next
previous
items
European Union flag
Home
Topics
Analysis and data
Countries
Newsroom
About us

File

Note: new versions of the publication are available!

EMEP/CORINAIR Emission Inventory Guidebook - 2007

B411vs2.2.pdf

File Published 12 Dec 2008
(1 of 25)
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-1
SNAP CODES:
040101
040102
SOURCE ACTIVITY TITLE:
P
ROCESSES
I
N
P
ETROLEUM
I
NDUSTRIES
Petroleum Products Processing
Fluid Catalytic Cracking
NOSE CODE:
105.08.01
105.08.02
NFR CODE:
1 B 2 a iv
1
ACTIVITIES INCLUDED
A basic refinery converts crude petroleum into a variety of sub-products. Principal products
of a petroleum refinery include:
Table 1.1: Refinery Principal Products
Product Type
Principal Products
Liquid Fuels
Motor Gasoline
Aviation Gasoline
Aviation Turbine Fuel
Illuminating Kerosene
High-Speed Diesel
Distillate Heating Fuel
Medium-Speed Diesel
Residual Oil
By-Product Fuels and Feedstocks
Naphtha
Lubricants
Asphalt
Liquefied Petroleum Gases
Coke
Sulphur (Product of Auxiliary Facility)
White Oils
Primary Petrochemicals
Ethylene
Propylene
Butadiene
Benzene
Toluene
Xylene
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-2
1 September, 1999
Emission Inventory Guidebook
The production of the latter group, primary petrochemicals, is, however, not included in this
chapter, even if these chemicals are produced at a petroleum refinery. Please refer to the
relevant chapters for sub-sector 040500 (chapters B451-B4522).
The petroleum refining industry employs a wide variety of processes. The types of processes
operating at any one facility depend on a variety of economic and logistic considerations such
as the quality of the crude oil feedstock, the accessibility and cost of crude (and alternative
feedstocks), the availability and cost of equipment and utilities, and refined product demand.
The four categories of general refinery processes are listed in Table 1.2.
Table 1.2: General Refinery Processes and Products
General Process
Products
Separation Processes
Atmospheric Distillation
Vacuum Distillation
Light Ends Recovery (Gas Processing)
Petroleum Conversion Processes
Cracking (Thermal and Catalytic)
Coking
Viscosity Breaking
Catalytic Reforming
Isomerization
Alkylation
Polymerization
Petroleum Treating Processes
Hydrodesulfurization
Hydrotreating
Chemical Sweetening
Acid Gas Removal
Deasphalting
Blending
Motor Gasoline
Light Fuel Oil
Heavy Fuel Oil
The major direct process emission sources of NMVOCs are vacuum distillation, catalytic
cracking, coking, chemical sweetening and asphalt blowing (U.S.EPA 1985a). Process-unit
turnaround (periodical shut-down of units) has also been reported as contributing to VOC
emissions (CEC 1991).
Fugitive emissions from equipment leaks are also a significant source of NMVOC emissions
from process operations at a refinery. Emissions from storage and handling are also classified
as fugitive emissions. To avoid confusion, fugitive emissions from equipment leaks will be
referred to as fugitive process emissions in this chapter.
Table 1.3 summarises significant sources of common pollutants from process and fugitive
process emissions sources at refineries.
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-3
Table 1.3: Significant Process Emissions Sources at Petroleum Refineries (U.S. EPA 1985)
Process
Particulate
NO
x
SO
x
CO
NMVOC
Vacuum Distillation
x
x
x
x
X
Catalytic Cracking
X
xx
X
X
Thermal Cracking
x
x
x
X
Sweetening
X
Blowdown Systems
x
x
x
x
X
X
- Possibly significant sources (depending upon the degree of contract)
x - Minor sources
This section is a summary of the main products possibly produced at a refinery and the major
processes that could be present, with an indication of the processes that are potentially
significant sources of emissions to the air. All of these processes are currently under SNAP
code 040101, with the exception of FCCs with CO boiler. However it is difficult to use this
code separately from other processes, particularly for simpler emission estimation methods,
which tend to encompass a wide variety of sources. It is therefore proposed that FCCs with
CO boiler also be inventoried under 040101.
It is important to note that fugitive process emissions are somewhat difficult to characterise
by their area (process vs. storage/handling vs. waste treatment), as they are estimated based
on equipment counts and are not usually classified as to type of use or area of the refinery.
Therefore fugitive process emissions for the entire refinery are inventoried under SNAP code
040101.
2
CONTRIBUTION TO TOTAL EMISSIONS
Table 2.1 summarises emissions from petroleum refining processes in the CORINAIR90
inventory. In a Canadian study (CPPI and Environment Canada 1991), the process/fugitive
process sources contributed 51.5% (process 4.7% (Only FCCU estimated) and fugitive
process 46.8%) of total VOC emissions for 29 refineries surveyed. Blending losses were not
estimated separately. The process/fugitive process sources would represent approximately
2.6% of total anthropogenic emissions.
Table 2.1: Contribution to total emissions of the CORINAIR90 inventory
(
28 countries
)
S
S
o
o
u
u
r
r
c
c
e
e
-
-
a
a
c
c
t
t
i
i
v
v
i
i
t
t
y
y
S
S
N
N
A
A
P
P
-
-
c
c
o
o
d
d
e
e
C
C
o
o
n
n
t
t
r
r
i
i
b
b
u
u
t
t
i
i
o
o
n
n
t
t
o
o
t
t
o
o
t
t
a
a
l
l
e
e
m
m
i
i
s
s
s
s
i
i
o
o
n
n
s
s
[
[
%
%
]
]
S
S
O
O
2
2
N
N
O
O
x
x
N
N
M
M
V
V
O
O
C
C
C
C
H
H
4
4
C
C
O
O
C
C
O
O
2
2
N
N
2
2
O
O
N
N
H
H
3
3
P
