Application of Optical Remote Sensing in the Petroleum and Petrochemical Industries

By
Mark Modrak, Wojciech Jozewicz, Geoffrey Swett1
ARCADIS
4915 Prospectus Drive
Research Triangle Park, NC 27713

Introduction
Numerous regulatory agencies and industrial firms have indicated an interest in fenceline monitoring
because of concerns expressed by citizen groups and/or a desire to gain a greater understanding of
emissions from tanks or process areas. One such method is open-path optical remote sensing (ORS).
This paper provides the following information:
Regulatory agency initiatives
Technology descriptions
Configuration of instruments
Cost information
DIAL technology
Results of actual use at refinery and petrochemical plants
Key conclusions and lessons learned

Regulatory Agency Initiatives with Optical Remote Sensing

Numerous regulatory agency and/or private company initiatives have resulted in the actual or planned
deployment of ORS technology at facilities in numerous states:
ConocoPhillips, Rodeo, California (San Francisco Bay Area) Citizen concerns about historic
releases by prior owner

Geoffrey Swett works in the Tucson, Arizona ARCADIS office.

Texas Petrochemical Company, Houston, Texas Texas Commission on Environmental Quality

and City of Houston voluntary agreement
BP, Texas City, Texas US Environmental Protection Agency (EPA) Consent Decree
ConocoPhillips, Wood River, Illinois Illinois and US EPA request as part of a permitting process
Chevron, Richmond, California Local land use control agency request as part of a permitting
process
Unnamed Houston Area Refinery City of Houston EPA grant to evaluate DIAL at Houston area
refinery2
In summary, a number of agencies and public interest groups have expressed interest in the technology,
and one should assume other facilities might be asked to implement some sort of ORS in the future.3

Technology Description
ORS instruments have been used as an alternative to conventional point monitors for measuring air
emissions for many years. ORS instruments use infrared (IR), laser, or ultraviolet (UV) light to measure
concentrations of chemical compounds of interest along the distance covered by the light signal.
A light signal is sent out to mirrors deployed in the field, and the signal is reflected back to the
instrument detector. Depending on the instrument and application, typical ORS instrument range varies
from 50 to 500 meters. The major advantage of ORS instrumentation over traditional point monitors is
their ability to provide an increased granularity of spatial information of the monitored area. The
increased granularity reduces the chance of emissions hot spots being undetected over the measured
area.
It is possible to accomplish increased granularity because of the development of the Radial Plume
Mapping method (RPM), which is capable of collecting concentration data along multiple beam paths in
the configuration. In this method, multiple retro-reflecting mirrors are deployed in the survey area. The
RPM can be applied using any scanning ORS instrument.

It is important to note that the City of Houston, in commenting on a proposed TCEQ Corrective Action Order
stated that . . . This project has clearly substantiated the value of fence-line monitoring and that it should be
employed at all facilities that are significant emission sources. (emphasis added) Source: Letter of March 23,
2009, from the Arturo Blanco, Chief, Bureau of Air Quality, City of Houston, to Terry Murphy, Enforcement
Coordinator, TCEQ.
3

Many facilities have fence-line gas chromatograph equipment that senses emissions in real time. Obviously these
are point measurements and do not capture all emissions along the length of a fence line. This paper does not
address facilities that have a fence-line gas chromatograph.

ORS instruments can be used in three basic modes of monitoring for the detection of air emissions:
1. Monitoring along and downwind the perimeter (fenceline) of the facility
2. Surface monitoring within the facility
3. Emergency response to predict impact of plume downwind from source
Depending on the mode, the ORS instrument can be scanning in a horizontal plane (Horizontal Radial
Plume Mapping [HRPM]) to deliver surface monitoring and to produce concentration contour maps. The
ORS instrument can also be scanning in a vertical plane deployed downwind of the survey area (Vertical
Radial Plume Mapping [VRPM]) to map the emissions plume downwind of the area of interest.
By including meteorological data collected concurrently with the ORS measurements, the VRPM method
can be used to calculate the downwind emission flux from the site. Because of the above, the ORS
monitoring methods appear to be particularly useful to determine emissions from petroleum and
petrochemical processing facilities.
This paper presents various types of instruments that can be used for ORS measurements conducted at
petroleum and petrochemical facilities. For each instrument, the basic principle of operation is
described, as well as its operational capabilities. The comparison of performance of various types of
instruments is also given. In addition, principles of and considerations for RPM setup are discussed.
Further, OTM10 and TO16 methods are analyzed in the context of RPM. Finally, actual industry
experience is reviewed.

