EPA 909/9-81-004
 ANALYSIS OF POTENTIAL METHODS
   TO DETERMINE VOLATILITIES OF
          HEAVY CRUDE OILS
              DECEMBER 1981
              FINAL REPORT
g.. - ,:; -x-- .:; .v., _,;;;:;., .y,.-;.-,-,,, ,-, yoy.v,,,. •^•'•'--';^ V: ::;.•. \; ;•; .;...: ::'•: :•:•••• ..;•;• . '•
  U. S. ENVIRONMENTAL PROTECTION AGENCY
               REGION IX
           215 FREMONT STREET
     SAN FRANCISCO, CALIFORNIA 94105

-------
                                       EPA 909/9-81-004
                                       December 1981
   ANALYSIS OF POTENTIAL METHODS
     TO DETERMINE VOLATILITIES
        OF HEAVY CRUDE OILS
                by:

          Robert J. Bryan
        ENGINEERING-SCIENCE
      Contract No. 68-02-3509
       Work Assignment No. 11
          Project Officer:

             Don Harvey
           Prepared for:

U.S. ENVIRONMENTAL PROTECTION AGENCY
             REGION IX
         215 FREMONT STREET
  SAN FRANCISCO, CALIFORNIA  94105
           December 1981
        ENGINEERING-SCIENCE
     125 West Huntington Drive
     Arcadia, California  91006

-------
                               DISCLAIMER

     This report has been reviewed by the U.S. Environmental Protection
Agency, Region IX, and approved for publication.  Approval does not
signify that the contents necessarily reflect the views and policies
of the U.S. Environmental Protection Agency, nor does mention of trade
names or commercial products constitute endorsement or recommendation
for use.
                                   ii

-------
                                CONTENTS
Figures	iv
Tables	 v

     1.  Introduction	 1
     2.  Background	 2
     3.  Approach	10
     4.  Discussion	11
              Alternative Methods for Crude Oil Volatility
                Cons idered	12
     5.  Conclusions and Recommendations	19

References	22
Non-Referenced Contacts	22
Appendices

     A.  Santa Barbara Modified RVF Approach	A-l
     B.  California Air Resources Board Vapor Composition Method....B-l
     C.  Chevron Vapor Composition Method	C-l
     D.  Evaporation Method	D-l
     E.  Equipment Costs	E-l

Report Documentation Page
                                  iii

-------
                                FIGURES


Number                                                              Page

 1       Extrapolation of RVP versus TVP nomograph from API 2518	5


                                                r   p  "i
 2       Hydrocarbon vapor concentration versus L14.7-P-I 0.68
         for crude oil standing storage tanks	6

 3       Emission concentration versus crude oil RVP, standing
         storage tanks	7

 4       Emission concentration versus crude oil TVP, standing
         storage tanks	8
                                 TABLES
          Measures of Heavy Crude Oil Volatility Versus
          Evaluation Criteria	18
                                   iv

-------
                               SECTION 1
                              INTRODUCTION

     This project, titled "Analysis of Potential Methods to Determine
the Volatilities of Heavy Crude Oils", was conducted under Contract No.
68-02-3509, Work Assignment No. 11.
     The purpose of the project was to screen alternatives for measuring
the vapor pressure of heavy crude oils considering the cost of method
development, equipment, and laboratory costs for using the developed
method, repeatability, and the degree of conformity with API Nomograph
results.  Four alternatives including (1) extrapolation of the API
nomograph, (2) measuring equilibrium vapor pressure, (3) using the Reid
Method with modifications, and (4) measuring weight loss with time were
to be initially considered.  A method for measuring evaporation loss
similar to (4) above was also to be studied for use in calculating
evaporative emissions for open tanks and basins.  Several other
approaches to determining heavy crude oil vapor pressure were added to
the list for preliminary evaluation.

-------
                               SECTION 2
                               BACKGROUND

     Compliance with the Clean Air Act Amendments of 1977 require that
a number of states, including California, prepare and submit by July 1,
1982 a plan detailing the control measures which will be implemented to
achieve attainment of the NAAQS for ozone by 1987.  All such plans will
depend heavily on comprehensive inventories of volatile organic compounds
(VOCs) as a basis for determining the level of control of such emissions
needed to attain compliance with the standard.
     Compilation of an accurate VOC inventory will be necessary, but
especially difficult, for those counties in California in which a
substantial portion of such emissions emanates from crude oil production
and associated storage and transfer.  Determination of emissions from
these sources requires the measurement of the volatility of the crude
oil at the source temperature.  This measurement in turn is dependent
upon a knowledge of the "true vapor pressure" (TVP) of the crude oil.
     Unfortunately, methods currently available for making such
measurements (principally the Reid Vapor Pressure [RVP] method) are
not satisfactory for measuring the TVP of crude oil because its normal
source temperature is usually too high and its RVP too low.  Therefore,
estimates of emissions from sources associated with the production of
crude oil are of questionable accuracy.   Further, it is not possible to
handle some high viscosity crudes at temperatures specified in the ASTM
Method for RVP.
     Several examples are cited to illustrate the problems associated
with determining the vapor pressure of heavy crude oil or with use of
vapor pressures to predict hydrocarbon emission losses from storage and
trans fe r.
     (1)  In Kern County, California,  an assumption is made for use in
          preparing the 1982 SIP revisions that all crude oil has  the
          same vapor pressure as kerosene (reference personal communica-
          tion between Larry Landis,  Kern County APCD,  and Stephanie
          Fullmer, Engineering-Science,  Inc.,  April 20,  1981) While

-------
     this may be useful for preparing countywide emission inven-
     tory estimates, the assumption is based upon a limited number
     of vapor pressure-temperature plots.  Further, there is a
     weakness in comparing a narrow boiling range product such as
     kerosene with unprocessed crude oils.
(2)  The Second Edition of API Publication 2517, "Evaporation Loss
     from External Floating-Roof Tanks"1), discusses the devel-
     opment of a product factor for use in their standing storage
     loss equation.  The basis for this factor was data from
     pilot tank tests conducted with an octane/propane mixture,
     gasoline, and crude oil.  It was found that after normalizing
     for differences in true vapor pressure and vapor molecular
     weight, crude oil losses were significantly lower than the
     octane/propane losses under the same conditions of seal
     configuration and wind speed.  This was attributed to evapor-
     ation at non-equilibrium conditions, caused by the slower
     rate of migration of light ends from the bulk liquid to the
     liquid surface in crude oil as compared to the octane/propane
     mixture because the light ends migration rate is strongly
     viscosity dependent.
     The average of crude oil losses to octane/propane losses was
     0.4, with a range of 0.3 to 0.6.   Therefore, it is obvious
     that factors other than the ability to obtain a reliable
     vapor pressure method for heavy crude oil must be considered
     in developing emission estimate procedures.
(3)  Data on crude oil vapor pressures, hydrocarbon emissions during
     marine loading of crude oils, and  composition of emissions were
     reported in a study conducted by Chevron Research Company
     for the Western Oil and Gas Association.2)  Vapor pressures
     were calculated from liquid compositional data using various
     vapor-equilibrium correlation procedures.   In all cases
     these vapor pressures were higher  than those predicted  by
     the RVP-TVP correlation of API Bulletin 2514,  "Hydrocarbon
     Emissions from Marine Vessel Loading of Gasolines",  1976.
     After recalculating on a methane  and ethane free basis, the

-------
     vapor pressures obtained from vapor-liquid equilibrium proce-
     dures compared favorably with the API correlation of TVP
     with RVP for the total (methane and heavier) crude oil.
     This suggests that substantial amounts of light ends are
     lost during the sampling and analysis of crude oil for vapor
     pressure by the RVP method even when substantial care is taken.
     Also, the RVP procedure tends to slightly de-emphasize the
     influence of methane and ethane.
     The emissions composition data obtained in the WOGA marine
     loading of crude oil study are also of interest.  On the
     average, propane and heavier components comprised 95 weight
     percent of the hydrocarbon emissions.  On a volume basis,
     the propane and components account for about 85 percent of
     the emissions.  Therefore, volumetric vapor concentrations
     and thus vapor pressure can be significantly affected by
     methane and ethane content of crude oil.  It follows that
     the value of vapor pressure as a predictor of emissions on a
     mass basis can be impaired by unknown variability in light
     ends because the mass concentration of hydrocarbon vapors
     present over crude oil is much less affected by light ends
     than is the volumetric concentration.
(4)  Several studies in recent years have shown that hydrocarbon
     vapor concentration is not well correlated with various
     measures of crude oil vapor pressure.  Engineering-Science
     (ES), in a study for the Western Oil and Gas Association^,
     prepared an extrapolated version of the API TVP-RVP nomograph
     (Figure 1).   This nomograph was used to determine TVP (accor-
     ding to the API terminology).   The RVPs, TVPs,  and a function
     of vapor pressure used in API Bulletin 2518 for prediction
     of breathing losses in fixed-roof tanks were all plotted
     versus measured hydrocarbon vapor concentrations.   These
     plots are reproduced in Figures 2, 3, and 4.   It is  interes-
     ting to note while there is a good deal of scatter in all
     the plots,  the RVP shows the highest correlation coefficient
     of the three parameters considered.   It should  be  recognized,

-------
           TRUE VAPOR PRESSURE  IN POUNDS PER SQUARE  INCH  ABSOLUTE


|l I I  I | I  I I I  |l|l|l|l| I |I  | I |  I  | I | I  |llll|llI I  | I I  I I  Jill I| I I I I |  I I  I  I   I
                             REID VAPOR PRESSURE

                       i  i i  i i  I  i—i—i—i  i  i I  i  | i
                  OOOOOOOOOO
                                                   iiiliiiiliiiiliiiiliiiiliilJiiiilii.iliiiiL
                     TEMPERATURE  IN  DEGREES  FAHRENHEIT                          ~^
                                                                                  C3 X
                                                                                  I— —t
                                                                                  *• 3D
                                                                                  — ^ »


Figure 1.   Extrapolation  of RVP  versus TVP  nomograph  from  API  2518.   g

-------

=«:
00
m
3D
3D
rn
3D
CO
m r^-
CTi co -b>
— I
<= -o
X 1 	
CD
m

TOTAL HYDROCARBON CONCENTRATION, lbs/1000 scf, THC
0
0
0
a
	 ' o
an
o
CJI
a
Figure 2 .
° « S S S SSSSS;











\
I •_
u





N



\
\
\











\
\
\










W
\
X\^,
—1 O
>• ao
a m
» r-
m ^
••IS
0 11
n -n
m o
-^ m
^7
ffl ^3
— Ul
a>



*

t

AK«H







x« i
X\4\-
Nf <^y


Hydrocarbon vapor concentration versus U
for crude oil standing storage tanks. L1
V
\

p 1
4 . 7 - P J










\
\
\





m
01















0.68
Ul
o





-------
                            TOTAL  HYDROCARBON CONCENTRATION, lbs/1000 scf, THC
m
so
«">
      rn
      c-o
      CO  *•
      •a
      CO





















V
>
•
•



















N
\
\
\
N





















X\
\





















V
v
N •
'K
,A











IB
\
' '- X


IB





'*v





















— IO
zai
» r-
°^
l-i O
TO
on
m
o-n
TI -n
mn
— z
m o
\ --
>s





\
\


















5


\
\
\





















       Figure 3.   Emission concentration  versus crude  oil  R\IP,  standing  storage tanks

-------

ft °
oo
m
aa
3D
m
-n
3 i ~
— « m
O
m ^
C/l TJ
—1 O
=c so «
00 X JJg
rn m
30 CXI
CXJ
m
•a
CO
Q> CJ1
-o
a>
Figur

N






TOTAL HYDROCARBON CONCENTRATION. Ibs/IOOO scf. THC
w

~
N




N






^\






&
\j
*W




5
';







•
\
\
N



in
-4
a» c

>
O 3D
» 30
aam
°C «
m — « A
3D —
30 <=>
0 Z
3D
0 0
•n m
m -n
5 ±2
K m
m
M
\- ~~ *
X. " c
^ *J C
N


1
3
n
D
xx

>. l
a c





\
\
n
3
B 4. Emission concentration versus crude oil TVP, standing storage tanks.

