PRESENTED AT: ISA Analysis Division, Houston, TX

Keywords

ULTRAVIOLET ABSORPTION SPECTROSCOPY, APPLICATIONS, SULFUR RECOVERY, SCOT CONTACTOR, NATURAL GAS, HYDROGEN SULFIDE

Abstract

Ultraviolet absorption spectroscopy has been used for decades in the refining and petrochemical industry. Online spectroscopic methods allow for rapid compositional analysis of process streams and provide the capability for real-time monitoring and control. Common applications have been the measurement of SO₂ and H₂S in sulfur recovery unit tail gas, H₂S in acid gas feed, continuous emissions monitoring for SO₂, and measurement of chlorine in numerous petrochemical processes. Improvements to the technology, both in terms of the spectrometers themselves and the sample systems, have made it possible to use spectrometers in new applications or to increase the measuring capabilities of existing spectrometers in conventional applications. Specific examples are reviewed, including the measurement of COS and CS₂ in sulfur plant tail gas, simultaneous determination of NH₃ and NO_x in catalytic cracking units, analysis of sour water stripper gases for ammonia and H₂S, as well as the simultaneous determination of H₂S, COS and Methyl Mercaptan in synthesis gas, amine contactor overhead and sales gas streams. The use of spectrometers in these applications have enabled real-time control of processes previously monitored by batch sampling, thereby reducing emissions and improving plant operations through better control.

Introduction

Ultraviolet and visible absorption spectrometers have been used for decades in process analysis and emissions monitoring. The UV method has shown great benefit from the fact that common stream components, like nitrogen, oxygen, carbon dioxide and water vapor have no characteristic absorption in the near UV region. As a result, UV absorption analyzers have no interference from these species. Historically, these analyzers have predominately been used for the measurement of a single species in a process stream.

As time has progressed, these analyzers have become increasingly advanced, incorporating improved optics, advanced micro-controllers and multi-component analysis capabilities. As a result, there are an increasing number of applications that may be addressed by such devices. The benefits provided by UV spectroscopic analyzers in a number of applications is presented herein.

Spectrometer Design

In all of the applications discussed in this paper, the spectrometer used was of a dual-beam multi-wavelength design, as shown schematically in Figure 1. The light sources used in this device are hollow cathode lamps, chosen specifically for their narrow emission lines at specific and reproducible wavelengths. The measure and reference detectors are photomultiplier tubes (PMT), chosen specifically for their excellent photometric accuracy and linearity [1]. A six-filter chopper wheel allows for multiple measuring wavelengths and for the simultaneous analysis of up to five species.

The dual-beam design was chosen to provide exceptional baseline stability. In this device, a separate reference and measure path is maintained for each measurement wavelength employed. The light emitted from the lamp is split into two separate beams through the use of a beam splitter. One beam passes through ‘clean’ air and impinges on the reference photomultiplier tube (PMT) while the other beam passes through the sample and impinges on the measure photomultiplier tube. Data are acquired from the two PMT’s simultaneously. Any drift in the light source is seen on both the reference and measure path. Taking the ratio of the signal strength as observed on reference and measure paths eliminates the effects of light source drift. This methodology allows for common-mode rejection of both short-term noise and long-term drift in the light sources, resulting in exceptional baseline stability. It is this stability which prompted Henderson to write [2] “Accurate measurement of the absorption coefficient at different wavelengths is best made with a double-beam spectrometer”.

Hollow cathode lamps were chosen as the light sources for the analyzers for a number of reasons. They emit characteristic emission lines that are determined solely by the materials used for construction of the cathode. The center wavelengths of the emission lines are known for high accuracy and precision, and due to the nature of the discharge in these lamps, the center wavelength does not drift due to temperature, pressure or any other external influence. The emission band is a fundamental property of the material. For example, the magnesium emission line which is used for the measurement of SO₂ occurs at precisely 285.214 nanometers, and is 0.016 nanometers wide at half maximum. The bands are also extremely narrow, which contributes to the linearity of the absorbance measurement. These narrow precise and reproducible emission lines are instrumental in achieving high resolution, wavelength accuracy and wavelength precision in the spectrometer, independent of variations in the filter bandpass.

FIGURE 1 – M900 SERIES SPECTROMETER

The filter wheel can hold up to six bandpass filters, thus allowing for the measurement of up to five species while using one filter for baseline stability corrections. The dual-beam nature of the spectrometer results in an automatic correction of the signals for variations in the light source, while the use of an additional wavelength provides baseline correction for variations in optical transmission of the cell, detector gain or amplifier drift. The net result is that each of the five measured species is corrected for any variations which may have occurred in the photometer itself. Even variations in the wavelength selective filters are removed by the dual-beam nature of the photometer.

