Testing the Composition of Natural Gas Using an Electronic Nose

Conventional electronic noses, or eNoses, produce an identifiable response pattern using a group of dissimilar but not specific chemical sensors. The concept of an eNose has drawn the attention of developers of neural networks and artificial intelligence algorithms in the last few years, yet physical sensors have restricted performance due to physical instability and overlapping responses. eNoses cannot quantify or isolate the chemistry of aromas.

The zNose® is a new type of electronic nose that is based upon ultra-fast gas chromatography. It simulates an almost infinite number of specific virtual chemical sensors, and generates olfactory images entirely based upon aroma chemistry.

The zNose® performs analytical measurements of volatile organic odors and vapors in near real time with part per-trillion-sensitivity, accuracy and precision. Satisfying the requirement for separation, speed and quantification of each chemical within an odor is done in fractions of seconds. Using a patented solid-state mass-sensitive detector, universal nonpolar selectivity, picogram sensitivity and electronically variable sensitivity are accomplished.

An integrated vapor preconcentrator combined with the electronically variable detector, allow the device to measure vapor concentrations spanning 6+ orders of magnitude. A portable zNose®, illustrated in Figure 1, is a beneficial tool for evaluating the quality of aromatic products and monitoring a wide range of biological and chemical processes.

Portable zNose® technology incorporated into a handheld instrument

Figure 1. Portable zNose® technology incorporated into a handheld instrument

How the zNose® Quantifies the Chemistry of Aromas

Figure 2 shows a simplified diagram of the zNose® system consisting of two parts. One section uses a capillary tube (GC column), helium gas, and a solid-state detector. The other section comprises of a heated inlet and pump, which samples ambient air.

Combining the two sections is a “loop” trap, which serves as a preconcentrator when positioned in the air section (sample position), and as an injector when positioned in the helium section (inject position). Operation is a two-step procedure. First, ambient air (aroma) is sampled and organic vapors collected (preconcentrated) on the trap. After sampling, the trap is moved into the helium section where the collected organic compounds are injected into the helium gas.

The organic compounds travel through a capillary column with varying velocities and thus, each chemical exits the column at characteristic times. As they exit the column, they are detected and quantified by a solid state detector. An internal high speed gate array microprocessor controls the collection of sensor data which is conveyed to a computer or a user interface using a USB or RS-232 connection.

Simplified diagram of the zNose® showing an air section on the right and a helium section on the left. A loop trap preconcentrates organics from ambient air in the sample position and injects them into the helium section when in the inject position.

Figure 2. Simplified diagram of the zNose® showing an air section on the right and a helium section on the left. A loop trap preconcentrates organics from ambient air in the sample position and injects them into the helium section when in the inject position.

In Figure 3, the aroma chemistry can be seen, and can be displayed as a sensor spectrum or a polar olfactory image of odor intensity vs retention time.

Calibration is performed using a single n-alkane vapor standard. A library of retention times of established chemicals indexed to the n-alkane response (Kovats indices) permits compound identification and machine autonomous measurement.

Sensor response to n-alkane vapor standard, here C6-C14, can be displayed as sensor output vs time or its polar equivalent olfactory image

Figure 3. Sensor response to n-alkane vapor standard, here C6-C14, can be displayed as sensor output vs time or its polar equivalent olfactory image

Chemical Analysis (Chromatography)

The time derivative of the sensor spectrum, as seen in Figure 3, produces the spectrum of column flux, normally referred to as a chromatogram.

Figure 4 illustrates the chromatogram response of n-alkane vapors (C6 to C14), which offers an accurate measure of retention times. Graphically defined regions illustrated as red bands in Figure 4 calibrate the system and provide a reference time base against which following chemical responses are indexed or compared. As an example, a response midway between C10 and C11 will have a retention time index of 1050.

Chromatogram of n-alkane vapors (C6 to C14)

Figure 4. Chromatogram of n-alkane vapors (C6 to C14)

Natural Gas

Natural gas is a combustible combination of hydrocarbon gases. While natural gas is formed mainly of methane, it regularly includes propane, ethane, pentane, butane, and even higher molecular weight hydrocarbons. The composition of natural gas can differ extensively.

