Technical Paper June 2013

Use of Lasers to Monitor Emissions

in the Fertiliser Industry

Hamish Adam, John Selby

Boreal Laser Inc.
12846 146 St., Edmonton, Alberta, Canada
hadam@boreal-laser.com

Lowry Harper
Lowry A. Harper Consulting Company
P.O. Box 772
Watkinsville Georgia 30677 USA

KEYWORDS
Tunable diode lasers, TDL, gas leak detection, ammonia, hydrogen fluoride

TABLE OF CONTENTS

Summary
1. Introduction to Laser Gas Detection
2. Laser Gas Analyzer Configurations
2.1 Multiple path fixed installation
2.2 Portable systems
3. Leak Detection Applications
4. Fertiliser Industry Examples
4.1 Ammonia and Urea Production
4.2 HF monitoring in phosphate plants
4.3 Monitoring magnitude and fate of NH3 from fertiliser application
5. Conclusions
References

 SUMMARY
Laser gas detection has become well established during the past decade in a variety of safety,
environmental and process monitoring applications which have challenged traditional gas detection
technologies. Laser gas detection is a high resolution measurement technique which gives a gas
specific response that greatly reduces the risk of both missed detection and false alarms. Being a
direct optical measurement it does not suffer from problems associated with reactive gases and
contamination in sampling systems. One second response times make the laser technique ideally
suited to critical safety and process control applications. Laser instruments use no consumables and
have no moving parts resulting in significantly reduced operational and maintenance requirements.
Open path laser measurements, when combined with meteorological data, yield well to analysis by
various flux calculation models. Laser gas detectors are available in both portable and fixed monitoring
configurations.
This paper describes the uses and benefits of laser gas detection in a variety of fertiliser related
applications: safety and process monitoring in ammonia and urea production, storage and
transportation; hydrogen fluoride monitoring in phosphate plants; monitoring the magnitude and fate
of releases of ammonia to the atmosphere during fertiliser application. Data from existing installations
and field trials are presented.

1. Introduction to laser gas detection

The laser gas detection method has several advantages over existing gas detection techniques
(Ballangrud, 1993; Silver, 1992; Bomse, 1995). For practical purposes, lasers generate light at a
single wavelength. Room temperature tunable diode lasers (TDL) emit light in the near infrared (NIR).
Many gases of interest absorb light in the NIR. Each gas has a unique absorption signature, or
spectrum, made up of a large number of individual absorption lines. A TDL can be tuned to select a
single absorption line of a target gas, which does not overlap with absorption lines from any other
gases. Therefore, laser gas sensing is very selective, and does not suffer from interference from other
gases. The single-wavelength nature of laser light also means that laser gas analysers obey the Beer-
Lambert law and exhibit a linear response over a wide dynamic range.
TDLs generate only a few mW of power. However, all this power is concentrated at one wavelength,
the wavelength where gas absorption occurs. So, high signal to noise ratios are achieved, and
response times are short, typically about 1 or 2 seconds. This also means that long path lengths can
be used, enabling accurate measurements of very low concentrations. Diode lasers provide all the
advantages of other semiconductor devices. They are small, solid-state devices. They operate at room
temperature and have long-term reliability (over 15 years). Currently, they can be readily obtained for
any wavelength throughout the NIR (0.7 to 2.5 micron). This restricts applications to measurement of
gases with relatively small molecular weights. Ongoing developments in semiconductor laser
technology will make commercial gas analyzers for use in the 2.5 to 10 micron region a reality of the
next year or two. This will dramatically increase applications potential as most gases have strong
absorption features in this mid infrared spectral region.
The laser light from NIR TDLs can be easily coupled into inexpensive and highly transmitting optical
fibres. The light from a single laser can then be switched into several fibres, enabling multiple point
measurement with a single laser gas analyzer (Tulip, 1998). This leads to lower cost per measurement
and easier installation in harsh and potentially explosive environments.

2
 2. LASER GAS ANALYZER CONFIGURATIONS

2.1 Multiple path fixed installation

Figure 1 presents a schematic representation and photographs a of GasFinderMC system, one
embodiment of a multiple channel laser gas analyzer that provides up to 8 independent measurement
paths.

Figure 1: GasFinderMC system for fixed leak detection.

Aiming'Scope'

Laser'
Display'

Diode' Remote'Head' Retro8'

Reector'
Calibra-on'
Mul8plexer'

Reference'Cell'
Control'
Electronics'

Detector'
Keypad'

Signal'
Processing'
Computer' and' Duct'Sensor'
Display'

The central control unit is a 19 rack mount unit that contains the laser, multiplexing and data
processing components. Fibre-optic cable carries the laser light to transceiver (transmit/receive)
heads, which direct the beam along a path to a reflector. The return light is collected on a non-biased
photo-detector. The photo current is returned to the central control unit on coaxial cable via a custom
high frequency barrier module. The use of this barrier module allows for paths to be monitored in
electrically classified areas. The analyzer system is available in open path, stack/duct, and
process monitoring configurations.
This laser gas detector has some features that enable advantages in cost, reliability and ease of use.
The wavelength modulation technique used to achieve maximum sensitivity for trace gas
concentrations (Bomse et al., 1992) normally requires a phase matching between measure and
reference electrical signals that takes time and expertise. This procedure must be repeated every time
the path length is changed. The current analyzer employs a detection technique that removes the
need for this procedure, by generating a reference signal from the measure signal, so that both are
always in phase (Tulip, 1997). What this means is that different path lengths can be measured without
requiring any adjustments to the analyzer.

3
 This analyzer also contains a patented, permanent and stable, reference cell that contains a sample of
the target gas. In operation, the laser wavelength is current tuned to be successively off and then on
the absorption line. When it is off the line, the reference cell measurement is equivalent to a zero
check. When it is on the line, the reference cell reading is a span check. The built-in reference cell also
ensures that the laser wavelength is locked on to the target gas absorption line and not some other
gas with an adjacent absorption line especially important when there is no target gas in the measure
path.
Built-in self-diagnostics and full data communications capability ensure high percentage analyzer up-
time. The absence of any moving parts, consumables or regular calibration requirements ensures low
cost of ownership.

2.2 Portable systems

Figure 2 shows GasFinder2, a laser gas analyzer using the same basic technology described above
that has been configured for portable use.

This configuration consists of an integrated transmitter/receiver unit and a remote, passive retro-
reflector array. The GasFinder2 is aligned with the retro-reflector using a two-axis instrument mount
assisted by a telescopic sight and an on-board visible aiming laser. GasFinder2 can operate with path
lengths from 1m to 1000 m, is battery operated and weighs less than 5kg. It takes less than 10
minutes to set up the instrument and commence measurements. Therefore GasFinder2 is ideally
suited for portable use, and temporary installations.
This portable instrument can also be adapted for multiple-path monitoring with a scanner or vehicle
mounted mobile operation in a fibre coupled configuration. Figure 3 shows a GasFinder2 mounted on a
precision scanning platform that is controlled by a host computer using motion control software. The
programmable scanner has 360 of horizontal movement and 120 of vertical movement with a step
precision better than 0.006 per step. The robust packaging of the system allows it to operate from -
30 C to 50 C in rain, snow, or fog provided that visibility remains good enough for sufficient laser
energy to traverse the measurement path. The system can tolerate a factor of 20 turn-down in
transmitted light energy from ideal conditions before a low light condition obtains. The scanning
system can be programmed to scan up to 36 different paths in succession. Typically, the system will
measure on one path for between 10 and 60 seconds before moving to the next path. Transition time
between paths is normally 2 or 3 seconds. In scanning mode, path averaged concentrations are
transmitted to a central logging computer.

4
 The scanning configuration has primarily been used for estimating emission rates of target gases from
distributed or area sources using a configuration such as shown in Figure 3.

Figure 3: Scanning configuration for flux monitoring

Laser detector on scanner measures Path Integrated
Concentration (PIC) over multiple paths in succession
Combine PIC data with meteorological data
Wind data most important. 3D ultrasonic best
Analyze data with available algorithms to:
Create concentration profiles which locate hotspots.
Provide emissions flux estimates
Calculate emission source strength
Retro-reectors'
IR'beam'execu>ng'' mounted'
a'single'"monitoring'event"'' above'the'
Wind!! (5'events'make'up'a'' ground'
Direc-on' Source' complete'"plume'traverse")'

Directly'measured''
plume'component'

Laser detector mounted on scanner

Precision: 0.006 TDL'scanner'

360 pan movement, 120 tilt movement

10/sec Ground'measured'retro-reectors'
Wireless control

Multiple path integrated gas concentration data are combined with meteorological data such as
turbulence (x, y, z), atmospheric stability (|L|), wind (u*) and terrain (zo) characteristics, and
then analyzed by one of several commercially available stochastic or deterministic dispersion models
to generate flux data.
In another configuration, the GasFinder instrument can be adapted for local point measurements and
mounted to an off-road vehicle for conducting ground based surveys. This configuration has been
most commonly used for mapping concentration profiles of greenhouse gases such as methane (CH4)
and carbon dioxide (CO2) (Adam et al, 2008).

5
 3. Leak detection Applications
A typical configuration for a multiple path laser line-of-sight (LOS) leak detection installation is shown
in Figure 4.

Figure 4: Typical multiple path LOS leak detection installation

Retro&'
Line&of&sight'(LOS)' Reector'
paths'provide' Wind!!
complete'perimeter'' Direc)on+
coverage'
1
Note that leak 1 creates a
cloud that is not detected by
any point sensors
Leak 2 is detected by 1 or 2
point sensors

Point'sensor'
2
Source'of'leak'
LOS$heads'

The primary advantages of the laser LOS system compared with point sensors traditionally used for
leak detection are:
The laser LOS system is gas specific, hence there are no false alarms resulting from the presence of
other gases in the measuring path.
The LOS system enables complete coverage of a process unit or facility perimeter resulting in greater
likelihood of intercepting plumes of leaking gas.
The laser technique responds instantaneously to the presence of target gas within the line of sight.
When target gas clears from the measurement path, the response returns immediately to zero, with
no hysteresis effects.
The laser LOS system provides reliable un-attended performance in all climate conditions. The laser
technique provides a broad dynamic measurement range of about 4 orders of magnitude. Superior low
end sensitivity enables the detection and repair of incipient leaks before they develop into major risk
hazards.
Laser instruments such as those described here are auto line centering and auto validating and
requires no consumables. This means that maintenance is minimal and cost of ownership is very low.
When compared with traditional infrared LOS detectors, the laser LOS technique has the additional
benefit that the very narrow bandwidth laser light is only minimally absorbed by atmospheric water
vapor. This means that the laser beam can penetrate the air further enabling maximum length
measurements paths of between 500m and 1000m depending on target gas.

6
 Laser LOS gas detection has typically been used in applications for which traditional gas detection
technologies were either deficient or non-existent. The systems described here have been successfully
used for detection of Ammonia (NH3), Hydrogen Fluoride (HF), Hydrogen Cyanide (HCN), Hydrogen
Sulfide (H2S), Methane (CH4) and other gases. The technique has become the accepted standard for
monitoring fugitive and stack HF emissions from primary aluminium smelters and is now installed in
over 100 such facilities worldwide. Laser LOS gas detection is recognized by the American Petroleum
Institute (API) as the best available technology for HF leak detection from refinery HF Alkylation units
and is used extensively in this process as well as in other chemical industries which make and use HF.
Open path gas detection using lasers is the method of choice for researchers investigating Greenhouse
Gas (GHG) emissions from area sources. The worlds hydrocarbon reserves are becoming ever more
sour in content leading to strong growth in the application of laser based systems for H2S leak
detection in sour gas production, transportation and processing. The recent adoption of an
International Cyanide Code by the gold industry has led to increased use of laser based HCN detection
systems in gold refining operations.

4. FERTILISER INDUSTRY EXAMPLES

In recent years, the fertiliser industry has also come to appreciate the benefits of laser technology,
particularly for detecting leaks and monitoring ambient concentrations of NH3 and HF. Reliable
detection of these gases has proven to be especially tricky for traditional electro-chemical gas
detectors, with cell saturation, large hysteresis effects and calibration drift being common challenges.
Detection solutions which require the air sample to be pumped from an intake site to the sensor are
susceptible to NH3 losses in sample tubes and plugging of such tubes especially during release
events. Conversely, laser-based open path monitoring of NH3 and HF emissions is easy and reliable,
with detection sensitivities that are ideally matched to typical alarm levels of 3 to 10 ppm for HF and
30 to 100 ppm for NH3. The wide dynamic measurement range of the laser technique enables reliable
detection of incipient leaks of both gases, with significant associated benefits. Firstly, potential leak
situations can be identified and fixed before they become dangerous. Secondly, there is an economic
benefit as costly product losses are limited. Figure 5 demonstrates the faster response and absence of
hysteresis in open path laser measurements compared with traditional techniques.

7
 Figure 5: Laser LOS sensors demonstrate faster response and no
hysteresis compared with traditional gas measurement techniques

4.1 Ammonia and Urea Production

Potential sources of ammonia leaks in ammonia and urea production are primarily associated with high
pressure equipment, storage facilities and loading and unloading operations. The final step in the
Haber-Bosch process involves a high pressure ammonia compressor, having a set of primary and
secondary seals. There is always ambient ammonia in the vicinity of this compressor because of bleed
past the seals. Ambient levels are typically between 18 25 ppm, which will poison electro-chemical
sensors but not LOS laser sensors. Reliable ammonia monitoring around the compressor is critical to
ensure fast response in the case of primary seal failure. A typical installation and data from a
compressor monitoring installation is shown in Figure 6.

8
 Figure 6: NH3 compressor room monitoring

Path%on%le*%is%20%cm%(blue%trace)%
Path%on%right%is%25m%(red%trace)%
Note:%NH3%spike%to%98%ppm%during%lter%change%

Ammonia is stored in either high pressure spheres or cooled down to below 33 oC and stored in large
tanks. The refrigeration compressor which recirculates NH3 coolant represents another potential leak
source. NH3 is transported by trucks, railcars or ships to a distribution network of ammonia terminals.
There are hundreds of such terminals worldwide. Storage facilities and trans-shipment at terminals
provide multiple opportunities for accidental releases of NH3. Laser based perimeter monitoring for an
NH3 leak has been implemented at several ammonia storage and terminal operations in North
America.
Ammonia is the primary feedstock in Urea production, so Urea plants also have several continuous
and discontinuous NH3 emission sources. A high pressure ammonia pump at the base of the urea
tower is a key area of concern for leaks. Piping, valves and safety rupture disks represent other high
risk leak sources. Liquid urea is turned into solid form in either granulators or prilling towers. Most
urea production in Europe and North America uses granulation, but prilling towers are still favoured in
many other countries because of lower operating costs. Prilling can produce substantial amounts of
urea dust and ammonia off gassing. A test of a laser analyzer for monitoring emissions from the top of
a prilling tower is proposed for a location in India during 2013.
A potential process monitoring application exists in nitric acid manufacture associated with ammonium
nitrate production. The laser technique can measure NH3, nitric oxide (NO) and nitrogen dioxide (NO2)
for improved process control and nitric acid production efficiency.

9
 4.2 HF monitoring in phosphate plants
The primary environmental effect of airborne fluorides from phosphate plants is that they accumulate
and cause damage in vegetation. A secondary effect is the impact of fluorides thus accumulated in
vegetation on herbivores (livestock and wildlife). In order to limit these environmental effects, some
jurisdictions have established ambient air quality standards for HF, which are typically in the range 1
to 4 parts per billion (ppb).
The main source of fluoride emissions from phosphate fertiliser facilities is the phosphogypsum formed
as a by-product of processing phosphate ore into fertiliser using sulfuric acid. Fluoride emissions from
phosphogypsum stacks, piles and associated ponds are comprised of HF and silicon tetrafluoride
(SiF4). Several environmental studies have been conducted to assess the ambient concentrations of
HF near phosphogypsum stacks. Although ambient concentrations in excess of typical ambient
standards are often detected, phosphate fertiliser plants tend to be located in jurisdictions without
such standards and so permanent ambient HF monitoring is uncommon. Nevertheless,
environmentally conscious phosphate fertiliser companies have used portable laser gas monitors such
as described in this paper to monitor levels of HF downwind of phosphogypsum stacks.
One example of permanent laser LOS HF monitoring at a phosphate facility is described by Ball and
Bauer, 1999. In this case, an existing phosphate fertiliser plant was required by local regulators to
install fence-line HF monitoring in order to comply with a plant expansion permit. The plant undertook
a series of open-path Fourier Transform Infrared (FTIR) measurements in order to speciate the
fluoride emissions from a phosphogypsum stack located in Wyoming, USA. This effort determined the
relative percentages of HF and SiF4 present in the total fluoride emissions from this area source.
Examples of data collected over the phosphogypsum and process water storage areas are given in
Figure 7.

Figure 7: HF emissions from Phosphogypsum pile. FTIR data in

black, TDL data in red.

FTIR HF data FTIR SiF4 data Laser HF data

160.0

140.0

120.0

100.0
ug F/m3

80.0

60.0

40.0

20.0

0.0
08 :30
09 :36
10 :52
10 :04
11 :10
12 :18
12 :24
13 :32
14 :42
14 :56
17 :38
17 :12
18 :36
19 :00
19 :30
20 :10
21 :44
21 :24
22 :10
23 :50
23 :32
00 :22
01 :14
01 :06
02 :00
03 :58
03 :58
04 :58
05 :58
05 :02
06 :06
06
0:
8
1
3
4
4
4
4
4
4
4
5
4
4
5
5
6
6
7
8
8
9
0
1
2
3
3
4
5
6
8
9
:0
:5
:3
:1
:5
:3
:1
:5
:3
:1
:5
:0
:4
:2
:0
:4
:2
:0
:4
:2
:0
:5
:3
:1
:5
:3
:1
:5
:3
:1
:5
:4
08

Time

10
 In this case, a permanent fence-line monitoring solution using laser based open path HF detectors was
adopted because the laser technology is much more robust and easier to operate than FTIR, provided
more accurate HF data and because the applicable ambient air quality standard was specific to HF.
This installation has now been operating continuously for over 12 years. Several other studies of HF
emissions from phosphate plants have been performed using portable versions of the GasFinder
technology.

4.3 Monitoring magnitude and fate of NH3 from fertiliser application

Early nitrogen (N) balance studies in various crops showed unexplained losses of applied N with much
of the explained losses being from NH3 volatilization. Varying amounts of estimated NH3 volatilization
from fertiliser were found using chambers or flux-boxes; however, these studies have been shown to
be erroneous since an enclosed environment destroys the emission processes. Early non-interference
techniques including flux-gradient and integrated horizontal flux used chemical absorption and/or
point-sensors to evaluate loss from fertiliser application (Harper et al., 1996; Harper and Sharpe,
1995). Apparent losses ranged from 20 to 40% of applied N. These type studies were expensive and
required highly trained personnel to determine accurate concentrations and climate data under field
conditions and convert the measured concentrations into emission rates. New transport technologies
using inverse dispersion analysis were developed (Flesch et al., 2004) allowing more rapid field
studies but the application of these techniques was initially restricted by the limitations of point
sensors. With the advent of accurate, portable, and reliable open-path laser spectrometers,
determination of emission rates from areal sources (e.g. fertiliser applications) became more accurate
since the concentration of an entire plume could be measured. The combination technique of open-
path lasers (Figure 8) and the inverse dispersion analysis technology (Figure 9) has been used
worldwide and tests for accuracy of the technique have been shown to be 98 5% (Harper et al.,
2010a).

Fig. 8. Instrumentation required for inverse dispersion analysis

and open-path laser determination of ammonia emissions from
an area source.

11
 Figure 9: NH3 & CH4 emissions from farms

Open%path)TDL)and)bLS)(simultaneous)measurements).! ! NH3$and$CH4$emissions$from$a$
large$ ca4le$ feeding$ opera6on$ (0.8$ x$ 1.5$ km).$ Note$ that$ the$ traceBgas$ emissions$ responded$ to$
dierent$environmental$variables$[ammonia$responded$to$clima6c$turbulence$and$temperature$
and$methane$(enteric$emissions)$responded$to$animal$feeding$periods].$$

This same technique used for field losses of NH3 has been used successfully for evaluating emissions
from water sources, large-scale animal feedlots, animal housing, field application of N fertiliser, and
grazing animals (Harper et al. 2010a, 2010b, 2011; Todd, et al. 2011; McGinn et al., 2009). Further
use of the technique has been shown to be successful in evaluating NH3 and CH4 from energy
production (Harper et al., 2010a), and landfills. Use of open-path laser instruments for measuring
representative NH3 concentration enhances the accuracy of the evaluation of emissions from sources
since a large proportion of an emanated plume concentration can be evaluated.