Optical Remote Sensing - Instrumentation

ORS utilizes a light source to detect and measure concentrations of chemical compounds along the
distance covered by the light signal. Depending on the source of light (wavelength), the following types
of ORS instrumentation could be used:
Open-Path Fourier Transform Infrared Spectrometer (OP-FTIR), IR light, 2 to 20 micrometers
wavelength
Open-Path Tunable Diode Laser Absorption Spectrometer (OP-TDLAS), near-IR light,
approximately 1.5 micrometers wavelength
Ultraviolet Differential Absorption Spectrometer (UV-DOAS), UV light, 245 to 380 nanometers
wavelength
The three types of ORS instrumentation are discussed below. In addition, we will address Differential
Absorption Light detection and ranging (Lidar) (DIAL) technology, which is currently in great demand by
community interest groups.

OP-FTIR
In the OP-FTIR spectrometer, a modulated IR light beam is transmitted from a single telescope to a
retro-reflecting mirror target, which is usually set up at a range of 100 to 500 meters. The returned light
signal is received by the single telescope and directed to a detector. The light is absorbed by the
molecules in the beam path as the light propagates to the retro-reflecting mirror and again as the light is
reflected back to the analyzer. Figure 1 is a picture of the installation of multiple retro-reflectors
deployed for OP-FTIR measurement.

a
Figure 1 Multiple retro-reflectors deployed for OP-FTIR measurement.
Thus, the round-trip path of the light doubles the chemical absorption signal. The advantage of OP-FTIR
monitoring is that the concentrations of a multitude of IR-absorbing gaseous chemicals can be detected
and measured simultaneously, with high temporal resolution. Typically, OP-FTIR instruments can collect
data at resolutions of 0.125 cm-1, 0.25 cm-1, 0.5 cm-1, 1 cm-1, 2 cm-1, 4 cm-1, and 8 cm-1. In fact, the
OP-FTIR is the most cost-effective ORS instrument for applications where it is necessary to measure
multiple compounds. Detection levels for typical refinery and petrochemical emission species is
provided in Table 1.

Table 1 - Detection Limits for Optimal OP-FTIR Setup with Clean, Fully Populated Retroreflector Arrays
MDLs

MDLs

200 meters

200 meters

1 minute

30 minute Avg.

(ppb)

(ppb)

Formaldehyde

0.8

1,3-Butadiene

1.2

Styrene

1.6

Benzene

30

Toluene

40

m-Xylene

22

o-Xylene

20

p-Xylene

32

Acetylene

1.2

Ethylene

0.6

Propylene

Methanol

1.4

Methyl tert-butyl ether

THC

10

1.8

Species

In one study conducted by ARCADIS, two OP-FTIR instruments (Unisearch, Inc. and IMACC, Inc.) were
evaluated. The OP-FTIR instruments were scanned to collect path-integrated methane concentration
data over multiple beam paths at the sites. Additionally, the OP-FTIR data was analyzed for the presence
of ammonia and volatile organic compounds (VOC) at the sites. The OP-FTIR detected ammonia,
methanol, and gasoline (primarily octane) at parts-per-billion (ppb) concentrations.
The major advantage of the OP-FTIR (over the other ORS instruments used in the study) is the ability to
detect multiple compounds from the same data set. Both OP-FTIR instruments were extremely stable
and reliable over the course of the four field campaigns, with little or no maintenance required (Modrak
et al., 2005a, Modrak et al., 2005b).
One of the disadvantages of the OP-FTIR is the effort needed for deployment of the instrument. During
the duration of the project, the OP-FTIR may need to be deployed in multiple locations at each site.
Although the instrument is mounted on a scanner, it requires at least two people to mount the
instrument to the scanner cart and deploy the scanner cart to the measurement area.
However, the effort needed for deployment of the OP-FTIR would not be an issue for applications where
the OP-FTIR instrument is deployed in a single configuration. Once the instrument has been deployed, it
5

must be aligned on each mirror used in the configuration. Alignment of the instrument is done using the
scanner joystick control, which makes the alignment process quick and simple.
In recent years, OP-FTIR manufacturers have developed smaller instruments that are substantially
lighter than the OP-FTIR instrumentation of the previous generation. The disadvantage of these lighter
instruments is that their detection limits are generally higher, and the optical range of the instruments is
shorter. These features would not allow the instrument to scan over beam paths greater than 100 to
200 meters. However, the smaller OP-FTIR instruments may be ideal for certain field applications.
Another disadvantage of the OP-FTIR is the need for liquid nitrogen to cool the instrument detector.
Therefore, it may be necessary to add liquid nitrogen to the OP-FTIR instruments at the beginning of
every day of data collection. For projects requiring 24-hour data collection, it would be necessary to add
liquid nitrogen to the instrument approximately every 12 hours. Although filling the instrument does not
require a large amount of effort, it does add some extra cost to the field operations.
It should be noted that large liquid nitrogen dewars are commercially available for use with OP-FTIR
instrumentation. The dewars can be filled with liquid nitrogen and are capable of automatically
providing liquid nitrogen to the OP-FTIR for long periods. Although the large dewars are ideal for
projects with permanent deployment locations, the large dewars are not an option for field campaigns
with multiple configurations and survey areas.
Post-field analysis of the OP-FTIR data is currently recommended to quantify path-integrated
concentrations. However, with the development of quantification software such as Non-Lin and
IMACCQuant, this process does not require a spectroscopy specialist. Recent advancements such as realtime IMACCQuant and ARCADIS Real-Time RPM Software have made it possible to quantify real-time
concentrations in the field.
However, concentration determinations made with real-time and post-field analyses should be carefully
compared. Examples of commercially available OP-FTIR equipment are shown in Figure 2.

Figure 2 - OP-FTIR Instruments from Unisearch Associates (left) and IMACC (right)

OP-TDLAS
The OP-TDLAS instrument is a fast, interference free, and sensitive technique, for making continuous
concentration measurements of many gases. Concentrations in the range of part per billion are suitable
for measurements over an open path up to 1000 meters for gases such as CO, CO2, NOX, NH3, and CH4.
The laser emits radiation at a particular wavelength when an electrical current is passed through it. The
light wavelength depends on the current.
The OP-TDLAS applies a small 4-inch telescope, which launches the laser beam to a retro-reflecting
mirror. The laser beam is returned by the mirror to the telescope, which is connected with fiber optics
to a control box that houses the laser and a multiple channel detection device.
OP-TDLAS instruments are available commercially as single channel (e.g., Boreal, Inc.) and multiple
channel (e.g., Unisearch Associates, Inc.) instruments. Multiple channel instruments use multiple
telescopes to collect path-integrated concentration data along multiple beam paths (up to 8 beams).
Unisearch OP-TDLAS and Boreal OP-TDLAS systems are shown in Figure 3.

Figure 3 - OP-TDLAS from Unisearch (left) and from Boreal, Inc. (right)
In a recent project, ARCADIS used OP-TDLAS instruments from Unisearch and Boreal to collect
path-integrated methane concentration data.
The major advantages of the OP-TDLAS instruments are that they are lightweight, can be easily deployed
by one person, and operate at ambient (rather than cryogenic) temperatures. The mirrors needed for
OP-TDLAS measurements are also smaller and more lightweight than the mirrors traditionally used with
OP-FTIR measurements. Another advantage of the OP-TDLAS is that it generates real-time, pathaveraged concentration data in the field, without the need for post-field data analysis.
One disadvantage of some OP-TDLAS instruments is that they cannot be mounted to a scanner. These
instruments consist of a control box (that houses the laser and a multiple channel detection device) and
stationary telescopes. Each telescope must be connected to the control box with fiber optic cables.
Depending on the topography and location of physical barriers at the survey area, the distance between
the control box and the telescopes may require a large amount of fiber optic cable, which can be
difficult to deploy. Instruments mounted on scanners are much easier to deploy.
Another disadvantage of the OP-TDLAS instruments is that they are not capable of measuring multiple
compounds from the same dataset. Although OP-TDLAS instruments can detect many compounds (such
as methane, CO, CO2, NOX, and NH3), it is necessary to employ separate lasers for analysis of multiple
gases.

UV-DOAS
The UV-DOAS instrument is useful for high sensitivity determination of many aromatic species (e.g.,
benzene-toluene-xylene [BTX] [Platt, 1994]). Additionally, the UV-DOAS instrument may be utilized for
determination of the concentration of unstable species like free radicals, nitrous acid, and others
(Cowen et al., 2004, Kelly et al., 2003, Myers et al., 2000). UV-DOAS, like all spectroscopic techniques,
makes use of the absorption of electromagnetic radiation by matter.
8

While the strong, structured UV absorption features of aromatic hydrocarbons have been known for a
long time, it only recently became possible to use these properties for the reliable, sensitive, and
selective measurement of monocyclic aromatics by UV-DOAS. UV-DOAS measurements of trace gases
can be a valuable complement to more traditional techniques like OP-FTIR.
The major advantage of the UV-DOAS instrument is the ability of sensitive detection of the BTX
compounds at low concentrations and with good time resolution. During one campaign conducted by
ARCADIS, the UV-DOAS detected benzene, toluene, and xylene at levels less than 5 ppb. The OP-TDLAS
instrument cannot detect the BTX compounds, and although the OP-FTIR is capable of detecting the BTX
compounds, the minimum detection level of the UV-DOAS instrument is much lower than the OP-FTIR.
Another advantage of the UV-DOAS instrument is that it does not require liquid nitrogen for operation.
Figure 4 presents a picture of the UV-DOAS system from OPSIS.

Figure 4 - UV-DOAS Instrument from OPSIS

One disadvantage of the UV-DOAS instrument used in the long-term study is that it requires
post-analysis of the data collected. It is recommended that a spectral validation of the detected
compounds be provided for studies performed using this instrument. In addition, when utilizing
UV-DOAS, particular care should be exercised to ensure proper alignment of the instrument.

Comparison of ORS Instruments

Background information on the selected ORS instrumentation is shown previously in Table 1. The table
lists the detectable compounds measured by each instrument, instrument limitations such as weather
and interfering species, and cost of instrumentation. As can be seen from Table 2, while all ORS
instruments have low detection limits on the order of a ppb, their capability to detect compounds
varies. OP-FTIR offers the widest detection capability.
9

Table 2 - Comparison of ORS Instruments

OP-FTIR

OP-TDLAS

UV-DOAS

Wavelength Range

Infrared
(2-20 micrometers)

Near Infrared
(approx. 1.5
micrometers)

Ultraviolet
(245-380 nanometers)

Detectable Compounds

Multiple

CO, CO2, NOX, NH3, CH4,

H2S and others

BTEX, NH3,, CO, CO2,

mercury, and other VOC

Detection Limits

Parts per billion

Parts per billion

Parts per billion

Limiting Weather Conditions

Heavy rain

Heavy rain, fog

Heavy rain, fog

Interfering Species

Carbon dioxide, water

None

Oxygen, ozone

Instrumentation Cost

$125,000

$75,000

$200,000

a. Cost includes ORS instrument, scanner, and retro-reflecting mirrors. Does not include installation cost, utility installation and
any associated data processing and reporting equipment.

Optical Remote Sensing Configuration of Instruments

The application of ORS instrumentation for measuring air emissions was enhanced by the development
of the RPM method (Hashmonay and Yost, 1999). The method, which can be applied using any of the
ORS instruments described above, collects path-integrated concentration (PIC) data along multiple
beam paths in the configuration. The ORS instrument can be deployed in a horizontal plane (HRPM) to
produce surface concentration contour maps or in a vertical plane deployed downwind of the survey
area (VRPM) to map the emissions plume downwind of the area of interest.
By including meteorological data collected concurrently with the ORS measurements, the VRPM method
can be used to calculate the downwind emission flux from the site. The RPM method is now an
approved EPA Other Test Method 10 (OTM-10) for characterizing fugitive emissions from area sources.
The RPM method can be used for characterizing emissions from a wide range of area sources (Thoma et
al., 2005).
Some of the important considerations when deploying VRPM method configuration include the fact that
it should be set up downwind of the survey area so that the prevailing wind direction is as close to
perpendicular to the plane of the configuration as possible. This is necessary to ensure that the VRPM
configuration is capturing as much of the emissions plume from the survey area as possible (there is no
need to consider prevailing wind direction when using the HRPM method).
When applying the VRPM configuration, the decision is made where to deploy the configuration based
on forecasted wind directions. However, the actual observed wind direction is not always the same as
the forecasted direction. This can be problematic when conducting a project involving multiple
configurations with limited project resources. When applying the VRPM configuration, it is

10

recommended to deploy the configuration in one location for the longest time period possible, to
account for periods of varying wind conditions.

Comparison of ORS to Point Sampling

Other measurement approaches can also be used to obtain emissions measurements for the area
sources. These include traditional point sampling instrumentation such as photo-ionization detectors
(PID), PID/flame ionization detectors (FID), Summa Canisters, various sorbent methods, and flux boxes.
Although these approaches are generally easier to deploy, less costly than ORS-based measurement
approaches, and do not rely on prevailing wind direction during the time of measurements, they only
provide concentration information from a single point in the survey area, greatly increasing the chances
of missing surface emissions hot spots or emissions plumes.
Even after collecting data from multiple points in the survey area, the point sampling approaches lack
the spatial and temporal data necessary to obtain a complete picture of the emissions from large area
sources. Additionally, the flux box approach may not accurately characterize surface emissions from the
site, as deployment of the flux box on the surface of the sampled area may not allow actual emissions to
escape in the vicinity of the deployment area. Another disadvantage of using the point sampling
approaches is that it is necessary to use dispersion modeling to obtain flux data from the site. Table 3
presents a summary of the advantages and disadvantages of the ORS and traditional point monitor
approaches.
Table 3 - Summary of the Advantages and Disadvantages of ORS and Traditional Point Monitor
Approaches
Measurement Method

Advantages

Disadvantages

ORS-Based Approach Using

Radial Plume Mapping

Measurements collected over a

larger area making it less likely to
miss major emissions areas

Instrumentation is more costly

Better spatial and temporal

resolution

Relies on prevailing wind

direction for emissions
measurements

Direct, measurement-based
emissions calculations
Traditional Point Monitors

Easy to deploy

Requires more time and effort to

deploy

Only provides data from a single

measurement point

Less costly to deploy

No reliance on prevailing wind
direction for measurements

Lack of spatial and temporal data

Possibility of missing major
emissions areas
Requires modeling to obtain
emissions calculations

Table 4 presents a cost comparison of traditional point monitoring approaches and ORS-based
approaches using RPM. The table includes approximate costs of instrumentation, number of target
11

compounds each approach is capable of detecting simultaneously, minimum detection limits (MDL),
spatial resolution of each measurement technique, and number of personnel necessary for deployment.
Table 4 - Summary Information on the ORS Instrumentation and Other Traditional Monitors

Measurement Approach
Scanning OP-FTIR with RPM

Equipment
Cost
a

MDL
(ppb)