The natural gas used by consumers is not pure methane. Raw natural gas is procured from three types of wells: gas wells, oil wells, and condensate wells. Natural gas can be dissolved in the crude oil (dissolved gas) or separated from oil during the formation (free gas). Gas wells usually yield raw natural gas by itself, while condensate wells yield free natural gas together with a semi-liquid hydrocarbon condensate.

Petroleum drilling rig collects natural gas above oil deposit

Figure 5. Petroleum drilling rig collects natural gas above oil deposit

Gas refinery for producing natural gas product

Figure 6. Gas refinery for producing natural gas product

Testing Natural Gas Samples

A sample of natural gas was taken from a local residence (Newbury Park, CA) using a one-liter tedlar bag. About 5 ml was extracted and tested by zNose® fitted with a one-meter db-624column. The GC-method file (Figure 3) shows the temperatures used and the instrument settings. A less than 1-minute sample time is followed by a short three second wait after injection and prior to ramping the column from 40 °C to 160 °C at 5 °C per second.

Figure 8 shows a full screen view of a 20-second analysis (time-stamped file) of a natural gas sample. The four main windows exhibit operator notes, derivative of sensor output vs retention time (column flux), sensor output vs retention time, and tabulated peak retention time and un-calibrated peak area counts.

GC method file

Figure 7. GC method file

Full screen display of 20 second analysis

Figure 8. Full screen display of 20 second analysis

Replicate Measurements (Speed and Precision)

Conducting five replicate measurements on the natural gas sample at one minute intervals shows the retention time stability and precision of the zNose®.

The top trace is an investigation of the tedlar bag filled with clean nitrogen prior to filling with natural gas. Reduction of peak amplitude with time is because of absorption on walls of tedlar bag.

Sensor Output vs Retention Time

The zNose® uses a unique GC detector which is non-specific and non-ionic, hence, it detects everything and does not miss anything. The sensor creates a frequency deviation proportional to the mass of analytes eluting from the GC column, hence it calculates concentration (odor intensity) and not column flux like other GC detectors do.

Since it is an integrator it has zero dead volume which is suitable for high speed chromatography. Odor intensity vs retention time also creates a distinctive olfactory image referred as a Vaporprint®.

Replicate measurements showing repeatability and precision of measurement

Figure 9. Replicate measurements showing repeatability and precision of measurement

SAW sensor output shows odor concentration vs retention time. Polar plot displays olfactory image (Vaporprint®)

Figure 10. SAW sensor output shows odor concentration vs retention time. Polar plot displays olfactory image (Vaporprint®)

Derivative of Sensor Output vs Retention Time

The derivative of sensor output frequency (odor intensity) produces a conventional chromatogram or column flux vs retention time. Vaporprint® images plotted on a logarithmic scale are used to improve the view ability of trace compounds and offer an entire view of the odor chemistry.

Derivative of odor intensity yields conventional chromatogram

Figure 11. Derivative of odor intensity yields conventional chromatogram

Kovats Indexing to N-Alkane Standard Response

The zNose® software is able to index retention times (Kovats indices) when natural gas peak retention times are compared to those of n-alkanes.

The benefit of this method is that it makes retention time dependent upon n-alkane vapor standards but independent of the instrument. This in turn allows results from various instruments to be compared and confirmed by autonomous laboratory measurements.

Referencing peak retention times to n-alkane response (C6-C14) shown as red bands (A) allows the software to express peak retention times as Kovats indices (B)

Figure 12. Referencing peak retention times to n-alkane response (C6-C14) shown as red bands (A) allows the software to express peak retention times as Kovats indices (B)

Peak Library – Compound Identification

After the peaks have been indexed to n-alkanes, a tentative identification based upon the Kovats indices can be made.

The ZNose® software enables the formation of an infinite number of such libraries specific to each user application. Current libraries can be edited and new compounds incorporated. For example, Figure 13 shows the library entry for one of the natural gas peaks, NG-1244. Identification based upon Kovats indices is thought to be useful but only tentative until confirmation by autonomous analysis e.g. with a GC-MS.

Click on any peak to identify odor using peak library based upon Kovats indices

Figure 13. Click on any peak to identify odor using peak library based upon Kovats indices

Low Molecular Weight Compounds

Delectability of low molecular weight compounds is based upon the sensor crystal’s temperature. In these measurements, a detector temperature of 20 °C was used, but detector temperatures can be as low as 0 °C.