P
e
e
t
t
r
r
o
o
l
l
e
e
u
u
m
m
P
P
r
r
o
o
d
d
u
u
c
c
t
t
s
s
P
P
r
r
o
o
c
c
e
e
s
s
s
s
.
.
0
0
4
4
0
0
1
1
0
0
1
1
0
0
.
.
5
5
0
0
.
.
1
1
0
0
.
.
7
7
0
0
0
0
0
0
.
.
2
2
0
0
.
.
1
1
-
-
F
F
l
l
u
u
i
i
d
d
C
C
a
a
t
t
a
a
l
l
y
y
t
t
i
i
c
c
C
C
r
r
a
a
c
c
k
k
i
i
n
n
g
g
0
0
4
4
0
0
1
1
0
0
2
2
0
0
.
.
4
4
0
0
.
.
1
1
0
0
-
-
0
0
0
0
.
.
1
1
-
-
-
-
0 = emissions are reported, but the exact value is below the rounding limit (0.1 per cent)
-
= no emissions are reported
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-4
1 September, 1999
Emission Inventory Guidebook
3
GENERAL
3.1
Description
3.1.1 Direct Process Sources
There are four main categories of processes in a petroleum refinery:
Separation Processes
Crude oil consists of a mixture of hydrocarbon compounds including paraffinic, naphthenic,
and aromatic hydrocarbons plus small amounts of impurities including sulphur, nitrogen,
oxygen and metals. The first phase in petroleum refining operations is the separation of crude
oil into common boiling point fractions using three petroleum separation processes:
atmospheric distillation, vacuum distillation, and light ends recovery (gas processing).
Conversion Processes
Where there is a high demand for high-octane gasoline, jet fuel and diesel fuel, components
such as residual oils, fuel oils, and light ends are converted to gasolines and other light
fractions. Cracking, coking and visbreaking processes break large petroleum molecules into
smaller petroleum molecules. Polymerization and alkylation processes rearrange the structure
of petroleum molecules into larger ones. Isomerization and reforming processes rearrange the
structure of petroleum molecules to produce higher-value molecules of a similar molecule
size.
Treating Processes
Petroleum treating processes stabilise and upgrade petroleum products. Desalting is used to
remove salt, minerals, grit, and water from crude oil feedstocks prior to refining. Undesirable
elements such as sulphur, nitrogen and oxygen are removed from product intermediates by
hydrodesulphurization, hydrotreating, chemical sweetening and acid gas removal.
Deasphalting is used to separate asphalt from other products. Asphalt may then be
polymerised and stabilised by blowing (see SNAP code 060310).
Blending
Streams from various units are combined to produce gasoline, kerosene, gas oil and residual
oil, and in some cases a few speciality items.
3.1.2
Fugitive Process Sources
Fugitive process emission sources are defined as NMVOC sources not associated with a
specific process but scattered throughout the refinery. Fugitive process emissions sources
include valves of all types, flanges, pump and compressor seals, pressure relief valves,
sampling connections and process drains. These sources may be used in the transport of
crude oil, intermediates, wastes or products.
Note that this category will actually include fugitive emissions from all such refinery sources,
rather than those sources only associated with process emissions.
3.2
Definitions
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-5
3.3
Techniques
See section 3.1 (above).
3.4
Emissions/Controls
3.4.1 Direct Process Emissions
Vacuum distillation, catalytic cracking, thermal cracking, sweetening, blowdown systems,
sulphur recovery, asphalt blowing and flaring processes have been identified as being
potentially significant sources of SO
2
and NMVOC from those sources included under SNAP
code 040101, with a relatively smaller contribution of particulate, NOx and CO. (U.S.EPA
1985a).
Vacuum Distillation
Topped crude withdrawn from the bottom of the atmospheric distillation column is composed
of high-boiling-point hydrocarbons. The topped crude is separated into common-boiling-
point fractions by vaporisation and condensation in a vacuum column at a very low pressure
and in a steam atmosphere. A major portion of the vapours withdrawn from the column by
steam ejectors or vacuum pumps are recovered in condensers. Historically, the non-
condensable portion has been vented to the atmosphere.
The major NMVOC emission sources related to the vacuum column include steam ejectors
and vacuum pumps that withdraw vapours through a condenser.
Methods of controlling these emissions include venting into blowdown systems or fuel gas
systems, and incineration in furnaces (SNAP code 090201) or waste heat boilers (SNAP code
030100). These control techniques are generally greater than 99 percent efficient in the
control of hydrocarbon emissions.
Note that the emissions from blowdown and vapour recovery systems have been included
under this SNAP code rather than under SNAP code 090100 (see below).
Catalytic Cracking
Catalytic crackers use heat, pressure and catalysts to convert heavy oils into lighter products
with product distributions favouring the gasoline and distillate blending components.
Fluidised-bed catalytic cracking (FCC) processes use finely divided catalysts that are
suspended in a riser with hot vapours of the fresh feed. The hydrocarbon vapour reaction
products are separated from the catalyst particles in cyclones and sent to a fractionator. The
spent catalyst is conveyed to a regenerator unit, in which deposits are burned off before
recycling.
Moving-bed catalytic cracking (TCC) involves concurrent mixing of the hot feed vapours
with catalyst beads that flow to the separation and fractionating section of the unit.
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-6