Spatial Resolution

Personnel
Needed for
Deployment

$125,000

1 to 100

Entire Survey Area

1 to 100

Entire Survey Area

Scanning OP-TDLAS with RPM

$75,000

Scanning UV-DOAS with RPM

$200,000

1 to 100

Entire Survey Area

2 to 3

$20,000 to
$40,000/week

1 to 100

Entire Survey Area

2 to 3

$10,000

1 to 100

Single Point

Cost of ORS/RPM-Based
b
Field Campaign
PID

PID/FID

$10,000

1 to 100

Single Point

Summa Canister

$350 per
d
sample

0.1 to 0.5

Single Point

Flux Boxes

$1,000

0.1 to 0.5

Single Point

Sorbent Methods

Varies

0.1 to 0.5

Single Point

a. Cost includes ORS instrument, scanner, and retro-reflecting mirrors. Does not include installation cost, utility installation and any
associated data processing and reporting equipment.
b. Cost includes creation of quality assurance documentation, conducting field campaign, data analysis, and reporting. The cost could
vary depending on the size of the site and number of survey areas.
c. Does not include the cost of hydrogen needed at the site to operate the instrument
d. Does not include the cost of a gas chromatograph needed to analyze sample
e. Cost includes materials for constructing box, but does not include the cost of sample analysis, which are typically collected with
summa canisters or FID instrumentation

It should be noted that the measurement approaches shown in Table 4 above can be performed as a
service by an environmental contractor, meaning that an initial capital investment for equipment is not
necessary. The approximate costs of hiring a contractor to conduct a 5-day field campaign using an
ORS/RPM-based approach are typically between $20,000 and $40,000, depending on the size of the site,
number of survey areas, and target compounds of the study. The cost includes creation of quality
assurance documentation, data collection, data analysis, and reporting.
In comparing the costs of an ORS-based measurement approach with traditional point monitoring
approaches, an ORS-based approach using the RPM method may be more cost-effective. In addition, the
ORS-based approach provides much better spatial and temporal resolution of concentration data,
allowing for characterization of emissions plumes and direct calculation of emissions fluxes. In order to
achieve the same level of temporal and spatial resolution (and speciation of the target analyte) using
traditional point monitors, the user would have to deploy a multitude of monitors simultaneously at the
site, resulting in substantially increased sampling and analysis costs.
12

DIAL-Based Approach for Total Site Emissions Measurements

DIAL is a type of light-based range-resolved system, similar to Radio Detection and Ranging (RADAR).
The laser source can be tuned-in to the absorption feature in the known spectral response region of the
target gas and then tuned-off of the region of the absorption feature. The concentration of the gas in
each atmospheric volume is determined by comparing the strength of the time-gated backscattered
light of both laser wavelengths. DIAL instrumentation can be used to measure gases with strong
absorption features in the UV and visible regions, such as N2O, NOX, and SO2, aromatics, or other
compounds with strong absorption features in the IR region, such as total hydrocarbons (THC) and
methane. As such, DIAL is particularly useful to detect and characterize fugitive emissions from
petroleum processing facilities. Figure 5 provides a typical depiction of the technology.

Figure 5 Typical DIAL Configuration (Source: Spectrasyne)

One of the key advantages of the DIAL technology is ability to scan greater distances than conventional
ORS. Measurements can be taken at a distance of 1 kilometer.
One of the key limitations of the technology is the difficulty in speciating compounds. The laser needs to
be tuned to the particular compound being looked for.
It can measure THC, but an assumption of the molecular weight must be made. ARCADIS has found that
the molecular weight of hydrocarbons at a refinery can vary significantly by process area (e.g., crude
tank THC vs. gasoline tank THC). DIAL is not able to differentiate the hydrocarbon molecular weights.