Response for the zNose® is thus restricted to compounds with boiling points over 0 °C. The lowest weight compound detected (274 counts) in the natural gas sample can be observed to possess an index of 479 or moderately below C5 (pentane). For reference background noise, levels in the zNose® are normally 1-10 counts.

Expanding time and amplitude scale shows response from low molecular weight compounds (C4 to C10)

Figure 14. Expanding time and amplitude scale shows response from low molecular weight compounds (C4 to C10)

Peak Files, Calibration Factors and Alarm Levels

The ZNose® software enables the user to effortlessly create individual peak response factors and alarm levels. Using chemical standards, the zNose® can be calibrated in vapor pressure (e.g. ppm), mass units (e.g. picograms), or user units.

Multi-point and single point calibration approaches are possible. In this figure, a virtual senor arrangement for the eight major peaks detected in natural gas are developed with a peak file. Scale factors (counts/ ppm) illustrated are based upon a presumed concentration of 100 ppm for each compound. Arbitrary alarm levels are added into the peak file as well. In effect, the peak file defines a virtual sensor arrangement for natural gas.

Creating a peak file allows scale factors and alarm values to be set for individual peaks which can be graphically set and displayed as red bands or virtual sensor regions

Figure 15. Creating a peak file allows scale factors and alarm values to be set for individual peaks which can be graphically set and displayed as red bands or virtual sensor regions

Calibrated Readings

Once scale factors established and a peak file has been defined, the software automatically shows peak readings in the preferred units, for example ppm. Referencing retention time to n-alkanes also offers an appropriate peak naming convention in cases where the actual compound is unidentified.

Using a peak file for natural gas peak concentrations are displayed directly

Figure 16. Using a peak file for natural gas peak concentrations are displayed directly

Virtual Chemical Sensors

It is often easier to display a collection of chemical sensor readings instead of a chromatogram. Based upon alarm settings, each sensor can generate alarms whenever concentrations surpass the alarm value. Working the zNose® in automatic measurement mode, the sensor array readings can be updated frequently, for example once per minute. Using isothermal GC techniques, it is possible to update sensor readings as frequently as every 15 seconds. Virtual sensors and high-speed chromatography are perfect for process control applications.

Display of virtual chemical sensor readings replaces the need to display chromatograms

Figure 17. Display of virtual chemical sensor readings replaces the need to display chromatograms

Summary

The zNose® provides a suitable solution for precision, speed and accuracy in the chemical analysis of all kinds of fragrances, odors and vapors. Natural gas as it is received in the regular home is far from just pure methane and examination of a random sample has revealed organic compounds covering the total hydrocarbon range up to almost C15 are present at comparatively high concentrations.

The ability to build virtual chemical sensor arrays to monitor these compounds may offer a valuable process-monitoring tool. Although the zNose® has only one physical sensor, the ability to conduct high-speed chromatography enables it to functionally perform as if it were a collection of specific chemical sensors. Virtual chemical sensors can be created in the hundreds.

The ability to complete numerous chromatographic analyses per day also offers a cost-effective solution to process development and sample screening in the laboratory. With a conventional long-column GC such projects could take many months, but with the zNose® it is mostly possible to finish development projects within a week. It is not hard to see how the zNose® pays for itself within 30 days.

This information has been sourced, reviewed and adapted from materials provided by Electronic Sensor Technology.

For more information on this source, please visit Electronic Sensor Technology.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Electronic Sensor Technology. (2019, July 18). Testing the Composition of Natural Gas Using an Electronic Nose. AZoM. Retrieved on September 18, 2019 from https://www.azom.com/article.aspx?ArticleID=13821.

  • MLA

    Electronic Sensor Technology. "Testing the Composition of Natural Gas Using an Electronic Nose". AZoM. 18 September 2019. <https://www.azom.com/article.aspx?ArticleID=13821>.

  • Chicago

    Electronic Sensor Technology. "Testing the Composition of Natural Gas Using an Electronic Nose". AZoM. https://www.azom.com/article.aspx?ArticleID=13821. (accessed September 18, 2019).

  • Harvard

    Electronic Sensor Technology. 2019. Testing the Composition of Natural Gas Using an Electronic Nose. AZoM, viewed 18 September 2019, https://www.azom.com/article.aspx?ArticleID=13821.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit