p                   UnlMdSuMs                           GRW4/0257.31
o      ^^"1*       EmNfemmenMit Prelection                      £PA-«»ff<-«-«Qn
                                                     June 1996
                   Research and
                   Development
P      ~~                       "       ~         •"—™"«  —«••

P      -ra- _ . «     m ********* m ^fm m +3>Sf\Jt            PB97-143051

p
p
P

p
p
p
                   METHANE EMISSIONS FROM THE

                   NATUBAL GAS INDUSTRY

                   "Volume 14: Glycol Dehjdrators
                   Prepared for
                   Energy Information Administration (V. S. DOE)
                   Prepared by

                   Nafionai Risk Management
                          Laboratory
                   Research Triangle Part, NC 27711
                               .

-------
                               f ICHWICAL REPORT DATA
t, *Ef> OUT NO.
EPA~iOO/R-96~080n
«. TITLE AND SUBTITLE
 .
Methane Emissions from the Natural Gas Industry,
 Volumes 1-15 (Volume 14: Glyeol Dehydrators)

               	PB37-143051
            S. REPORT' OATE  *"
             June
            S. PERFORMING ORGANIZATION CODE
           , Campbell, M. Campbell, M. Cowcill. D. Ep-
person, M,Hall, M.Harrison, K. Hummel,D .Myers,
T. Shires, B. Stapper, C. Stapper, J. Wessels, and *
            A. PERFORMING ORGANIZATION REPORT NO.
                 96-263-081-1?
I, PERFORMING ORGANIZATION NAME AND ADDRESS
Radian International LLC
P. O.  Box
Austin,  Texa»  18720-1088
                                                     10. PROGRAM ELEMENT NO,
            11. CONTBACT/BRAHT WO.
                 251-1171
            68-D1-0031
12. S?ONSOntN@ AGENCY NAME AND ADDRESS
 EPA? Office erf Research and Development
 Air PottuMon Prevention and Control Division
                   Park, NC  177E
                                                     <3. TYPE Of REPORT AND PERIOD COVERED
                                                     Final;  3/91-4/96
            14. SPONSORING AGENCY CODE
             EPA/600/13
                       project officer is D. A.Kirchgessner, MD-63,919/541-4021.
Cosponsor GM project officer is R. A. Lott, Gas Research Institute, 8600 West Bryn
Mawr Aye., Chicago, IL         (*)H. Williamson        7).	
             15-volume report summarizes the results of a comprehensive program
to quantify methane (CH4) emissions from the U. S. natural gas industry for the base
fear. The objective was to determine CH4  emissions from .the wellhead and ending
downstream at the customer's meter. The  accuracy goal      to determine these
smlssions within-+/-0.5% of natural gas production for a 90% confidence interval. For
tihe 1S82 base year, total CH4 emissions for the "0, S. natural    industry was 314
+/- 105 Bscf (6.04 -f/- 2.01 Tg).  This is equivalent  to 1.4 +/- 0. 5% of gross natural
gas production,  and reflects neither emissions reductions (per the voluntary Ameri-
Gas  Association/EPA Star Program)  nor incremental increases (due to increased
gas usage) since 1991, Results from this program were used to  compare greenhouse
             from the fuel eyele for natural gas, oil, and coal using the global war-
ming potentials (GWPs) recently  published by the Intergovernmental Panel on Climate
Change (IPCC).  The analysis showed  that natural gas contributes less to potential
global warming than coal or oil,  which  supports the fuel switching strategy suggested
   the IPCC and others. In addition, study results are being used by  the natural gas
Industry to reduce operating costs while reducing emissions.
                            KEY WORDS AND DOCUMENT ANALYSIS
                OESCniPTORS
                                         h.IDENTIFIERS/OPEN 6NOEO TERMS
                        c. COSATI Field/Croup
Pollution
Emission
Greenhouse Effect
Natural Gas
Gas Pipelines
Methane
Pollution Prevention
 Stationary Sources
 Global Warming
13 B
14G
04A
21D
15E
07C
18, DISTRIBUTION STATEMENT
 Release to Public
W. SECURITY CLASS /Thir Keportf
Unclassified
21. NO. OF PAGES
     4?
2O. SECURITY CLASS (Thitpagt)
Unclassified
                        22, PRICE

-------

      The U. S,  Environmental Protection Agency is charged by Congress with pro-
      tecting the Nation's land,  air, and water resources.  Under a mandate of national
      environmental iaws> the Agency strives to formulate and implement actions lead-
      ing to a compatible balance between human activities and the ability of natural
      systems to support     nurture life.  To meet this mandate, EPA's research
      program is providing data and technical support for solving environmental pro-
      blems today     building a science knowledge base necessary to manage our  eco-
      logical resources wisely,  understand how pollutants affect our health, and pre-
      vent or reduce environmental risks in the future.

      The National Risk Management Research Laboratory is the Agency's center for
      investigation of technological and management approaches for reducing  risks
      from threats to human health and the environment. The focus of the Laboratory's
      research program is on methods for the prevention and control of pollution to air,
      land, water,  and subsurface resources; protection of water quality in public water
      systems; remediation of contaminated      and groundwater; and prevention
      control of indoor air pollution. The goal of this research effort is to catalyze
      development and implementation of innovative. Cost-effective environmental
      technologies; develop scientific and engineering information needed by EPA to
      support regulatory and policy decisions; and provide  technical support and infor-
      mation transfer to ensure effective implementation of environmental regulations
      and strategies.

      This publication has been  produced as part of the Laboratory's strategic long-
      term research plan. It is  published and made available by EPA's Office of Re-
      search and Development to assist  the user community and to link researchers
      with their clients,


                                E. Timothy Oppelt, Director
                                 National Risk Management Research Laboratory
                                EPA

           This report has been peer and administratively reviewed by the U.S. Environmental
           Protection Agency, and approved for publication. Mention  of trade names or
           commercial products does not constitute endorsement or recommendation for use.

           This document is available to the public through the National Technical Information
           Service. Springfield, Virginia 22161.
PROTECTED UNDER INTERNATIONAL COPYRIGHT
ALL RIGHTS RESERVED.
NATIONAL TECHNICAL INFORMATION SERVICE
U.S. DEPARTMENT OF COMMERCE

-------
                                  EPA-600 /R- 86-080n
                                  June
       THE NATURAL GAS INDUSTRY,
             14; GLYCOL
              fTOAL
                Prepared by:

                Duane Myers

           Ridian International LLC
                 N,
              P.O. Box
           Austin, TX 78720-1088


            DCN: 95-263-081-15


                   For

      GRI Project Manager: Robert A. Lott
         GAS
          Contract No. 5091-251-2171
          8600 West Bryn Mawr Ave.
             Chicago, IL

                   and

  HPA Project         David A. Kirchgessner
U.S. ENVIRONMENTAL PROTECTION AGENCY
           Contract No. 68-D1-0031
  National Risk Management Research Laboratory
                     Park, NC  27711

-------
                                  DISCLAIMER

LEGAL NOTICE:  This report             by Radian             LLC as an account
of work sponsored by Gas Research Institute (GRI) and the U.S. Environmental Protection
Agency (EPA). Neither EPA, GRI,         of GRI, nor any             on behalf of
either:

a.     Makes any warranty or representation,  express or implied, w-th respect to the
       accuracy,              or          of the                     in    report, or
       that the use of any apparatus, method, or process disclosed in this report may not
              privately              or

b.             any liability with       to the use of, or for                  from the
       use of, any information,  apparatus, method, or process disclosed in this report.

NOTE: EPA's Office of Research and Development quality assurance/quality  control
(QA/QC)             are applicable to      of the                    by    project.
Emission data and additional count data are from industry  or literature sources,  and are not
subject to EPA/ORD's QA/QC policies. In all           and       were reviewed by the
panel of experts listed in Appendix D of Volume 2.

-------
                            RESEARCH SUMMARY
 Title
Contractor
Principal
Investigator

Report Period


Objective
Technical
Perspective
Results
 Methane Emissions from the Natural Gas Industry,
 Volume  14: Glyco! Dehydrators
 Final Report

 Radian International LLC

 GRI Contract Number 5091-251-2171
 EPA Contract Number 68-D1-0031

 Duane B. Myers
March 1991 - June 1996
Final Report

This report describes a study to quantify the annual methane emissions
from glycol dehydrators and acid gas recovery units (AGRs), which are
significant sources of methane emissions within the gas industry.

The  increased use of natural gas has been suggested as a strategy for
reducing the potential for global warming.  During combustion, natural
gas generates less carbon  dioxide (COj) per unit of energy produced man
either coal or oil.  On the basis of the amount of CO3 emitted, the
potential for global warming could be reduced by substituting natural gas
for coal or oil.  However, since natural gas  is primarily  methane, a potent
greenhouse  gas, losses of natural gas during production, processing,
transmission., and distribution could reduce the inherent  advantage of its
lower CO2 emissions.

