ASTM E260 Standard Practice for Packed Column Gas Chromatography
13. Methods of Qualitative Analysis
13.1 Identification of compounds by gas chromatography alone cannot be absolute, and the results must be considered with care. Elution of a compound is dependent upon carrier gas flow rate, column temperature, support size, amount and type of liquid phase, column dimensions, instrument dead volume, and column pressure drop. These parameters must be stable to obtain reproducible results. The recommended format for a gas chromatographic method is given in Section 15.

13.2 Tentative identification of a compound can be made by comparing its adjusted retention time against those of known standards using exactly the same chromatographic parameters.
13.2.1 The retention time is the time interval measured from the point of injection to maximum peak height of the sample. Adjusted retention time, t'R, is derived by subtracting the time required for an unabsorbed gas, like air, or methane, tM, to traverse the column (also called the gas holdup time) from the retention time, tR.

NOTE 7 - On some solid adsorbent columns, such as molecular sieves, there is no nonadsorbed component.
t'R = tR - tM

13.2.2 Retention times are affected by all chromatographic parameters. As a result, direct comparison of retention times of the same components on different instruments or between laboratories should be done with caution. Use of relative retention time is an easy practical technique for providing elution data. The retention of a component is expressed relative to the retention of a known reference standard. The reference standard should possess structural or chemical similarity to the compounds being analyzed.

13.3 The retention of a given weight of compound is usually independent of its concentration if the compound does not overload the column producing skewed peaks. The retention of the compound is also independent of other substances present if there is no appreciable overlap with another compound. Substances that exhibit positive, or Langmuir-type, skewing (tailing) during elution will produce a decrease in retention as the concentration increases; while negative, or anti-Langmuir-type fronting will produce an increase in retention time with increased concentration.
13.3.1 The logarithm of the retention time of members of a homologous series run isothermally is usually a linear function of the number of carbon atoms of a molecule. Using this characteristic, two or three reference compounds can provide sufficient information to prepare a plot of the logarithm of the retention time versus carbon number, and they can identify other members of the series. Retention time, adjusted or not, is of little value in comparing the results from various instruments. The use of Kovats retention index, based on the relative retention of a compound to the retention of normal paraffins, provides a more reliable means of comparing the results obtained from different instruments (see Practice E 355).

13.4 Absolute compound identification or characterization, must be made with ancillary techniques such as mass or infrared spectrometry, nuclear magnetic resonance, chemical analysis of the effluent, or spot tests for functional groups.
13.4.1 The samples for the analyses in 13.1-13.4 may be obtained by trapping components as they emerge from the chromatograph. A trap, glass capillary, or U-tube, is cooled with ice or dry ice, and placed in the effluent stream of the column. Several collections may be required to obtain a sufficiently large sample.

13.4.2 The collection of effluent is easiest with nondestructive detectors, see 6.7. In the case of destructive detectors, a split is made for the collection just before the detector.

13.4.3 Apparatus is also available so that the effluent from the gas chromatographic column can be analyzed directly by mass spectrometers or infrared spectrophotometers.

14. Methods of Quantitative Analysis
14.1 Gas chromatography can be used to determine quantitatively the composition of complex samples. There are several factors that must be considered before the sample is analyzed. The recommended format for gas chromatographic methods is given in Section 15.
14.1.1 The Chemistry of the Sample - The chemistry of the sample, if known, allows a chromatographer to select more accurately a column compatible with the sample and to anticipate potential interferences from reaction by-products.

14.1.2 The Choice of a Detector - A detector must be chosen with the needed selectivity and sensitivity. If components will be analyzed at low levels, an electrolytic conductivity electron capture, nitrogen phosphorus, microcoulometric, ionization, or flame photometric detector should be selected. The detector may be limited to these lower concentrations and not applicable to high concentrations.

14.1.3 Initial Separation of Components - Next, a column must be chosen that will resolve the components of interest in the sample within a reasonable amount of time. First, a rough separation should be achieved with known standards. Next, actual samples should be analyzed to determine if there are any interferences. A second column, or an ancillary technique (GC/mass spectrometry, GC/infrared spectrometry, etc., should be used to verify that additional components are not eluting with the component of interest. Each new sample adds the possibility of an interference eluting with the component of interest; therefore this should be checked often. If an interference is detected, the chromatographer must change the method to remove it. The several options for doing this are as follows:
14.1.3.1 Select a column stationary phase with a greater selectivity for either the interference or the component of interest.