13

The results of DIAL studies at refineries in Europe and Canada have been very controversial in the
industry. In most cases, the data has indicated that estimated annual hydrocarbon emissions were
significantly higher than predicted by AP-42. In some cases, the estimates were ten times higher than
AP-42 would have predicted. However, the results have not been extensively peer reviewed.
Figure 6 presents a summary of multiple DIAL surveys at petrochemical facilities, with measured
hydrocarbon emissions reported as a percentage of throughput on a weight basis.

Figure 6 - Comparison of Emissions from Various Facilities (from Chambers and Strosher, 2006)
As can be seen in Figure 6, crude and product tank farm emissions dominate the hydrocarbon emissions
at the facilities where DIAL has been deployed.
A key issue in many recent DIAL studies is flare emissions. This debate, coupled with EPAs recent flare
enforcement initiative, indicates that the accurate measurement of flare emissions may be an issue in
the future.
Flare emission quantification can be done with either DIAL or FTIR. When the FTIR technology is used,
the flare itself is the source of infrared radiation.
One final issue with DIAL is the high cost of the technology. There are no domestic suppliers, so the
equipment must be mobilized from Europe. For a typical 28 to 35 day-long field campaign at a major
facility, DIAL would cost about $500,000 compared to $150,000 to $175,000 for an FTIR study.

14

Long-Term Field Experience

ORS has been deployed on a permanent basis at two facilities, a Gulf Coast petrochemical manufacturer
and a California refinery. Other facilities have used ORS technologies for short-term monitoring
campaigns.
The refinery has had more than 12 years of experience with the tools. The system was added as a permit
condition with the local land use authority. The primary purpose was to provide an early warning system
of upset conditions that would impact ambient air quality in the surrounding communities.
The original 1997 system was mounted on the facilitys north and south fences. It consisted of a
UV-DOAS and a monostatic FTIR with retroreflectors. The systems never were reliable and had problems
with corrosion on mirrors and water vapor interference. The systems could detect higher
concentrations, but not the low ppb ambient concentrations needed to detect unusual releases.
The system did not identify any releases that the refinery did not know about from its operating and
control systems.
In 2001, a different UV system was added, and it was much more reliable and effective.
In 2006, the original FTIR system was replaced with a Midac bistatic FTIR system. This system eliminated
the use of reflectors, so the mirror corrosion challenge was eliminated.
The newer equipment had a level of detection of about 5 ppb, much lower than the original systems.
Measurements of methane and nitrogen oxide matched expected ambient concentrations.
The system has not achieved the objective of being an early warning system for releases that the facility
was unaware of. In fact, the only unknown significant release ever detected was from a tank at an
adjacent terminal owned by a separate party.
The petrochemical facility is a butadiene manufacturer, which is near a residential neighborhood. Due to
agency and public concerns, the facility voluntarily put in place a fenceline monitoring system. It was
installed in early 2006 and became operational on July 1, 2006.
The system consisted of a monostatic FTIR set up with the reflectors at 50-yard increments up to 400
yards. The system is focused continuously on the 400-yard reflector, unless an increased level of
butadiene is detected. Then the system automatically rotates to the shorter reflectors until the highest
concentration of butadiene is detected. At that point, operations and maintenance are notified and an
investigation of potential sources of a release is searched for using a hand held IR spectrometer camera.
A key component of the overall system is in the internal data management and communications tools
that the facility uses. The FTIR feeds into a compliance management system that continuously analyzes
the FTIR and other operational data and notifies responsible individuals of a potential release. Every
potential release is investigated using root-cause methodologies.
15

The ORS installation occurred at the same time the facility was undergoing numerous improvements in
its operations, including improved flare gas recovery, continuous total organic carbon (TOC) analysis of
cooling tower water, and upgraded operations and maintenance procedures. Therefore, it is difficult to
attribute its improved performance to just the ORS program.
The ORS system was one component of a systematic approach to improving the plants equipment and
operational performance. The result of the systematic approach was dramatic, with fugitive emissions
being reduced by more than 45% and flare hydrocarbon emissions being reduced by 90%.
Because of this improved performance, the plant gained credibility with regulators, nearby residents,
and community activists. However, the improvement was not significant enough to satisfy fully the
regulators who recently negotiated a commitment by the company to expend an addition $20,000,000
in incremental expenditures on projects to enhance environmental performance at the plant.