5. CONCLUSIONS
Laser line-of-sight (LOS) gas detectors have advantages over established gas detection techniques for
leak detection. They do not suffer from interferences and hence do not generate false alarms. They
provide fast response and can measure a wide range of concentration values. They provide reliable
operation in all climate conditions. Multiple measurement points can be coupled to a single gas
analyzer using optical fibres. This leads to lower installation cost per measurement point. Absence of
moving parts and consumables reduces long-term cost of ownership. As a result of these benefits,
laser gas detectors are replacing traditional gas analyzers and detectors in a growing number of
applications across a broad spectrum of industries. This is especially the case for gases, which have
historically been difficult to detect using traditional techniques such as HF and NH3.
In recent years, laser LOS gas detectors have displaced traditional technologies for safety and
environmental monitoring in several agricultural and fertiliser industry applications, most notably in
ammonia production and transportation. The benefits of the laser technology have been
demonstrated. Acceptance and use of laser gas detection in the fertiliser industry will continue to grow
in the future.

12
 REFERENCES
1. Ballangrud, A.M. (1993). Diode laser spectroscopy for gas monitoring of environmental pollution and for
industrial process and emission control. Dr. Scient. Thesis, University of Oslo, June 1993.
2. Silver, J.A. (1992). Frequency-modulation spectroscopy for trace species detection: theory and comparison
among experimental methods. Applied Optics, Vol 31, No. 6, pp 707-717.
3. Bomse, D. (1995). Diode Lasers: Finding Trace Gases in the Lab and the Plant. Photonics Spectra, 29(6):88.
4. Tulip, J. (1997). Gas Detector. U.S. Patent No. 5,637,872.
5. Bomse, D., Stanton, A.C. and Silver, J.A. (1992). Frequency Modulation and Wavelength Modulation
Spectroscopies: Comparison of Experimental Methods Using a Lead Salt Diode Laser. Appl. Opt. 31:718.
6. Tulip, J. (1998). Gas Detector for Plural Target Zones. U.S. Patent No. 5,748,325.
7. Tulip, J. (2000). Gas Detector with Reference Cell. U.S. Patent No. 6,121,627.
8. Adam, H., Parker, C., Bauer, J., Tulip, J. (2008). Portable Laser Gas Detector Systems for Landfill CH4 and
CO2 Monitoring. AWMA Annual Meeting, Portland, OR, June 2008.
9. Ball, M., Howe, D., Bauer, J. (1999). Measurement and Speciation of Fluoride Emissions at a Phosphate
Fertiliser Manufacturing Plant Using Open-Path FTIR and TDL, AWMA Annual Meeting, June 1999
10. Flesch, T.K., Wilson, J.D., Harper, L., Crenna, B. and Sharpe, R. (2004). Deducing ground-air emissions from
observed trace-gas concentrations: A field trial. Am. Meteorol. Soc.43:487-502.
11. Harper, L. A., Bussink, D. W., van der Meer, H. G., and Corre, W. J. (1996). Ammonia transport in a
temperate grassland: I. Seasonal transport in relation to soil fertility and crop management. Agron. J.
88:614-621.
12. Harper, L.A., Denmead, O. and Flesch, T. (2011). Micrometeorological techniques for measurement of
greenhouse gas emissions from ruminant animal production. Anim. Feed Sci. Tech. 166167:227 239.
13. Harper, L.A., Flesch, T., Weaver, K. and Wilson, J. (2010). The effect of biofuel production on swine farm
ammonia and methane emissions. J. Environ. Qual. 39:1984-1992.
14. Harper, L.A., Flesch, T. and Wilson., J. (2010). Ammonia emissions from broiler production in the San
Joaquin Valley. J. Poul. Sci. 89 :18021814.
15. Harper, L.A. and Sharpe, R. R. (1995). Nitrogen dynamics in irrigated corn: Soil-plant nitrogen and
atmospheric ammonia transport. Agron. J. 87:669-675.
16. McGinn, S.M., Beauchemin, K.A., Flesch, T.K., Coates, T. (2009). Performance of a dispersion model to
estimate methane loss from cattle in pens. J. Environ. Qual. 38, 17961802.
17. Todd, R.W., Cole, N., Rhoades, M, Parker, D. and Casey, K. (2011). Daily, monthly, seasonal and annual
ammonia emissions from southern high plains cattle feedyards. J. Environ. Qual. 40:16.

13