of gasoline
VOCs  and/or benzene present in the soil pores depends upon the blend  of
gasoline present.  Using  the synthetic blend  given in Appendix  A, and
assuming that there is liquid gasoline and air present in  the  soil pores,
the maximum concentration of gasoline VOCs that can  be present in the soil
gas at 68°F is 361,000 ppm-v.   The maximum concentration of benzene that
can be present in the soil  gas at  68°F is 3600 ppm-v.

(3)  Using  Figure  3-1 to estimate  the emission rate  of gasoline VOCs,
locate the  pumping rate of  the well on the x  axis, and the gasoline soil
gas concentration on the y  axis.   The intersection of these two points will
fall on  or  near a  curve having a specific emission rate.   This curve can be
used to  estimate  the emission  rate under the prescribed pumping rate and
contamination levels.  Figure  3-2  should be used  to  estimate benzene
emission rates in  the same manner.
                                    3-4

-------
                              s-e
                               GASOLINE CONCENTRATION

                                 IN SOIL PORES (ppm-v)
m


i

H
m
a
z H w


ir

  m

  ^X^
  ^^^
  o
  z
  73

  m
             •o
             C

             2
             Z
             Ci


             5
             m

-------
m
1
m
a
c


m
> m
m
O


70


5
m
tn

•n

O
73

03
m

Z
N
rn

Z
m
                                      9-e
                                    BENZENE CONCENTRATION

                                      IN SOIL PORES (ppm - v)
              -a
              C


              -o
KJ
O
O

-------
3.4  VARIATION OF EMISSION RATES WITH TIME

When a vacuum extraction well begins pumping, a certain amount of time is
required to establish the flow field in the soil.  The soil type and
moisture greatly affect the amount of time required.  For example, the time
required to establish a flow field in moist soil will be longer than for
dry soil.  Additionally, more time is required to establish a flow field in
a clay soil as opposed to sand and gravel.  As the flow field radiates
outward from the well, the emission rate of the contaminants tends to
increase, as indicated in Figure 3-3.  This is because the radius of
influence encompasses more contamination as it grows outward (Terra Vac,
1987).

In general, the amount of time required to reduce the VOC soil gas
concentrations to negligible levels is approximately several weeks to a few
months.  In any soil contaminated with a mixture of VOCs, the higher vapor
pressure components such as isobutane, n-butane, isopentane, and 1-pentene
will volatilize first.  With time, the residual in the soil will become
rich in the less volatile compounds, resulting in vapor concentrations and
emission rates that will decrease with time (Johnson et al., 1988).  This
trend is also illustrated in Figure 3-3.

After the extraction well has been pumping for some time, the air in the
soil will tend to flow towards the well along preferential flow paths.  The
contamination will be removed from the area around these flow paths fairly
quickly; however, contamination will still exist beyond the flow paths.
Researchers have found that pulsing the well (turning it off and on, or
systematically increasing and decreasing the pumping rate) disturbs these
preferential flow paths.  This encourages the air to move through more of
the soil, thus removing more contamination.  For pulsed systems, the
highest emission rates for VOCs occur immediately after the air flow rates
are changed  (Rainwater, 1988).  This is illustrated in Figure 3-4.
                                    3-7

-------
Ill
s
DC
g
CO
CO
LLJ
                   TIME (WEEKS -^-MONTHS)
   PROPAGATION OF
     FLOW FIELD
                                   FIGURE 3-3
                       CHANGE IN EMISSION RATES WITH TIME
                        FOR VACUUM EXRACTION SYSTEMS
                            3-8

-------
UJ
<
DC

O
(/)
CO

UJ
                       SYSTEM IS "OFF"

                  TIME (WEEKS -^-MONTHS)
                                   FIGURE 3-4
                       CHANGE IN EMISSION RATES WITH TIME
                     FOR PULSED VACUUM EXTRACTION SYSTEMS
                             3-9

-------

-------
                             4.0 AIR STRIPPING
4.1  OVERVIEW

Air stripping is widely used to remove VOCs from ground water.  The removal
efficiency of an air stripper depends upon the volatility of the contami-
nants, but is generally in the range of 99 to 99.5 percent for most VOCs
found in gasoline (CDM, 1987a).

Air stripping involves pumping contaminated water from the ground and
allowing it,to trickle over packing material in an air stripping tower.  At
the same time, clean air is circulated past the packing material.  When the
contaminated water  (which coats the packing material in a thin film) comes
into contact with the clean air, the contaminants tend to volatilize from
the water into the  air.  The contaminated air is then released into the
atmosphere.

A typical site investigation will generally result in several ground water
contaminant concentrations, each sampled at a different location.  In order
to estimate the maximum emission rate using the graphs presented in Section
4.3, the maximum ground water concentration should be used.   If an average
ground water concentration is used instead, the estimated emission rate
will represent a long-term average of the actual emission rate  (perhaps
over the first six  months of operation).

The procedure presented in Section 4.3  for estimating emission  rates is
commonly used by industry and  regulatory officials.  It relies  on informa-
tion pertaining to  the design  of the air stripper  (such as pumping rate and
removal efficiency),  along with field measurements of the contaminant       /
concentration  in the  ground water.  The procedure was checked against
examples published  in the literature to ensure  that  realistic estimates
were  obtained.
                                     4-1

-------
 Emissions from air strippers tend to  be  less  than the  emissions  from
 excavated soil piles and vacuum extraction systems; however,  they tend  to
 be  longest in duration.   Air stripper emission  rates depend  in part  upon
 the pumping rate  and removal efficiency  of the  system.  .For  systems  pumping
 at  less  than 100  gpm and having removal  efficiencies between 85  and  99.99
 percent,  VOC emissions will  range from 0.5-4  Ib/hr.  Benzene emissions will
 generally be between 0.1-0.5 Ib/hr.

 4.2 FACTORS THAT AFFECT AIR EMISSIONS FROM AIR STRIPPERS

 (a)  Solubility of Contaminants

 The solubility of the contaminants largely determines  the concentration of
 the contaminant in the ground water.   The  solubility of many of  the
 compounds typically found in gasoline can  increase tremendously  in the
 presence  of additives.   It should be  noted that the synthetic gasoline
 described in Appendix A  contains  no additives.

 (b)  Removal Efficiency

 The removal efficiency is largely determined by the design of an air
 stripper,  and  to  a lesser extent, the volatility of the compound being
 removed.   Typically, air  strippers can remove more than 99 percent of the
 volatiles found in gasoline.  Three removal efficiencies (85, 95 and 99.99
 percent)  are used in Section  4.3  to represent the entire range of removal
 efficiencies that might be encountered under operating conditions.

 (c)  Pumping Rate

 The pumping rate  of the ground water extraction well and the contaminant
 concentration determine the pollutant loading to the air stripping system.
 The pollutant loading and the removal efficiency then determine the
 emission  rate from the system.  Reducing the pumping rate can lower the
pollutant loading to a system, and hence reduce the emission rate as well.
 It should be noted, however, that reducing the pumping rate in an effort to
 lower emissions can significantly increase the time and cost of cleanup.
                                    4-2

-------
4.3  AIR EMISSION ESTIMATION PROCEDURES

The following procedure should be used to estimate an emission rate for an
air stripping system:

(1)  Determine the pumping rate of the well in gallons per minute (gpm).
Appendix D contains the information required to convert other systems of
units to gpm.

(2)  Determine the maximum concentration (in mg/1) of the gasoline and/or
benzene in the ground water.  These should be available from the site
investigation.

The maximum concentration of gasoline or benzene that can be measured in
the water is limited by the solubility of the compounds.  The solubility of
gasoline in water is determined by the specific blend of hydrocarbons.  The
concentration of benzene in water is affected by its percent composition in
gasoline.  Using the blend given in Appendix A, the maximum concentrations
that can be measured in the ground water are 131 mg/1 for gasoline and 65
mg/1 for benzene.

(3)  Determine the removal efficiency of the air stripping system.  The
removal efficiency is largely determined by the air-water ratio and packing
height of the air stripping tower (Radian, 1987).

(4)  Select the figure corresponding to the removal efficiency of the air
stripper and the contaminant being considered.  Three graphs corresponding
to removal efficiencies of 99.99, 95, and 85 percent have been provided for
both gasoline and benzene, as follows:
      Figure  4-1:
      Figure  4-2:
      Figure  4-3:
      Figure  4-4:
      Figure  4-5:
      Figure  4-6:
Gasoline, removal efficiency: 99.99%
Benzene, removal efficiency: 99.99%
Gasoline, removal efficiency: 95%
Benzene, removal efficiency: 95%
Gasoline, removal efficiency: 85%
Benzene, removal efficiency: 85%
                                     4-3

-------
 Op
 OD
 uz
-33
       20
                                                                      4 Ib/hr
                                                                    100
                                  PUMPING RATE (gpni)
                                               FIGURE4-1
                          ESTIMATED AIR STRIPPER EMISSION RATES FOR GASOLINE VOCs
                                       REMOVAL EFFICIENCY = 99.99 %

-------


z
0  i
u  5
z  O
^J  OH
N  
-------
Op
81
18
OO
U
        20
                                                                            4 Ib/hr
3 Ib/hr
2 Ib/hr
1 Ib/hr

.5 Ib/hr

.1 Ib/hr
                                    PUMPING RATE (gpm)
                                                 FIGURE 4-3
                           ESTIMATED AIR STRIPPER EMISSION RATES FOR GASOLINE VOCs
                                         REMOVAL EFFICIENCY = 95.00 %

-------
Z st
O Ml


§!

1|

y ^
Z D

81
UJ O

UJ n
N u
Z Z
UJ —
CO
                                              .5 Ib/hr
                                              .4 Ib/hr
                                              .3 Ib/hr
                                              .2 Ib/hr
                                              .1 Ib/hr



                                              .05 Ib/hr


                                              .01 Ib/hr
      PUMPING RATE (gpm)
                   FIGURE 4 - 4

ESTIMATED AIR STRIPPER EMISSION RATES FOR BENZENE

           REMOVAL EFFICIENCY = 95.00 %

-------
00
           z
                                                                              4 Ib/hr
                                          PUMPING RATE (gpm)
                                                        FIGURE4-5
                                  ESTIMATED AIR STRIPPER EMISSION RATES FOR GASOLINE VOCs
                                               REMOVAL EFFICIENCY =85.00 %

-------
20
 40          60

PUMPING RATE (gpm)
80
                                                  .5 Ib/hr
                                                  .4 Ib/hr
100
                           FIGURE4-6
         ESTIMATED AIR STRIPPER EMISSION RATES FOR BENZENE
                  REMOVAL EFFICIENCY = 85.00 %

-------
Select the figure most closely corresponding to the removal efficiency
of the air stripper and the contaminant being considered.

(5)  Using the figure selected, locate the pumping rate of the well on
the x axis, and the concentration of the contaminant in the water on
the y axis.  The intersection of these two points will fall on or near
a curve having a specific emission rate.  This curve can be used to
estimate the emission rate under the prescribed pumping rate and
contamination levels.

(6)  If the pumping rate of the air stripper is greater than 100 gpm,
equation B-12 in Appendix B can be used to calculate the air emission
rate.
4.4  VARIATION OF EMISSION RATES WITH TIME
At a given site, contaminant concentrations can vary from one pumping
well location to another.  At a given location! of pumping well, the
concentration of the contaminant also can vary considerably with time,
which corresponds to variations in the emission rate from the air
stripping tower.

For these reasons, the exact variation of emission rate with time is
extremely difficult to predict; however, the overall trend is that
indicated by Figure 4-7.  In general, the amount of time required to
remove volatile compounds adequately from ground water is measured in
years.  The variation in emission rate with time is characterized by a
somewhat rapid decrease during the first few months, followed by a
leveling off period (Radian, 1987).
                                    4-10

-------
Ill
<
DC
Z
o
UJ
                         TIME (YEARS)
                                     FIGURE 4-7
                        CHANGE IN EMISSION RATES WITH TIME
                                 FOR AIR STRIPPERS

                             4-11

-------

-------
                            5.0  REFERENCES
American Petroleum Institute.  September 1985.  Subsurface Venting of
    Hydrocarbon Vapors from an Underground Aquifer.  Health and
    Environmental Sciences Department.  API Publication Number 4410.