-------
     however, that there are a number of factors other than vapor
     pressure which influence vapor concentration in a tank head
     space.  These include approach to equilibrium, presence of
     leaks, and immediate tank history as to product characteristics.
(5)  A special problem exists with unstabilized crudes.   Crude oil
     from wells can be subjected to a number of treatments to
     separate gas, water, and contaminants from the crude oil.
     Operations are performed in oil/gas separators, heater
     treaters, free water knockout tanks, and wash tanks to name
     several of these process vessels.  Some of these tanks may be
     pressurized,  others are at or close to atmosphere pressures.
     In some situations the crude oil is hot enough for  flashing
     to occur.  In oil field tank operations conducted prior to
     custody transfer (the last transfer from the field  to a means
     of transportation), then, it is quite possible for  the
     concentration of light ends, and thus the vapor pressure, to
     change quite  rapidly along the processing steps,  and even with
     time in any given tank.   This obviously complicates the
     problem of collecting representative crude oil samples.
     Further,  when the crude  oil is held at temperatures sig-
     nificantly above ambient  temperature and turn-over  is rapid,
     there  exists  a significant potential for temperature strati-
     fication  in the  tank,  thus raising  a question as  to what
     temperature is to be selected for reporting  true  vapor pressure
     or other  measure of volatility.

-------
                               SECTION 3
                               APPROACH

     This project was based upon'review of literature and direct contact
with knowledgeable persons in the field.  No laboratory or field work
was conducted.  The project was divided into two phases.  Phase I
covered the screening of alternative methods for measuring the vola-
tilities of heavy crude oils other than the four approaches specified
in the Work Assignment.  The principal alternatives reviewed during
Phase I were methods which had been the subject of a staff study by
the California Air Resources Board.  Verbal suggestions were obtained
from several petroleum industry sources which generally addressed the
need for some technique to account for the influence of methane and
ethane when measuring vapor pressure of crude oils.  A report was
submitted on the Phase I effort.  Following submission of this report,
a specific written procedure was submitted by Chevron Research Company
for the determination of vapor composition and vapor pressure of heavy
crude oils.  This method was also considered.
     The methods selected for further study were evaluated using the
following criteria:
     0  Cost of method development
     0  Equipment and laboratory costs for using the developed method;
     0  Repeatability of the method
     0  Possible sources of error
     0  Degree of expected conformity with API Nomograph results
        Problems with use as a predictive parameter for hydrocarbon
        emissions
     0  Problems of field and laboratory use
     As a result of this analysis, the methods were rated as  candidates
for further developmental work.
                                   10

-------
                               SECTION 4
                               DISCUSSION

     This project is concerned with potential methods of determining
the volatilities of heavy crude oils.  Volatility is a term relating to
the potential of a substance to evaporate.  Vapor pressure is probably
the most common measure of volatility but measures of evaporation such
as distillation have also been used.  Vapor pressure, for our purposes,
is defined as a measure of the force that tends to volatilize any
volatile liquid.  This force is conventionally measured as pressure
under specified conditions.  The term "true vapor pressure" has a
special meaning in multi-component mixtures because composition of the
mixture will be changed as vaporization occurs and vapor pressure will
be lowered concurrently.  Thus, true vapor pressure of mixtures must
be measured without vaporization taking place.  This is difficult to
carry out experimentally with a high degree of accuracy.   True vapor
pressure is also termed as the "bubble point" pressure.
     Because of the experimental difficulties in directly determining
true vapor pressures, the Reid Vapor Pressure (RVP) is most often used
to express the vapor pressure of petroleum mixtures.  The RVP is the
absolute pressure in pounds per square inch determined at 100°F and
V/L = 4 (ratio of vapor volume to liquid volume, as defined in ASTM
Designation : D323-72) by using the apparatus and procedures as stan-
dardized under the auspices of the American Society for Testing Materials.
Thus, the RVP represents the vapor pressure of a sample which has had
its composition changed during the vaporization required to saturate
the vapor space of the measuring apparatus.
     Correlations have been developed relating true vapor pressure to
Reid vapor pressure.  These are prepared in the form of nomographs
included in a number of American Petroleum Institute (API) publica-
tions. ' For gasoline and finished petroleum products, the alignment
charts include RVP, temperature, the slope of the ASTM distillation
curve at the 10 percent point, and the "true vapor pressure".   For
crude oil, however, there is no provision for consideration of the
                                   11

-------
distillation curve slope.  This alignment chart  is limited to crude
oils with RVPs above 2 psi and storage temperatures below 140°F.
There is still another problem having to do with the use of ASTM Method
D323-72 for RVP with some crude oils.  This is the requirement that
the sample container and its contents be cooled  to 32 to 40°F before
the container is opened and the contents transferred to the liquid
chamber of the RVP bomb (a term for the apparatus).  Some crude oils
cannot be poured at these temperatures.  The alternative methods for
crude oil volatility considered in this project  take into account one
or more of the problems discussed previously.  They include modifying
the TVP-RVP nomograph to take into account distillation slope or gravity,
methods of minimizing transfer loss, extending the applicable temperature
and vapor pressure range, and determining the methane and ethane influence
on vapor pressure.
Alternative Methods for Crude Oil Volatility Considered
     Each of the alternative approaches considered as a measure of
heavy crude oil volatility is discussed below.   Some of the approaches
are in the form of fairly detailed methods and some are conceptual
only.  Those in the former category are attached in the appendix to
this report.   Following the narrative discussion of the methods, a
matrix chart rating the methods against certain evaluation criteria is
given.
     1.   Extrapolation or Modification of the API Nomograph
         This approach would require the acquisition of more data on
         true vapor pressures versus Reid vapor pressures covering a
         variety of crude oils.   Possibly the minimum RVP could be
         lowered and the temperature limit raised.   This would extend
         the range of applicability but would not solve sampling, crude
         oil transfer,  and light ends problems.  It is possible that
         sufficient data could be obtained that ASTM distillation curve
         slope or API gravity could be included to  refine the TVP-RVP
         nomograph.  The Chairperson of the API Committee on Evaporation
         Loss, Karen Hanzevack of Exxon Research, stated (personal
                                   12

-------
communication with Robert Bryan of Engineering-Science,  Inc.,
July  17,  1981)  that  the Committee is not satisfied with  the
existing  crude  oil nomograph and is collecting existing  data
on a  wider  range of  crude oils.  The intent would be  to  add a
grid  of lines for crudes of different API gravities as a
substitute  for  the single correlation line in the current
chart.  The Chairperson further gave the opinion that no
measure of  volatility better than the RVP method has  yet been
developed.  She believes field work is necessary to confirm
the value of any revised procedure for vapor pressure deter-
mination.
The WOGA  sponsored marine crude oil loading study also com-
ments on  this matter.  Their report^) states that an  alignment
grid which includes  both temperatures and slope of the ASTM
distillation curve or API gravity would improve the present
API nomograph for obtaining TVP from RVP.
The method development and equipment costs would be relatively
low using this approach if existing data were to be used to
construct a new TVP-RVP nomograph for heavy crude oil as the
actual test procedures would not be changed.  The limitations
on sampling and handling heavy crude oils would remain.
Direct Measurement of True Vapor Pressure
Essentially no support was found for pursuing this approach.
The method requires  use of an air-free sample and a V/L  ratio
approaching zero.  There is no standardized method for the
measurement of true  vapor pressure.   On a theoretical basis,
the method is very sensitive to methane and other light ends
content of the stock being measured.   Thus the sampling
procedure would be even more critical than with the RVP method.
Methods development cost would be rather high.   Equipment
costs may not be too high as no compositional data are speci-
fically required.  Laboratory costs  may be high depending
upon the care needed to obtain accurate and repeatable results.
                          13

-------
3.  Modified RVP Method (Santa Barbara Approach)
    This method involves collecting the crude oil sample directly
    in the liquid container portion of the vapor pressure apparatus
    thus eliminating the transfer of sample after collection.  It
    is run at stock storage temperature.  This eliminates the
    problem implicit in ASTM D323-72 requiring cooling of sample
    to 32-40°F before transferring it from the sample container to
    the test apparatus liquid chamber.  Thus, samples which cannot
    be poured at 32 to 40°F still can be run using the modified
    procedure.  The complete procedure as supplied by the Santa
    Barbara County Air Pollution Control District is given in
    Appendix A of this report.
    The modified RVP procedure has several advantages and dis-
    advantages.  Already mentioned is elimination of the sample
    transfer problem.  However, this also eliminates the oppor-
    tunity for partial air saturation of the sample.  Therefore,
    the developmental work should include parallel determinations
    by the standard RVP procedure where possible to determine the
    influence of this change.  In general, a lesser degree of
    sample air saturation would tend to lower the vapor pressure
    measured as some of the air in the air chamber would dissolve
    in the sample during conduct of the test.
    Perhaps a more serious problem with this method is that related
    to air saturation of the sample.  In the Reid method, there is
    provision for partial air saturation by means of shaking the
    sample container after cooling to 32 to 40°F.  The sample
    container has an air space so the shaking action results in
    the partial saturation.  In the Santa Barbara modification,
    the sample is collected in the analysis container and there is
    no means for conducting the air saturation step.  Thus,  upon
    connecting the air chamber to the sample container and heating
    to the analysis temperature, there is some possibility that
    additional air will be dissolved depending upon the degree of
    saturation of air in the crude oil at time of sample collection.
                              14