For additional information on the design of UV spectrometers, and factors that influence their performance, the reader is directed to papers presented at previous ISA Meetings [3.4]. The spectrometer described herein has been used in a range of applications, some of which are now presented.

Applications

During the past half-century, photometric analyzers based on ultraviolet/visible (UV/Vis) absorption have found increasingly wide application throughout the process industries. UV/Vis photometric analyzers were among the first plant stream analyzers used for direct monitoring of process stream composition and closed-loop control [5]. Since the implementation of the Clean Air Act amendments, UV analyzers have also seen extensive use in continuous emissions monitoring. The largest application has likely been the online monitoring of sulfur dioxide emissions, but other large applications include the monitoring of chlorine, chlorine dioxide, mercury and aromatic hydrocarbons.

Many of these applications have been addressed in the technical literature, or in manufacturers’ application datasheets. However, there still exist many potential applications where other analytical techniques have been the method of choice, or for which there were no readily available monitoring technologies. Some examples of these applications, which are discussed in further detail in this paper, are:

• The online measurement of COS and CS₂ in sulfur plant tail gas,
• Sulfuric Acid Process Control,
• Emissions Monitoring in Nitric Acid Plants,
• Emissions Monitoring in Refinery NOx applications,
• Sales gas analysis for reduced sulfur.

SULFUR PLANT TAIL GAS

Ultraviolet absorption spectrometers have seen extensive use in the sulfur recovery industry since the 1960’s. The analysis of H₂S and SO₂ in sulfur plant tail gas had originally been performed by gas chromatographs, although an elaborate sample handling system was needed to deal with the high levels of moisture and entrained elemental sulfur. Since the 1970s, this has been one of the largest applications of UV process photometers [6].

In the modified Claus sulfur recovery process, a hydrogen sulfide rich stream is partially oxidized, to produce an effluent stream containing large quantities of H₂S and SO₂ in a nitrogen/CO₂ background. The stream is also laden with sulfur and water vapor, and contains carbonyl sulfide (COS) and carbon disulfide (CS₂) as byproducts of the combustion/quench process. The H₂S and SO₂ laden stream is allowed to flow over an appropriate catalyst, where these species react to form elemental sulfur which is removed by condensation. The reaction stoichiometry dictates that the H₂S concentration should be maintained at twice that of the SO₂. The overall sulfur recovery efficiency is to a large part determined by how well this ratio is maintained, as well as by efficiency losses due to entrained sulfur vapor and to the other reduced sulfur species (COS and CS₂).

While UV analyzers have long been used to monitor the H₂S and SO₂ concentrations in sulfur plant tail gas, it is only recently that these analyzers have been used for the simultaneous analysis of COS and CS₂. The COS and CS₂ produced in the Sulfur Recovery Unit reaction furnace, should in theory be hydrolyzed to H2S on the first catalyst bed (assuming that this bed is operated at appropriate temperatures). If this hydrolysis reaction proceeds to completion, all of the sulfur in the stream becomes available to participate in the Claus reaction and sulfur recovery efficiency is maximized (which also implies that sulfur emissions are reduced). By monitoring the COS and CS₂ in the sulfur plant tail gas, the activity of the first catalyst bed can be evaluated. As the catalyst bed becomes de-activated, the concentrations of COS and CS2 will increase. This can be especially important in plants that have low concentrations of H2S and high hydrocarbon contents in the acid gas feed. Such plants frequently bypass some of the acid gas around the reaction furnace, which can lead to “sooting” and rapid deactivation of the first bed.

One of the major factors that can affect the ability of a tail gas analyzer to measure COS and/or CS₂ is cross-interference of sulfur, sulfur dioxide and/or hydrogen sulfide onto these species. It becomes imperative then to test that this cross-interference testing be performed during production testing of an analyzer. Typical gas run results of a COS/CS2 equipped tail gas analyzer are presented in Table 1. The analyzer is ranged for 0 to 7500 ppm SO₂, 0 to 15,000 ppm H2S, and 0 to 5000 ppm for both COS and CS₂.