Camp Dresser & McKee Inc.  March 2, 1987a.  Cross-Media Contamination
    Issues; U.S. Environmental Protection Agency, Office of
    Underground Storage Tanks.

Camp Dresser & McKee Inc. and PEI Associates.  November 1987b.  Draft
    Seminar Proceedings on the Underground Environment of an UST Motor
    Fuel Release.  U.S. EPA Contract Number 68-3-3409.

Camp Dresser & McKee Inc.  April 7, 1988a.  Research for Abatement of
    Leaks from Underground Storage Tanks Containing Hazardous
    Substances.  U.S. EPA Contract No. 68-03-3409.

Camp Dresser & McKee Inc.  June 13, 1988b.  Final Report: Phase 1 of
    Modeling Vapor Phase Movement in Relation to UST Leak Detection.
    U.S. EPA Contract No. 68-03-3409.

Camp Dresser & McKee Inc.  1986.  Interim Report: Fate and Transport
    of Substances Leaking from Underground Storage Tanks, Volume
    1-Technical Report.  Office of Underground Storage Tanks, U.S.
    EPA Contract No. 68-01-6939.

Fancy, C.H.  Interoffice memorandum.  State of Florida Department of
    Environmental Regulation.  October 20, 1987.
Freeze, R. Allan and John A. Cherry.  1979.
    Cliffs, New Jersey: Prentice Hall.
    Groundwater.  Englewood
Geoscience Consultants Limited.  February 29, 1988.  Draft Final
    Report: Background Hydrocarbon Vapor Concentration Study for
    Underground Fuel Storage Tanks.  U.S. EPA Contract No. 68-03-3409.
Johnson, Mark.  Personal communication.
    November, 1988.
Camp Dresser & McKee Inc.
Johnson, Paul C., Marian W. Kemblowski, and James D. Colhart.  1988.
    Practical Screening Models for Soil Venting Applications.
    NWWA/API Conference: Petroleum Hydrocarbons and Organic
    Chemicals in Ground Water - Prevention, Detection and
    Restoration,  pp. 521-546.

Krishnayya, A.V., M.J. O'Connor, J.G. Agar, and R.D. King.  1988.
    Vapour Extraction Systems - Factors Affecting Their Design and
    Performance.  NWWA/API Conference: Petroleum Hydrocarbons and
    Organic Chemicals in Ground Water - Prevention, Detection and
    Restoration,  pp. 547-569.
                                  5-1

-------
 PEI Associates,  Inc.  .February 1988.   Draft  Internal Working
    Document-LOCl  Research Report on Mobilization,  Immobilization, and
    Transformation of Motor Fuel  Constituents and Organic Chemicals  in
    the  Environment.  U.S.  EPA Contract No.  68-03-3409.

 Perry, Robert H.,  (late  editor),  and D. Green, j. Maloney,  (editors).
    1984•  Perry's Chemical Engineers' Handbook,  sixth edition.
    McGraw-Hill  Inc.

 Radian Corporation.   May 1987.  Air Strippers, Air  Emissions and
    Controls.  Draft  report prepared for U.S. EPA.

 Rainwater, Ken,  Billy j. claborn, Harry w. Parker,  Douglas Wilkerson,
    and  Mohammad R. Zaman.   1988.  Large-scale Laboratory Experiments
    for  Forced Air Volatilization of Hydrocarbon Liquids in Soil.
    NWWA/API Conference: Petroleum Hydrocarbons and Organic Chemicals
    in Ground Water - Prevention, Detection and Restoration.
    pp.  501-520.

 Stephanatos, Basilis N.  1988.  Modeling the Transport of Gasoline
    Vapors by an Advective-Diffusive Unsaturated Zone Model.  NWWA/API
    Conference: Petroleum Hydrocarbons and Organic Chemicals in Ground
    Water - Prevention, Detection and Restoration,   pp. 591-611.

Terra Vac Inc.   November 20, 1987.  Demonstration Test Plan: In-Situ
    Vacuum Extraction Technology; Groveland Wells Superfund Site.
    EERU Contract No.  68-03-3255.
                                 5-2

-------
                               APPENDIX A
       PHYSIOCHEMICAL PROPERTIES OF COM SYNTHETIC GASOLINE BLEND
                          (Source:  COM,  1988b)
 Petroleum fuels such as automotive gasoline,  jet fuel,  diesel,  and
 home heating oil are composed of a wide variety of hydrocarbons.   Many
 petroleum products also contain additives, which not only enhance the
 performance,  but may also modify the physical and chemical properties
 of the fuel.  Additives are  generally used in small quantities,
 however they occasionally compose as much as  10-12% of  the fuel.

 This Appendix contains  tables that list the physiochemical properties
 of a synthetic  gasoline blend at different temperatures.   By comparing
 the typical  constituents of  other fuels to the constituents of the
 synthetic  gasoline,  generalizations can be made as  to air  emission
 rates that could be  expected at  sites where fuels other than gasoline
 have leaked  into the ground.

 In general, automotive  gasoline  has  a higher vapor pressure than other
 fuels such as diesel, home heating oil  and jet  fuel.  At the same
 temperature then, these  fuels will evaporate at  lower rates than
 automotive gasoline.  Consequently,  the air emission rates of these
 compounds would also be  lower than those expected for gasoline.  Of
 these  fuels, jet  fuel tends to be more volatile than both diesel and
 home heating oil, and diesel tends to be more volatile than home
 heating oil.  For these  reasons, air emission rates for jet fuel would
 tend to be higher than emission rates for diesel.  Emission rates for
both jet fuel and diesel would tend to be higher than the emission
 rates for home heating oil.
                                 A-l

-------
Typical categories of fuel additives include lead scavenging agents,
octane enhancers, anit-oxidants, metal deactivators,  and corrosion and
rust inhibitors, to name just a few.  For the most part however, only
the octane enhancers and lead scavenging agents are found in
significant quantities (they commonly make up 2-5% of the
composition), and these are only found in automotive gasolines.  In
general the lead scavenging agents have low vapor pressures, and hence
the air emission rates for these compounds would be negligible.  The
octane enhancers include MTBE (methyl t-butyl ether), t-butyl alcohol,
methanol, and ethanol.  These compounds have high vapor pressures, and
hence the air emission rates for these compounds would be significant.
The contribution of these compounds to the total emission rate for
gasoline may not be significant however, because they make up a small
portion of the total composition of the gasoline.
                     List of Tables for Appendix A
Table
A-l
A-2
A-3
A-4
Gasoline Blend
CDM Synthetic
CDM Synthetic
CDM Synthetic
CDM Synthetic
Temperature
32° F (0°C)
50°F
68°F (20°C)
86° F (30°C)
A-3
A-4
A-5
A-6
                                  A-2

-------
CHEMICAL PROPERTY ESTIHftTION FOR SYNTHETIC
6ASGL1NE AND CONSTITUENTS AT 0 OE6REES C.
REPRESENTATIVE PERCENT BRAH HDL. LIB, PHftSE AIR DIFFUSION LIQUID
CHENICftL COMPOSITION HEIGHT MOL FRACT, COEFFICIENT DENSITY
(6H/NOL) (CMA2/SECI (BM/CN"3)
Isobutane
n-8utane
Isopentane
n-Pentane
n-Octane
Benzene
Toluene
Xylene (•)
n-Hexane
2-Hethylpentane
Cyclohexane
n-Heptine
2-Methylhexane
Methyl cyclohexane
2,4-Oiiethylhexane
Ethylbenzene
1-Pentene
2,2,4-Triiethylhexane
2,2,5,5-Tetraiethylhexane
1,4-Diethylbenzene
1-Hexene
1,3,5-Triiethylbenzene
C12-»liphatic
Total
Teiperature ' 273.15
Pressure * 760
6j5 Constant * 0.0623

2
1
14
3
1
3
5
7
9
8
3
1.5
5
1
e
2
1.5
2
1.5
5
1.5
5
10
100
58.12 0.0326 0.0805
58.12 0.0163 0.0805
72.15 0.1840 0.0722
72.15 0.0394 0.0722
114.23 0.0083 0.0528
78.11 0.0364 0.0800
92.14 0.0515 0.0728
106.17 0.0625 0.0590
86.18 0.0990 0.0628
86.18 0.0880 0.0628
84.16 0.0338 0.0636
100.20 0.0142 0.0583
100.20 0.0473 0.0583
98.19 0.0097 0.0589
114.23 0.0664 0.0546
106.17 0.0179 0.0648
70.14 0.0203 0.0733
128.26 0.0148 0.0537
142.29 0.0100 0.0510
134.22 0.0353 0.0525
84.16 0.0169 0.0636
120.20 0.0394 0.0555
170.00 0.0558 0.0463
102.20 1.0000
deg. K Molecular Height of Air = 2B.96
u Hg Average Nolecular Height of
•• Hg*i*3/iol«K Baseline Vapors • 68.34
NoUtulw Height of

Basoline-ftir Mixture = 35.50
0.5876
0.6064
0.6418
0.6465
0.7161
0.9061
0.8844
0.8761
0.6759
0.6707
0.7975
0.6986
0.6942
0.7869
0.7141
0.8823
0.6620
0.7306
0.7327
0.8745
0.6909
0.87S5
0.8711
0.7539
gi/iol
qi/iol
gi/iol
PURE CHEMICAL
VAPOR PRESSURE
In Hg)
1174.26
774.09
259.30
183.40
2.86
26.34
6.72
1,63
45.32
67.27
27.85
11.42
17.59
12.11
7.20
1.89
235.47
3.20
1.68
0.14
57,64
0.38
0.01

PARTIAL PRESSURE PURE CHEMICAL CONCENTRATION OVER
OVER BASOL1NE VAPOR DENSITY LIQUID 6ASOLINE
(it Hg) (6M/M*3I Ippi)
38.3215
12.6310
47.7170
7.2320
0.0237
0.9594
0.3459
0,1017
4.4888
5.9218
0.9414
0.1621
0.8326
0.1170
0.4779
0.0338
4,7757
0.0473
0.0168
0.0048
0.9742
0.0152
0.0005
126.1422
Vapor Density of Basoline-
Air Mixture «
Weighted Average Air
Diffusion Coefficient -


4006.65
2641.24
1098.34
776.84
19.16
120.79
36.36
10.14
229.31
340.32
137.60
67.17
103.49
69.83
48.25
11.79
969.60
24.07
14.00
1.07
284.77
2.71
0.10

1583.72 gi/r3
0.0642 c«*2/sec

50423.0
16619.8
62785.6
9515.8
31.2
1262.4
455.1
133.8
5906.4
7791.8
123B.7
213.3
1095.5
154.0
628.8
44.5
6283.8
62.2
22.0
6.3
1281.9
20.0
0.7
165976.6
Weighted average
liquid density =
Heighted average
gas density -

BOILINB
POINT
(deg. K)
261.25
272.65
300.95
309.15
398.75
353.25
383.75
466.25
341.85
333.35
353.85
371.55
363.15
374.05
382,55
409.25
303.05
399.65
410.55
456.85
336.55
437.85
489.15

0.736
506

HENRY'S LAN
CONSTANT
(dil.)
. 24.08
18.18
23.20
19.94
27.35
0.07
0.07
0.06
22.72
24.10
2.29
25.55
41.04
4.67
32.58
0.07
6.64
33,29
109.49
0.08
5.78
0.04
27.20
18.68111
gt/ci*3
S»,V3


-------
CHEMICAL PROPERTY ESTIKATIOK FOR SYNTHETIC
GASOLINE AHD CONSTITUENTS AT 10 DEGREES C.
REPRESENTATIVE PERCENT BRAK KOL
CHEMICAL COHPOSITION HEIGHT
. LIB. PHASE
HDL FRACT.
(6H/HOL)
Isobutine
n-Butjne
Isopentane
n-Pentane
n-Qctane
Benzene
Toluene
lylene (•)
n-Hexane
2-Hethylpentane
Cyclohexane
n-Keptane
2-Hfthylnexane
Itethylcyclohtxane
2,4-Diitthylhixane
Etbylbsnzene
1-Pentene
2,2|4-Triiethylhexane
2,2,5,5-TetraKthylhexane
1,4-Diethylbenzene
1-Hfxine
1,3,5-Triiethylbenzene
C12-aliphatic
Total
2
1
14
3
1
3
S
7
9
8
3
1.5
S
1
8.
2
1.5
2
1.5
5
1.5
5
10
100
58.12
56.12
72.15
72.15
114.23
78.11
92.14
106.17
86.18
86.18
84.16
100.20
100.20
98.19
114.23
106.17
70.14
128.26
142.29
134.22
84.16
120.20
170.00
102,20
Teiperature « 2S5. IS dig, 1C
Pressure » 760 a:
Gis Constant * 0.0623 »

Kg
Hg*iA3/iol*K



0.0326
0.0163
0.1840
0.0394
0.0083
0.0364
0.0515
0.0625
0.0990
0.0880
0.0338
0.0142
0.0473
0.0097
0.0664
0.0179
0.0203
0.0148
0.0100
0.0353
0.0169
0.0394
0.0558
1,0000
Molecular Height
Average Sol Kul »r
Sasoline Vapors
HolKuIar Height
AIR DIFFUSION
COEFFICIENT
(CH*2/SEC)
0.0857
O.OB57
0.0769
0.0769
0.0563
0.085!
0.0776
0.0629
0.0669
0.0669
0.0677
0.0620
0.0620
«.0627
0.0581
0.0690
0.0780
0.0572
0.0543
0.0559
0.0677
0.0591
0.0493

of Air « 28.96
Height of
68.92
Df
Easolinc-Air Nixture • 38.88
LIQUID
DENSITY
(GH/CNA31
0.572B
0.5931
0.6311
0.6364
0.7096
0.8957
0.8758
0.6701
0.6676
0.6620
0.7884
0.6914
0.6867
0.7789
0.7071
0.8748
0.6513
0.7240
0.7264
0.8683
0.6821
0.8718
0.8656
«,7457
gi/inl
gi/iol
PURE CHEMICAL
VAPOR PRESSURE
In Kg)
1647.77
1112.75
392.47
283.84
5.63
45.53
12.43
3.26
75.70
109.55
47.51
20.66
30.94
21.45
13.30
3.77
359.48
6.18
3.40
0.32
94.87
0.84
0.03

Vapor Denlty of
Air nature •
Weighted Average
PARTIAL PRESSURE
OVER 8ASOLINE
(ii Hg)
53.7745
18.1571
72.2231
11.1925
0.0468
1.45E3
0.6397
0.2039
7.4976
9.6446
1.6060
0.2933
1.4642
0.2071
0.8833
0.0673
7.2909
0.0915
0.0340
0.0113
1.6036
0.0333
0.0016
S8S.4252
Gasoline-
1673.38
Air
Diffusion Coefficient * 0.0684
gi/iol


PURE CHEMICAL CONCENTRATION OVER BOILING
VAPOR DENSITY LIQUID GASOLINE POINT
(6M/«A3I
5423.76
3662.69
1603.70
1159.80
36.43
201.40
64.86
19; 61
369.48
534.70
226.44
117.23
175.57
119.26
86.03
22.66
1427.99
44.92
27.37
2.43
452.20
5.74
0.27


gt/iA3
ciA2/sec

(ppi)
70755.9
23890.9
95030.4
14727.0
61.5
2182.0
841.7
268.3
9865.2
12690.2
2113.2
385.9
1926.5
272.6
1162.2
88.6
9593.3
120.3
44.7
14.9
2110.1
43.8
2.1
24S191.2
Heighted average
liquid density *
Keighted average
gas density *

(deg. Kl
261.25
272.65
300.95
309.15
398.75
353.25
383.75
466.25
341.85
333.35
353.85
371.55
363.15
374.05
382.55
409.25
303.05
399.65
410.55
456.85
336.55
437.85
489.15


0.727
736

HENRY'S LAN
CONSTANT
(dil.)
32.60
25.22
33.88
29.77
52.00
0.11
0.13
0.12
36.61
37.86
3.77
44.60
69.63
7.97
58.08
0.14
9.78
62,14
214.06
0.19
9.18
0.09
77.43
32,02673

gi/ciA3
gi/iA3


-------
                      CHEItlCftL PROPERTY ESTIHATION FOR SYNTHETIC
                      6ASOLINE AND CONSTITUENTS AT 20 DEGREES C.
en
REPRESENTATIVE PERCENT
CHEH1CAL COMPOSITION
Isotmtan*
n-Butane
Isopentane
n-Pentane
n-Octane
Benzene
Toluene
lylene (i)
n-Hexine
2-Nethylpentane
Cyclohexane
n-Heptine
2-ltethyUexane
Hethylcyclohexane
2,4-Diiethylhexane
Ethylbenzeiie
l-Pentene
2,2,4-Triiethylhexane^
2,2,5,5-Tetruethylnemne
1,4-Diethylbenzene
Hexene
1,3,5-Triiethylbenzene
C12-»liphatic
Total
Teiperiture > 293,15 deg
Pressure • 740 u
6i> Constant > 0.0623 11
2
1
K
3
1
3
5
7
1
8
3
1.5
5
1
e
2
1.5
2
l.S
5
1.5
5
10
100
GRAN HOI.
HEIGHT
IGH/HOU
58.12
58.12
72.15
72.15
114.23
78.11
92.14
106.17
86.18
86.18
84.16
100.20
100.20
98.19
114.23
106.17
70.14
128.26
142.29
134.22
84.16
120.20
170.00
102.20
. K
H9
Hg»iA3/ioltK
US. PHASE AIR DIFFUSION LIQUID
HOL FRACT. COEFFICIENT DENSITY
(CHA2/SEC> (6H/CIT3)
0.0326
0.0163
0.1840
0.0394
0.0083
0.0364
0.0515
0.0625
0.0990
0.0880
0.0338
0.0142
0.0473
0.0097
0.0664
0.0179
0.0203
0.0148
0.0100
0.0353
0.0169
0.0394
0.0558
1.0000
Molecular Height
Average Holecular
Gasoline Vapors
Holecular Height
0.0911
0.0911
0.0817
0.0817
0.0598
0.0905
0.0824
0.0668
0.0711
0.0711
0.0719
0.0659
0.0659
0.0666
0.0617
0.0733
0,0829
0.0608
0.0577
0.0574
0.0719
0.0628
0.0524

of Air > 28.96
Height of
69.48
of
Basoline-Air Kixture * 43.58
0.5570
0.5790
0.6200
0.6260
0.7030
0.8850
0.8670
0.8640
0.6590
0.6530
0.7790
0.6940
0.6790
0.7707
0.7000
0.8670
0.6400
0.7173
0.7200
0.8620
0.6730
0.8650
0.8600
0.7373
51/10!
51/10!
gt/iol
PURE CHEHICAL
VAPOR PRESSURE
(u Hg)
2252.75
1555.33
574.89
424.38
10,46
75.20
21.84
6.16
121.24
171.50
77,55
35.55
51.90
36.21
23.32
7.08
530.80
11.30
6.47
0.70
149.97
1.73
0.08

PARTIAL PRESSURE PURE CHEHICAL CONCENTRATION OVER
OVER GASOLINE VAPOR DENSITY LIQUID GASOLINE
til Hg) (6H/HA3) (ppi)
73.5178
25.3788
105.7925
16.7346
0.0869
2.7390
1.1237
0.3852
12.0080
15.0985
2.6218
0.5047
2.4563
0.3497
1.5492
0.1264
10.7654
0.1671
0.0647
0.0246
2.5348
0.0681
0.0042
274.1020
Vipor Density of Basoline-
Air HUture «
lighted Average Air
Diffusion Coefficient «


7162.14
4944.83
2268.97
1674.92
65.37
321.31
110.05
35.78
571.57
808.51
357.04
194.86
284.50
194.50
145.75
.41.09
2036.57
79.30
50.35
5.11
690.40
11.36
0.70

1811.60 91/1*3
0.0726 ciA2/sec

96733.9
33393.1
139200.7
22019.2
114.3
3604.0
1478.6
506.8
15800.0
19866.4
3449.7
664.1
3232.0
460.2
2039.4
166.3
14165.0
219.9
85.1
32.4
3335.3
89.7
5.5
360660.5
Helghted average
liquid density *
Heighted average
gas density »

BOILING
POINT
(deg. K)
261.25
272.65
300.95
309.15
398.75
353.25
383.75
466.25
341.85
333.35
353.85
371.55
363.15
374.05
382.55
409.25
303.05
399.65
410.55
456.85
336.55
437.85
489.15

0.718
1042

HENRY'S LAH
CONSTANT
(dii.)
43.04
34.04
47.93
43.00
93.30
0.18
0.21
0.22
56.64
57.24
5.94
74.13
112.83
13.00
98.40
0.25
13,94
109.70
393.73
0.40
14.01
0.18
198.23
54,34632
gi/MA3
gi/iA3


-------
CHENICAL PROPERTY ESTIHflTION FOR SVHTHETIC
GASOLINE AND CONSTITUENTS AT 30 DECREES C.
REPRESENTATIVE
CHEHICAL
Isobutane
n-Butane
Isopentane
n-Pentane
n-Octane
Benzene
Toluene
Xylene (il
n-Hexane
2-Hethylpentane
Cyclohexane
n-Heptane
2-Hethylhexane
Hethylcyclohexane
2,4-Diuthylhexane
Ethylbenzene
1-Pentene
2,2,4-Triiethylhexane
PERCENT 6RAH HDL
COMPOSITION HEIGHT
(6H/HOLI
2
1
14
3
1
3
5
7
9
8
3
1.5
5
1
8
2
l.S
2
2,2,5,5-Tetraiethylhexane l.S
1,4-Diethylbenzene
1-Hexene
1,3,5-Triiethylbenzene
C12-aliphatic
Total
Teaperajure « 303. 15
Pressure * 760
Gas Constant = 0.0623

5
1.5
5
10
100
deg. K
it Hg
n Hg«i*3/iol*K

SB. 12
58.12
72.15
72.15
114.23
78.11
92.14
106,17
86.18
86.18
84.16
100.20
100.20
98.19
114.23
106.17
70,14
128.26
142.29
134.22
84.16
120.20
170.00
102.20


. LIQ. PHASE
KOL FRACT.
0.0326
0.0163
0.1840
0.0394
0.0083
0.0364
0.051S
0.0625
0.0990
0.0880
0.0338
0.0142
0.0473
0.0097
0.0664
0.0179
0.0203
0.0148
0.0100
0.0353
0.0169
0.0394
0.05S8
1.0000
Molecular Height
Average Holecular
AIR DIFFUSION
COEFFICIENT
(CHA2/SEC)
0.0966
0.0966
0.0867
0.0867
0.0634
0.0960
0.0874
0.0708
0.0754
0.0754
0.0763
0.069?
0.0699
0.0706
0.0655
0.0777
0.0879
0.0644
0.0612
0.0630
0.0763
0.0666
0.0556

of Air * 28.96
Height of
Gasoline Vapors - 70.03

flclecular Height
of
Gasoline-Air Hixture = 49.94
LIQUID
DENSITY
(GH/CH"3)
0.5400
0.5640
0.6084
0.6151
0.6962
0.8740
0.8580
0.8578
0.6502
0.6437
0.7693
0.6764
0.6711
0.7622
0.6927
0.8591
0.6282
0.7104
0.7135
0.8556
0.6636
0.8580
0.8543
0.7286
gi/iol

gi/iol
PURE CHEHICAL
VAPOR PRESSURE
In Hg)
3009.32
2120,42
818.02
615.42
18.46
119.33
36.66
11.05
187.12
259.27
121.78
58.54
83.54
58.65
39.07
12.61
760.91
19.65
11.67
1.42
228.71
3.33
0.18

Vapor Density of
Air Mixture =
Heigh ted Average
PARTIAL PRESSURE PURE CHEHICAL CONCENTRATION OVER
OVER GASOLINE VAPOR DENSITY LIQUID GASOLINE
(M Hg) (6H/IT3I (ppi)
98.2082
34.5996
150.5330
24.2677
0.1532
4.346S
1.8869
0.6909
18.5328
22.8249
4.1170
0.8311
3.9535
0.5664
2.5953
0.2254
15.4324
0,2906
0.1166
0.0500
3.8659
0,1315
0,0101
388.2295
Gasoline-

Air
Diffusion Coefficient -
gi/iol


9251.90
6519.05
3122.04
2348.77
111.51
493.05
178.70
62.05
853.05
1181.93
542,17
310.30
442,80
304.61
236.11
70.84
2823.16
133,31
87.81
10.06
1018.20
21.19
1.64


2007.64 g«/«A3

0.0770 cn'2/ssc

129221.4
45525.8
198069.8
31931.1
201.6
5719.1
2482.7
909.0
24365.2
30032.8
5417.1
1093.6
5201.9
745.3
3414.B
296.5
20305.8
382.3
153.5
65.8
5086.7
173.0
13.3
S1082B.4
Weighted average
liquid density =
Heighted average
;as density -

BQILIHG
POINT
(deg. K>
261.25
272,65
300.95
309.15
398.75
353.25
383.75
466.25
341.85
333.35
353.85
371.55
363.15
374.05
382.55
409.25
303.05
399.65
410.55
456.85
336.55
437.85
489.15


0.709

1436

HENRY'S LAH
CONSTANT
(dii.)
5S.60
44.88
65.95
60.29
159.16
0.28
0.35
0.37
84.53
83.68
9.03
118.05
175.61
20.36
159.40
0.43
19.33
184.41
686.71
0.80
20.66
0.33
463.43
91.48287

g«/ciA3

g«/BA3


-------
                                APPENDIX B
                     AIR EMISSION ESTIMATION EQUATIONS
This appendix contains the development of the equations used to estimate
the air emission rates for soil excavation, vacuum extraction and air
stripping.
B-l  Soil Excavation

The transport processes of advection and diffusion contribute to the
overall flux of vapors from an excavated soil pile.  Although research by
Fukuda (1955) has indicated that wind eddies can increase the advection of
vapors through soil, this movement has been shown to be small when compared
to other transport processes (CDM, 1986).  For this reason, it is assumed
that the flux of vapors is controlled by the process of diffusion.

Molecular diffusion in one dimension is described by the following form of
Pick's First Law (PEI, 1988):
                                                          (B-l)
where: J       = the flux of vapors [lb/(ft -hr)J
               = the effective diffusion coefficient [ft2/hr]
               = the concentration gradient in the vertical direction
                 [lb/ft4]
        ef f
The effective diffusion coefficient can be defined in the following manner
(PEI, 1988):
                 D
                  ef f
                        D .   T 0 .
                         air    air
(B-2)
                                    B-l

-------
where: D
        air
       ©air
the diffusion coefficient [ft /hr]
tortuosity,  (0 < T < 1) [dimensionless]
the air filled porosity (expressed as a fraction of 1)
[dimensionless]
The air diffusion coefficient is adjusted by the tortuosity and air filled
porosity in order to account for diffusion through a porous medium.  The
path length of the diffusing molecule increases as it wanders through the
air filled pores of the soil (COM, 1987b).

It follows that the more tortuous the path, the longer it takes the
molecule to diffuse through soil, effectively reducing the air diffusion
coefficient by a greater amount (PEI, 1988).  There are many expressions
that can be used to calculate the,tortuosity, however the Millington and
Quirk (1961) expression that is used here has a theoretical basis, and
tends to be preferred (PEI, 1988):
                            1/3
                                  F ©air V
                                   total
                                        J
                                                                   (B-3)
where: ©,
        total
                                  < D
       0
        air
the total porosity, (0 < ©total
the air filled porosity
 total ~  water
where ®water = the moisture content of the soil
Different soil types typically have different values of total porosity and
moisture content.  For the formulations presented here, a soil type having
a medium grain size (such as a sand) was selected.  The soil was assumed to
have a total porosity of 0.35, and moisture content of 0.08.  These numbers
correspond to average values that might be expected for a moist sand, and
were obtained from graphs and tables given in Freeze and Cherry (1979) and
COM (1986).
                                    B-2

-------
The steady state form of the concentration gradient can be written as:

= (c         — c
    soil gas    atmosphere
                                                               (B-4)
where: C
        'soil gas
        "atmosphere
          concentration of the contaminant in the soil pores
          [lb/ft3]
                           l_3 T
          concentration of the contaminant in the air above the
          soil pile [lb/ft3]
          the depth of the soil over which the concentration
          gradient exists [ft]
It is assumed that the wind is sufficient to mix the contaminants that
diffuse out of the soil with clean air, thereby reducing the concentration
of the contaminants in the air above the soil.  Hence, Catmos here = 0.
This has the effect of maintaining a steep concentration gradient across
the soil-air interface, and consequently a maximum flux of vapors is
maintained.
The steady state concentration gradient takes time to establish, however
the thinner the soil layer thickness being considered, the less time is
required.  The assumption of a thin soil layer thickness again allows for
the calculation of a maximum flux of vapors.  For the calculations
presented here, a soil layer thickness of 0.5 inches was assumed.

When liquid gasoline and air are present in the soil pores, concentrations
in the vapor phase are controlled by the gasoline composition and the vapor
pressures of the constituents (Stephanatos, 1988).  These concentrations
are temperature dependent, as is the air diffusion coefficient.  For
temperatures of 32°F, 50°F, 68°F, and 86°F the values of the maximum
gasoline VOC and benzene concentrations in (mg/1) are:
                                   32°F
                                  50°F
68°F
86°F
         gasoline concentration    757
                                  1091
1532
2099
         benzene concentration     4.4
                                  7.3
11.7
17.9
                                    B-3

-------
Values for the air diffusion coefficients for different gasoline
constituents are tabulated in Appendix A.