-------
    The error will be in a negative direction, but is not determinate.
    The modified method is designed to conduct the vapor pressure
    measurement at stock storage temperature.  This eliminates the
    need for a correlation nomograph.  On the other hand, individual
    determinations would be necessary for each temperature of
    interest.
    Because the test proposed uses the same V/L ratio as the
    standard RVP method, the influence of light ends should not
    change greatly.
    Fairly extensive field testing of this method would be required
    in the developmental stage.  Equipment and laboratory costs
    should not differ greatly from the standard RVP method.
4.  Vapor Composition Method
    The general principal of this approach involves analysis of
    the headspace vapor in equilibrium with a liquid crude oil
    sample.  The vapor pressure is assumed to be a product of the
    mol fraction of hydrocarbon vapor in the headspace times
    the absolute pressure.  The method eliminates interference from
    water vapor and dissolved air.   It does assume that  an accurate
    measurement of  the hydrocarbon present in the vapor  space can
    be made,  however.   Normally this would be done by an appropriate
    gas chromatographic method.
    Two methods involving this approach have been reviewed.   One,
    from the  California Air Resources  Board, involves collection
    of a sample in  a  300 cc cylinder,  which prior to  the vapor
    analysis  step,  is  connected to  a 30 cc evacuated  chamber.   The
    two are connected,  creating a vapor space.   After the  connected
    vessels are held  for 30 minutes at the analysis  temperature,
    the vapor  space is  opened to atmosphere  long enough  for  the
    vapor  space pressure to  rise to one atmosphere.   Aliquot
    samples of  the  vapor-air mixture are  thus withdrawn  and
    analyzed by gas chromatograph.  As can be seen, the
                              15

-------
    determination is made at a V/L ratio or  1/11  (30/330).  A
    preliminary draft of this method  is given in  Appendix B.
    A second vapor composition method was submitted by Chevron
    Research Company.   In this method a sample is collected by  the
    water displacement  method.  Prior to sample collection the
    sample container is filled with water on the  clean side of  a
    flexible impervious membrane.  The membrane is free to move so
    as to allow the entire volume of  a rigid container to be
    filled.  The sample is collected  from a  sampling tap on the
    sampling side of the container by displacing  an exact volume
    of water into another container.  This latter container is
    removed and the remaining water drains from the clean side  of
    the membrane.  At the same time the sample side of the container
    is pressurized with nitrogen.  The container  is then allowed
    to come to equilibrium at the desired analysis temperature.
    Aliquot vapor-N2 samples are then withdrawn for gas chromatograph
    analysis similar to the GARB approach.  The method as submitted
    suggests a V/L ratio of 4/1.  This mthod is included as Appendix
    C to this report.
    Both of the vapor composition methods can provide results on a
    methane or methane  + ethane-free basis;  thus  reducing the
    variable influence  of non-reactive hydrocarbons on the vapor
    pressure.  Both specific methods require a reliable gas
    chromatographie analysis step.  The Chevron method, as submitted,
    is restricted to tap or line sampling.
    The vapor composition methods would require rather extensive
    field and laboratory testing and would increase the cost of
    analysis because of the gas chromatograph step.   Not all
    control agencies have access to such equipment in-house.

5.  Evaporation Method

    This  method involves determination of  the loss of  mass over
    time  of a sample held at a specified temperature.   While  it is
                              16

-------
one approach at obtaining a direct measure of evaporation loss
instead of one which uses a measure of the potential to
evaporate (such as vapor pressure), substantial questions on
design of the test conditions would have to be answered.
These include method of achieving test temperature — oven or
bath, rate of air or nitrogen sweep across the liquid surface,
surface to volume ratio of the evaporation container, length
of test, and quantity of stock to be used in the test.  Finally,
some new criterion would have to be developed as to whether
the stock tested would have to be stored in a controlled tank.
The method might be more applicable to open storage or waste
ponds than it would be to tanked storage.  This method is
described more fully in Appendix D.  This method would require
extensive developmental work, but would be low in equipment
and laboratory costs.
The various measures of heavy crude oil volatility are sum-
marized in Table 1 against the evaluation criteria specified
for use in this study.
                          17

-------
                           TABLE 1.  MEASURES OF HEAVY CRUDE OIL VOLATILITY VERSUS EVALUATION CRITERIA
Evaluation Criteria
Methods
Coat of
Method
Development
Equipment
and Lab
Costs
Repeatability
Conformity to
Sources of Error API Nomograph
oo
     Extrapolate API
     Nomograph
Gather additional
data which in
part should be
available.
No new c os t s.
    Equilibrium (True)
    Vapor Pressure
    Method

    Modified RVP
    Vapor Composition
    Method
    Evaporation
    Method
Extensive field
testing.
Extensive field
testing.
Extensive lab and
field testing.
Extensive lab and
field testing.
Also need to cor-
relate with loss
tests.
Unknown to fair.
Takes into account
gravity.
Modest.
Modest.
Fairly high -
need G.C.
Poor
Fair to good,
Possibly good.
Modest if lab
has balance and
oven.
Poor
Sampling - Loss of
GI and G£ in
various amounts.
Needs to be run
correctly re-
garding H20 and
barometric
pressure.

Extremely sensi-
tive to Ci and C2
loss.

No opportunity for
air saturation.
Accuracy of C.C.
analysis possible
problem.  Elimin-
ates water and
diss. air problem.

Conditions have to
be carefully con-
trolled.  How to
relate to losses.
Could be good de-
pending on
sampling and
can par ability of
new curve data.
Not good.  GI and
C2 effect.
Should measure
all GI and C-±
to gut good
correlation.

Probably not
good because of
unknown influ-
ence of GI and
in API method.

Unknown to poor

-------
                               SECTION 5
                    CONCLUSIONS AND RECOMMENDATIONS

     The results of our analysis of methods potentially useful for
measuring the volatilities of heavy crude oil were reviewed to deter-
mine whether any single candidate method clearly outranked the others
according to the rating criteria.  This appeared to us not to be the
case.  We did eliminate two methods from further consideration.  These
were (1) the Equilibrium (True) Vapor Pressure Method, and (2) the
Evaporation Method.  The former was eliminated because of the per-
ceived experimental difficulties, the extreme sensitivity to methane
and ethane content, and the lack of any information suggesting that
any of the problems with use of the current RVP method would be elimin-
ated.  The evaporation approach was eliminated because of the significant
departure from current measures of volatility, particularly as related
to predictive loss equations for tank storage of petroleum.  The method
does appear to be useful for sumps, pits and open oil tanks.
     Further review of the analysis results suggested that several rather
specific objectives should be achieved by any new method.  These include:
     (1)  It should permit sampling and testing of higher viscosity
          crude oils.
     (2)  It should permit determination of a volatility parameter for
          storage temperature above 140°F and RVPs (as currently
          measured) below 2 psi.
     (3)  It should reduce substantially the variable influence of
          methane and ethane present in the crude oil.
     (4)  It should be rugged enough to be conducive to good repeatability.
     (5)  The equipment and laboratory costs should permit reasonably
          widespread use.
     Our recommendations were also influenced by the information that
the American Petroleum Institute  Committee on Evaporation Loss in the
                                   19

-------
process of collecting currently available data relating RVP to TVP
measurements.  This could result in the publication of a revised RVP
versus TVP nomograph utilizing a broader data base and which includes
an alignment grid for crude oils of different API gravities.  The
current nomograph has a single correlation line representing all crude
oils instead of a grid.  Therefore, we do not recommend at this time
any supplemental work to extrapolate the current API nomograph.
     The two remaining methods in consideration are the Santa Barbara
modification of the RVP method and the vapor composition method.  While
the modified RVP method does address the sampling and transfer problems
with the current RVP method, it does not solve the problem caused by
light ends and possibly exacerbates the air saturation effect.  Therefore,
we recommend selection of the vapor composition method as a subject for
a methods development and evaluation program.  Where restrictions on
ASTM D323-72 for determination of RVP permit, intercomparisons between
vapor composition results and RVP should be included in the research
program.
     We suggest that the following parameters be considered in any
experimental design developed for a reasearch program:
     0  Crude type - covering a range of API gravities, viscosities,
        light ends content
     0  Sampling procedure - dip, water displacement, with and without
        transfer from sample container to analysis container
     0  Vapor to liquid ratio
     0  Temperature of analysis
     0  Specific analytical procedure - standard RVP, vapor composi-
        tion, etc.
     0  Number of replicates
     Because the light ends content of some crudes can change quite
rapidly after withdrawal from a well in a production field, it will be
                                   20

-------
important to consider exact location of sample withdrawal.  It may be
desirable to carry out analyses of liquid composition in at least a
portion of the crude samples.
     We have prepared a preliminary estimate of the level of effort to
conduct the research and development effort suggested.  Our estimate is
based upon examination of ten crude oil types, two methods of analysis,
three V/L ratios, two temperatures, and three replications.  This
results in about 450 samples.  For this program we estimate an overall
technical effort in the range of 2,000 to 3,000 hours.  Other direct
charges would include equipment, travel, per diem, expendable supplies,
computer costs, communication, and reproduction.
    A preliminary cost estimate has also been made for equipment needed
by users to conduct both the standard RVP method and the vapor composition
method.  This estimate is given in Appendix E.
    For sumps, ponds, and open tanks, we suggest further development of
the procedure described in Appendix D.
                                   21

-------
                               REFERENCES
1.  "Evaporation Loss from External Floating-Roof Tanks," API Publication
    2517, Second Edition, February 1980, American Petroleum Institute,
    Washington, D.C.

2.  "Hydrocarbon Emissions During Maring Loading of Crude Oils," Chevron
    Research Corporation for Western Oil and Gas Association, August 1977.

3.  "Hydrocarbon Emissions from Fixed-Roof Petroleum Tanks," for Western
    Oil and Gas Association by Engineering-Science, Inc., July 1977.

4.  "Evaporation Loss in the Petroleum Industry - Causes and Control,"
    API Publication No. 2513, reaffirmed 1973, American Petroleum
    Institute, Washington, D.C.
                        NON-REFERENCED CONTACTS

1.  Richard Burr, U.S. Environmental Protection Agency, OAQPS, RTP,
    NC.

2.  Harry Metzger, California Air Resources Board,  Sacramento, CA.

3.  Abe Moore, South Coast Air Quality Management District,  El Monte,
    CA.

4.  T. H. Gouw, Chevron Research Co., Richmond, CA.

5.  Sam Fett, Union Oil Research Co., Brea, CA.

6.  John English, Santa Barbara County Air Pollution Control District,
    Santa Barbara, CA.

7.  Robert A. Farnham, Chevron Research Co., Richmond,  CA.
                                   22

-------
             APPENDIX A




SANTA BARBARA MODIFIED RVP APPROACH

-------
         SANTA BARBARA VAPOR  PRESSURE  FOR CRUDE  OILS






SCOPE




This method is designed  to determine a vapor  pressure  of  crude oils




at the temperature at which they  are being stored.






EQUIPMENT




Equipment and apparatus  are those used in the Reid Method (ASTM D323)




with the modification shown in Figure  1.






PROCEDURE



Sampling:



   1.  Evacuate a clean, dry  liquid chamber by suitable means  to




       1 mm Hg or less absolute and close the valves.



   2.  Submerge the evacuated liquid chamber  31  to 5'  below the



       surface of the crude oil in the storage tank and open the




       top valve to allow the crude to enter  and fill  the chamber.



       Reclose the chamber while  submerged and remove  from the  tank.



       Clean the crude from the outside of the liquid  chamber.



   3.  Place the sealed  liquid chamber in an  ice water bath or



       crushed ice to transport it to  wherever the test is to be



       conducted.



   4.  Determine the temperature  of the crude in the storage tank.






MEASUREMENT OF THE VAPOR PRESSURE



   1.  Prepare the air chamber as outlined in the modified ASTM



       procedure Section 6.3 attached.  The test is run at the



       nearest 5°F above the temperature  of the  tank sampled.
                               A-l

-------
                         - 2 -






2.   Quickly assemble the liquid and air chambers and return to



    the bath at the test temperature.   Do not open the inter-



    connecting valve until the sample  is at the test temperature



    (about 20 minutes).   Agitate the sample as described in the



    ASTM procedure, following that procedure from Section 7.4



    to completion.   Record the gauge reading as the vapor



    pressure and indicate the temperature at which the measure-



    ment was made.
                            A-2

-------
Figure 1
  A-3

-------
         Designation: 0 323 - 72
                                                             American National Standard 211 44-1973
                                                                         Approved Jung 4 1973
                                                              By American National Standards Institute
         Standard Method of Test for
         VAPOR PRESSURE OF PETROLEUM
         PRODUCTS (REID METHOD)1
This Standard is issued under the fixed designation D 323. the number immediately following the designation indicates the
year of original adoption or. in the case of revision, the year of last revision A number in parentheses indicates the year of
last rcapproval
 1.  Scope
          is method covers the determination
 of tnTSpksolutc vapor pressure (Note I) of vol-
 atile CTOBL oil and volatile nonviscous petro-
 leum  prtmcts, except liquefied  petroleum
 gases (Note1
   NOTE 1—BcSbf the external atmospheric pres-
 sure is counter«fi\by the atmospheric pressure
 initially present tnwkair chamber, the "Reid va-
 por pressure" is arrBaWute pressure  at 100 F in
 pounds per square in&pRie "Reid Vapor Pres-
 sure" differs from  the tnwapor pressure of the
 sample due to some small *KBRle vaporization and
 the presence of water vapor aSEw m the confined
 space.
    NOTE 2—For  determination oTtjai^vapor pres-
 sure of liquefied  petroleum gases rBSj^fc should
 be made to Method D 1267.
    NOTE 3—The values stated in
 umis are to be regarded as the standard.
  ^Applicable Documents
       ISTM Standards-
   D 2?5fampling Petroleum and  Petroleum
     Produ
   D 1267 TesT^^yapor  Pressure of Lique-
     fied  PetroleurH^&£.)  Gases  (LP Gas
     Method)1
  E I Specification for A
 3.  Summary of Method
   J. 1  The gasoline chamber of the vapor pres-
            tus is filled with the chilled sample
 andl&ficctcd  to  the air chamber at 100 F or
 other leTHhuture. The apparatus is immersed
 in  a constan^jkjpperature  bath <100 ±  02 F)
 and is shakenpl&dically until equilibrium is
 reached. The "rrBimmeter reading"  corre-
 sponding to the prnsQt  read  on the gage
 attached to the apparanJBhaiitably corrected
 (Table  1) if the  air cbambejBks initially at a
 temperature other than I00"|ni$ the Reid
 vapor pressure.
  3.2 This  method provides  Tor 'What air
saturation of products with Reid vap
sure below 26 Ib (Sections  I to 9. and
        saturation for products above  26 Ib
                    and  17) and for narrower
                            i for the measure-
 ment of  the vap"o?$S^jSiSfc4y^lJon Baso~
 lines (Sections 16 and 17)

 4. Apparatus
   ^fcjGie construction of the required appa-
 rat^SEafasbed in the Appendix.  For sam-
 ples havin'Pttfcpfcatressures below 26 Ib. use
 the  gasoline  cnnlkCjwUn  one opening
 (AI.I.2)  and for  iiii|ili  lj^jiii|gryaBpr pres-
 sures above 26  Ib, use the gaffS8jggȣfacr
 with two openings (Al.1.3).          ^"w

 S. Handling of Samples
       The general provisions  in 5.2 and  5.6
  IPJwAwilj to all samples  for vapor pressure
 den erfiifcat ions,  except as specifically  ex-
          fe samples  having vapor  pressures
 above 26^thL
-------
                                                                  D323
  1 \quid

    .quid
i<   test
'i   erature
                             is opened
                             $omple  Transfer— The,  Reid  vapor
                                         ion  shall be the first test
                       run on*e. In the instances of transfer
                       of liquids mS^fteer sample containers or of
                       withdrawal oTfsrates  for  other tests, the
                       transfer connectionol^^J  shall be used.
                         5.6 Care of Sample&&ti$K shall  be put
                       in a cool place  as soon as pfc&J&kafter they
                       have been obtained  and held TntSSuuil the
                       test  has been completed. Sa
                       containers shall  not be considered  for
                       shall be discarded and new samples obtain

                       6. Preparation for Test
                              '>  Saturation of Sample in  Sample
                                       the sample at a temperature
                       of 32 too^take the container from the
                       water coohng^ab|Unseal it, and examine it
                       for  its liquid  comci^%hicri  shall be between
                       70 and 80 percent oftnQUiainer capacity.
After the correct liquid
sured. reseal the container,
1v  and rgturn il_lA-thc-
  6.2 Preparation  of
Completelv immerse the
                                                        been as-
                               Chamber—
                                    cham-
 beranTtne sample transfer connection in the
 water cooling bath for a sufficient time to al-
 low the chamber and  connection to reach the
 bath temperature (32 to 40 F).
   6.3 Preparation  of Air Chamber 0HHE
^ggfff)—After  purging and rinsing the air
 chamber  and  pressure  gage in accordance
 with 7.5, connect the gage to the air chamber.
 Immerse  the air  chamber to  at  least  1  in.
 above its  top in the water bath maintained
^jfgfaf, for not less than 10 nun just be-
 fore coupling u to the gasoline chamber.  Do
 not  remove the air chamber from  the bath
 until the  gasoline  chamber  has  been filled
 with sample as described in 7.1
      Preparation of Air Chamber (Ambient
            Procedure)—*.* an  alternative
to fetet the air chamber  to ambient or
other -tempSraihfewhich may-be-determined
with an accuraBjj^^jjeasr I "Pin the-follow- •-
ing manner Afteroi^M^and.Tinsunnheair.
chamber and pressure*l9^in accordance
with 7.1, conned the gageTSiralur chamber.
Insert ihc-thcunometcr Jnto/
supporting it by means of a looselylwfcL(nol
airtight) stopper in the opening  orNw air
  Camber Adjust the position of the thermom-
         >t u is aligned as closely as possible
with HfSfexis of the air chamber, and with the
ihermomefeVbulb located in the air chamber,
about 9 in. frxWthc opening. Leave the ther-
mometer in posntai  until the  temperature
reading has remamecratasiani within I  F for
a period of 5 mm or more^sjbcfore coupling
the air chamber to the gasoTrH^jchambcr. At
this time, record the thermomeieTO^gjling as
the "initial air temperature "

7. Procedure
      Sample  Transfer—With everything  in
redWiess. .remove the chilled sample con-
laineraom the  bath, uncap it, and insert the
chilled nvsfer  connection and air tube (see
Fig.  1). Qwcly empty the  chilled gasoline
chamber and^hue it over the sample delivery
tube of the tran»LConnection. Invert the en-
tire system rapidly^kJhat the gasoline cham-
ber is finally in an upBohi position with the
delivery tube extending t&rithin '/« in. of the
bottom of the gasoline charrnr. Fill the gaso-
line chamber to  ovcrflowine.^hhtlv tap the
gasoline chamber against the  woq^.bcnch  to
ensure that  the sample is free of
If any sample is displaced, refill the
to overflowing.
  7.2  Assembly  of Apparatus—Without de-
lay, and as quickly asDOssible^ttachjJjeju^
chamber to the  •••CUecharnDerNoi more
                                           than 20 s shall be consumed in completing the
                                           assembly of  the apparatus after .filling  the
                                           gasoline  chamber, using the following  se-
                                           quence of operations.	
                                                              ~"~          "    line
                                                                                _       remove the air cham-
                                                                  ber from the I^^P water bath (6.3).
                                                                    7.2.3 Connect the air chamber to the <
                                                                      chamber.
                                                                    7J Introduction of Apparatus into Bath—
                                                                                                         is
                                                                                     [us. Immerse the assem-
                                                                  bled apparatus in the bath, maintained at  _
                                                                  4BBH, in an inclined position  so that the
                                                                  connection of the«nMint and air chambers is
                                                                                                                   liquid
                                                                                            liquid
                                                                                            the  test
                                                                                            temperature

                                                                                            liquid
                                                                 199
                                                               A-5

-------
                                                              with  a'sultable  solvent  and  dry
                                                              with  air  and heat  as  may  be  required,
                                                              Care  should  be taken  to  be  sure  that
                                                              the  Bourdon  tube  of the  pressure
                                                              gauge  is  thoroughly clean and  dry.
liquid
 below the water level and may be observed
 closely for leaks. If  no leaks  are  observed.
 immerse the apparatus to at least I  in. above
 the top of the air chamber  Observe  the appa-
 ratus for  leaks throughout the  test. When at
 any lime a leak is detected, discard the test
   NOTE 4—Liquid leaks are more difficult to delect
 than vapor  leaks, and because the  much-used cou-
 pling device is normally in the liquid section or the
 apparatus, give it particular attention
   7.4 Measurement  of  Vapor . Pressure—
 After the assembled vapor pressure  apparatus
 has been immersed in the bath  for 5 mm, tap
 the pressure  gage lightly,  and observe the
 reading. Withdraw  the  apparatus  from the
 bath,  invert it, shake it vigorously,  and re-
 place  it in the bath  in the shortest possible
 time to avoid cooling the apparatus. At inter-
 vals of not less than 2 mm, repeat this agita-
 tion and gage  observation ar least five times.
 until the last two consecutive  gage readings
 are constant,  to  ensure  equilibrium. These
 operations  normally  require 20 to 30 mm.
 Read the  final gage  pressure to the nearest
 0.05 Ib  for gages with intermediate  gradua-
 tions of 0.1  psi and to the nearest  0.1 Ib for
 gages with graduations of 0.2 to 0.5 psi, and
 record this value as  the "uncorreeled vapor
 pressure"  of the sample under test. Immedi-
 ately remove the pressure gage and check its
 reading against  that  of the manometer, re-
 cording the value  found  as the  Reid vapor
 pressure (under the procedure  of 6.3). or as
 the "manometer reading" to be  used in the
 calculations of Section 9 (under the procedure
 of 6.4)
_   7.5 Preoaranon  of Apparatus  for Next
 fesi—Disconnect  the  air chamber^
 chamber, and pressure
             in the Bourdon lube by repeated
        eaTthrusts. This may be accomplished
 in  Ilirnowing manner:  hold the gage be-
 tween thlaklms of the hands  with the right
 hand on  uWtface side and  the  threaded
 connection of^e gage  forward. Extend the
 arms forward  annupward at an angle of 45
 deg with the coupnaLpf the gage-pomting in
 the same direction. ^WP the arms downward
 through an arc of aboOW35 deg so that the
 centrifugal force aids grav^kjn removing the
 trapped liquid. Repeat  this^fctration three
 times to expel all  liquid. Purgc"^uressure
 gage by directing a small jet of air^W* its
                                                                             tease of crude oil.  the Bourdon
                                                                               .with a volatile  solvent after
D323