14.1.3.2 Choose a different type of detector that would detect the component of interest but not the interference. Examples would be water not being detected by a flame-ionization detector, or hydrocarbons not detected by an electrolytic conductivity or electron capture detector.

14.1.3.3 Consider other types of chromatographic separation such as capillary gas chromatography for more efficient separation of volatile compounds, liquid chromatography for separation of non-volatile compounds, or another appropriate separation technique.

14.1.4 Detector Sensitivity and Linearity - Once the chromatographic separation has been optimized, the detector can be optimized and calibrated. Gas flows should be adjusted to the optimum levels to get peak sensitivity at the concentration range of the components of interest. The detector must also be clean and leak-tight. (See the manufacturer's manual for suggested procedures.)
14.1.4.1 The linearity of the detector over the desired concentration range of the component(s) of interest is determined using prepared standards. This step will determine what the response is to increasing amounts of component. The peak area or height should be plotted versus the concentration for about five concentrations near the expected sample concentration. There should be a linear correlation. Nonlinearity may be caused by reactivity, adsorptivity, thermal sensitivity, or excessive column bleed. If the latter is the cause, change to a more thermally stable column or one of different polarity. Column reactivity can be characterized by skewed, misshaped peaks. This can be corrected by installing a fresh column of the same type that does not have reactive sites. Test mixtures can be used to demonstrate nonreactivity. Other sources of adsorptivity or reactivity with the sample are the injection port, connecting lines to the detector, or glass wool. Each of these sources can be detected by carefully troubleshooting the system.

14.1.4.2 The detector performance should be checked periodically throughout the analysis. This can be done by injecting one of the linearity standards and comparing it to the linearity plot.

14.1.5 Peak Area or Height Measurement - Many types of peak area and height measurement techniques exist. The oldest methods for calculating the peak area are manual measurement with a ruler of the peak area using one of the following equations:
peak area = wh x h
where:
wh = peak width at half height, and
h = peak height
or
peak area = ½ wb x h
where:
wb = peak width at the base of the peak, and
h = peak height.

Another precise measurement defines the peak area as retention distance (in millimetres) times the peak height (also in millimetres). For peak height, this distance is simply measured from the baseline to the apex of the peak. However, these techniques now, for the most part, have been replaced by electronic integration, which is much faster. The proper use of these devices is crucial for accurate quantitative analysis. The instruction manual for the particular integrator should be studied and understood thoroughly before attempting to use electronic integration for peak area or peak height measurement.

14.1.6 Data Handling:
14.1.6.1 All manufacturers supply an integral electrometer to allow the small electrical current changes to be coupled to recorders/integrators/computers. The preferred system will incorporate one of the newer integrators or computers that converts an electrical signal into clearly defined peak area counts in units such as microvolt-seconds. These data can then be readily used to calculate the linear range.

14.1.6.2 Another method uses peak height measurements. This method yields data that are very dependent on column performance and, therefore, not recommended.

14.1.6.3 Regardless of which method is used to calculate linear range, peak height is the only acceptable method for determining minimum detectability.

14.1.7 Calibration - It is essential to calibrate the measuring system to ensure that the nominal specifications are acceptable and particularly to verify the range over which the output of the device, whether peak area or peak height, is linear with respect to input signal. Failure to perform this calibration may introduce substantial errors into the results. Methods for calibration will vary for different manufacturers' devices but may include accurate constant voltage supplies or pulse generating equipment. The instruction manual should be studied and thoroughly understood before attempting to use electronic integration for peak area or peak height measurements.

14.2 Types of Calculations:
14.2.1 Each method of quantitative analysis has advantages and disadvantages. The four methods of quantitative analysis are as follows:
14.2.1.1 Internal standardization,

14.2.1.2 External standardization,

14.2.1.3 Normalization, and

14.2.1.4 Corrected area.

14.2.2 Internal Standardization - In this technique, a pure component (the internal standard) is added to a sample in a known amount. The peak area, or height, of all components of interest is compared to the peak area, or height, of the internal standard. These comparisons are referred to as response factors:
RF = AC/AIS x W IS/WC
where:
RF = response factor,
AC = peak area of component,
AIS = peak area of internal standard,
WIS = mass of internal standard, and
WC = mass of component.