SULFURIC ACID PROCESS CONTROL

Sulfuric acid plants convert relatively high concentrations of sulfur dioxide to sulfuric acid via a catalytic reaction between the sulfur dioxide and oxygen to form sulfur trioxide, and subsequent absorption of the sulfur trioxide in a dilute sulfuric acid contactor.

Sulfur dioxide may be produced in a number of ways, including

• Combustion of elemental sulfur
• Sulfuric acid regeneration
• Roasting of Sulfur Bearing Ores

Typically, the sulfur dioxide concentration at the entrance to the catalytic stage is between 4 and 12 mole %. The process is operated with excess oxygen, and the catalyst bed promotes the formation of sulfur trioxide from sulfur dioxide and oxygen. Optimum overall system behavior requires a balance between the reaction velocity and equilibrium, but this optimum depends also on the SO₂ concentration in the raw gas and on its variability with time. Thus each process is more or less specific to a source of SO₂.

TABLE 1 – TESTING RESULTS OF H2S, SO2, COS, CS2 TAIL GAS ANALYZER

Sulfuric acid (H₂SO₄) is obtained from the absorption of SO₃ into sulfuric acid with a concentration of at least 98%, followed by the adjustment of the strength by the controlled addition of water.

A particularly interesting application for a hot-wet UV analyzer arose at a Spent Acid Regeneration Facility. The application was unique in that the purpose for monitoring is two-fold. The primary reason is process-based. The process may be optimized to maximize the production of SO₃ if the SO₂/oxygen ratio is known. Also, the facility involved also incinerates hazardous waste in the furnace as an alternative fuel source to natural gas. The Resource Recovery and Conservation Act (RCRA) permit requires that the gas going into the converter be at least 3% SO₂. If the converter inlet concentration falls below 3% SO₂ for more than 15 minutes, the hazardous waste feeds to the furnace are automatically shut off. The hot wet UV analyzer replaced an older dry-basis infrared device, and has performed very reliably in this application, dramatically reducing the need for sample system maintenance.

In addition to monitoring the SO₂ concentration, the analyzer is capable of measuring NO_x (as NO₂). This capability can be extremely important in some facilities. The high temperatures in the combustion furnace can lead to NO_x formation, primarily as nitric oxide or NO. However, the presence of large quantities of excess air and the porous catalyst promotes the formation of nitrogen dioxide (NO₂). The NO₂ may be absorbed into the sulfuric acid, resulting in an off-color and off-specification product. The analyzer is capable of measuring both the SO₂ concentration (in the mole percent range) and ppm levels of NO₂.

EMISSIONS MONITORING IN NITRIC ACID PLANTS,

UV analyzers are capable of measuring both ammonia and NOx species simultaneously. This has unique advantages in a nitric acid plant. In such a facility, ammonia is reacted with air on platinum/rhodium alloy catalysts in the oxidation section. The reaction produces nitric oxide and water vapor via:

4NH₃ + 5O₂ -> 4NO + 6H₂O,

although side reactions also occur, such as:

4NH₃ + 3O₂ -> 2N₂ + 6H₂O

4NH₃ + 4O₂ -> 2N₂O + 6H₂O

The exothermicity of the reaction may be used to preheat reaction gases and/or to make steam. The heated waste gas is discharged to the atmosphere through a gas turbine for energy recovery.

The process is run oxygen rich, such that nitric oxide is oxidized to nitrogen dioxide. The gas stream is contacted with water, where the NO2 is absorbed with the subsequent release of additional quantities of nitric oxide:

3NO₂ + H₂ O -> 2HNO₃ + NO

In an effort to minimize emissions, the effluent gas stream from the acid contactor is reacted with additional ammonia in a selective non-catalytic reduction (SNCR) system:

6NO + 4NH₃ -> 5N₂ + 6H₂O.

There is typically some bypass or slip of ammonia through the catalytic reactor. This results in ammonia being present in the tail gas, where it can react with the acid gases. The resultant salt formation as the gases cool can be problematic for many analyzers, since the salt deposits in the sample system. Furthermore, conventional chemiluminescent NO_x analyzers require a catalytic converter to measure NO₂, since it must first be reduced to NO. Since a larger percentage of the NO_x present in a nitric acid plant is NO₂, loss of converter efficiency can lead to severe errors in the measurement of NO_x by the chemiluminescent method.

This combination of factors, combined with the ability of UV analyzers to simultaneously measure ammonia, has resulted in a number of nitric acid plants implementing UV ammonia/NO_x process and emission monitoring analyzers.

One such unit was installed at Agrium's Kennewick Fertilizer Operations in late '99 (then Prodica). The meter is used as part of the implementation of a Reasonably Available Control Technology (RACT) requirement. The unit is used to monitor two acid plant stacks with NO_x emissions that might be in the 300 ppm range. The UV style meter was selected principally because of the expected low drift and low maintenance requirements. The application was on two atmospheric pressure stacks each downstream of a selective catalytic reactor (SCR) unit. Stack emissions were expected to be in the range of 300 ppm NOx (75% to 97% NO) with less than 30 ppm NH₃ slip. The combination of low levels of NH₃ and NO_x in the stack was expected to cause ammonium nitrate salt deposition if the sample lines were cooled below 100 degrees Fahrenheit. The operating temperature of the UV analyzer (300°F) assured no ammonium nitrate salt formation with no associated calibration drift.

The results of this installation have been:

• typical meter drift over two months with no correction for NO is 1 to 2 ppm in 800 ppm span gas applications and atmospheric air blank applications.
• typical meter drift over two months with no correction for NO₂ is 3 to 7 ppm in atmospheric air blank applications.
• typical meter drift over two months with no correction for _O2 analysis at 2.5% is less than 0.05 %.

The analyzer has provided quick stack monitoring results. One analyzer can monitor two stacks with 99% of the full change seen after sample switching between stacks in 56 seconds. Response time to changes in stack analysis is comparable.

After 13 months of service, the meter has had a single maintenance event - one bulb had to be replaced. The meter is automatically checked against a calibration gas of NO_x once per stack per day. The meter calibration factor is only changed if the meter drifts 10 ppm for either a 800 ppm span gas or a blank atmospheric gas sample. This typically happens about once per two months.

In a similar application at another facility, a single analyzer is used to monitor both inlet and effluent gas from a de-nox reactor. The analyzer was again chosen to replace a chemiluminescence-based NO_x analyzer. A driving factor for this decision was the fact that it was a hot wet extractive analyzer, with virtually no sample conditioning system, converters or pumps. In the experience of this facility, 90% of problems with analyzers have been with the sample systems.

Online performance has been very good. On average the facility runs three weeks between manual calibrations with a zero-drift of about 8 ppm and a span drift of about 4 ppm. Most of the downtime is for cal gas change out and lamp replacement. Over four years, the facility has about 10 hours downtime per year. They have passed all of their NO_x Relative Accuracy Test Audits (RATA) tests, with the most recent relative accuracy results being better than 5% of the mean value for the test.

NOx MEASURMENT IN POWER BOILERS

The measurement of NO_x emissions levels is of primary importance due to the adverse effects that result from elevated concentrations of these species in ambient air. These effects include acidic deposition on natural resources, increased levels of photochemical smog with the associated impacts on public health, and reduced yields of crop plants. Currently, electrical power generation, particularly from coal-fired boilers, accounts for approximately 40% of the total NO_x emissions in North America. The majority of boilers use coal or oil as a fuel source and thus produce emissions of both SO₂ and NO_x.

These facilities are required to perform continuous on-line monitoring of the emissions levels for SO₂ and NO_x, and have specific performance requirements. New regulations have increased the number of facilities affected. All such facilities require that the accuracy of the monitoring system are periodically proven during a Relative Accuracy Test Audit (RATA).

Typical Relative Accuracy Test Audit results for two UV NO_x analyzer systems on gas-fired boilers are presented in Table II. Both analyzers pass the Relative Accuracy Test Audit with better than 2% accuracy, far better than the required 20% accuracy relative to the reference method. The same analyzers were configured for simultaneous SO₂ analysis in the event that the boilers are fired with oil. Given that the reference method showed 0 ppm SO₂ during the test (as did the CEMS), no relative accuracy numbers could be given.

TABLE 2 – NO_x RATA TEST RESULTS – REFINERY POWER BOILER

SALES GAS ANALYSIS FOR REDUCED SULFURS.

Transportation of natural gas from the wellhead to the customer normally involves a series of contracts between producers, brokers, transporters, distributors and end-users. H₂S concentration is just one of the many important specifications required in such contracts. De-regulation of the gas industry has intensified pressure on producers and transporters to provide on-spec gas. As a result, one of the most common applications for on-line H₂S analyzers is monitoring of gas quality at sour gas metering stations. Sour metering stations are designed for continuous H₂S detection, block valve control and sour gas return.