    don tube for at least 5 mm Thorough!)
    \thc air chamber of residual  sample  b\
      Jhe  air  chamber  with warm  water
        I F) and allowing it lo drain (Noie  6)
        kis purging at least Five times  After
lhorouu\ removing the previous sample
from  Um^tasolinc chamber,  immerse the
chamber iw\ice bath for the next lest
  NOTE 5—IB
lube musl be
each test
  NOTE 6—If  the"'tegl&Jyof the air  chamber is
done in a bath, be sureTJfey8>CsmaII and unnonce-
able films of floating sampreitj^Bj^HUvthe bottom
and top openings of  the chanfrWJSs^ped as thex
pass through ihe'surface of water

8. Precautions
  8.1  Gross errors can be obtained in vapor
pressure measurements if the  prescribed pro-
cedure is not followed carefully. The following
list emphasizes  the importance of  strict ad-
herence to the precautions given in the proce-
dure:
  8.1.1 Cheeking the Pressure Cage—Check
all gages against a  manometer after each test
in order to ensure  higher precision of results
(7.4).  Read all  gages while ihe gage is in a
vertical position
       Air Saturation of Sample—Open and
               container once after the con-
               id  a temperature  of 32 to  40
F. ShaHK^kkCpntamer vigorously  to ensure
equilibnunlH&ibSample with the  air  in the
container I       ^—_._
  8.1.3 Checking JorTeaks—Check all appa-
ratus  before and during each lest for  liquid
and vapor leaks (A 1.1.6 and Note 4).
       Sampling—Because initial  sampling
       handling of samples will greatly affect
             s. employ the utmost precaution
and Bgg^pst meticulous care to avoid  losses
througnBraknoration and  slight  changes  in
compositraUnuons  5 and  7.1). In no case
shall any parr^HMkRe'd apparatus itself  be
used as the samtS|Mp^iner previous to ac-
     1 conducting the
         urging the
purge fn^gessure gage, the gasoline cham-
ber, and the'fifebarnber to be sure that they
are free of residuaT^fcBle.frhis is most con-
veniently done at the enW^^BiauaiiS test.)
(See 7.5).
  8.1.6  Coupling the Apparatus—Carefully
                                                           200
                                                            A-6

-------
liquid
  with
  observe the requirements of 7.2
    8.1.7 Shaking  the  Apparatus—Shake  the
  apparatus "vigorously" as directed in  7 4 in
  order to ensure equilibrium
    8.1.8 Temperature Control— Carefully con-
  trol the temperature at the time of air satura-
  tion and the temperature  of the  100 F bath
  (Appendix, A1.3  and  A1.4,  respectively). Be
  certain  that  the temperature of the air  in the
  airj:hambcr at the time of coupling with the
       Re chamber (7.2) has remained constant
  within I F for a period of 5 mm or more.

      Jculation
          ange  in Pressure of  Water Vapor
 and ^ISfiJ the ambient  temperature proce-
 dure  deflipjL in 6.4. calculate  the  "Reid
 vapor presrlHsSjfthe sample under lest by
 applying to inwjgfefjioineier reading" the
 correction given irSfegjhJfor the change in
 pressure of the wateSj^gjg^and air in the
 chamber on healing frn mfffJi^mSTttitJ t tem-
 perature" to  100 F.
   9.2  Recording KesuJts-
^•^B^M^MBftBfP*).  record the result
 observed in 7.4 as the  "•• vapor pressure"
 in poundsBE reference to temperaiurefl|
                                  from the application of the
                                                  vapor pres-
                      MODIFICATIONS FOR PRODUCTS HAVING
                             REID VAPOR PRESSURES
                                 ABOVE 26 LB

                  10. General
                         With products having vapor  pressures
                  ovelH6 Ib (Note 7), the procedure  described
                  in Sea»ns 5 to 8 is hazardous and inaccurate.
                  Conseqowty,  the  following sections  define
                  changes m^taparalus  and procedure for the
                  determinationl^vapor pressures above 26 Ib.
                  Except as -speci^^ly  stated, all the require-
                  ments of Sections T^g 9  and Section 17 shall
                  apply.
                     NOTE 7—When the questi^Lariscs. the air satu-
                  ration method shall be used ur^knnine whether or
                  not a product-has a vapor nressurtiVbove 26 Ib.

                  II. Appmratus
                  _  11.1 Bomb, as described  in the
 D323

       the gasoline chamber with two openings
        pressure Cage Calibration—A dead-
              (AI.7) may be used in  place of
 the merctft&Qanometer (A 1.6) for checking
 gage readmgs^^e 26 Ib. In 7.4p -8.1.1. and
 9.1. and Table Iw^jcihc words "manom-
 eter"  and "manomelerS^duig" appear, in-
 clude  as an alternate "dcad-wSSfehl tester" and
 "calibrated gage reading,"

 12.  Handling of Samples
   nSl^^j^j.4. and 5.5 shall not apply.

 the  sample contarn^^^p^iwhich  the vapor
 pressure sample is takensWa^^^be less than
 1 pt liquid capacity.          ^^^^^fea^.

 13.  Preparation for Test
 &3-I  6.] and 6.2 shall not apply.
 IB.2 Any  safe method of displacement of
 thonest sample from the sample  container
 thaCwsures filling the gasoline chamber  with
 a'chflBd,  unweathered sample  may  be  em-
 ployed:@jhe  following 13.3 to  13.5, together
 with SecrokU, describe displacement  by self-
 induced preotte.
   13.3  Maiivfflb the sample container at a
 temperature suffiunlly high to maintain su-
 peratmospheric pMBure  but not substantially
 over 100 F.      ^1^
   13.4  Completely ^nrnerse  the  gasoline
 chamber, with both vamtopen. in  the water
 cooling bath  for a sufficiCTklength of time to
 allow it to reach the bath Iguxraturc (32 to
40 F).                     ^».
   I3.S  Connect  a suitable  Jcc^jJcd  coil to
the outlet valve of the sample conuj^v.
  NOTE 8—A suitable ice-cooled coil c^ae pre-
pared by immersing a spiral of approxiinaifi^^& ft
of Vj-in. copper tubing in a bucket of ice waicr^^

   . Procedure
       7.1  and 7.2 shall not apply.
      ^Connect the '/«-in. valve of the chilled
gasolTwfchambcr to the ice-cooled coil. With
the  >/i3^h^aTve~ of "the gasoline  chamber
closed,  oplwhe outlet valve of the sample
container ana*^^V«-in. valve of the gasoline
chamber.  Open  ofc^asoli ne  chamber  Vi-in.
valve slightly-and-alWfcAe gasoline -chamber-
to  fill  slowly. Allow the^Mole to overflow
until the overflow volume is^B^n]  or more.
Control this operation so that no^Bicjable
                                                            201
                                                            A-7

-------
                                               D323
 CO
 po
 Who1
 rupt
 the .
 must
 and the
p in  pressure occurs at the gasoline cham-
 '/4-in. valve. In the order named, close the
 line chamber V2-m. and '/«-m. valves: and
  close all other valves  in the sample sys-
  Disconnect the gasoline chamber and the
     coil.  (Caution: Safe  means for dis-
     hquid and vapor escaping during this
    Deration  must  be provided.  To  avoid
    because of the  liquid-full condition of
         chamber,  the  gasoline  chamber
      ,uickly  attached to the air  chamber
        . valve opened.)
   14.3 1
 chamber li
 oline cham
 s shall be
 bly  of the a
 chamber, usi
 erations: (I)
 or remove  tl
 bath, (2) con
 line chamber, an
 bcr Vj-in. valve.
   14.4 If a dead-
 of the  mercury m
 calibration  factor
 established  for  the
 the "uncorrecled va
 corrected vapor
 found as the "calibrat!
 used in the  calculation
 ble 1 in place of the "m
        ^d iatcly  attach  the  gasoline
         ie air chamber and open the gas-
          '/j-in. valve. Not more than 25
           med in completing the asscm-
           ratus after filling the gasoline
            he following sequence of op-
              the  initial air temperature
             r chamber from  the  water
              e air chamber to the gaso-
               ) open the gasoline cham-
                :ht tester is used instead
                meter (11.2). apply the
                   unds per square inch
                  issure  gage at  or near
                   pressure" to the "un-
                     recording the value
                    gage reading" to be
                    if Section 9 and Ta-
                     mcter reading."
 15. Precautions
   IS.I The precaution 8. HI

   MODIFICATIONS FOR A VIA:
      OF ABOUT 7 LB
                        hall not apply.
f GASOLINE
KESSURE
16. General
   16.1 The  following   paragra&s  define
changes in apparatus  and proccdiAfor the
determination of the vapor pressuremf avia-
tion gasoline.  Except as specificallv^ktated
herein, all the requirements set forth
tions 1 to 9 and 17 shall apply.
  16.1.1  Ratio of Air and Gasoline  ChaZ
hers — The ratio of the volume of the air cham-
    to  the  volume of the gasoline chamber
   ill  be between the  limits of 3 95 and 4.05
          A I in the Appendix)
    LI .2  Water  Cooling  Bath— The water
     jg bath shall be held at  a temperature of
        F (See A 1. 3)
         Checking the Pressure  Cage — The
          be checked  at  7 Ib against a  mer-
         mn before each vapor pressure mcas-
         to ensure that it  conforms to the re-
          s of Section A 1. 2. This preliminary
          I  be made in addition to the  final
           nson specified  in 7.4
              Chamber  Temperature — The
             6.3 shall be followed:  the provi-
sions of 6.4Hhall not apply
        all
       The
                iwmg criteria should be used
                 acceptability  of results (95
                bitity—Duplicate results by
                   should  be considered sus-
                 oy more  than the following
                      gaso-
                              Re pea lability
                            (Same Operator
                            and Apparatus)
                                  0 I
                                  0.2
                17. Preci
                  17.1
                for judging
                percent confide'
                  17.1:1 Repe,
                the same opera
                peel if they diffi
                amounts:

                           Range
0 to S Ib
5 to 16 Ib (except aviat
  line)
16 to 26 Ib
Above 26 Ib
Aviation gasoline (approxn
  7lb)
   17.1.2 Reproducibili
muted by each of two
considered suspect if they
the following amounts:
                                                                      telv
                                 0 3
                                 04
                                 0 I
                                                                    The results sub-
                                                                   ratones should be
                                                                        by more than
                           er
                                                     Range

                                         0 to 5 Ih
                                         5 to 16 Ih (excepi aviation gaso-
                                           line)
                                         16 to 26 Ib
                                        . Above 26 Jb
                                         Aviation gasoline (approximate)}
                                           7 Ib)
                             Reproducibility
                            (Different Oper-
                               ator and
                              Appjratus)

                                0 35
                                0 .1
                                        202
                                             A-8

-------
                          D 323
  TABLE 1  Corrections  to  be  Subtracted  from
"Minometer Readings" for Calculating  Reid Vapor
Pressure
Barometric Pressure,'
Initial Air
Temperature,"
degF
32
40
50
60
70
80
90
100
110


760
2.90
2.60
2.20
1.80
1.40
0.95
0.50
0.00
-0.55


700
2.70
2.45
2.10
1.70
1.30
0.90
0.50
0.00
-0.55
'mm Hg


600
2.45
2.20
1.90
1.55
1.20
0.85
0.45
0.00
-0.50
  * I-or other temperatures and pressures, the correc-
tions may  be calculated  by means  of the following
equation:
Correction -
        \(F - P,)(i - I00)/460 +  /) -.(/•,«,  - />,)
where:
   / -  air chamber  temperature at  beginning  of
        lest, deg F.
  P "  barometric pressure, psi, at time of lest (if a
        barometer is not  available, the normal baro-
        metric pressure may be used),
  P, -  vapor pressure of water, psia, at / deg F, and
Pirn -  vapor pressure of water, psia, at 100 F - 0.95.
  Calculated corrections  arc to be rounded off  to the
nearest  0.05 psi.
  Example—The pressure gage gives an "uncorrected
vapor pressure" reading  of 11.6 psi. When the gage
is compared to  a mercury column, a "manometer
reading" of 11.5 psi  is obtained. For an "initial air
temperature" of 80 F and atmospheric pressure  of
700 mm Hg, the  correction shown in Table I is 0.90
psi. Because the  "initial  air temperature" is  below
100 F, this correction of  0.90 psi is subtracted  from
the  "manometer  reading" of 11.5 psi giving, a "Reid
vapor pressure"  of 10.60  psi.
                             Chilled Gosolmt Chambtr
                                                  (d)
                                      fcsrtion of System for
                                         Sample Transfer
     (b)
 Sealing Closure
OrpHcri ly Samplt     PtocedOwl
     r Cofwcnon
      to)
Sample Container
Pnor to Transfer
  of Sample       transfer Connection     Dekvtry Tubi
   FIG.  I   SinpiifM Sketekcs OwtiMof MetlM4 of Trusfenug Sup
                     from Opcs-Typc CMUUCTI (V. bk - 6.25 ••).
                      203
                        A-9

-------
                                                                       D323
                                                               APPENDIX
                                             Al.  APPARATUS FOR REID VAPOR PRESSURE TEST
  iquid.

liquid.
  iquid
liquid -
  iqriri.
    A 1.1 Reid Vapor Pressure Bomb, consisting of
  two chambers—an air chamber (upper section) and
  a  gasoline chamber (lower section)—shall conform
  to the following requirements
    NOTE Al: Caution—To maintain the correct vol-
  ume ratio between the air chamber and the gasoline
  chamber, the units shall not be interchanged  wifA-
  out recalibrating to ascertain thai the volume ratio
  is wiiftin satisfactory limns

    A 1.1.1  Air Chamber—The upper section or air
  chamber, as shown in Fie. Al. shall be a cylindrical
  vessel  2  ±  '/§ in. in  diameter and  10 ± '/•  in. in
  length, inside dimensions, with (he inner surfaces of
  the ends slightly sloped to provide complete drain-
  age from either end when held in a vertical position
  On one end of the air chamber, a suitable gage cou-
  pling with an internal diameter not less than V,. in.
  shall be provided to  receive the '/.-in. gage connec-
  tion. In the other end of the air chamber an opening
  approximately  '/,  in. in diameter shall be provided
  for coupling with  the gasoline chamber. Care  shall
  be taken that the connections lo the end openings
  do not prevent the  chamber from  draining com-
  pletely.
    Al.1.2 Gasoline Chamber (One-Optning)—Tt\t
  lower section or gasoline chamber, as shown in Fig.
  Al. shall be a cylindrical vessel of the same inside
  diameter as the air  chamber and of such volume
  that the ratio of the volume of the an chamber to
  the volume of the gasoline chamber shall be be-
  tween the limits of 3.8 and 4.2 (see Now A2). In
  one end of the gasoline chamber an opening approx-
  imately V, in. in diameter shall be provided for
  coupling  with the air  chamber.  The inner surface
  of the  end containing the coupling member shall
  be sloped to provide complete drainage when in-
  verted. The other end of  (he  gasoline  chamber
  shall be completely closed.

   Non A2—The ratio  for units  to be used  foi
  aviation gasoline testing shall be between 3.95 to
 4.0S.	.	
    A LI.3 G*tmtme Chamber (Two-Opening}— For
^Smpiin^fSm  closed vessels, the lower  section or
 gasoline chamber, as shown in Fig. Al shall be es-
 —...IK.  .u- .—uj^jhejjjKm chamber de-
                 except tbaiT '/.-in. valve shall be
                      mAf-ih^^Mfcv chamber
TH
TrT
 MdaVr-in.  straigm-ihrough.  full-opening  valve
 shall be introduced in the  coupling between the
 chambers.  The volume of the_HOTMK  chamber,
                                T by the valves.
mliiuiuu limy  uit uiuuij
shall fulfill the volume ratio  requirements  as  set
forth in Al.1.2.
meats as set lortb in Al.1.2.
   NOTE A3—In determining capacities for the two- .
               chamber (Fig. AI), the capacity of
	^    __   jnber shall  be considered as that
below  me Vr-m. valve closure. The volume above
the '/.-in. valve closure including the portion of the

                                           204
                                                                       coupling permanently  attached to the gasoline
                                                                       chamber  shall be considered as a part  of the air
                                                                       chamber capacity

                                                                         A1  1.4 Method of Coupling Air and'*
                                                                       Chambers—My  method of coupling  the air and
                                                                               chambers mav h^     nv^ >lrovif1rH thfll
                                                                            -
                                                       TnToTTWiraDTrcoupling be on the
                                                                       no giwvfle is losi during the coupling operation.
                                                                       that no compression effect is caused by (lie an of
                                                                       coupling,  and thai the assembly is free from leaks
                                                                       under the conditions of the tests. To avoid displace-
                                                                       ment  of •••*«:  during assembly
                                                                       thatthe male fiuT
                                                                        ng tlTeassernolyoaTuTTSoTeTcrewcoupiing. a vent
                                                                       hole may be used to ensure atmospheric pressure in
                                                                       the air chamber at the instant of sealing
                                                                          Al.1.5 Volumetric Capacity of Air and <
                                                                       Chambers — In order to ascertain if the volume ratio
                                                                       of the chambers is between the specified  limits  of
                                                                       3.8 to 4.2 (see Note A2). measure a quantity of wa-
                                                                       ter greater than will be needed to fill the
                                                                       and air  chambers. The
                                                                       completely  filled with
                                                                       between the original volume and the remaining vol-
                                                                       ume IS the volume of the aaaahae chamber Then
                                        after connecting the
                                        air chamber shall be filleono tne seat of the gage
                                        connection with more of the measured water, and
                                        the difference in volumes shall be the volume of the
                                        air chamber
                                          A 1.1.6 Checking for Freedom from Leaks—Be-
                                        fore placing new apparatus in service and as often
                                        as necessary thereafter, the assembled vapor pres-
                                        sure apparatus  shall be checked for freedom from
                                        leaks by filling with air to 100-psi gage pressure and
                                        completely immersing in a water bath. Only appa-
                                        ratus which stands this test without leaking shall be
                                        used.
                                          A1.2 Pressure Cage—The pressure gage shall be
                                        a Bourdon-type  spring gage of test gage quality 4 Vi
                                        to S'/i in in diameter provided with a nominal 'If
                                        in. male thread connection with a passageway not
                                        less than s/« in. in diameter from the Bourdon lube
                                        to the atmosphere. The range and graduations of
                                        the pressure gage used shall be governed by  the
Cage to be Used
                                                                         Reid Vapor
                                                                         Pressure.Ib
                                         4 and under
                                         3to 12
                                         10 to 26
                                         10 to 36
                                         30 to JJ
                                         30 and higher
                                                       Scale
                                                       Range.
                                                         psi
                                                                 Oto 3
                                                                 Oto 13
                                                                 0(o 30
                                                                 Olo45
                                                                 Oto 60
                                                                 0 to 100
                                                                Maximum
        I
        3
        5
        5
       10
       10
                                                                                                           ations.
                                                                                                             psi
                                                                                        O.I
                                                                                        O.I
                                                                                        0.2
                                                                                        0.2
                                                                                        0.25
                                                                                        0.5
                                                                                                         liquid

                                                                                                         liquid


                                                                                                         liquid

                                                                                                         liquid


                                                                                                         liquid
                                                                                                         liquid
                                                                                                         liquid

                                                                                                         liquid
                                                                                                         liquid
                                                                      A-10

-------
                                                                         D323
.iquid
vapor pressure of the  sample being tested, as fol-
lows:
Only  accurate gages  shall be  continued in use.
When the gage reading differs from the manometer
(or dead-weight tester  when testing gages above 26
Ib) reading  by more than  I  percent  of  the  scale
range of the gage, the gage shall be considered inac-
curate. For example, the calibration correction shall
not be greater  than 0.15 psi for a 0 to 15-psi gage or
0.3 psi for a 0 to 30-psi  gage.
   NOTE A4—Gages 3'/»  in. in diameter  mav  be
used in the 0 to 5  Ib range.3
   A 1.3  Water Cooling Bath—A. water  cooling bath
shall be provided  of such dimensions that the sam-
ole comameii_and MWffiie chambers may be com-
                      plcieiy immersed. Means for maintaining the bath
                      at a temperature of 32 to 40 F shall be provided.

                        NOTE AS—Solid carbon dioxide shall not be used
                      to cool samples  in storage or in the preparation of
                      the air saturation step. Carbon dioxide is apprecia-
                      bly soluble in gasoline, and its use has been lounrl to
                      be the cause of erroneous vapor pressure data.
                        A 1.4  Water  Baih—Jhe water bath shall  be of
                      such dimensions that  the vapor pressure apparatus
                      may be immersed to at least 1 in. above the top of
                      the air chamber. Means for maintaining the bath at
                      a constant temperature  of 100  ± 0.2 F shall  be
                      provided.  In  order to check this  temperature, the
bath thermometer shall  be immersed to  the 98 F
mark throughout the vapor pressure determination.
   A 1.5 Thermometers:
   A 1.5.1  For 100 F Air Chamber Procedure— An
ASTM Reid Vapor Pressure Thermometer No.  18F
having a range from 94 to  108 F and conforming to
the requirements in Specification E 1.
   A 1.5.2  For Water Bath— Use the  ASTM Ther-
mometer I8F described in Al.5.1.
   Al.5.3  For Air  Chamber— When the  ambient
temperature procedure is employed, a thermometer
conforming  to the following requirements shall be
used:  Length, approximately 12 in.; range,  —40 or
-30 F to +120 or +130 F; graduated in 1 F divi-
sions; total irnmcrsion. scale error not greater than
1 F.
   A 1. 6 Mercury Manometer — A  mercury mano-
meter.  having  a range suitable  for  checking  the
pressure gage  employed  shall be  used. The  ma-
nometer scale may  be graduated in steps of 1 mm.
0.1 in., or 0.1 psi.
   A 1.7 Dead-Weight Tester— A dead-weight tester
may be used in  place of the  mercury manometer
(AI.6) for checking gage readings above 26 Ib.
                                                      ' Suitable gapes are available from the Fisher Scientific
                                                   Co. (Special Order), Pittsburgh. Pa. and U. S. Gauge Co.
                                                   (Catalog No. 510SP). Sellersville. Pa.
                                                                                Coupling
                                                                                1/2 O.Q
                                                                                                liquid
                                                    Coiptng.l/ZLO.
                                                         Air Chombar

                                                            M»<«; Al Dl

                                                          FIG. Al   V
                                                   Goiellnt Chombtr
                                                    (O«t Opining)
                        By publication of tUs standard no position it taken with respect to the validity of any patent rights in connection thert-
                      MiA. and ine American Society for Testing and Materials does not undertake to insure anyone utilizing tni standard
                      against liability for infringement of any Letters-Patent nor assume any suck liability.
                                                                       205
                                                                       A-ll

-------
          APPENDIX B

CALIFORNIA AIR RESOURCES BOARD
   VAPOR COMPOSITION METHOD

-------
                  Determination of Crude Oil Vapor  Pressure
                    by Gas Chromatographic  Vapor Analysis
                          (Draft Proposed Method)
 SCOPE
 This method is  proposed  to  determine  the  approximate  true  vapor  pressure
 (bubble point pressure), exclusive of the partial  pressure of methane,
 for any crude oil  at  temperatures high enough  for  the oil  to flow.
 The practicality and  validity of the  method  for  crude oil  will be
 demonstrated in the laboratory.  If subsequent testing  of  the method
 with other liquids is successful, the method's scope  may be expanded.
 The method will be applicable only if the total  vapor pressure,  including
.methane, is less than 14.7  psia.

 SUMMARY OF METHOD
 Oil will be collected, by methods described  elsewhere,  in  a 300  cc  cylinder
 equipped with leak-tight valves at both ends and a rubber  septum at one
 end.  The entire volume  of  the cylinder will be  full.  The sample will
 be kept on ice pending analysis.

 Jn the laboratory, an empty chamber of exactly 30  cc  volume will be attached
 to the lower sample cylinder valve and evacuated.   The  cylinder  valve will
 then be opened, creating a  vapor space in the  sample  cylinder.   The
 assembly, with the 30 cc chamber at the bottom,  will  be placed in a
 temperature bath until it has reached the desired  analysis temperature
 throughout.  Aliquots of equilibrium  vapors  will then be withdrawn  through
 the septum at the upper  cylinder valve with  a  syringe which is at the
 test"  temperature.   The aliquots will be analyzed by gas  chromatography
                                         B-l

-------
                                     -2-
for the mass concentrations of hydrocarbons of each carbon number.
The total non-methane vapor pressure will be calculated according to the
perfect gas relationship from the sum of each mass concentration
divided by the appropriate molecular weight.

EQUIPMENT
1.  300 cc stainless steel sampling cylinder equipped with vacuum-tight
    valves at each end and with a rubber sampling septum.  See Figure 1.
    The total volume between the two valve closures shall be between
    295 and 305 cc.
Z. ' Vacuum chamber with two openings, one fitted with a vacuum tight valve,
    the other with a connector compatible with the lower sample cylinder
    .valve. The volume of the chamber between the valve closure and the opening
    of the connector (points 1 and 2 in Figure 1) shall  be 30.0 cc.
3.  Vacuum pump capable of achieving 1.0 mm Hg.
4.  Temperature bath capable of maintaining within 0.3°F of the desired
    temperature.
5.  Gas chcomatograph meeting the following specifications: - (to be
    reconmended by AIHL).
6.  (Chromatographic column to be recommended by AIHL).
7.  Syringe for chromatograph sample injection.

SAMPLING PROCEDURE
Samples will be collected in a 300 cc sample cylinder according to methods
approved by the Air Resources Board.   (A draft method is  enclosed.)   The
sample cylinder will  be completely full.   The sample  will  be kept in ice
until  the headspace analysis is  performed.
                                       B-2

-------
                                     -3-
LABORATORY PROCEDURE
A.  Gas Chromatograph Preparation (to be recommended by AIHL).
B.  Sample Preparation
    1.  Clean the outlet of valve B (Figure 1).
    2.  Attach  the 30.0 cc vacuum chamber to valve B and evacuate it to
        1.0 mm Hg or less.  Close valve D and verify that the chamber is
        vacuum-tight.
    3.  Close valve C, disconnect it from the vacuum line and place a rubber
        stopper over the valve outlet to prevent any leak through
        the valve.
    4.  Open valve B and immerse the entire assembly, with the 30 cc
        chamber down, in the temperature bath.  Also immerse the syringe,
        enclosed to prevent contact with the Bath fluid.  Agitate the
        cylinder at five minute intervals.
    5.  After 30 minutes, open slightly valve A to allow the total pressure
        to rise to one atmosphere.  Close the valve.
    6.  Insert the sample syringe, with the plunger slightly out, through
        the septum and depress the plunger completely (to eject any oil
        which enters the syringe needle during insertion).
    7.  Withdraw the desired aliquot volume and inject the aliquot into the
        Chromatograph.  Steps 6 and 7 should be done quickly to prevent
        cooling of the aliquot.
    8.  At ten minute intervals, withdraw and inject two additional aliquots
        per steps 5-7.
                                     B-3

-------
                            -4-
    CALCULATIONS
    1.  Determine the total mass in the aliquot of each group of compounds
       with the same carbon number.
    2.  Calculate the total vapor pressure in the aliquot as:
           p s RT  VMi
    Where:
    R  s gas constant
    •
    V  = aliquot volume
    T  = test temperature
    ft. - mass of compounds with carbon number i; all measurable hydrocarbons
       •except methane
      - molecular weight of carbon number i compounds
P&csr./r^jm,,.:-,
rf  iJI'VV   .:,:-  '.  :::•: '  ' -^ !  '' '   '•  "."
 1   7           .  f   -•,-,..-!.•;•••'
 <  -..    .-   '' •  •     -•"-.:.-•,   ''4
    '  " ' •   ~L'," •    .'-, v .':'' .,'•;')  J  .
                • - --. *;u. ';-,tj :..•;.    :-;  .
                                *
                           B-4

-------
I
01
                                              cc
                                             >le
                                         Cylinder
                               Chamber
                                                   I
                                                   I
                                                   2
4^>
-------
           APPENDIX C




CHEVRON VAPOR COMPOSITION METHOD

-------
Chevron
         Chevron  Research Company
Ifc^^^    A Standard Oil Company of California Subsidiary
*^^P    576 Standard Avenue, Richmond, California
         Mail Address PO Box 1627, Richmond, CA 94802

A. L. Grossberg                            .  ,   Q   loo,
V,ce-PreS,den,                             JulV  9,  1981
      Mr.  Robert J.  Bryan
      Supervising Engineer
      Engineering Science
      125  West  Kuntington Drive
      P.O.  Box  358
      Arcadia,  California  91006

      Dear Mr.  Bryan:

      Recently,  Mr.  R. A. Farnham of  my  staff contacted you  in
      response  to your request to WOGA for industry assistance  with
      your EPA  contract involving vapor  pressures of heavy crudes.
      At that time,  he mentioned he had  conceived a method which  he
      believed  to be superior to other methods under considera-
      tion.  We  are  enclosing a report outlining the details of that
      method.

      We recommend that the method outlined in the report should
      receive your careful consideration for further development.
      Although  the method has not been tested in the laboratory,  it
      has  the potential of solving the difficult sampling and sample
      transfer  problems associated with  other methods listed in your
      letter to  WOGA.  The method also has other advantages:

      1.   It provides useful information on vapor composition for
           calculating total hydrocarbon  and reactive hydrocarbon
           emissions.

      2.   The results are obtained at the  temperature of interest.

      If you have any questions regarding  the above,  please contact
      Mr.  R. A.  Parnham at (415) 237-4411,  Ext.  4882.
                                   Sincerely  yours,
      Encl. - Report,  "Apparatus and Method
                for  Determining Vapor
                Composition and Vapor Presssure
                of Heavy  Stocks at 4:1 Ratio
                of V/L,"  above date

      cc:  Mr. R. N. Harrison


                                     C-l

-------
                       CHEVRON  RESEARCH  COMPANY
                         RICHMOND,  CALIFORNIA


APPARATUS AND METHOD FOR                                 JULY 9, 1981
DETERMINING VAPOR COMPOSITION
AND VAPOH PRESSURE OP HEAVY  STOCKS
AT 4:1 RATIO OF V/L

Author - R. A. Farnham
Sampling and analyzing heavy  stocks  for  vapor  pressure Is difficult
and in many cases impossible  when using  the  Reid  Vapor Pressure
Test.  The apparatus and method  proposed herein  resolves the problems
associated with the above.  Also, the  results  of  the  proposed method
provide additional data which are directly applicable to emission
calculations which pose the most frequent need for  testing of heavy
stocks.

The method using the apparatus described is  not  affected by the pres-
ence of water or dissolved air in the  sample.  The  apparatus and
method can be easily modified to give  results  at  different ratios of
V/L.  The method involves use of the equipment shown  on Attachment I.

Preparation for Sampling

Vessel A is first assembled as shown.  Valve 2 is closed; the rest are
open.  The vessel is inverted so Valve 3 is  uppermost.   Water or other
suitable liquid is introduced through  Valve  4  until liquid overflows
from Valve 3»  All valves are then closed, and the  vessel is ready for
sample introduction.

Sampling

Vessel B (which is empty) is  attached  to Vessel  A by  screwing the
close Nipple 5 into Valve 3-  Valve  1  is then  connected to the vessel
to be sampled and then opened.  Vessel B should  be  positioned so that
it is vertical and Valve 6 is uppermost.  Valve  6 should be opened and
the air vented slowly until liquid appears in  the gauge glass.
Valve 6 should then be closed quickly.   Then close  Valve 1 and discon-
nect from source.  Close Valve 3 and remove  Vessel  B.

Analysis

In the laboratory connect a 10-psig  pressure gauge  to Valve 2 and a
nitrogen source regulated to  5 psig  to Valve 1.   Open Valve 3 and
slowly pressurize with N2 to  3 psig.   Close  Valves  3  and 1.  Immerse
Vessel A into a water or an oil bath at  desired  temperature.  If after
equilibrium is reached the pressure  gauge reads  less  than 5 psig,
Encl. - Attachment I  (RE  8163W
                                    c-2

-------
                                  -2-
pressurize slowly with nitrogen  to  5  psig.   After equilibrium is
peached,  remove nitrogen  source  and withdrawn vapor sample from
Valve 1 for injection to  chromatograph.   Read pressure gauge and tem-
perature  of bath.

Calculation and Reporting Results

Prom chromatograph analysis  calculate the following:

1.  Average Molecular Weight  of  Hydrocarbon  (HC)  in Vapor, Lb/Lb Mole,
   . [A]

2.  Molecular Fraction of HC  in  Vapor,  [B]

3.  Weight Fraction Reactive  HC  of  Total  HC  in Vapor,  [C]

Using above data and pressure  gauge reading,  the  vapor pressure of the
hydrocarbon at 4:1 V/L and bath  temperature  can be  calculated.

          VP)T = [B] x [Gauge  Reading + Barometric  Pressure]

Calculation of Emissions  Using Results

Emissions can be calculated from a  source at  bath temperature and
containing stock below its bubble point if the SCFH of vapor emitted
from the  source and the fractional  approach  to equilibrium [F]  are
known.

Total HC  Emissions, Lb/Hr =  [A] 	VP)T	 x  SCFH x rpi
                                 Barometric Pressure   379

Reactive  HC Emissions, Lb/Hr  = [C]  x  Total HC Emissions

Special Cases

If the stock to be sampled is  at high temperature,  a mineral oil can
be used in place of water.  If cooling of the sample during sampling
is a problem or if very precise  V/L is required,  mineral oil can be
used and  the apparatus containing oil can be  heated to bath tempera-
ture (source temperature)  before sampling.   If heating is  required,
Valve 3 should be uppermost and  partially open during  heating .


: cd
                                    C-3

-------
                          ATTACHMENT I


        APPARATUS FOR DETERMINING VAPOR COMPOSITION AND
              VAPOR PRESSURE FOR HEAVY STOCKS  AT
                        4:1 RATIO OF V/L
For Pressure Gauge
         Identical
          Vessels
              Thermal
               Relief
                Valve
Sample Inlet and  Vapor Withdrawal


  Capped or Plugged
                                               Flexible Neoprene  or
                                               Equivalent "Sack"
                                                Screw Fit
                                               'Bolted Threaded Flanges
                                             •Total Volumn = A
                                              (of Combined Vessels)
                                              A*l Liter
       Inert Liquid  Fill
                                             JInert Liquid Fill Vent
                                             [inert Liquid Withdrawal
                              "T\  Close Couple
                                7  Nipple
                           Vessel  B
                      Volumn  =  1/5  Volumn A
                                  C-4
                   /. &.. Cs&r7T,r~
                   EVRO* RESEARCH COMPANY
                 PROCESS ENGINEERING DEPARTMENT
                 ENOR.  RES. AND DEV. DIVISION
                                                        D7V O
                                                                . DIP o i

-------
    APPENDIX D




EVAPORATION METHOD

-------
                       Characterization of Crude Oil




                   Procedure for Gravimetric Method






     The weights of several petri dishes with lids were measured and




recorded.  In order to preserve a homogeneous sample, the crude was




vigorously shaken before pouring to an approximate depth  of 1/8" in three



petri dishes.  The dishes and lids containing the sample were again




weighed and placed in a laboratory oven at  100°F.  The oven was main-




tained at 100°F  throughout the test.  A small Thomas pump purged the




oven (approx. val.  66 liters) at the rate of 2 liters/min.  This small



flowrate did not disturb the evaporation while  keepingthe percent of



hydrocarbons at an insignificant value in the oven vapor space.  The



samples were removed from the oven  and weighed at 1/2 hr. intervals



at the start of the test.  After approximately 2-1/2 hrs., the weight loss




rate dropped to a value that allowed  1  hr. weighing intervals to be sufficient.




The test was conducted for approximately seven hours.  Upon completion



of the test, weight loss and percent weight loss were calculated and



plotted versus time.
                                  D-l

-------
                                       COMPARISON OF
                            GRAVIMETRIC METHOD AND REID METHOD
     SAMPLE SOURCE
CONOCO #66
EDGINGTON REFINERY #2
CONOCO #35
STD CARPINTERIA OFFSHORE
UNION SHIPPING TANK
GETTY OIL LOYD LEASE
ARCO RRU #6
                                          PERCENT WEIGHT Loss
                                               AT 100° F
AFTER 1 HR
98,9
98,3
96,4
89,4
88,1
85,3
81.7
AFTER 2 HR
97,7
96,9
93,2
86,5
85,6
81,6
77,9
AFTER 6 HR
94,4
94,4
87,1
82,7
82.2
77.4
73.0
REID VAPOR PRESSURE
      AT 100° F
   PER UNION OIL

       0,4
       0,4
       0,6
       4,0
       2,9
       5,2
       3.8

-------
       100.00
        35.00
        90.00
HT.
           Knot
          .MB-
    o


    U)
        7S.BH
          #
                                  #
                                          #
                                                 #
                                        #
I  EDEINETQN REFINERY 12

•  5TRNDRRO CflRPENTCRIR  'OFFSHORE*

«  GETTY OIL  'LLOYD LER5E'
                                                                       #
#            #
                                                                        eg
                                                                         •

                                                                        a-

                ui

                                                                                                                   GO
                                                              TIME  CHRJ

-------
HT.
0
I
      7J.I
•
i  CONOCO tEE
•  OKOCOI3S
*  UNION 5HIPPINE TBNK
X  RRCO RRUIE

E9
ea
as

g
••

s
IM

g
in

S
X

ca
IA

S
•
10

g g
i^ m
                                                            TIKE:  (HRJ

-------
   APPENDIX E




EQUIPMENT COSTS

-------
                               APPENDIX E
                            EQUIPMENT COSTS

     Costs are estimated for specialized and laboratory equipment to
conduct the standard Reid Vapor Pressure (RVP) test and the vapor
composition test.  In the case of the RVP test, a container is required
for each sample collected at the same time.  The RVP "bomb" apparatus
is used only during the analysis, but several bombs may be needed to
avoid delay in analysis caused by the need for cleaning the apparatus.
Sample containers used for the vapor composition method are considerably
more expensive than those used for RVP testing but are close in cost to
the RVP bombs.  Unit cost estimates are given in all cases.
     Only significant items of equipment are included in our estimate.
No attempt was made to estimate the number of sample containers or RVP
bombs needed.  Costs are summarized in the attached table.
                                  E-l

-------
                       SUMMARY OF EQUIPMENT COSTS
         Cost Item
    Reid
Vapor Pressure
                   Vapor
                Composition
Laboratory Equipment

(1)  Bath, Constant Temperature
     (Thelco 261)
(2)  RVP Bomb per ASTM D323-72
     (VWR 51864-029)
(3)  Pressure Gage
     0-5 psi +0.1 psi
     per ASTM D323-72
     (VWR 51869-022)
(4)  Gas Chromatograph,
     single column, FID,
     with linear temperature
     program (Varian 1440)
(5)  Gas Chromatograph Data
     System (Varian CDS-III and
     chart recorder)
(6)  Flow meter, valves, flow
     controller, filters,
     regulators for gas chromato-
     graph

Field Equipment

(1)  Cooling Bath per ASTM D270-
     2546
(2)  Sample Containers, RVP
     samples
(3)  Sample containers, Chevron
     vapor composition procedure
$545.00

 345.00 ea.

 156.95 ea.
                S5030.00



                 4845.00


                  930.00
 200.00 ea.

  50.00 ea.
                  600.00 ea.
                                  E-2

-------
 50272-101
  REPORT DOCUMENTATION
          PAGE
1. REPORT NO.
 EPA 909/9-81-004
  4. Title and Subtitle
    Analysis of Potential Methods to Determine the
    Volatilities of  Heavy Crude  Oils
  7. Author(s)
    Robert  J.  Bryan
  9. Performing Organization Name and Address

    Engineering-Science, Inc.
    125 West Huntington Drive
    Arcadia, CA  91006
  12. Sponsoring Organization Name and Address

    U.S. Environmental  Protection  Agency, Region IX
    215 Fremont Street
    San Francisco, CA   94105
3. Recipient's Accession No
                                                5. Report Date
                                                   December  1981
                                                8. Performing Organization Rept. No
                                                   9211.00
                                                 10. Proiect/Task/Work Unit No.
                                                   W. A. 11
                                                11. Contract(C) or Grant(G) No
                                                (a 68-02-3509

                                                (G)
                                                13. Type or Report & Period Covered
                                                   Final Report
                                                14.
  IS. Supplementary Notes
  16. Abstract (Limit- 200 >ords)
                   This  report covers the  investigation of possible  alternative
              methods  to measure  the volatility of heavy crude oil.   The work was
              restricted to a literature search and inquiries made  to Informed govern-
              ment and industry groups.   The  current method for determining vapor
              pressure at storage  temperature for crude oil and refined petroleum
              products involves determination of Reid Vapor Pressure and use  of a
              correlation nomograph.  The  technique is  not applicable to some heavy
              crudes.  Also, there can be  an  undue influence from methane and ethane.

                   Alternative methods investigated include developing data to extend
              the temperature and  vapor pressure range  of the correlation nomograph,
              modifying  the Reid Vapor Pressure Method, use of a vapor composition
              approach,  and determining evaporation losses under controlled conditions.
              Sampling and analytical problems  were evaluated for the alternatives.
              A recommendation was made  to conduct further studies on the vapor
              composition method.
 17. Document Analysis a Descriptors

    Crude  oil,  vapor pressure, Reid  Vapor Pressure, volatility, evaporation loss.
    b. Identifiers/Open-Ended Terms
   c. COSATI Field/Group
  8. Availability Statement
                                                          19. Security Class (This Report)
                                                            Unclassified
                                                         20. Security Class (This Page)
                                                            Unclassified
                                                          21. No of Pages
                                                          22. Price
(S«eANSI-Z39.18)
                                         See Instructions on Reverse
                                                                                   OPTIONAL FORM 272 (4-77)
                                                                                   (Formerly NTIS-35)
                                                                                   Department of Commerce

-------