1 September, 1999
Emission Inventory Guidebook
Aside from combustion products from heaters, emissions from catalytic cracking processes
are from the catalyst regenerator. These emissions include NMVOC, NOx, SOx, CO,
particulates, ammonia, aldehydes, and cyanides.
In FCC units, particulate emissions are controlled by cyclones and/or electrostatic
precipitators. CO waste heat boilers may be used to reduce the CO and hydrocarbon
emissions to negligible levels.
TCC catalyst regeneration produces much smaller quantities of emissions than is the case for
FCC units. Particulate emissions may be controlled by high-efficiency cyclones. CO and
NMVOC emissions from a TCC unit are incinerated to negligible levels by passing the flue
gases through a process heater firebox or smoke plume burner.
SOx from catalyst regeneration may be removed by passing the flue gases through a water or
caustic scrubber.
Thermal Cracking
Thermal cracking units break heavy oil molecules by exposing them to higher temperatures.
In viscosity breaking (visbreaking), topped crude or vacuum residuals are heated and
thermally topped in a furnace and then put into a fractionator. In coking, vacuum residuals
and thermal tars are cracked at high temperature and low pressure. Historically, delayed
coking is the most common process used, although fluid coking is becoming the more
preferred process.
Emissions from these units are not well characterised. In delayed coking, particulate and
hydrocarbon emissions are associated with removing coke from the coke drum and
subsequent handling and storage operations. Generally there is no control of hydrocarbon
emissions from delayed coking, although in some cases coke drum emissions are collected in
an enclosed system and routed to a refinery flare.
Sweetening
Sweetening of distillates is accomplished by the conversion of mercaptans to alkyl disulfides
in the presence of a catalyst. Conversion may then be followed by an extraction step in which
the disulfides are removed.
Hydrocarbon emissions are mainly from the contact between the distillate product and air in
the air-blowing step. These emissions are related to equipment type and configuration, as
well as to operating conditions and maintenance practices.
Asphalt Blowing
Please refer to SNAP code 060310 for inventory methods for asphalt blowing.
Sulphur Recovery
Please refer to SNAP code 040103 for inventory methods for sulphur recovery plants.
Flaring
Please refer to SNAP code 090203 for inventory methods for flaring in a refinery.
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-7
Blowdown Systems
Many of the refining process units subject to hydrocarbon discharges are manifolded into a
collection unit (i.e., blowdown system), comprising a series of drums and condensers,
whereby liquids are separated for recycling and vapours are recycled or flared with steam
injection. Uncontrolled blowdown emissions consist primarily of hydrocarbons, while vapour
recovery and flaring systems (see SNAP code 090203) release lesser NMVOC and greater
combustion products including SO
2
, NOx and CO.
3.4.2 Fugitive Process Emissions
Fugitive process emissions sources include valves of all types, flanges, pumps in hydrocarbon
service (packed or sealed), compressor seals, pressure relief devices, open-ended lines or
valves, sampling connections, and process drains or oily water drains.
For these sources, a very high correlation has been found between mass emission rates and
the type of stream service in which the sources are employed. For compressors, gases passing
through are classified as either hydrogen or hydrocarbon service. For all other sources,
streams are classified into one of three stream groups: gas/vapour streams, light liquid/two
phase streams, and kerosene and heavier liquid streams. It has been found that sources in
gas/vapour service have higher emission rates than those in heavier stream service. This
trend is especially pronounced for valves and pump seals.
Of these sources of NMVOC, valves are the major source type. This is due to their number
and relatively high leak rate.
Normally, control of fugitive emissions involves minimising leaks and spills through
equipment changes, procedure changes, and improved monitoring, housekeeping and
maintenance practices.
Applicable control technologies are summarised in Table 3.1.
Table 3.1: Control Technologies for Fugitive Sources (U.S. EPA 1985a)
Fugitive Source
Control Technology
Pipeline Valves
monitoring and maintenance programs
Open-Ended Valves
Instillation of cap or plug on open end of valve /line
Flanges
monitoring and maintenance
Pump Seals
mechanical seals, dual seals, purged seals, monitoring and maintenance
programs, controlling degassing vents
Compressor Seals
mechanical seals, dual seals, purged seals, monitoring and maintenance
programs, controlling degassing vents
Process Drains
Traps and covers
Pressure/Relief Valves
Rupture disks upstream of relief and/or venting to a flare
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-8
1 September, 1999
Emission Inventory Guidebook
4
SIMPLER METHODOLOGY
The simplest inventory methodology is to combine the crude oil throughput of each refinery
with either a single emission factor, or two emission factors (one for process and one for
fugitive process emissions) for each refinery. The first approach would be the easiest to use if
very limited information is available. However, the second approach would allow the user to
in some way reflect the type of processes and related controls at the refinery as well as
accounting for the sophistication of the fugitive emissions inspection and maintenance
programs typical of the region and/or that particular refinery.
It is strongly recommended that the detailed methodology be used for petroleum refineries.
5
DETAILED METHODOLOGY
The detailed methodology requires each refinery to estimate its process emissions for each
process, using detailed throughput information and emission factors. Site specific emission
factors or data would be preferable, wherever possible.
Remote sensing, using Fourier
transform techniques, is making it possible to measure total refinery emissions, although it
may be difficult to identify the individual source strengths.
The state-of-the-art technology for estimating fugitive process emissions is to use an emission
testing program to classify equipment into groupings and then estimate emissions using
emission factors or algorithms (see section 16, Verification Procedures). However, this is a
very expensive and time-consuming proposition and is considered beyond the resources of
most inventory personnel. The methodology proposed below is a compromise between a
testing program vs. estimates of the number of each type of equipment that might be in a
refinery based on either its throughput or production data.
Fugitive process emissions, which are considered to be the major source of NMVOCs at a
petroleum refinery, are inventoried using emission factors for each type of equipment and
stream, based on a count of the number of sources, a characterisation of the NMVOC content
of the stream in question and whether the refinery conducts an inspection and maintenance
program.
The U.S. EPA has published a detailed protocol for equipment leak emissions estimates
(U.S.EPA 1993). In the average emission factor method, the following unit-specific data is
required:
1.
the number of each type of component in a unit (valve, connector etc.);
2.
the service each component is in (gas, light liquid, or heavy liquid);
3.
the NMVOC concentration in the stream (weight fraction) and;
4.
the number of hours per year the component was in service.
The equipment is then grouped into streams, where all of the equipment within the stream has
approximately the same NMVOC weight percent. Annual emissions are then calculated for
each stream using equation 1 as follows:
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-9
NMVOCs = AEF * WFnmvoc * N
(1)
where:
NMVOCs
= NMVOC emission rate from all equipment in the stream of a
given equipment type (kg/hr)
AEF
= applicable average emission factor for the equipment type
(kg/hr/source)
WFnmvoc
= average weight fraction of NMVOC in the stream and
N
= the number of pieces of equipment of the applicable
equipment type in the stream.
If there are several streams at the refinery, as is usually the case, the total NMVOC emission
rate for an equipment type is the sum of emissions from each of the streams. The total
emission rates for all of the equipment types are summed to generate the process unit total
NMVOC emission rate from fugitive process sources.
6
RELEVANT ACTIVITY STATISTICS
For the simpler methodology, the crude oil throughput of each refinery is required.
For the detailed methodology, specific data will be required on the throughput for each
process area. For fugitive process emissions estimates, each emission source must be counted
by type and process stream, and the NMVOC content for each stream must then be
characterised. The number of annual hours of operation for each stream is also required.
Finally it must be determined if an inspection and maintenance program is conducted at the
refinery.
7
POINT SOURCE CRITERIA
All refineries are to be inventoried as point sources.
8
EMISSION FACTORS, QUALITY CODES AND REFERENCES
8.1
Simpler Methodology
Total hydrocarbon emission factors based on an inventory of Canadian refineries in 1988
(CPPI and Environment Canada 1991) were 0.05 kg/m3 feedstock for process emissions and
0.53 kg/m3 for fugitive process emissions. Of the latter, valves accounted for 0.35 kg/m3
feedstock. Data was not available for NMVOC only. This inventory was based on a survey of
individual refineries, in which some reported their own emission estimates and some reported
base quantity data for which emissions were estimated using a variety of techniques.
The use of CONCAWE derived VOC emission factors, based upon a hypothetical 5 Mt/yr
refinery, as follows was recommended as a default method for the Corinair 1990 project
(CEC 1991). The emission factor for fugitive process emissions is 0.25 kg/t crude (0.21
kg/m
3
crude assuming specific gravity of 0.85 (BP 86)).
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-10
1 September, 1999
Emission Inventory Guidebook
Sixty percent of these emissions are reported to be from valves. CONCAWE also indicates
that average fugitive emissions in the same refinery with a maintenance and monitoring
programme is 0.01% by weight (.085 kg/m3) of refinery throughput (CONCAWE Report
87/52 1987). The CORINAIR90 default emission handbook also reports a U.S.EPA factor of
0.18 kg/Mg crude (U.S.EPA 1985b) for process unit turnaround, and estimates that Western
European refineries would emit half of this for turnaround, or 0.09 kg/Mg.
It is apparent that detailed emission inventory data is required for several refineries in