To investigate mis, Gas Research Institute (GRI) and the U.S.
Environmental  Protection  Agency's Office of Research and Development
(EPA/ORD) cofunded a major study to quantify methane emissions from
U.S. natural gas operations for the 1992 base year. The  results of this
study can be used to construct global methane budgets and to determine
the relative  impact on global wanning of natural gas versus coal and oil.

The annual  emissions rates for glycol dehydrators  for each industry
segment are as follows: production,  3.42 ± 192% Bscf; gas processing,
1.05 ± 208% Bscf; transmission, 0.10 ± 392% Bscf, and storage, 0.23 ±
167% Bscf.  AGR methane  emissions  are 0.82 ± 109% Bscf.
                                         m

-------
                  Based on data from the entire program, methane emissions from natural
                  gas operations are estimated to be 314 ± 105 Bscf for the 1992  base
                  year.   This is about 1.4 ± 0.5% of gross natural gas production.  The
                  overall project also showed that the percentage of methane emitted for an
                  incremental increase in natural  gas sales would be significantly lower
                  than the baseline case.

                  The program reached its accuracy goal and provides an  accurate estimate
                  of methane emissions that can be used to construct  U.S. methane
                  inventories  and analyze fuel switching  strategies.

Technical         Glycol dehydrators are used to  remove water from natural gas streams. A
Approach         lean (low water content) glycol stream  is contacted  with the wet natural
                  gas and the glycol absorbs most of the water.  The  glycol also absorbs
                  small  amounts of methane and other natural gas constituents which may
                  then be emitted to the atmosphere when the glycol is regenerated.  AGRs
                  work in much the same way as glycol dehydrators.  A Sean (low acid  gas
                  content) amine is contacted with natural gas containing carbon dioxide
                  and/or hydrogen sulfide.  The amine preferentially absorbs the carbon
                  dioxide and hydrogen sulfide but also absorbs  some methane, which may
                  then be emitted to the atmosphere.

                  The techniques used to determine methane emissions were developed  to
                  be representative of annual emissions from  the natural gas industry.
                  However, it is impractical  to measure every source continuously  for a
                  year.   Therefore, emission  rates for glycol dehydrators and AGRs were
                  determined  by developing annual emission factors for typical units in
                  each industry segment and extrapolating these  data based on activity
                  factors to develop a national estimate, where the national emission  rate is
                  the product  of the emission factor and activity  factor.

                  Emission  factors were developed by using process simulation software to
                  model  the glyco! dehydrator and AGR process operations.  Information
                  from site  visits and other research programs  was used to develop the
                  characteristics of representative  units used in the process modeling.  An
                  emission factor was developed for glycoi dehydrators that reported the
                  amount of methane emitted per  unit of  natural  gas throughput and for
                  AGRs  that reported the amount  of methane  emitted  annually for  a typical
                  unit.

                  The development of activity factors for each industry segment are
                  presented  in a separate  report.  In general, the gas throughput for each
                  industry segment was determined from  surveys conducted across  the
                  entire industrv.
                                          IV

-------
Project           For the 1992      year the                                  for the
                 U.S.        gas         is 314      ± 105 Bscf (±        This is
                 equivalent to 1.4% ± 0.5% of gross natural gas production.         from
                              were used to compare greenhouse gas          from the
                 fuel cycle for       gas, oil, and coal      the global warming
                           (G WPs) recently          by the Intergovernmental Panel on
                 Climate Change (IPCC).  The analysis showed that natural gas
                 contributes less to potential global  warming than coal or oil, which
                 supports the fuel switching strategy suggested by IPCC and others.

                 In addition, results from this  study are being used by the natural gas
                 industry to reduce operating costs while reducing emissions.  Some
                 companies are also participating  in the Natural Gas-Star program, &
                 voluntary program sponsored by EPA's Office of Air and Radiation   in
                 cooperation with the American Gas Association to implement cost-
                                             and to report reductions to the EPA,  Since
                 this program  was begun      the 1992          year, any           in
                 methane           front this         are not         in this study's
                 Robert A. Lott
                 Senior Project Manager, Environment and Safety

-------
                            TABLE OF

                                                                          Page

 1.0               	1

 2,0   INTRODUCTION  .	,  	2

 3.0   DESCRIPTION OF GLYCOL DEHYDRATORS IN THE NATURAL GAS
      INDUSTRY  	3

      3.1   Operation Overview	,	3
      3.2   Field Gas Production .,,..,.,,	6
      3.3   Gas Processing Plants	6
      3.4   Gas Transmission	7
      3.5   Gas		7

4.0   ACTIVITY		g
      4.1    Industry Gas        Gas Throughput     Dehydrator Counts	8
            4,1.1  Production    Transmission	9
            4.1.2  Gas Processing	 10
            4.1.3         .	 10
      4.2    Other Dehydrator Characteristics	 10
            4.2.1              	 10
            4.2.2  Stripping Gas and Vapor Recovery	 11
      4.3    AGR Activity Factors	 12

5.0   EMISSION FACTORS  	 13

      5.1    Test Description		 14
      5.2    Results  of Emission Estimates	 16
      5.3    Calculated Emission  Factors	 18
      5.4    AGR Emission Factor	,23
      5.5    Emission Factor Summary	 23

6.0   ANNUAL                        	,	24

7.0			.,	25
      APPENDIX A - Emission Source       	  A-l
                                      vt

-------
                              LIST OF FIGURES

                                                                         Page

3-1   Block Process Flow Diagram of r Typical Glycol Dehydrator	4

5-1   Methane          - Glyool Regenerator Effect on Methane
      Composition	,.,.,.,	19

5-2   Methane          - Glycol            Effect on Glycol
                       	.,,,,,,	20

5-3   Methane          - Glycol      • ,-ator Effect on Flash Tank
      Pressure	 21
                              LIST OF TA1LES
                                                                         P
4-1   Estimated Annual Dehydrator Througl^ut  	,  . ,	9

4-2   Activity Factors for Flash Tank Populations .	,	  11

4-3   Dehydratots Using Stripping Gas or Vapor Recovery		  11

5-1   Test Matrix for Studying the Effect of Process Parameters on
      Methane         from Glycol             	  15

5-2   Effects of Process Parameters on Methane Emissions from Glycol
      Regenerators  .,,,,,,...,	, , ,	  17

5-3   Summarv of Glvcot Defavdtator    AGR Emission Factors	 23

-------
 1.0        SUMMARY

           This report is one of several volumes that provide background information
 supporting the Gas Research Institute and U.S. Environmental Protection Agency Office of
 Research and Development (GRI-EPA/QRD) methane emissions project  The objective of
 this comprehensive program is to quantity the methane emissions from the gas industry for
 the 1992 base year to within ± 0.5% of natural gas production starting at the wellhead and
 ending immediately downstream of the customer's meter.

           This report describes the characteristics of glycol dehydrators that affect
 methane emissions and summarizes the basis of the national estimate of emissions from mis
 source.  Also included in this category are methane emissions from acid gas removal
 (AGR) units in  gas processing plants, since AGRs are similar to glycol dehydrators in
 design and characteristics that affect emissions.

           The annual emissions for glycol dehydrators for each industry segment are as
follows:  production,  3.42 ± 192% Bscf; gas processing,  1.05 ± 208% Bscf; transmission,
0.10 ± 392% Bscf; and storage, 0.23 ± 167% Bscf. AGR methane emissions are 0.82 ±
 109% Bscf.

-------
 2.0        INTRODUCTION

            Dehydrator activity factor" demographics were developed on the basis of data
 from several surveys.  The percentage of glycol dehydrators (as opposed to molecular sieve
 or other types) was established to be about 95% of the total population of 41,700, for a
 count of 39,615 glycol dehydrators nationwide.1  Initially, the count of dehydrators in each
 industry segment was used as the activity factor.  At the suggestion of the Industry
 Working Group (industry members who serve as project advisors),  the activity factor basis
 was changed to dehydrator gas throughput.  The final  activity factors used by this project
 are documented in Section 4. An emission factorb was developed using information from
 field measurements, as well as a computer simulation using ASPEN/SP® software.  The
 emission factor results are reported in Section 5.  The  estimated annual methane emissions
 from dehydrators from each industry  segment are given in Section 6.
    "An activity factor is a count of the total industry population of a particular type of
source.  It is the total number of sources in the entire target population or source category.
    bAn emission factor for a source category is a measure of the average annual emission
per source.  It is the summation  of all measured or calculated emissions from sampled
sources  divided by the total number of sources in the category that were sampled.

-------
 3.0       DESCRIPTION OF GLYCOL DEHYDRATO1S  IN THE NATURAL  GAS
           INDUSTRY
           This        describes the glycol            found in the       gas industry.
 as -well as           in           in various          of the industry.

 3.1        OperatioB Overview

           Dehydrators  are designed to remove water from the natural gas vapor stream,
                  and          the          of hydrates, which are
           that can       flow           and         in valves and ever, pipelines.
 There are several types of dehydrators.        from      molecular
 to liquid absorption dehydrators.

