December 2013, Vol. 240 No. 12


Natural Gas Odorization Monitoring For Safety And Consistency

David Amirbekyan, Regional Application Engineer, Applied Analytics and Nick Stylianos Field Service Engineer, Applied Analytics

Editor’s note: This is the first of two parts on measuring odorants online in natural gas pipelines. It concentrates on the background and history of odorants, and identifies a need for analytics beyond the minimum standards set forth by Code of Federal Regulations (CFR), Title 49, Part 192.625.The second part will focus on the application of UV-VIS absorption spectroscopy for odorant detection.

In the fracking era, natural gas is playing an increasingly important role in domestic energy markets. The continuing surge in natural gas production from the boon of shale reserves will come hand-in-hand with a greater investment in gas logistics, including distribution networks, process efficiency and environmental regulation. One of the most important current areas of improvement is the odorization of natural gas, a critical safety measure, which allows humans to detect gas leaks by smell.

Odorization is a highly advanced field, with local distribution companies (LDCs) and downstream operations leading in expertise. Upstream operators are not typically required to odorize, but safety regulations increase along the logistics chain toward consumers. Depending on the quality standards they face, different natural gas companies use different odorization control technologies, especially in the realm of odorant measurement methodology.

Since the objective of odorization is human detection, U.S. law requires validation procedures performed by human sniffers; due to the discrete and subjective nature of this method, sniffer data is increasingly being supplemented with continuous odorant measurement technology. These analyzers provide objective odorant trend data and differentiate between different odorizing agents.

One technology that seems poised to become an industrial standard for odorant monitoring is the diode array UV-Vis (ultraviolet-visible) spectrophotometer, an optical device that produces extremely reliable results and is affordable to implement. This article reviews the difficulties of effective natural gas odorization, and how they can be mitigated by this monitoring technology.

Brief History Of Odorization
The first documented instance of gas odorization brings us back to Germany in 1880, when a scientist named Von Quaglio added ethyl mercaptan to give water gas a distinguishable smell. The danger of carbon monoxide poisoning from town gas was an impetus for other early odorization enterprises. However, many at the time depended on the odor of naturally occurring sulfur compounds to detect gas leaks – a highly unsafe practice, as gases varied widely in sulfur concentration and were sometimes odorless.

Germany began small-scale odorization by 1918, with the U.S. soon following suit. Initially, the odorant chemicals suitable for this use were limited. During the rise of automobiles and the advent of World War II, when the synthetic chemical industry flourished, many new odorant chemicals and odorizing technologies were developed.

The effort to standardize and regulate odorization was galvanized by the 1937 disaster in New London, TX, where the high school was using odorless natural gas for heating and cooking. When an undetected leak was ignited by a spark from a sanding machine, the resulting explosion killed 239 people and the tragedy spurred enormous change in the industry.

In 1951, odorization guidelines appeared in the text of the American Society of Mechanical Engineers (ASME) B.31.8 “Standard for Natural Gas Transmission and Distribution Piping Systems.” This document paved the way for the regulations enacted in the 1970s CFR Title 49 Part 192.625, which continues today to define the rules of the U.S. natural gas pipeline system. Many countries have adopted these rules or similar versions, including Europe’s G280 (DVGW).

Current Regulations
The following is the CFR Title 49 Part 192.625 (italics) with additional comments:

(a) A combustible gas in a distribution line must contain a natural odorant or be odorized so that at a concentration in air of one-fifth of the lower explosive limit, the gas is readily detectable by a person with a normal sense of smell.

Figure 2 depicts section (a) graphically.
Figure 2: Odor detection threshold for sniff tests

(b) After Dec. 31, 1976, a combustible gas in a transmission line in a Class 3 or Class 4 location must comply with the requirements of paragraph (a) of this section unless:

(1) At least 50% of the length of the line downstream from that location is in a Class 1 or Class 2 location;

(2) The line transports gas to any of the following facilities, which received gas without an odorant from that line before May 5, 1975;

(i) An underground storage field

(ii) A gas processing plant

(iii) A gas dehydration plant

(iv) An industrial plant using gas in a process where the presence of an odorant:

(A) Makes the end product unfit for the purpose for which it is intended

(B) Reduces the activity of a catalyst

(C) Reduces the percentage completion of a chemical reaction

(3) In the case of a lateral line, which transports gas to a distribution center, at least 50% of the length of that line is in a Class 1 or Class 2 location

(4) The combustible gas is hydrogen intended for use as a feedstock in a manufacturing process.

Section (b) lists exceptions to odorization regulations. An important example is an industrial plant that requires a feedstock of unodorized natural gas in order to produce the desired product. Due to pipeline classification, these sites sometimes only have access to odorized natural gas, and must install an absorption bed to remove the present odorant before use.

The pipeline classes are summarized as:

Class 1: Any location within 220 yards of the pipeline that contains 10 or fewer dwellings.

Class 2: Any location within 220 yards of the pipeline that contains more than 10, but fewer than 46 dwellings.

Class 3: Any location within 220 yards of the pipeline that contains 46 or more dwellings.

Class 4: Any location within 220 yards of the pipeline with multistory dwellings or dense populations.

(c) In the concentrations in which it is used, the odorant in combustible gases must comply with the following:

(1) The odorant may not be deleterious to persons, materials or pipe.

(2) The products of combustion from the odorant may not be toxic when breathed nor may they be corrosive or harmful to those materials to which the products of combustion will be exposed.

(d) The odorant may not be soluble in water to an extent greater than 2.5 parts to 100 parts by weight.

After decades of testing by various private and government agencies, a group of sulfur compounds known as mercaptans (or thiols) were proved superior to other odorizing agents in terms of producing an immediate and distinct gas smell without posing health risks, damaging equipment or sacrificing desirable properties of the natural gas.

(e) Equipment for odorization must introduce the odorant without wide variations in the level of odorant.

(f) To assure the proper concentration of odorant in accordance with this section, each operator must conduct periodic sampling of combustible gases using an instrument capable of determining the percentage of gas in air at which the odor becomes readily detectable. Operators of master-meter systems may comply with this requirement by:

(1) Receiving written verification from their gas source that the gas has the proper concentration of odorant; and

(2) Conducting periodic ‘‘sniff’’ tests at the extremities of the system to confirm that the gas contains odorant.

The following section describes how odorant is added to the natural gas system in accordance with this code.

Natural Gas Distribution And Odorization
Natural gas is extracted from different types of wellheads and shuttled through gathering lines to a gas processing station. The station removes impurities and brings the natural gas up to national and contractual standards before selling the processed gas to a transmission company, which compresses the gas in order to deliver it along its transmission lines to various customers.

The transmission lines break up into smaller distribution lines (also called “mains”), which are located at city gates. Here, the ownership of the high-pressure gas stream is transferred from the transmission company to the LDC, which reduces the pressure of the stream and supplies the gas to residential and commercial customers.

Odorization is always required in distribution lines, and is often required in transmission lines depending on location. The transmission company must implement odorization where its system feeds into Class 3 or Class 4 piping; also, before transferring gas to the LDC at the city gate, the odorant level must be measured and “topped off” if necessary.
Figure 3: Natural gas supply chain schematic
How Odorant Is Traditionally Measured
Pipeline companies check for the presence of odorant at a given point by using the human sniff test. Service technicians are dispatched over long distances to sniff test the odorant level and log leak calls, while statistical models are created from the sniff data to understand the flow of odorant through the distribution network.

Pipeline companies check for the presence of odorants by gathering sniff test data. Sniff tests are performed at the beginning, middle and ends of their pipeline system: Service technicians check for the presence of odorant at every service visit or log leak call, and companies develop models to sample a statistically significant fraction of their distribution system. The sniff test is performed in accordance with CFR Title 49 Part 192.625, whereby the natural gas must be detectable by a person with a “normal” sense of smell at a concentration in air equal to one fifth of the minimal concentration at which the gas mixture could ignite (known as the lower explosive limit (LEL).