differing regions in order to develop meaningful emission factors. Major factors affecting
regional differences include crude characteristics, product demand (and hence refinery
processes) and regulatory requirements.
Emission factors for non-combustion process sources of other contaminants were not
identified, other than as provided in Table 4 of SNAP sector 040100.
8.2
Detailed Methodology
The more detailed methodology involves the use of process-specific emission factors based
on the throughput of the unit and fugitive process emission factors based on equipment
counts. It is important to note that the factors presented below must be used with caution, as
they do not account for regional differences in crude, product demand and regulatory
requirements. Wherever possible, site-specific emission estimates based on monitoring
should be considered.
8.2.1 Process Emission Factors
Table 8.1 lists emission factors for refinery processes based on tests conducted in the 1970’s,
noting that overall, less than 1 % of the total hydrocarbons emissions are methane (U.S.EPA
1985a). Note that new emission factors are being developed by the U.S. EPA, with some
speciation. These are due for release soon (Zarate 1992).
Table 8.1: Emission Factors for Petroleum Refineries (U.S. EPA 1985a and 1995)
Process
Particulate
Sox
(as SO
2
)
CO
THC
a
NOx
(as NO
2
)
Aldehydes
NH
3
Quality
Fluid catalytic cracking units
Uncontrolled
kg/10
3
liters fresh feed
0.695
(0.267-0.976)
b
1.143
(0.286-
1.505)
39.2
0.630
0.204
(0.107-
0.416)
0.054
0.155
B
ESP and CO boiler
kg/10
3
liters fresh feed
0.128
(0.020-0.428)
1.413
(0.286-
1.505)
Neg
c
Neg
0.204
d
(0.107-
0.416)
Neg
Neg
B
Moving-bed catalytic cracking units
kg/10
3
liters fresh feed
0.049
0.171
10.8
0.250
0.014
0.034
0.017
B
Fluid coking units
Uncontrolled
1.5
NA
e
NA
NA
NA
NA
NA
C
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-11
Process
Particulate
Sox
(as SO
2
)
CO
THC
a
NOx
(as NO
2
)
Aldehydes
NH
3
Quality
kg/10
3
liters fresh feed
ESP and CO boiler
kg/10
3
liters fresh feed
0.0196
NA
Neg
Neg
NA
Neg
Neg
C
Blowdown systems
Uncontrolled
kg/10
3
liters refinery feed
Neg
Neg
Neg
1.662
Neg
Neg
Neg
C
Vacuum distillation column
condensers
Uncontrolled
kg/10
3
liters refinery feed
Neg
Neg
Neg
0.052
Neg
Neg
Neg
C
Uncontrolled
kg/10
3
liters vacuum feed
Neg
Neg
Neg
0.144
Neg
Neg
Neg
C
Controlled
Neg
Neg
Neg
Neg
Neg
Neg
Neg
C
a
Overall, less than 1 percent by weight of the total hydrocarbon emissions are methane
b
Numbers in parenthesis indicate range of values observed
c
Negligible emission
d
May be higher due to the combustion of ammonia
e
NA, Not Available.
The VOC emission factors listed in Table 8.2 were used to estimate emissions from processes
in the United Kingdom (Passant n.d.).
Table 8.2: United Kingdom VOC Emission Factors (Passant n.d.)
Process
Emission Factor
Quality
Catalytic Cracker
Uncontrolled
628 g/m
3
feed
D
Controlled
negligible
D
Fluid Coking
Uncontrolled
384 g/m
3
feed
D
Controlled
Negligible
D
Vacuum Distillation
Uncontrolled
51.6 g/m
3
feed
D
Controlled
negligible
D
Asphalt Blowing
Uncontrolled
27.2 kg/Mg asphalt
D
Controlled
0.54 kg/Mg asphalt
D
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-12
1 September, 1999
Emission Inventory Guidebook
8.2.2 Fugitive Process Emission Factors
Emissions factors for fugitive process emissions of NMVOC are expressed as losses per
equipment unit per day. As previously discussed, the methods for estimating mass emissions
from process equipment leaks range from the use of emission factors with equipment counts
to comprehensive field measurement techniques. These methods have evolved from a
number of studies of the organic chemical and petroleum refining industries for the U.S.
EPA.
The U.S. EPA (1993) has provided NMVOC emission factors (Table 8.3). These are
considered to give a good estimate of fugitive process emissions from a refinery with no
inspection and maintenance program.
Table 8.3: Process Fugitive Average NMVOC Emission Factors for Petroleum
Refineries (U.S.EPA 1993)
Source
Emission Factor (kg/hr-source)
Quality
Valves
Gas
0.0268
Light Liquid
0.0109
Heavy Liquid
0.00023
Open-ended Lines
All Streams
0.0023
Connectors
All Streams
0.00025
Pump Seals*
Light Liquid
0.114
Heavy Liquid
0.021
Compressor Seals
Gas
0.636
Sampling Connections
All Streams
0.0150
Pressure Vessel Relief Valves
Gas
0.16
* The light liquid pump seal factor can be used to estimate the leak rate from agitator seals
These emission factors, however, are in most cases considered to overestimate NMVOC
emissions from sources in more modern facilities. The U.S. EPA allows a 75% reduction in
emissions estimated by using these emission factors if an approved I and M program is
conducted at the petroleum refining facility.
Passant (1993) used the VOC emission factors presented in Table 8.4, which were referenced
to U.S.EPA 1988.
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-13
Table 8.4: Process Fugitive Emission Factors for Petroleum Refineries (Passant 1993)
Source
Emission Factor (kg/hr)
Quality
Valves
gas
0.0056
D
light
0.0071
D
heavy
0.0023
D
Pump Seals
light
0.0494
D
heavy
0.0214
D
Compressor Seals
all streams
0.2280
D
Pressure Relief Seals
all streams
0.104
D
Flanges
all streams
0.00083
D
Open-ended Lines
all streams
0.0017
D
Sample Connections
all streams
0.015
D
Although the derivation of the emission factors in table 8.4 is not given, it would appear that
these are actually average synthetic organic chemical manufacturing industry (SOCMI)
(1988) uncontrolled emission factors. These sets of factors are thought to be biased on the
high side for petroleum refineries due to the inclusion of ethylene plants, which operate at
15,000 to 40,000 psig.
9
SPECIES PROFILES
9.1
Applicability of Generalised VOC Species Profiles
In both North America and Europe, VOC species profiles have been published based on
measurements or engineering judgement. There is a need to produce generalised profiles for
use by models, the alternative being to obtain refinery specific data. Generalised profiles can
be generated at the most detailed process level, however, there are several important
influences which should be considered in attempting to specify such generally applicable
data. Some of these influences are:
Meteorological and Climatological effects: Ambient temperature and temperature ranges may
have important influences on the emitted species profiles. Due to the logarithmic
behaviour of vapour pressure, higher temperatures tend to favour the loss of the lower
molecular weight compounds from storage vessels and some process streams.
Variability of Raw Materials: The type of crude oil being processed can influence the fraction
of more volatile and more easily emitted compounds.
Process Variability: Different refineries will have process differences. Where the species
profiles are based on individual operations, process differences can be allowed for.
However, overall average refinery profiles will differ between refineries.
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-14
1 September, 1999
Emission Inventory Guidebook
Maintenance and Regulation: Equipment maintenance and the enforcement of regulations
will have significant effects on the overall emitted species distribution depending on
which processes or operations are impacted by maintenance practices or by regulation.
The broad application of generalised profiles should be done with some caution. Where such
profiles are necessary, consideration should be given to stratification of the data according to
some of the major factors of influence such as climate, country, raw material etc.
9.2
Simplified VOC Speciation
For some applications, where there is no process detail, or where refineries are grouped as a
single emission source, there is a need for a single overall species profile to characterise the
emissions for modelling or other purposes. Process specific species profiles can be combined
into a single overall refinery profile by appropriately weighting the individual profiles
according to their relative contribution to the total refinery emissions.
Consideration should be given to stratification of the data according to some of the major
factors of influence such as climate, country, raw material (crude) etc.
CONCAWE reports (Report 2/86) that refinery emissions are essentially saturated, with the
saturated hydrocarbon content lying between 80 and 90% by weight. The balance of 10 to
20% is unsaturated and/or aromatic hydrocarbons, the actual values depending on the nature
of the refinery processes installed. Several overall refinery species profiles are available, such
as those reviewed by Veldt (1991) for application to the EMEP and CORINAIR 1990
emissions inventories. On the basis of this review, this chapter proposes an overall species
profile for petroleum refining by mass fraction.
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-15
Table 9.1: CONCAWE Petroleum Refinery Speciation Profile
Species
CONCAWE (%)
Quality
Methane
0
(Unknown)
Ethane
5
Propane
20
n-Butane
15
i-Butane
5
Pentanes
20
Hexanes
10
Heptanes
5
>Heptanes
5
Ethene
1
Propene
1
Butene
0.5
Benzene
2
Toluene
3
o-Xylene
0.7
M,p-Xylene
1.3
Ethylbenzene
0.5
TOTAL
100
The Air Emission Species Manual (AESM) for VOC (U.S. EPA 1994) provides an overall
refinery species profile (Profile 9012: Petroleum Industry - Average, Data Quality E - based
on engineering judgement) as summarised in Table 9.2.
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-16
1 September, 1999
Emission Inventory Guidebook
Table 9.2: US EPA Petroleum Refinery Speciation Profile
Species
CAS Number
Wt (%)
Quality
Methane
74-82-8
13
E
Ethane
74-84-0
6.05
E
Propane
74-98-6
19.7
E
n-Butane
106-97-8
7.99
E
i-Butane
75-28-5
2.89
E
Pentanes
(109-66-0)
21.4
E
Hexanes
(110-54-3)
8.02
E
Heptanes
(142-82-5)
1.87
E
Octanes
(111-65-9)
2.13
E
Nonanes
(111-84-2)
1.01
E
Decanes
(124-18-5)
1.01
E
Cyclo-hexane
110-82-7
0.08
E
Cyclo-heptanes
2.27
E
Cyclo-octanes
0.66
E
Cyclo-nonanes
0.11
E
Propene
115-07-01
1.75
E
Butene
106-98-9
0.15
E
Benzene
71-43-2
0.38
E
Toluene
108-88-3
0.44
E
Xylenes
1330-20-7
0.19
E
Formaldehyde
50-00-0
8.88
E
Total
100.02
The suggested applications are:
Blowdown system - vapour recovery./Flare
Blowdown system - without controls
Wastewater treatment - excl. Separator
Vacuum distillation - column condenser
Sludge converter - general
Fluid coking - general
Petroleum coke - calciner
Bauxite burning
Lube oil manufacturing
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-17
9.3
Detailed Process VOC Speciation
The most detailed speciation of VOC is achievable at the process level using the U.S. EPA
AESM (U.S.EPA 1994). Such a detailed method is generally only applicable on an
individual refinery basis where estimates of the emission contributions from the various
process streams and operations are available. The generalised profiles, which are available for
individual processes and operations, as well as fugitive process emissions, are summarised
below. These profiles are based on United States data, and in many instances on data from
California.
U.S. EPA petroleum refinery species profiles applicable to petroleum refinery process and
fugitive process emissions are presented in Tables 9.3 through 9.8.
Table 9.3: Profile 0029: Refinery Fluid Catalytic Cracker.
CAS Number
Name
Wt %
Quality
Isomers of hexane
13.00
C
74-82-8
Methane
36.00
C
50-00-0
Formaldehyde
51.00
C
TOTAL
100.00
Table 9.4: Profile: 0031 Refinery Fugitive Emissions - Covered Drainage / Separation
Pits.
CAS Number
Name
Wt %
Quality
Isomers of hexane
12.20
C
C-7 cycloparaffins
16.90
C
C-8 cycloparaffins
5.20
C
Isomers of pentane
10.10
C
74-82-8
Methane
2.90
C
74-84-0
Ethane
1.70
C
74-98-6
Propane
5.90
C
106-97-8
N-Butane
14.30
C
75-28-5
Iso-Butane
4.50
C
109-66-0
N-Pentane
12.00
C
110-54-3
Hexane
11.90
C
71-43-2
Benzene
2.40
C
TOTAL
100.00
Used for:
Fugitive hydrocarbon emissions - drains - all streams
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-18
1 September, 1999
Emission Inventory Guidebook
Table 9.5: Profile: 0039 Description: Refinery Fugitive Emissions - Compressor Seals -
Refinery Gas.
CAS Number
Name
Wt %
Quality
Isomers of hexane
1.00
D
Isomers of heptane
0.10
D
Isomers of pentane
8.60
D
74-82-8
Methane
13.30
D
74-84-0
Ethane
5.60
D
74-98-6
Propane
16.00
D
115-07-01
Propene
8.80
D
106-97-8
N-Butane
23.20
D
106-98-9
Butene
1.20
D
75-28-5
Iso-Butane
10.00
D
109-66-0
N-Pentane
7.60
D
110-54-3
Hexane
4.60
D
TOTAL
100.00
Used for:
Compressor seal - gas streams
Compressor seal - heavy liquid streams
Table 9.6: Profile: 0047 Description: Refinery Fugitive Emissions - Relief Valves -
Liquefied Petroleum Gas
CAS Number
Name
Wt %
Quality
74-84-0
Ethane
4.10
D
74-98-6
Propane
90.40
D
115-07-01
Propene
5.10
D
75-28-5
Iso-Butane
0.40
D
TOTAL
100.00
Used for:
Vessel relief valves
Pipeline valves - gas streams
Pipeline valves - lt liq/gas streams
Pipeline valves - heavy liqd streams
Pipeline valves - hydrogen streams
Open-ended valves - all streams
Flanges - all streams
Vessel relief valves - all streams
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-19
Table 9.7: Profile: 0316 Description: Pipe / Valve Flanges
CAS Number
Name
Wt %
Quality
C-7 cycloparaffins
0.20
C
C-9 cycloparaffins
0.10
C
Isomers of pentane
7.80
C
74-82-8
Methane
28.60
C
74-84-0
Ethane
5.80
C
74-98-6
Propane
11.50
C
115-07-01
Propene
0.10
C
106-97-8
N-Butane
18.30
C
75-28-5
Iso-Butane
7.40
C
109-66-0
N-Pentane
7.70
C
(10-54-3)
Hexanes
5.00
C
(42-82-5)
Heptanes
2.20
C
(11-65-9)
Octanes
2.20
C
(11-84-2)
Nonanes
1.10
C
(24-18-5)
Decanes
1.10
C
110-82-7
Cyclohexane
0.10
C
1330-20-7
Isomers of Xylene
0.20
C
71-43-2
Benzene
0.10
C
108-88-3
Toluene
0.50
C
TOTAL
100.00
Used for: Pipeline - valves / flanges
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-20
1 September, 1999
Emission Inventory Guidebook
Table 9.8: Profile: 0321 Description: Pump Seals – Composite.
CAS Number
Name
Wt %
Quality
C-7 cycloparaffins
1.10
C
C-8 cycloparaffins
0.10
C
C-9 cycloparaffins
0.80
C
74-82-8
Methane
3.30
C
74-84-0
Ethane
1.20
C
74-98-6
Propane
3.70
C
106-97-8
N-Butane
8.10
C
75-28-5
Iso-Butane
0.80
C
109-66-0
Pentanes
17.70
C
(110-54-3)
Hexanes
16.50
C
(142-82-5)
Heptanes
12.60
C
(111-65-9)
Octanes
14.80
C
(111-84-2)
Nonanes
7.00
C
(124-18-5)
Decanes
7.00
C
(110-82-7
Cyclohexane
0.50
C
1330-20-7
Isomers of Xylene
1.30
C
71-43-2
Benzene
0.50
C
108-88-3
Toluene
3.00
C
TOTAL
100.00
Used for: Pump seals - with/without controls
Pump seals - light liq/gas streams
Pump seals - heavy liqd streams
Sampling/purging/blind changing
10
UNCERTAINTY ESTIMATES
See next section on: Weakest Aspects/Priority Areas for Improvement in Current
Methodology
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-21
11
WEAKEST ASPECTS/PRIORITY AREAS FOR IMPROVEMENT IN
CURRENT METHODOLOGY
More measurements of emissions from petroleum refineries should be done: based on testing
programs in the United Kingdom, currently available emission factors have underestimated
emissions typically by 30%.
Emission factors must be developed that can account for regional differences in the major
sources of NMVOCs in refineries (see above and this part of section 040104). There are also
difficulties in determining what the data really represents, as there is a wide variation in the
definition of total hydrocarbons, hydrocarbons, non-methane hydrocarbons, VOCs and
NMVOCs. There is a need to identify a standard method or definition of speciation of
NMVOCS towards which all expert panels could work.
12
SPATIAL DISAGGREGATION CRITERIA FOR AREA SOURCES
No spatial disaggregation (of national emissions estimates) should be required since refineries
are to be inventoried as point sources. However if data is not available on individual
refineries, total regional or national crude processing data could be disaggregated based on
refining capacity.
13
TEMPORAL DISAGGREGATION CRITERIA
No temporal disaggregation is possible if the simpler methodology is used.
If the detailed methodology is used, then individual refineries can indicate the temporal
aspects of shutdowns.
None of the computational methods used to estimate fugitive leaks are based on parameters
that show seasonal or diurnal changes. Therefore it is not possible to disaggregate fugitive
process emissions at this time.
14
ADDITIONAL COMMENTS
In the European community, CONCAWE (1992) reports that the complexity of refineries has
increased with the installation of additional conversion units (e.g. thermal crackers, catalytic
crackers and hydrocrackers) as the demand for fuel oil production decreases and the demand
for a higher yield of gasoline and other light products. This is shown on the table in which
CONCAWE uses a system of refinery classifications that are based on increasing complexity.
PROCESSES IN PETROLEUM INDUSTRIES
Activities 040101 & 040102
pr040101
B411-22
1 September, 1999
Emission Inventory Guidebook
Table 14.1:Concawe Petroleum Refinery Classification System
Year
No. of Refineries
Reporting
Type I
Type II
Type III
Type IV
No.
%
No.
%
No.
%
No.
%
1969
81
34
42
31
38
16
20
-
-
1974
110
49
45
40
36
21
19
-
-
1978
111
50
45
36
32
25
23
-
-
1981
105
38
36
44
42
23
22
-
-
1984
85
18
22
47
54
20
24
-
-
1987
89
12
13
53
60
19
22
5
5
1990
95
9
9
53
56
22
23
11
12
Notes:
Type I:
Simple (non-conversion refinery: composed of crude oil distillation, reforming, treatment of
distillate products, including desulphurization and/or other quality improvement processes (i.e.
isomerization or specialty manufacturing).
Type II:
Type I plus catalytic cracking and/or thermal cracking and/or hydrocracking.
Type III:
Type II plus steam cracking and/or lubricant production within the refinery fence.
Type IV:
Refineries not in above categories, e.g. those producing only bitumen, lubes, etc. which import
their feedstocks from other sources.
This classification system could be adopted for use in developing generic emission factors for
application in the simpler inventory method. It could also be useful in developing generic
speciation profiles.
15
SUPPLEMENTARY DOCUMENTS
There are no supplementary documents.
16
VERIFICATION PROCEDURES
There are more sophisticated and accurate methods to estimate fugitive process emissions, as
developed by the U.S.EPA (1993). All of these methods involve the use of screening data,
which are collected by using a portable monitoring instrument to sample air from potential
leak interfaces on individual pieces of equipment. A screening value is a measure of the
concentration, in ppmv, of leaking compounds in the ambient air near the equipment in
question. The EPA has detailed what is involved in an acceptable screening program in the
protocol for equipment leak emissions estimation manual (U.S.EPA 1993).
PROCESSES IN PETROLEUM INDUSTRIES
pr040101
Activities 040101 & 040102
Emission Inventory Guidebook
1 September, 1999
B411-23
The approaches to estimating equipment leak emissions based on screening data are:
1.
Screening Ranges Approach
2.
EPA Correlation Approach and
3.
Unit -Specific Correlation Approach.
In the screening value approach, it is assumed that components having screening values
greater than 10,000 ppmv have a different average emission rate than components with
screening values less than 10,000 ppmv.
The EPA Correlation approach offers an additional refinement by providing an equation to
predict mass emission rate as a function of screening value.
In the last approach, mass emissions rates are determined by bagging a specific type of
equipment. The associated screening value can then be used to develop a leak rate/screening
value correlation for that equipment in that process unit.
All of these methods are described in detail in the protocol document (U.S.EPA 1993).
As previously discussed, remote sensing monitoring programs can also provide verification of
emissions estimates based on emission factors. However it is often difficult to differentiate
between different refinery sources, and so this method would more often be used to verify
total refinery emissions (i.e., more than just process and fugitive process emissions).
17
REFERENCES
British Petroleum Company, 1986. “Diary 1986”, London, England.
Canadian Petroleum Products Institute (CPPI) and Environment Canada, 1991.
“Atmospheric Emissions from Canadian Petroleum Refineries and the Associated Gasoline
Distribution System for 1988.” CPPI Report No. 91-7. Prepared by B.H. Levelton &
Associates Ltd. and RTM Engineering Ltd.
Commission of the European Community (CEC), 1991a. “CORINAIR Inventory. Default
Emission Factors Handbook.” Prepared by the CITEPA under contract to the CEC-DG XI.
Commission of the European Community (CEC), 1991b. “CORINAIR Inventory. Part 6.
VOC’s Default Emission Factors (Total NMVOC and CH4)(Updating).” Working Group: R.
Bouscaren, J. Fugala, G. McInnes, K.E. Joerss, O. Rentz, G. Thiran, and C. Veldt. (July)
Concawe, 1992. Report No. 1/92.
Passant, N.R., 1993 “Emissions of volatile organic compounds from stationary sources in the
United Kingdom: A review of emission data by process.”
Poten & Partners Inc., 1988. “Oil Literacy.” Poten & Partners Inc., New York.
Tsibulski, V., 1993. Private communication. Scientific Research Institute of Atmospheric Air
Protection, St. Petersburg.

PDF document icon B411vs2.2.pdf — PDF document, 97 KB (99912 bytes)

Permalinks

For references, please go to https://www.eea.europa.eu/publications/EMEPCORINAIR5/B411vs2.2.pdf or scan the QR code.

PDF generated on 20 Aug 2025, 12:53 AM

EEA Page URL QR
Document Actions
The EEA is an agency of the European Union

Engineered by: EEA Web Team

Software updated on 26 September 2023 08:13 from version 23.8.18

Software version: EEA Plone KGS 23.9.14

Code for developers

Systems Status

Legal notice

Creative commons license

CMS login