Detection of an off-specification H₂S concentration calls for immediate block valve closure. Metering stations are designed with sufficient downstream bottle capacity to prevent off-specification sour gas from entering mainlines or sales taps. The capacity of the bottle is designed in direct relationship to the response time of the H₂S analyzer system. Using a fast response H₂S analyzer can reduce the requirement for bottle pipeline capacity. Also reduced is the return meter run size that is directly related to the volume of gas that is to be returned to the gas plant.

While UV analyzers have been providing rapid H₂S response in pipeline gas for almost two decades [7], recent advances have extended the capabilities of these devices. Most notably, additional species can be measured, online time has been increased, and sensitivity has been improved. Current devices are capable of measuring both carbonyl sulfide (COS) and methyl mercaptan as well as the hydrogen sulfide. Measurement of the former species may be important in that it can hydrolyze to H₂S, while the mercaptan measurement is, in some cases, indicative of the total sulfur content of the gas. Full-scale ranges as low as 5 ppm can now be accommodated on UV H₂S analyzers, and the analyzer response speed is less than 30 seconds.

FIGURE 2 – UV AND TAPE ANALYZER RESULTS – TYPICAL OPERATION

Testing of the UV analyzer was performed at a gas plant in Northern Canada. Data presented in Figure 2 shows the response of the ultraviolet absorption-based analyzer and that of a lead acetate tape analyzer. One notes that there is reasonable agreement between the two analytical methods, although the tape analyzer overshoots on the initial response to calibration gas. This is most likely an artifact of the predictive algorithm used to speed response times. Subsequently, the plant purposefully upset the process by adjusting the amine re-circulation rate. In this case, the increase in H₂S concentration is more gradual, and the two analyzers track well with each other.

Data presented in Figure 3 depicts the results obtained during an actual process upset. The UV analyzer is capable of measuring the carbonyl sulfide (COS) and methyl mercaptan (MeSH) as well as the hydrogen sulfide concentrations (H₂S). H₂S results from the tape analyzer are presented for comparative purposes. One notes that the COS started to increase before the H₂S process upset occurred. This situation can arise when there is insufficient contact between the amine and the process gas. The final concentration of COS in the produced gas is often determined by the adsorption rate and not by the equilibrium composition. As a result, changes in contact time or contact efficiency (as may be caused by foaming) will likely affect the COS composition before any effect on H₂S composition is seen.

FIGURE 3 – UV AND TAPE ANALYZER RESULTS – PROCESS USET

Conclusions

Several applications of UV spectroscopy have been presented. In each, the UV method has been used either in
place of an existing analytical technique or an improvement to the standard UV method has been described.

In sulfur plant tail gas, new UV analyzers have allowed for the online measurement of carbonyl sulfide and
carbon disulfide, allowing for the continuous monitoring of catalyst bed performance.

In sulfuric acid plants, the use of hot wet extractive analyzers has reduced maintenance requirements for CEM
and process monitoring equipment. In addition, the new analyzers allow for multi-range SO₂ analysis and NO_x
analysis in such plants.

The UV method may also be used for simultaneous measurement of ammonia and NO_x in nitric acid plants.
Again, the use of a hot wet extractive system has greatly reduced sample system bias and lowered plant
maintenance requirements.

Dry extractive analyzers have also been used for simultaneous SO₂ and NO_x measurement, and are seeing
extensive use in the petrochemical and refining industry.

The UV method has also proven very beneficial in the analysis of hydrogen sulfide, carbonyl sulfide and methyl
mercaptan in natural gas streams. This measurement provides faster response speeds and may give advance
warning of process upsets.

References

1. Bennet, H. E., “Accurate methods for determining photometric linearity”, Applied Optics, 5:1265, 1966.
2. Henderson, Brian, “Optical Spectrometers”, Handbook of Optics II, edited by M. Bass et al., McGraw Hill, New York, , 1995
3. Adam H. and Harris P., “The design and practical application of UV process photometers”, L’analyse Industriel, Paris, France, 1996.
4. Adam H. and Harris P., “The design and practical application of UV process photometers Part II Otpical Components and Configuration”, ISA 96, Chicago, Oct 1996
5. Saltsman, R. S., “Ultraviolet/Visible (UV/VIS) Absorption Analyzers”, Analytical Instrumentation, edited by R. E Sherman, ISA Press, Research Triangle Park, NC, 1996.
6. Saltzman, R.S., and Hunt E. B. Jr, Analysis Instrumentation, 10:55, 1972.
7. Babiuk, Ken F., “Advances in fast-response H2S Analysis”, 1982.