Some sniffers test for the absolute minimum detection threshold, while others test for the readily detectable value. All sniffers make use of a special machine that reduces the sampled gas to a lower pressure and rotates an air mixing dial to get the desired 1/5 LEL concentration. As already mentioned, U.S. law stipulates that these tests must be performed by humans, despite the subjective nature of human judgment and the unhealthy aspects of inhaling potentially impure natural gas.

Odorization Difficulties

While odorization is an established science with a wealth of operating knowledge, there are many factors that introduce uncertainty and require constant vigilance by the responsible party to ensure odorization compliance. These challenges include:

• Odorant fade: Over long distances of piping, odorant may disappear for a number of reasons.

1. Naturally occurring odors present in the gas may react with the injected odorant and create unanticipated results. An example would be naturally occurring methyl mercaptan oxidizing the injected tert-butyl mercaptan blend, resulting in an under-odorized natural gas supply.

2. New pipelines often absorb odorant and must be first pre-conditioned, or pickled, to reduce odorant fade. This procedure delays new pipeline readiness and does not completely eradicate wall adsorption of the odorant.

3. The presence of air, rust or water can catalyze the oxidation of the odorant.

4. The phenomenon of “odorant masking” occurs when the correct amount of odorant is injected but other chemical species present in the mixture mask the smell.

• Odorizer shutdown or malfunction: the system that injects odorant into a pipeline system fails or drifts away from its prescribed injection rate.

• Odorant blend suitability: numerous types of odorants exist for different climates and stream specifications; there is no “one size fits all” solution.

• Uneven odorant distribution: even when odorant is properly injected at a constant rate, observations indicate that the odorant concentration distribution travels in slugs.

To detect and correct these safety issues in real time, the odorization industry will likely need to cut dependence on human sniff testing because of its labor-intensive nature, non-continuous data and subjective human error.

The industry invests heavily in training employees to detect odor, but error is unavoidable: the ability of a technician to perform the sniff test can be compromised by odorant fatigue (in long-time employees), improper operation of the sniffing device, sickness, and variations in smell perception due to foods eaten, hormones and stresses. Human negligence due to distraction or recording error can also reduce the validity of the method.

[inline: Figure 4: Odorization system (courtesy of Welker)]

On the other hand, the use of automated odorant measurement technology provides continuous monitoring at any number of unattended sampling points, even in remote locations. For transmission companies operating hundreds of miles of high- pressure piping, the use of these analyzers drastically decreases travel expenses and lightens the burden placed on sniffers.

Moreover, using real-time, quantitative analytics that differentiate between the different odor-producing species provides for much greater understanding of the distribution network and faster response to safety risks. The prospect of eliminating unsafe odorization inconsistencies establishes a real need for odorant monitoring in parallel to current sniff testing.

Editor’s note:
Part 2 will introduce the diode array UV-Vis spectrophotometer and proper sample conditioning. The advantages of this technology will be discussed and specific odorant measurement applications will be examined in detail.

• Code of Federal Regulations, Title 49 Part 192.625

• ASME B31.8: Gas Transmission and Distribution Piping Systems

• J.H. Kostro & Associates,

• U.S. Energy Information Administration, Annual Natural Gas Consumption 1949-2014

• GASODOR® S-FREE by Symrise,

• “MPI-Mainz-UV-VIS Spectral Atlas of Gaseous Molecules,” Keller-Rudek, H., Moortgat, G.K., Sander, R., & Sörensen, R., Max-Planck-Institui für Chemie

• “Measuring Odorants in Natural Gas Pipelines,” Applied Analytics Application Note No. AN-011, October 18, 2013

• “Multi-Component Analysis,” Applied Analytics Technical Note No. 203, September 20, 2013

• Usher M. J. (1999). Odor fade –Possible Causes and Remedies, CGA Gas measurement School London, Elf Atochem North America, Inc., Philadelphia

David Amirbekyan
works as regional application engineer at Applied Analytics, where he develops business and key analytical applications worldwide for various industries. Amirbekyan holds a bachelor’s degree in chemical engineering from Rensselaer Polytechnic Institute, Troy, NY.

Nick Stylianos is a field service engineer at Applied Analytics, servicing and commissioning analytical systems worldwide.

Related Articles


{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}