Substituting the definitions for the effective diffusion coefficient and
the concentration gradient allows Equation B-l to be rewritten in the
following form:
              ©air
                         \!/3
                                          "soil gas
(B-5)
Equation B-5 allows the calculation of the flux in units of [lb/(hr-ft )].
In order to obtain the emission rate in units of [Ib/hr], the flux must be
multiplied by the surface area of the soil pile, as follows:
              ER = J x A
(B-6)
where: ER = emission rate [Ib/hr]
       J  = vapor flux [lb/(hr-ft2)]
       A  = surface area of the soil pile [ft2]

In order to simplify this process, it is assumed that the soil pile can
have one of two basic shapes: a horizontal layer, or cone.  For the
horizontal layer it is also assumed that there is no flux of vapors out of
the sides of the pile.  One of the following two equations can be used to
calculate the surface area of a horizontal pile:
     Surface area = (total volume)/(height)
(B-7)
     Surface area - (length) x (width)
(B-8)
The following equation can be used to calculate the surface area of conical
soil piles:
     Surface area = 3.142 x r x /r2 + h2
(B-9)
Here, r is equal to one-half of the diameter at the base of the soil pile.
                                    B-4

-------
If the height of the pile is approximately equal to the radius at the base
of the soil pile/ the equation for the surface area can be simplified
considerably:
     Surface area = 1.11 x (diameter)'
(B-10)
B-2  Vacuum Extraction

The following equation is used frequently by industry to estimate air
emission rates from a single vacuum extraction well (Johnson, 1988):
                                           7
     ER = (Q x Cso il gas x MW x 1.581 x 10  )
(B-ll)
                     gas

where: ER = emission rate [Ib/hr]
       Q  = pumping rate [cfm]
       C  = soil gas concentration [ppm-v]
       MW = molecular weight of contaminant [Ib/lb-mole]

The constant (1.581x10~7) has units of [(lb-mole min)/(ft3 ppm-v hr)] and
was derived in the following manner:
                       60min
                                              _
           10- ppm-v    1 hr    379.5 ft3

 (Source: Perry's Chemical Engineers' Handbook, 1984)
API (1985) gives a similar form of this equation to calculate emission
rates.
                                    B-5

-------
B-3  Air Stripping

Emission rates are a function of influent concentrations and the rate of
flow to the air stripper.  The form of the equation that can be used to
calculate emission rates is (Radian, 1987):
       ER
     (Q x C x RE x 5.042 x 1CT4 )
                                                           (B-12)
where: ER » the emission rate in [Ib/hr]
       Q  = the groundwater pumping rate in [gpm]
       C  = the concentration of the contaminant in the groundwater in
            [mg/1]
       RE - the removal efficiency, expressed as a fraction of 1 (for
            example, a removal efficiency of 95% =0.95)
and (5.042 x 10"4) is a constant having units of
[(Ib liters min)/(mg gal hr)] and is derived in the following manner:
2.2 Ib
                1000_liters   60_min
       10  mg    261.8 gal     1 hr
                                         5.042 x 10"
                                   B-6

-------
                              APPENDIX C
                  SOIL GAS CONCENTRATION MEASUREMENTS
To convert soil gas concentrations given in //g/1 to ppm-v (parts per

million-volume) the following equation should be used (Geoscience

Consultants Limited, 1988):
    ppm-v   = C
                    (RT) x (1)  x  1000 liters
                                                  (C-l)
                           MW
                                          m
where:   C = soil gas concentration of contaminant [/yg/liter]
         R = gas constant                               .
           = .06236 (mm Hg m3)/(mole °K)
         T = temperature [°K]
           = 273.15 + °C
                      (°F-32)(f)
  = 273.15 +
P = atmospheric pressure [mm Hg]
  = 760 mm Hg approximately
MW= molecular weight of contaminant
  = 78.11 g/mole for benzene
  = 102.2 g/mole for gasoline
To calculate soil gas concentrations from total soil concentrations:


Assume that there is equilibrium partitioning of the chemical
(gasoline or benzene) between the soil air, soil moisture and the soil

organic carbon.  Under this assumption there will be no sorption of
the chemical vapors onto the dry soil.  The total soil concentration
is therefore composed of the concentration in the soil gas, soil
moisture and the amount sorbed onto the soil organic matter.  This
relationship can be expressed as follows:
    c      = c    4- r         j- c
    ^total    gas T  moisture T  soil
or by rearranging terms:
                                                  (C-2)
    C    = C      — C         — C
     gas    total    moisture    soil
                                                  (C-3)
                                 C-l

-------
The following relationships can be used to determine the contaminant
concentrations in the gas, moisture and soil:
        Cgas [atm]
           H [atm-m /mol]
    "moisture
             [g/m3]
             MW [g/mol]
                                                           (c-4)
where: H  = Henry's Law Constant
       MW = molecular weight of the contaminant
      "soil
    "moisture
              [mg/1]
        -   Koc
                                                           (C-5)
where: f   - organic carbon content of the soil, (0 < foe < 1)
       K°c = organic carbon normalized soil/water partition
             coefficient
       soii
                                        MW[ g/mol]
       Cgas  [atm]
                                                            (C-6)
               H [atm-m /mol]
An estimate for K   can be obtained from the following expression:
    log K
log K   - .21
where:  K
octanol/water partition coefficient
Experimental values of KOW are tabulated  in the literature.  The
fraction of organic material  in the  soil  can be measured  in  the field,
or it can be roughly approximated.   The molecular weight;  and Henry's
Law constants  for  the synthetic gasoline  blend are  listed in Appendix
A.
                                  C-2

-------
      APPENDIX D
UNITS CONVERSION TABLE

To convert



To convert



To convert







To convert




in
cm
m

in2
cm
m2

ft3 /sec
gal/min
liters
mm
liters
sec

gal/sec
ftVntin
ft3 /sec
liters
mm
liters
sec

to



to



to







to





ft



ft2



ft3 /min







gal/min





Multiply by



Multiply by



Multiply by







Multiply by




.083
,£
.033
3.3

.069
.0011
10,89

60.
.134

.035

2.1

60.
7.48
448.8

.26

15.6
         D-l

-------

-------

-------

-------