The amount of the component can be calculated from the weights of the sample and internal standard, the response factor, and the peak areas (or heights) as follows:
% ConcC = WIS/WS x A C/AIS x 1/RF x 100 %
where:
ConcC = concentration of component in sample,
WIS = mass internal standard,
WS = mass sample,
AC = peak area of component,
AIS = peak area of internal standard, and
RF = response factor.

This technique provides a correction for the relatively high variability of syringe injection and, therefore, yields a more precise method of analysis. Neither the quantity of solution injected, nor change in detector response, will alter the area ratio of the analyte and the internal standard. To achieve optimum performance, the internal standard must meet the following criteria.

14.2.2.1 The internal standard must elute in an area of the chromatogram that is free of sample components, or possible sample components.

14.2.2.2 The internal standard must not react with the sample or any of its components.

14.2.2.3 The internal standard and the sample must be homogeneous. A cosolvent may be used to produce a homogeneous mixture.

14.2.2.4 The internal standard must be easily and accurately added.

14.2.2.5 The internal standard must be pure.

14.2.2.6 The internal standard should elute near the component of interest.

14.2.2.7 The concentration of the internal standard, relative to that of the analyte, should be such that these two peaks are within 50 to 100 % of full scale deflection with the same electronic attenuation and sensitivity setting in order to allow manual measurements and calculations of parameters, if desired.

14.2.2.8 The most common use for the internal standard technique in chromatography is to correct for quantitative variations in the injection, particularly when using syringes. For this purpose, the internal standard need not be chemically related to the analyte, but must possess the criteria cited above and may be added in the final solution.

14.2.2.9 In certain applications, an internal standard with functional groups similar to the analyte may be desirable. For instance, those with a labile proton can be expected to exhibit similar adsorption isotherm behavior and to undergo similar physico-chemical transformations during such processes, as extraction from a complex matrix or derivatization, or both. Likewise, similar electronegative functional groups are likely to behave similarly towards an electron capture detector.

14.2.3 External Standardization:
14.2.3.1 This method compares peak areas or heights of components in a sample chromatogram to those in a standard solution injected separately. It is critical that accurate amounts of sample and standard be injected for the method to be valid. Generally, the solvent flush injection technique (see 10.3.1.2) or a sample valve of fixed volume is preferred.

14.2.3.2 The advantages of this method are as follows:
(a) Nondetected components do not bias the results.
(b) It can be used where several known components must be determined in a very complex sample.
(c) It can quantitate relatively reactive components.
(d) A single sample can be analyzed where maximum accuracy is not required.
(e) Nonlinearity has a minimal effect if the external standard is near the concentration of the sample.

14.2.3.3 The critical part of this method is the injection. The volume of sample in the injection syringe and standard must be accurately measured, allowing no bubbles in the slug of sample or standard solution. If the sample and standard have different densities, a correction must be made. Densities are easily determined by filling a 50-µL syringe to about 30 µL, wiping the needle, weighing it, expelling the sample, wiping the needle again, and reweighing it.

14.2.3.4 The peak areas or heights of the component in the sample and the standard compound are measured and the concentration calculated as follows:
% ConcC = AC/AES x W ES/WS x % ConcES
where:
ConcC = concentration of component,
AC = peak area of component in sample,
AES = peak area of external standard,
WES = mass of external standard injected,
WS = mass of sample injected, and
ConcES = concentration of external standard in solution.

14.2.4 Normalization - This calculation assumes that every component elutes and that each has similar response factors. It is a fast procedure that requires no weighing. The sample is injected, and the peak areas or heights of all components are measured. The concentration of the component of interest is calculated as follows:
% ConcC = A C/AALL x 100
where:
ConcC = concentration of component of interest,
AC = area of component, and
AALL = sum of areas of all components.

Severe errors result if the components have different response factors or do not all elute.

14.2.5 Corrected Area - This method corrects for differences in response but still assumes that all components elute and are observed by the detector. Response factors are used to correct for response differences as follows:
% ConcC = AC/(A x RF)ALL x RFC x 100 %
where:
ConcC = concentration of component of interest,
AC = peak area of component,
(A x RF)ALL = sum of peak areas times their respective response factors relative to a standard, and
RFC = response factor of component to the same standard.