Conclusion and Lessons Learned

Our experience with ORS technologies has taught us the following lessons:
Carefully define your project goals and objectives up front, including what compounds need to
be measured and for what purpose
Align your technology selection with your project goals
Ensure you implement a quality assurance/quality control program. Analysis of spectroscopic
data is often complex and -despite software advances- the value of a seasoned spectroscopist
should not be under estimated.
ORS technologies have moved from the experimental stage in the 1990s to a reliable and valuable
technology for helping to evaluate emissions in certain situations. The value of the technology is in
identifying sources of hazardous air pollutants in situations where risk assessments indicate an
unacceptable risk. For facilities that have not been identified as posing an unusual risk by the residualrisk rule, site-specific modeling, or ambient monitoring, ORS adds little value at the current time in
reducing risk.
For those facilities that could potentially pose a risk and opt to use ORS, it is recommended that a
systematic and integrated approach be deployed to risk reduction. The facility should have a systematic
approach to identifying sources, investigating root causes and making corrective actions, including
changing work practices to leverage the value of the tool.
ORS is only valuable if it is a part of a systematic effort to change operational and maintenance behavior
to address a known risk.

Acknowledgments
We wish to thank the numerous industry participants who were very generous with their time.
16

Literature
Chambers, A. K.; Strosher, M. Refinery Demonstration of Optical Technologies for Measurement of
Fugitive Emissions and for Leak Detection; ARC Contract Report No. CEM 9643-2006, March 2006.
Cowen, K., I. MacGregor, K. Hand, J. Carvitti, M. Rectanus, T. Kelly, and K. Riggs. Environmental
Technology Verification Report: OPSISAB LD500 Continuous Emission Monitor for Ammonia; U.S.
Environmental Protection Agency ETV Program, Battelle, Columbus, OH, September 2004.
Hashmonay, R.A., and M.G. Yost, Innovative approach for estimating fugitive gaseous fluxes using
computed tomography and remote optical sensing techniques, J. Air Waste Manage. Assoc., 49, 966972, 1999.
Kelly, T., Z. Willenberg, K. Riggs. Environmental Technology Verification Report: OPSIS AB Hg-200
Mercury Continuous Emission Monitor; U.S. Environmental Protection Agency ETV Program, Battelle,
Columbus, OH, September 2003.
Modrak, M. T.; Hashmonay, R. A.; Varma R.; Kagann, R. Evaluation of Fugitive Emissions at a Brownfield
Landfill in Ft. Collins, Colorado Using Ground-Based Optical Remote Sensing Technology; EPA-600/R05/042; U.S. Environmental Protection Agency, Research and Development, Work Assignment No. 0025, March 2005a.
Modrak, M. T.; Hashmonay, R. A.; Varma R.; Kagann, R. Evaluation of Fugitive Emissions at a Brownfield
Landfill in Colorado Springs, Colorado Using Ground-Based Optical Remote Sensing Technology; EPA600/R-05/041; U.S. Environmental Protection Agency, Research and Development, Work Assignment
No. 0-025, March 2005b.
Myers, J., T. Kelly, C. Lawrie, K. Riggs. Environmental Technology Verification Report: OPSIS Inc. AR-500
Ultraviolet Open-Path Monitor; U.S. Environmental Protection Agency ETV Program, Battelle, Columbus,
OH, September 2000.
OTM-10 http://www.epa.gov/ttn/emc/tmethods.html
Thoma, E.D., R.C. Shores, E.L. Thompson, D.B. Harris, S.A. Thorneloe, R.M. Varma, R.A. Hashmonay, M.T.
Modrak, D.F. Natschke, and H.A. Gamble; Open-Path Tunable Diode Laser Absorption Spectroscopy for
Acquisition of Fugitive Emission Flux Data; Journal of Air and Waste Management Association, (55), 658668, 2005.

17