           Glycol dehydrators  are liquid absorption units that absorb water in a liquid
 glycol stream.  Approximately  95% of glycol dehydrators use triethylene glycol  (TEG),
 with      of the remainder      ethylene       (EG).  (TEG and EG     wry different
 properties for water removal but are       for methane             The
 usually        of two primary          the absorber and the regenerator.  Figure 3-1
 shows a typical block flow diagram for a glycol dehydrate* The lean liquid glycol usually
 flows downward in an absorption tower, counter-current to the natural gas. The glycol
 absorbs  most of the water from the  natural gas, but it also absorbs other materials present in
 the gas stream.  The dried        gas      the top of the tower. The water-rich glycol
      the bottom of the tower  and flows to the regenerator. The                the
glycol to drive off water vapor, and the       vapor is       vented        to the
atmosphere through the regenerator  vent stack. The lean glycol is     returned  to the
absorber.  Glycol has a high affinity for water and a relatively low affinity for non-aromatic
hydrocarbons, which makes it a very good absorbent fluid for drying  natural gas. However,
the glycol does                     of        and      hydrocarbons  from the
gas.  The hydrocarbons are        to the                 with the      vapor      the
           vent

-------
Wet
Gas
                                             Dry Gas to
                                              Pipeline
                                                             Water Vapor and
                                                            Light Hydrocarbons,
                                                              Including CII4
                                                             (to Atmosphere or
                                                            to Control System)
                             Dry
                            Glyco!
Wet Glycol
%
                           and Gas
                        (High Pressure)
             Absorber
           (High Pressure)
                                                                         Combustion
                                                                            Gases
                Glycol
                  Gas
                Pump
                                                           Glycol
                                                         Reboiier/
                                                        Regenerator
                                                            Fla§h Tank
                                                              (Optional)
                                                            at Intermediate
                                                              Pressure
                                                                                                      Firebox
                                                                                                                    Fuel
                                                                                          ReboMer
                                                                                         (Atmospheric
                                                                                          Pressure)
                     Figure 3-1,  Block Process Flow  Diagram of a Typical Glycol Dehydrator

-------
           All giycol dehydrators have pumps to         the glycol.       pumps in the
      are                   that greatly increase the                        the glycol
 unit These pumps are powered by          (wet) line gas, and the               gas is
 dumped     the rich glycol        and        off in the regenerator. For the purposes of
     study, the                    were considered         sources, even        the
 methane they use is vented through the regenerator.   Gas-assisted pumps are          in a
 separate report,2 and are not included in this analysis of dehydrator emissions.

           Some glycol dehydrators have additional equipment.  Two common additions
 are flash tanks  and regenerator vent emissions control equipment.  The flash tank is placed
 in the rich glycol loop between the absorber and the regenerator. The glycol line pressure
 is dropped in the      tank, causing most of the      hydrocarbons to          the vapor
 phase.  The flash gas is               to the           burner as fuel.  The
          from the            vent can be significantly reduced by      a      tank.

           Regenerator vent control devices      been         on           to
          of benzene, toluene, ethylbenzene, and xylenes  (BTEX) and volatile organic
 compounds (VOC) to the atmosphere.  These            are absorbed fern the gas
 and driven off with the  water in the regenerator vent  Control devices usually condense the
 water and hydrocarbon (containing BTEX and heavier VOC), then decant the hydrocarbon
 for sale and the water for disposal.  The methane in the vent is not condensed and is
 usually  vented,  but it can be flared or used as fuel in the regenerator burner.  Many glycol
 dehydrator operators have installed some type of vapor recovery system on the regenerator
 still vent, although the controls are primarily targeted for BTEX and not         control.

           Some dehydrators use stripping gas to the regenerator,           the
outlet or from the flash      is introduced into the            to help     the water and
otter         compounds out of the glycol by           the vapor flow rate to the
rebofler still.  Methane in the stripping gas       directly through the regenerator into the
            vent

-------
 3.2        Field Gas Production

            Field production removes water in two       First, a                 vessel
 removes the liquid        (free water and oil) from the IK viral       This liquid phase water
 is                    the oil to preserve the purity of the oil. The g?s      the top of the
 separator often remains  saturated with absorbed water and is treated again by field
 dehydrators to dry the gas to low parts per million (ppm) levels of water to prevent
 corrosion and plugging of the gathering lines.

            Many field dehydrators are small glycol units with very little instrumentation
 and without flash tanks.  Comparatively few production units have regenerator vent
         controls,          more          are           controls as new environmental
 regulations      effect.   Many production           glycol        driven by gas-pressure
 letdown.       production      use  TEG as the absorption fluid,

33        Gas Processing Plants

           Dehydration  is fundamental to gas processing plants, especially those that use
refrigerated or cryogenic liquids recovery methods.   However, if water is present, the cold
temperatures promote the formation of hydrates.  Therefore, gas processing plants use
molecular sieve  beds or  glycol dehydrators  upstream of the liquids recovery section.

           Some plants  do not use a typical dehydrator configuration with an absorber.
Rather, they inject the glycol directly into the gas stream and allow contact to occur in the
pipeline.   The entire        then       through a separator, where the dry      rich liquid
glyco!, and condensed  hydrocarbon       are            The rich glycol        to a
regenerator and  is recycled to the injection point.       injection-type            use
ethylene  glycol (EG) as the           liquid.

-------
            Plants that use a typical absorber tower may or may not have a flash tank or
 vent recovery equipment.  Some plants may route die vent gases to a plant flare system.
 Most plant glycol pumps are powered by electricity instead of gas.

            AGE units have the same basic equipment as a glycol dehydrator:  an absorber
 tower, a pump to circulate the liquid, and a reboiler to regenerate the absorber liquid.
 AGRs typically use an aqueous solution of one of a variety of amine  compounds (e.g.,
 monoethanolamiffle, diethanolamine)  to remove carbon dioxide and hydrogen sulfide from
 natural gas.

 3.4         Gts Transmission

            Production gas is typically dry when it enters a gas  transmission system, having
 passed through field production and gas processing plant dehydrators.  There usually is no
 need to dry gas being transported through the pipeline, although some pipeline gas is
 dehydrated.

 3.5        Gas Storage

           Oas stored ysd£raToiisd for distribution dunn° **e3k ussge KSV nick *m wnter
 and need to be dehydrated.  Dehydrators used to dry stored gas are typically me same
 design as production field dehydrators but tend to be much larger and better maintained.
 These large storage dehydrators are more likely to include flash tanks and some type of
vent recovery system than are production field dehydrators.

-------
 4.0        ACTIVITY FACTORS

            This section briefly summarizes activity factor calculations for dehydrators and
 AGRs. A more detailed discussion is presented in the Volume 5 on activity factors.3  The
 results account for the 90% confidence limits calculated for each activity factor.

            The overall activity factor for each industry segment is the total segment gas
 throughput.  The overall activity  factor is multiplied by the emission factor (given in
 Section 4) to obtain the  annual methane emissions.

           Other characteristics of glycol dehydrators are used in the calculations overall
 activity factor and emission factor. These include:

           •        Number  of dehydrators;
           •        Dehydrator throughput;
           •        Fraction  of dehydrators with flash tanks;
           •        Fraction  of dehydrators with stripping gas; and
           •        Fraction  of dehydrators with vent vapor recovery.

More specific information for  each characteristic is given in the following sections.

4.1        Industry Segment Gas Throughput and Dehvdrator Counts

           The overall activity factors are the amount of gas dehydrated annually in each
industry segment.  The estimated  annual glycol dehydrator  throughputs for each industry
segment are listed  in Table 4-1.

-------
       TABLE 4-1.  ESTIMATED ANNUAL PEHYPRATOR THROUGHPUT

           .,   .    ;.	Gas T&rougjBgat (MMscgyear)

 Production                                            12.4 x 10 6 ± 61.9%

 Processing                                            8.63 x 10 6 ± 22.4%
 Transmission                                           1.09 x 10 6 ± 144%

 Storage                                               2.00 x 10 * ± 25.0%
 Total Gas Industry                        	24.12 x 10  6± 33.5%


The total industry segment throughputs were calculated in several different ways be

discussed below.
4.1.1      Production and Transmission


          The activity factors for production and transmission were calculated using the
equation:
                  AF= P x CP x CU x 365 days/year                            (1)

                  P     =     Population of dehydrators in each industry segment (see
                               Appendix A)
                               Production:         37,824 ±21.1%
                               Transmission:       201 ± 119%

                  CP    =     Average  gas throughput capacity per dehydrator1
                               (NlMsefd)
                               Production:         2.00 ±28.1%
                               Transmission:       14.8 ± 29.5%

                  CU    -     Capacity utilization—ratio of actual gas throughput to
                               capacity  (see Appendix A)
                               Production:         0.45 ± 32%
                               Transmission:       1.00 ± 0%

-------
 4.1.2      Gas Processing

           The gas           activity                     from the reported gas plant
 throughput and        type from the  Oil and Gas Journal        survey of gas plants.4  It
                  all gas       using  a                    use glycol dehydration and gas
 plants using a cryogenic process use some type of dry-bed dehydration (which has
 negligible methane emissions). The fraction of gas processed by glycol dehydrators was
 determined to be 0.495 (or 8.63 Tsctfyear) of a total of 17.44 Tscf/year.

 413      Storage

           The        activity factor  was          from the        of gas removed from
                    annually (2.4 Tsef) as         in A.G.A. Gas Facts,5  It
    most gas removed                          is dehydrated by glycol; 2.0 Tscf/year ±
25% was     as the activity factor.

42        Other Dehydrator Characteristics

           Fractions of dehydrator populations with flash tanks, stripping gas, and vapor
recovery systems were also used in the emission calculations.  These characteristics and the
field data can be found in the Activity  Factor report.3

4,2.1       Flash Tanks

           The         of glycol dehydrators with            was           by combining
the       of site surveys with the        of a survey            by the        Mid-
Continent Oil                     (TMOGA).S  The fractions      in the emission factor
calculations are listed in Table 3-2.
                                          10

-------
      TABLE 4-2.  ACTIVITY FACTORS FOR FLASH TANK POPULATIONS
                                                      Ilasfe Tacks
 Production                                        0.265 ± 8.35%
 Processing                                        0.667 ± 10.1%
 Transmission                                      0.669 ± 9.70%
 Storage                                           0.520 ± 33.6%
4.2.2      Stripping Gas and Vapor Recovery

          The fractions of glycol dehydrators that use stripping gas in the regenerator or
have a vapor recovery system that eliminates methane emissions were estimated from the
results of site surveys. The fractions used in me emission factor calculations are listed in
Table 4-3.

  TABLE 4*3.  DEHYDRATORS USING STRIPPING GAS OR VAPOR RECOVERY
Production                      0.0047 ± 116%            0.012 ±73.1%
Processing                       0.111 ± 186%               0.10 ± 0%*
Transmission                     0.074± 118%             0.148 ± 80.3%
Transmission	0.080 ± 118%             0.160 ± 80.8%
*For the emissions calculations it was assumed that 10% of ga? processing dehydrators have
vent controls, although none were observed during the site visits.
                                      11

-------
43        AGR Activity Factors

           The number of amine-based AGRs in gas processing service has been reported
to be 371 in a report for GRI by Purvin & Gertz, Inc.7  Confidence limits were not given in
the report; therefore, they were assumed to be ± 20%.  Assuming an average AGR gas
throughput of 36,5 MMscfd ± 20% (equal to a gas processing dehydrator throughput), the
AGR activity factor is  1.354 x 104 MMscf/year.  Another survey reported that  18% of the
AGR reboilers vent directly to the atmosphere and would be a source of methane
emissions.8
                                         12

-------
SJ                    FACTORS

           Estimates of methane emissions from dehydrators were developed using
estimates from computer simulation and some field <*ata measurements,  ASPEN/SP* (from
Simulation Sciences, Inc.) process simulation software was used for several case studies.9

           Glyco! dehydrators have numerous                  affect
from the regenerator vent.  Using a compiler simulation model and varying the key
dehydr&tor parameters,  die following characteristics of glycol dehy dravors that affect
emissions were examined:

           •        Overall unit
                    —     Size of the unit (MMscf of gas processed/day)
                    —     Gtycol type
                    —     Glycol circulation me
                    —     Lean glycol percent water
                    -     Regenerator reboiler temperature
           *        Inlet gas information
                          Methane composition
                   Flash tank information
                   -     Use of t flash tank
                          Pressure
                   —     Temperature
                   Stripping  gas use
                   V«ot weowry/eotttroi equipment
          The size of the unit affects how much methane is contacted, how much glyeol is
circulated, and therefore how much methane is absorbed.  Several types of glycol can be
                                         13

-------
 -..    .                ™    .   .                ...                  ,        _.         -.
 T.T lr°l.-^T}  %^**% A J5_r^= ****** M-f^r «&a *^ &kAV iia^^iji, w~^i&Aa~£*Vf **•  *_«*••**«* OA «Aa^'h?%' ^*J w^Ak? £«M*> ** O^AAA^B iofiii. «£sS.,&A.&u *• T
 for methane.   The glycol circulation rate affects the contact time, and therefore how much
 methane  is absorbed by the glycol.  The lean glycol water concentration is a measure of
 how well the regenerator has restored the glycol before it returns  to the absorber.  A high
 concentration of water in the lean glycol reduces its ability to absorb water from the gas
 stream.

            The inlet methane composition, gas temperature, and gas pressure affect the
 methane  partial pressure in the absorber.  This changes the amount of methane relative to
 other materials that can be  absorbed by the glycol.

            Many characteristics that were judged to have a negligible  effect on the amount
 of methane absorbed were eliminated from consideration.  Examples are the number of
 trays in the reboiler still, the inlet BTEX composition, and the inlet  water composition.

5.1        Test Description

           A matrix approach was used to study the effect of process  parameters on
methane emissions from a glycol regenerator.  The process parameters include:

           •        Methane  composition;
           •        Glycol circulation rate;
           •        Lean glycol water content;
           •        Flash tank temperature and pressure;
           •        Gas flow rate; and
           •        Gas temperature and pressure.

           The test matrix is  shown  in Table 5-1.
                                           14

-------
   TABLE 5-1. TEST MATRIX FOR STUDYING THE        OF                      ON METHANE
                                           GLYCOL
•. - '-,^m ';*i^vfam^Mm
Methane Composition (wt%)
Glycoi Circulation Rite (gph)
Leal Olyeol (% water)
Tank (pM§)
Flash Tank Temperature fF)
Oas Flow Rate (MMscfd)
Gas Temperature (°F)
Gas Pressure (psig)
'"'" " '*Y;tf i' :'.;*V::: : :" '": ::!:^i;':
:%-iwii|i|P
iiiiiiii



15




Sliiil*^
S5
4.75
0.5
30
70
0.9
90
600
^;':^:^::S-'r''':':^::r-:::^':i::
^I^J^lWv:-;-^
87.5
7.14

45




iisii
iHiliii
90
9.48
1
60
110
1
95
800
giiii^
92.5
1L88

75




'-Ipi
,:;:;;;: ^f.WfllP^^:;.
95
14.28
1.5
90
150
I.I
100
1000
rS^ii^l
lipi
V«itie
l'''lim'i*'r''':';''' "'"'" "'



120




f3S!pRii
iiMliiiir



No tan c

10«


1 Glycoi circulation rate is also mcrestsed by a factor of ten.

-------
 input information lor a base case denyorator was chosen to represent Radian s  oest estimate
 of average dehydrator parameters based on the company's experience with permitting
 dehydrators and performing dehydrator studies for GRI and other private clients.

            Initially, the base case was run to determine the emissions and to establish  the
 number of theoretical  stages for the glycol dehydrator.  (The number of theoretical stages
 for a dehydrator is the number of absorber trays, with the gas and glycol at equilibrium,
 required to dry the gas to pipeline specification.) Then, low and high values were studied
 for each parameter.  During the evaluation of one parameter, the other process parameters
 were kept at the base  case values. A few supplemental cases were also  studied.

            After running the initial  tests, the matrix was expanded  for the parameters that
 showed the most variability.  More tests were performed on the methane composition,
 glycol recirculation rate, and flash tank pressure.  A run was also performed at a gas flow
 rate ten times the base case value. The glycol recirculation rate was correspondingly
 increased by a factor of ten. The emission rate for  this case  was found  to be exactly one
 order of magnitude larger than the base case (0.0837 to 0.837 tons/yr), which indicates that
 the emission rate is linear with the flow rate, assuming that the glycol-to-gas ratio remains
 constant.

 5.2        Results of Emission Estimates

           Table 5-2 presents the results of the emission estimates  generated from the
ASPEN/SP* model runs.9  The glycol circulation rate remained proportional to the gas  flow
rate  to maintain a constant glycol-to-gas ratio.  Emission  rate was found  to be directly,
linearly proportional to the gas flow rate if the glycol-to-gas ratio was held constant.  The
other variables also produced nearly linear relations.
                                           16

-------
      TABLE S-2. EFFECTS OF PROCESS PARAMETERS ON METHANE EMISSIONS FROM GLYCOL
                                          REGENERATORS
v; - ^ ;...;;;;;;J||i^
Methane Composition (voi%)
Methane Emissions (tons/yr)
Qlycol Circulation Rate (gph)
Methane Emissions (tons/yr)
Lean Olycol (% water)
Methane Emissions (tons/yr)
Wash Tank Pressure (psig)
Methane Emissions (tons/yr)
Flash Tank Temperature (°F)
(tons/yr)
(las Flow Rate (MMscfd)
(tons/yr)
Gas Temperature (°F)
Emissions (tons/yr)
Gas Pressure (pstg)
Methane Emissions (tons/yr)
liiiiP
iliiili



IS
0,0261




'^•'•:-;";'/''-'^???r?rv £ ; v"' :-::
S5
0.0701
4.75
O.Q4191
OJ
0.0841
30
0.0442
70
0.092
0.9
O.OS37
90
0,0832
600
0.0837
•^JM^itiiffif" '
K\^^it&^
S7.5
0.0767
7.14
0.0626

45
0.0635




•Bill
90
0.0837
9.48
0.0837
1
0.0837
60
0.0837
no
0.0837
1
0.0837
95
0.0837
800
0.0837
Medium
liiii
92.5
0.0911
11.88
0.104

75
0.104





95
0.0999
14.28
0.125
1.5
0.0832
90
0.125
150
0.0753
1.1
0.0837
100
0.0841
1000
0.0837
iipi
MMi



120
0.168




;::>gx.:-:;:-:;:;-^-i:; .- ..'-vj.;. .,/:',;£,.•,. :



No tank
1.12

10b
0.837


1 Results not valid since the dry gas water content is greater than 7 Ib H2O/MMscf.
b Olycoi circulation rate is also increased by a factor of ten.
Number of absorber trays is fixed at 1.41,

-------
            Figures 5-1, 5-2, and 5-3 show the relation of methane composition, glycol
 circulation rate, and flash tank pressure on methane emissions.  The single largest effect on
 the total emission rate was the presence or absence of a flash tank.  A flash tank can reduce
 methane emissions by a factor of ten.  One parameter not modeled was the addition of
 stripping gas.  When stripping gas is added to the  regenerator, all of it should exit as
 exhaust through the regenerator vent.  This parameter has a major effect on dehydrator
 methane emissions.

 53        Calculated Emission  Factors

           The variables accounted for in the emission factor calculations were presence of
 a flash tank, use of stripping gas,  presence of a vapor recovery  device on the still vent, and
 dehydrator gas throughput. Based on field observations  from other GRI programs and on
 input from industry advisors, a dehydrator capacity utilization factor  and glycol
 overcirculation factor were included.

           These data were used to produce a national emission factor estimate for the
 average dehydrator in  each industry segment using the average dehydrator capacity  for each
 segment.  Emission factors were derived for dehydrators in the production, gas processing,
 and transmission and transmission segments by the same basic method.

           A thermodynamic computer simulation  was used to determine the most
 important variables affecting emissions from dehydrators.  The important variables are gas
throughput, existence of a flash tank, existence of stripping gas, existence of a gas-driven
glycol pump, and existence of vent controls routed to  a burner.  Other variables (e.g..
reboiler temperature) were determined to be relatively unimportant from an emissions
standpoint.

           Throughput, since its effect is linear,  is handbd by establishing an emission rate
per unit of gas throughput for all dehydrators.  Emission rates per unit of throughput are
                                           18

-------
       0.12
       0.10
1
       0.04
       0.02
       0.00
             84
Si
SS
m
94
                             Methaae Composition (mo!e%)
    Figure 5-1. Metbane         - Gtycol Regenerator Effect on        Composition

-------
        
-------
  0.1$
  0,10
  0,05
  0.00
                                         11
13
                                  Tank Pressure
figure 5-3.                   - Giyeol Rsfweritor       on Hash Tank Pressure
                                   21

-------
then established for the other important emission-affecting characteristics.  Gas-assisted
pumps are ignored here and handled in a separate source analysis.2  The stripping gas rate
was determined from observations at one site from the GUI Glycol Dehydrator Sampling
and Analytical Program.10 The emission factor is then:


          EF = [ (Fpr x EF^) + (FOT x EF^) + (Px x EFSG) ] x  Fwc x OC           (2)
          Fpj       —     Fraction of the population WITH flash tanks
                                 Production:          0.265 ± 8.35%
                                 Gas processing:      0.667 ± 10.1%
                                 Transmission:       0.669 ± 9.70%
                                 Storage:             0.520 ± 33.6%

          FOT       ~     Fraction of the population WITHOUT flash tanks
                                 Production:          0.735 ± 2.99%
                                 Gas processing:      0.333 ± 20.1%
                                 Transmission:       0.331 ± 19.6%
                                 Storage:             0.480 + 36.3%

          FSG       =     Fraction of the population WITH stripping gas
                                 Production:          0.0047 ± 1 16%
                                 Gas processing:      0.111 ± 186%
                                 Transmission:       0.074 ± 118%
                                 Storage:             0.080 ±118%

          FNVC     =     Fraction of the population WITHOUT combustion vent controls
                                 Production:          0.988 ± 0.87%
                                 Gas processing:      0.900 ± 10%  (estimated)
                                 Transmission:       0.852 ± 14.0%
                                 Storage:             0.840 ± 15.2%

          EFj-j-     =     Total methane emission rate scf per 1 MMscf throughput per
                          dehydrator with a flash tank
                                 All:                3.57 + 102% /- 58%

          EFN-j.     =     Total methane emission rate scf per 1 MMscf throughput per
                          dehydrator WITHOUT a flash tank
                                 All:                175.10 + 101% /- 50%
                                         22

-------
          EFSG     =                              rate per 1 MMscf throughput per
                         dehydrator that has         gas
                              All:               670 + 40% / - 60%
          OC      =     Glyeol              factor—number of times the industry
                         rule-of-thumb of 3       giycol/pound water
                              Production:        2.1 ± 41%
                              Others:            1.0 ± 0%
All of the emission factors (EFs) in these equations, such as EF^, EF,^ and EF^, were
derived from the modeling described in Section 5.2.

5.4       AGR Emission Factor
          The ACR         factor was          by                      for a
typical unit He                         were 965 sef CH4/MMscf gas throughput.
Assuming an average AGR gas throughput of 36.5 MMsefd (equal to a gas processing
dehydrator throughput1) and a fraction of AGRs venting methane to the atmosphere of
0.18,* the methane         for a typed AGR       be

S.S       Emissioji factor tSmuBMPC

          The emission fectors for each dehydrator industry segment and for AGRs are
listed in Table 5-3.
           TABLE 5-3.            Of GLYCQL DEHYDRATOR AND
                          AGR            FACTORS
 ftoduction                                         275.6 ± 154%
 C^s                                               121.6 ±
                                                   93.72 ±
                                                   117.2 ±160%
 AGRs                                        6083          ±
                                      23

-------
 6.0        ANNUAL METHANE EMISSIONS

           Annual methane emissions from glycol dehydrators in each industry segment
 and from AGRs were calculated by multiplying the activity factor by the emission factor.
 The results are as follows:

           •    Production:
               275.6 scf CH4/MMscf x 12.4x10 6 MMscf =  3.4 Bscf ± 192%
           •    Gas Processing:
               121.6 scf CH4/MMscf x 8.63x10 6 MMscf =  1.1 Bscf ± 208%
           •    Transmission:
               93.72 scf CH4/MMscf x 1.09x10 6 MMscf =  0.1 Bscf ± 392%
           •    Storage:
               117.2 scf CH4/MMscf x 2.00x10 6 MMscf =  0.2 Bscf ± 167%
           «   AGRs (Production and Gas Processing):
              6083 scfd/unit x 371 units x 365 days  = 0.8 Bscf ± 109%

           The estimate for annual methane emissions from glycol dehydrators is 4.8 Bscf.  The
estimate of annual methane emissions from AGRs is 0.8 Bscf.
                                           24

-------

 I.         Wright       & Co,  Natural Gas Dehydration:        and Trends, GRI-
           94/W99, Gas                  Chicago, IL, January 1994,

 2.         Myers, D.B. and M.R. Harrison. Methane Emissions from the Natural Gas
           Industry, Volume 15:  Gas-Assisted Gfycol Pumps.  Final Report, GRI-
           M/025733  and EPA-600/R-96-OSOQ,   Gas Research         and U.S.
           Envtoninental Protection Agency, Jane 1996.

 3.         Stapper, B.E. Methane Emissions from the Natural Gas Industry, Volume 5:
           Activity Factors.  Final Report, GRI-94/0257.22 and EPA-600/R~96-08Qe.  Gas
           Research        and U.S. Environmental Protection Agency,  June  1996.

 4.         Bell, L. "Worldwide  Gas Processing.''  Oil and Gas Journal.  June  13, 1994,
           pp. 63-91.

 5.         American Gas Association.  Gas Facts 1994:  A Statistical Record of the Gas
           Industry.  Arlington, VA, 1994.

6.         Texas Mid-Continent  Oil and Gas Association (TMOGA) and Gas Processors
                      (GPA),            Survey, 1991.

7.         Tanndull, C.C. and C. Calvin.  Business Characteristics of the Natural Gas
           Conditioning Industry. Topical Report, GRI-93/0342, prepared by Purvin &
           Gate, Inc.,  Gas         Institute, Why 1993.

S,         Radian Corporation.  Investigation of U.S. Natural Gas Reserve Demographies
           anrf Gas Treatment Processes* Topical Report, Gas Research Institute, January
           1991.

9.                Coiporation,  ASPEN/SP*                   project files, 1993.

10.        Radian Corporation. Glycol Dehyarator Emissions:  Sampling and Analytical
           Methods and Estimation Techniques.  GRI-94/0324,  Gas Research Institute,
           Chicago, IL,       1995.
                                        25

-------
   APPENDIX A




Eaussfee Sewce Sfcetfe
       A-l

-------
                                                P-6
                                 PRODUCTION SOUR-rt

 SOURCES:                           Glycol Dehydrators
 COMPONENTS:                      N/A
 OPERATING MODE:                 Normal Operation
            TYPE:                    Vented
 ANNUAL EMISSIONS:                3.42 Bscf ± 192%

 BACKGROUND:

 Glycol dehydrators remove water from a gas stream by contacting the gas with giycol and then driving the
 water from the giycol by heating in the glycol reboiler and into the atmosphere.  The giycol also absorbs a
 small amount of methane, and      methane can be driven off to the atmosphere through the reboiler vent,

 EMISSION  FACTOR:   (275,57 scCMMscf gas processed  ± 154.48%)

 A thermodynamic computer simulation was used to          the most important emission-affecting variables
 far dehytfestors. Tlhc variables are; gas throughput,  existence  of a          existence of stripping gas,
 existence of a gas driven pump, and  existence of vent controls routed to a burner. Throughput, since its
effect is linear, is handled by establishing an emission rate per unit of gas throughput.  Emission rates per unit
of throughput are then established for tbe other important emission affecting characteristics.  Gas driven
      ass ignored here and        in a               analysis (see Methane Emissions from the Natural
Ga$ Industry,  ¥olume IS:  Gas-Assisted Giycol Pumps) (I).  The emission factor is then:

     EF     -    [(FrTxEFFT) + (FwxEFOT) + (FSGxEFsa)]x  F^ x OC

                      [       x 3.57) + (0,735 x 175.10) +         x 670) j x        x 2.1

                 Fpr   =      Fraction of the population WITH flash tanks
                                   ± 8.35%
                 FOT  =      Fraction of the population WITHOUT flash tanks
                              0.735 ± 2.99%
                 FSG   =      Fraction of the population WITH stripping gas
                              0,00473 ± 115.78%
                 FNVC  =      Fraction of the population WITHOUT combustion vent controls
                                    ±
                 EFpj.  =     Total methane emission rate scf per i MMscf throughput with a flash tank
                             3.57 +102%/-58%
                 EFffr =       Total methane emission rate scf per I MMscf throughput WITHOUT a
                                  tank
                              175.10 -H01%/-5G%
                 EFSC  =      Incremental methane emission rate per 1  MMscf throughput per dehydrator
                                 has snipping gas
                             670 +40%/-60%
                 CXI*   =      Overcirculation factor For giycol—number of times the industry mle-of-
                             thumb of 3 gallons glyeol/lb water
                             2.1  ±41%
                                              A-2

-------
 EF DATA SOURCES:'

             1.    Methane Emissions from the Natural Gas Industry,  Volume 14: Gfycol Dehydrators (2)
                  establishes emission affecting characteristics of dehydrates,
             2.    GRI/EPA site visit data          fl» fm and F^ for         sites (19 PROD
             3.    An analysis of a          database          TMOGA's 1019 dehydrators and GRI/EPA
                  site      444 dehydrators           Fw and F^j, for production ckbydrators.
             4.    ASPEN computer           were used in combination with measured data to
                  fPn, and EFW from the dehydrator vent
             5.    Sampling data from die GRI Giycoi Sampling and Analytical Program  for one dehydrator
                  was used to determine EF^ {Gfycol Dettydraiar Emissions:  Sampling  and Analytical
                  Methods and Estimation Techniques) (3), The upper bound was calculated by assuming
                  that a!! of the measured noncondensable vent gas was due to stripping  gas that was 100%
                  methane.  The lower bound was calculated as the rale*of-thumb stripping gas rite recom-
                  mended by a glycoi dehydrator manuf&eturer.
             4    Overcirculation factor determined      data from the GR] Giycoi Sampling aid
                  Analytical Program data for ten dehydrators.


 EF               275J7           gts processed ± 154.4S%

             Basis:

             The accuracy is propagated through die EF calculation from each term's accuracy:

             1,    ASPEN has been demonstrated to match actual dehydrators within ±20% within the
                  eaknbtted confidence intervals obtained from site data,

             2,    Individsml EF confidence satervals were calculated irom the data      in the calcalanon.

             3.    Data from she visits has been assigned confidence intervals based      the spread of die
                  444 dehydrators from GRI/EPA  site data,

ACTIVITY FACTOR:  (1X4 Tsc^ytar f*s tkrwigbprt  in the proAKtioa segmeat)

The amount of gas processed by glyco! dehydrators in die production segment was calculated from the
estimated number of giycol dehydrators in production aid the average throughput capacity for production
dehydntms (Wright Killers &. Co., 1994),  A capacity  utilization factor was estimated based on observations
at several      m die  CM Giyco! Dehydrstor Sampling and Analytical Program.

At DATA

The report: Natwal Gas Deky&wlion  Status and Trends (4) by  Wright Killen tor the Gas  Research Institute,
provides data and describes die  raethodology used to develop an estimate of the gas dehydrator count for the
U.S.  The  count also estimated the number in several industry segments: production, transmission,  and gas
processing.
                                               A-3

-------
 Basis:
             I.   A GRJ       by Wright Kilfcn & Co. found 41700            in the U.S. gas industry
                  for 1993.  Wright       also     a TMOGA/GPA          on dehydrators to split the
                  population into the following industry segments:
                        Production:             25270
                        Processing:             7923
                        Transmission:           8507
                        TOTAL:                4J70Q
                  The study also found that 95.0 % of the dehydrators were glycol for a total of 39,615
                  (versus molecular sieve or other types).

             2.   Site visit data on 24 transmission compressor stations shows: 2/17 = 0.118 per
                  transmission compressor station, and 17/6 = 2,83 per storage compressor station.  The  site
                  visit numbers would lead to an estimate of 1293 total transmission and storage
                  dehydrators.  Site visit data on 11 gas plants show 1.41 dehydrators per plant,  or  1,024 in
                  gas plants.

                  Subtracting processing,  transmission, and storage glycol dehydrators from the total of
                  39,615 yields 37824 glycol dehydrators in production.

             3,   Average capacity of production dehydrators was reported to be 2 MMscfd by Wright
                  Killea,

Information on  actual dehydrator throughput as compared to       capacity is, in general, difficult to obtain
especially for production field units. Data from several sites in the GRI Glycol Dehydrator Sampling and
Analytical Program and other          information from various site visits indicate mat capacity utilization
may be    Aan 50%,  so a value of 4S% was chosen for the AF calculations.


AF PRECISION:  12.4 Tscf/year ± 61.87%

             Basis:
             The 90% confidence limits for total glycol dehydrators were established  in the Wright Killen
             report.  The confidence limits for the segments other than production were based on  site visit
             data. Confidence limits for the capacity utilization was based on engineering judgement.

ANNUAL METHANE EMISSIONS;  (3.4171 Bscf/yr ± 191,90%)

The annual methane emissions were determined by multiplying the dehydrator emission factor by the activity
factor.

REFERENCES

I.           Myers,  D.B. and M.R. Harrison.  Methane Emissions from the Natural Gas Industry,  Volume
             15: Gas-Assisted Glycol Pumps,       Report, GRI-94/0257.33 and EPA-6Q0/M-96-QgQo. Gas
             Research  Institute and U.S. Environmental Protection Agency, June  1996.

2,           Mjsrs,  D.B.  Methane Emissions from the Natural Gas Industry,  Volume 14: Glycol
             Dehydrators, Final Report, GRI-94/0257.31  and EPA-600/R-96-080n.  Gas Research Institute
             and U.S. Environmental Protection Agency, June  1996.
                                                A-4

-------
3,           Radian Corporation. Gtycol Dehydraiat- Emissions: Sampling and Analytical Methods and
             Estimation Techniques  GRI-94/0324, Gm Research Institute, Chicago, IL, March 1995,

4.           Wright Kiien & Company.  Natural Gas Dehydration:        ana' Trends, FiotI Report.  GRI-
             94/0099, Gas Research Institute, Chicago, Chicago, IL, January 1994,
                                                A-5

-------
                                               T-6
                                TRANSMISSION SOURCE SHEET

SOURCES;                          Glycol Dehydrators
OPERATING MODI:                       Operation
            TYPE:                   Unsteady, Vented
COMPONENTS;                             Vents
ANNUAL                            OJO Bsef ± 392%

BACKGROUND:

Glycol dehydrators remove water from a gas stream by contacting the gas with glycol and then driving the
water from the glycol and into the atmosphere. The glycol  also absorbs a small amount of methane, and
some methane can be driven off to the atmosphere through the reboiler vent.

EMISSION  FACTOR:  (93.72 scfflMMsef gas processed  ± 287.99%)

A thermodynamie computer simulation was     to determine the most important emission-affecting variables
for dehydwtors.  The variables are; gas throughput, existence of a flash tank, existence of stripping gas,
existence of a gas driven pump, existence of vest controls routed to a burner.  Throughput, since its effect is
linear, is handled by establishing an emission rate per gas throughput. Emission rates per throughput are then
established for the other important         affecting characteristics. The emission factor is then:

            EF - [ ( tn x EF^ ) * { fm x EWm ) + ( F^ x EFjo  ) ] x Fwc  x OC

            EF = [      x 3.57) + (0.331 x 175.10} + (0.0741 x 670) ] x  0.152  x 1.0

                 Fpj =        Fraction, of die population WITH flash
                                   ±
                 Fw =        Fraction of the population WITHOUT
                              0.331 ± 19.6%
                 Fm -        Fraction of the population WHU stripping gas
                              0.0741 ± 118.26%
                 FNVC=        Fraction of the population WITHOUT combusted vent controls
                             0.852 ± 14.0%
                 EFyy^       Total CH4 emission rate per 1 MMscf throughput for dehydrator that has a
                              flash tank
                              3.57 scf/MMscf (+102% / -58%)
                 EFfff"       Total CH4 emission rate per 1 MMscf throughput for dehydrator that does
                             NOT have a flash tank
                              175.1 scf/MMscf (+10!%  / -50%)
                 EFSG=       Incremental emission rate per J MMscf throughput  for dehydrator that has
                             stripping gas
                             670          (+40% / -60%)
                 OC =       Overcirculation factor for glycol—number of times the industry rule-of-
                             thumb of 3 gallons glycol/lb water
                              1.0 ± 0%
                                              A-6

-------
IF DATA SOURCES;
             I.    Methane Emissions from the Natural Gas Industry, Volume 14: Giycol Dehydrators (1)
                  establishes emission affecting characteristics of dehydrators.
             2.    Site visit data establishes the F^ and fmc for multiple sites.  Wyoming ADQ data also
                  verifies FNVO though it implies a higher F, and thus a higher  overall EF.
             3.    TMOGA/GPA survey of 1019 dehydrators established Fro and F^, and TP for
                  dehydrators.
             4.    ASPEN computer simulations were used to determine EF^,, and EF^ from the dehydrator
                  vent.
             5,    Sampling data from the GRI Giycol Dehydrator Sampling and Analytical  Program for one
                  dehydrator was used to determine EF^ (1). The upper bound was calculated by assuming
                  that all of die measured noncondensable vent gas was due to  stripping gas that was  100%
                  methane.  The lower bound was calculated as the rale-of-thumb stripping  gas rate
                  recommended by a glycol dehydrator manufacturer.
EF ACCURACY: 93.72 scflMMscf ± 207.99%
            Basis:
            The accuracy is propagated through the EF calculation from each term's accuracy:
            1.   ASPEN has been demonstrated to match actual dehydrators  within ±20% within the
                 calculated confidence intervals obtained from site data.
            2.   Individual EF confidence intervals were calculated based upon die spread of the site
                 averages.


ACTIVITY FACTOR:  (1.M6 Tseffyear gas thnraibpat in the transmission segment)

Toe amount of gas processed by  giyeol dshydrttors in me transmission segment was calculated from the
estimated number of glycol dehydrstors in transmission  service and die average throughput capacity for
transmission dehydrators  [Wright Kilkn & Co., 1994 (2)].  See Source Sheet P-6 for a detailed discussion of
the breakdown of glycol dehydrators into industry segments.  The capacity utilization factor for transmission
was assumed to be 1.

AF ACCURACY: 1.086 Tscfyear ± 143.85%
            Basis:
            1.   Uncertainty based on confidence limits from the site visit data.

ANNUAL METHANE EMISSIONS:  (0.1018 Bscftyr ± 391.75%)

The annual methane emissions were determined by multiplying the dehydrator emission factor by the activity
factor.

REFERENCES

1.           Myers, D.B. Methane Emissions from the Natural Gas Industry, Volume 14: Gfycol
            Dehydrators, Final  Report, GRI-94/Q257.31 and EPA-600/R-96-080n.  Gas Research Institute
            and U.S. Environmental Protection Agency, June 19%.

2.           Wright Kilten & Co.  Natural Gas Dehydration:  Status and Trends, Final Report, GR1-
            94/0099, Gas Research Institute, Chicago, IL, January 1994.
                                               A-7

-------
                                               S-2
                                   STORAGE SOURCE SHEET

SOURCES:                          Glycol Dehydrators
OPERATING MODI:                Normal Operation
            TYPE;                   Unsteady, Vented
COMPONENTS:                              Vents
ANNUAL                            0,23 Bscf ±

BACKGROUND:

Glycol dehydrators remove water from a gas stream by contacting the gas with glycot and then driving the
water from the glycol and into the atmosphere. The glycol  also absorbs a small amount of methane, and
some methane can be driven off to the atmosphere through  the reboiler vent.

            FACTOR;  (117.18           ± 159,76%)

A thermodynantie computer simulation was used to determine the      important emission-affecting variables
for dehydrators.  The variables are: gas throughput, existence of a      tank, existence of stripping gas,
existence of a gas-assisted pmp,         of vent              to a burner.  Throughput, since its effect is
linear, is handled by establishing an         rate per gas throughput.  Emission      per throughput are then
established for die other important emission affecting characteristics.  The emission factor is then:

            EF = [ ( ¥n x EFn ) + ( FOT x EFm ) + ( F^ x EF^ ) j x F^c x OC

            EF = [ (0,520 x 3.5?) + (0.480 x 175,10) + (O.OSO x 670) J x  0.840 x  1.0

                 ¥„ =         Fraction of the population WITH
                                   ±
                 ¥m =        Fraction of the population WITHOUT
                                   ±
                 Fjg =         Fraction of the population WITH stripping gas
                              O.OSO ± 118.44%
                 **NVC=        Fraction 01 ifae population WITHOUT combusted vcni controls
                              0.840 ± 15.24%
                 EFpf™        Total CH4 emission rate per 1 MMscf throughput for dehydrator that has a
                              flash tank
                              3.57 (+102% / -58%)
                 E¥m"       Total CH4 emission rate per 1 MMscf throughput for dehydrator     does
                             NOT     a      tank
                              175.10 (+101% 1-50%)
                 EFsc=        Incremental emission rate per 1 MMscfd throughput  for dehydrator that has
                             stripping gas
                             670 (+40% / -60%)
                 OC =       Overcirculation factor for glycol—number of times the industry rule-of-
                             thumb of 3 gallons glycol/lb water
                              1.0±0%
                                              A-8

-------
 EF DATA

             I.   Mertswe imissims from the Netmal Gas Industry, Volume 14; Glycol DetytA-atars (1)
                 establishes emission affecting  characteristics of dehydrators.
            2.   Site visit data establishes the fm and Fwc for multiple      Wyoming AOQ data also
                 verifies F^, though it        a higher F, and    a higher overall EF.
            3.   TMOGA/OPA  survey of 10! 9 dehydatora established F,r and fm and TP for dehydrates.
            4.   ASPEN computer simulations were used to determine EF3?, and Ef^ from the dehyclrator
                 vent
            5.   Sampling data  from the  GRI Giycol Dehydrator Sampling and Analytical Program for one
                 dehydrator was used to determine EFSG (1). The upper bound  was calculated by assuming
                 that all of the measured  noDcondeiisable vent gas was due to stripping gas that was 100%
                 methane. Hie  lower bound was calculated as the rule-of-thumb stripping gas rate
                 recommended by a glycol dehydrator manufacturer.

 EF ACCURACY:  117.18 ± 159.76%
            Basis:
            He accuracy is propagated  through the Ef calculation from each term's accuracy:
            1,   ASPEN has been demonstrated to match actual dehydrators within ±20% within the
                 calculated confidence  intervals obtained from site data.
            2.   Individua! EF confidence intervals were calculated based upon die      of the site
                 averages.


 ACTIVITY FACTOR: (2.00 Tscf/year §•$ tfcnmgfcpBt  is the storage MgBent)

 The meant of ps processed by glycol dehydrators in the storage segment was calculated from die
 amount of gas withdrawn from underground storage, A total of 2.4 Tscf was withdrawn in 1992, and it is
 assumed that most stored gas is dehydrated.

 AF ACCURACY; 2.00 Tse%ear ± 25%
            Bask:
            L   Uncertainty based on estimate of confidence limits.

ANNUAL METHANE EMISSIONS:  (0.2344 Bscf ± 16456%)

The annual methane emissions were determined by multiplying the dehydrator emission factor by the activity
factor,

REFERENCES

 1.           Myers, D.B.  Meitime Emissions from tke tfatmrol Gas JraAsoy, Fo/«we 14: Gfycol
            Detytfaaoiyt Final Report, GRI-94/02S7.31 and EPA-600/R-96-OSOn. Gas Research
            and U.S.  Environmental Protection Agency, toe 1996.
                                              A-9

-------
                                              GP-2


 SOURCES:                           Glycol Dehydrators
 COMPONENTS:                      Reboiler Vent
 OPERATING MODE:                 Normal Operation
 EMISSION TYPE:                    Unsteady, Vented
 ANNUAL EMISSIONS:                1.05 Bscf ±

 BACKGROUND:

 Glycol dehydrators remove water from a gas stream by contacting the gas with glycol and then driving the
 water from the glycol and into the atmosphere.  The glycol also absorbs a small amount of methane, and
 some methane can be driven off to the atmosphere through the reboiler vent.

 EMISSION FACTOR:   (121.55 scfMMscf ± 201.96%)

A ifaenBodynamic computer simulation was used to determine the most important emission-affecting variables
for dehydrators.   The variables are: (gas throughput, existence of a      tank, existence of stripping gas,
existence of a            pump, existence of vent controls routed to a burner).  Throughput, since its effect
is linear, is handled by establishing an emission rate per gas throughput.  Emission rates per throughput are
     established for tie other important         affecting characteristics. Gas dri¥en pumps are ignored here
and handted in a        source analysis. The emission factor is then:

            EF - [ ( Fn  x EF^ ) •«• ( FNT x EFn ) + ( FK x EFM ) ] x  Fwc x OC

            EF = [ (0.667 x 3.57) + (0.333 x 175.10) + (0.1 II x 670) J x  0.900 x  1.0

                 FJT =        Fraction of the population WITH flash
                              0.667 ± 10,13%
                 fm =        Fraction of the population WITHOUT  flash tanks =  l-F^
                              0,333 ±20.12%
                 Fsc "        Fraction of the population WITH stripping  gas
                              0.111 ±186%
                 FNVC~        Fraction of the population WITHOUT  combusted  vent controls
                              0.90 ±  10%
                 EFp,.3        Total CH4 emission rate per 1  MMscf throughput  for dehydrator that has a
                              flash tank
                              3.5? (+102% / -58%)
                 Ef w=        Total CH4 emission rate per 1  MMscf throughput for dehydrator that
                              NOT have t flash tank
                              175.10 (+i01%/-50%)
                 EFj^        Incremental emission rate per I MMscfd           for dehydrator that has
                              stripping gas
                              670 (+40% / -60%)
                 OC =        Overcirculation factor for glycol—number of      the industry rule-of-
                              thumb of 3 gallons glycol/lb water
                              1.0 ± 0%
                                              A-10

-------
 IF DATA

             1.    Methane Emissions from the Natural Gas Industry,  Volume 14: Gfycol Detydraters (I)
                           emission affecting charactcristJcs of dehydrators.
             2.    Site visit data establish die F£C, and f^ for multiple     (? PROC     with
                  dehydrators).
             3,    TMO6A/GPA  survey of 20? fis plant dehydratiws  established Fw and F,® and TP far
                  dehydrates for the processing segment
             4.    ASPEN computer simulations wen used to determine EF1P> and EFW tan the dehydrator
                  vent.
             5,    Sampling data  from the GR1 Glycol Dehydrator Sampling and Analytical Program for one
                  dehydrator was used to determine EF^ {Gfycol Dehydrator Emissions: Sampling and
                  Analytical Methods and Estimation Techniques) (2). The upper bound was calculated by
                  assuming that all of the measured noncondensable vent gas was due to stripping gas that
                  was 100% methane. The lower bound was calculated as the rale-of-thumb stripping gas
                  rate recommended by a glycol dehydrator manufacturer.

 IF ACCURACY  121.55 sefMMsef ±201, 96%
             Basis:
             The accuracy is rigorously propagated through the EF calculation ftom each term's accuracy:
             1,'    ASPEN h^ been ckmoD^ted to match actoal dehydi^ors within  ± 20% withii) the
                  calculated confidence satervais obtained inn site data.
            2,    Individual EF confidence intervals were calculated from the other data based upon the
                  spread of the il site averages,


 ACTIVITY  FACTOR: (8.63 Tsc£^««r gas ffcttsegfepat in the gas pocessiag segweot)
    glycol dehydrttor througlspot is estimated from the fraction of gas processed by refrigerated processes (as
opposed to dry bed dehydration for cryogenic processes).  The estimate was obtained from the Oil & Gas
Jownal (3) annual Cits Processing Survey,  Of a total of 17,44 Tscf, 8.63 Tscf were determined to be
dehydrated by glycoi,

AF ACCURACY:  8.63 Tscfyear ± 22.45%
            Basis:
            1.   Uncertainty based on estimate of confidence limits for Oil and Gas Journal survey.

AF DATA SOURCES;
            1.   OU & Gas Jownoi* (3) annual Gas Processing Survey.

ANNUAL METHANE               (1.0490  Bscf  ± 20&20%)

The annual methane         were determined by multiplying the dehydrator         factor by the activity
factor.

REFERENCES

I,          Myers, D.B. Metkame Emi$$ian$ Jhm the Natural Gar Iwfcaffy, Volume 14: Glycol
            DOydfaton, Final Report, GRI-94/02S7.31 and EPA-600/R-M-OS§n.  Gas Research Institute
            and U.S. EnviiwmieBttJ Protection Afency, Jane 1996.
                                              A-II

-------
2.           Radian Corporation. Gfycol Defydrator Emissions:  Sampling and Analytical Methods and
             Estimation Techniques.  GRI-94/D324, Gas Research Institute, Chicago, 1L,        1993,

3.           Oil «ft Gas Journal, 1992 Worldwide Gas Processing Survey Database,  1993.
                                               A-1

-------
                                            GP-3
                                PROCESSING SOURCE SHEET

SOURCES:                          Acid Gas Removal (AGR) Units
OPERATING MODE:                Norms] Operation
EMISSION TYPE:                   Unsteady, Vented
ANNUAL EMISSIONS;               0.82 Bscf ± 199%

BACKGROUND:

AGR units remove acid gas (H2S and CQJ from  a natural gas stream by contacting the gas with material
(usnaiiy amines) and then driving the absorbed components from the solvent.  The amines can also absorb a
small amount of methane, and some methane can be driven off to the atmosphere through the reboiier vent to
the atmosphere.

EMISSION FACTOR:  (
-------
 ACTIVITY FACTOR: (371 active AGR units in the U.S.)
 The number of AGR units in the U.S. have all been assumed to be in the processing segment.  The activity
 factor was extracted from the Purvin & Gertz survey.
 AF DATA SOURCES:

             I.    Purvin & Gertz, toe. Business Characteristics of the Natural Gas Conditioning Inc'ustry,
                  1993 (4).

 AF ACCURACY: 371 ±  20%
             Basis:
             1.    The accuracy is based upon engineering judgement.  The survey should have excellent
                  accuracy (± 5%), but the upper bound at 90% confidence was revised upward to 20% to
                  be conservative.
ANNUAL METHANE EMISSIONS:  (0.8237 Bscf ± 108.85%)

The annual methane emissions were determined by multiplying an emission factor for an average dehydrator
by the population of AGRs in the segment.

REFERENCES

1.           Myers, D.B.  Methane Emissions from the Natural Gas Industry,  Volume 14: Glycol
            Dehydrators,  Final Report, GRI-94/0257.31  and EPA-^00/R-96-080n.  Gas Research Institute
            and U.S. Environmental Protection Agency, June  1996.

2.           American Petroleum Institute.  1982 Survey of Gas Processing Units Database.  Washington,
            DC, 1982.

3.           Radian Corporation,  investigation of U.S. Natural Gas Reserve Demographics and Gas
            Treatment Processes," Topical Report, Gas Research Institute, January 1991.

4.           Tannehill, C.C.  and C. Galvin.  Business Characteristics of the Nahtral Gas Conditioning
            Industry.   Topical Report, GRI-93/0342, prepared by Purvin & Gertz, Inc., Gas Research
            Institute,  May 1993.
                                               A-14

-------
_t>
1"

C » «•»
 ^ 35
      §
                Reproduced by NTIS

                         Technical             Service
                Springfield, VA 22161
             repon' was printed specifically for your order
      from H forty 3 atillioit titles mwiJa&le tit our collection.


For ecxjnomy and effioency, NTIS      not        stock of its
collection of technical reports  Rattier, most documents are printed for
each order. Documents that are not in electronic format are reproduced
from master archival copies and are tie best possible reproductions
          If you haw any                    Ms document or any
      you have       with NTIS,       call our Customer Service
Department at      487-4660.

About NTIS
NTIS  collects scientific,          engineering, and
information — then  organizes,             and               that
information in a variety of formats  — from microfiche to oniine services,
The NTIS coHecnon of nearly 3 million titles  includes reports describing
research conducted or sponsored  by                  and their
contractors;            and  business  information;  U.S.  military
publications;  audiovisual products; computer software and electronic
databases developed by federal agencies; training tools;  and technical
reports prepared by research organizations  worldwide.  Approximately
100,000 tmw       are added and indexed into  the  NTIS collection
annually.
   For more information about NTIS products and services, call NTIS
    at f?03) 487-«i50 and        the free NTIS Catalog of Products
         and Services, PR-827LPG, or visit the NTIS Web site
                      http://www.ntis.gov
                     'four /nd/""pensad/e resource for government-sponsored
                                info, matioij—U.S. and worttiwide

-------