ASTM E516 Testing Thermal Conductivity Detectors Used in Gas Chromatography
5. Sensitivity (Response)
5.1 Definition:
5.1.1 Sensitivity (response) of the TCD is the signal output per unit concentration of a test substance in the carrier gas, in accordance with the following relationship (7):
S = A Fc/W
where:
S = sensitivity (response), mV•mL/mg,
A = integrated peak area, mV•min,
Fc = carrier gas flow rate (corrected to detector temperature), mL/min, and
W = mass of the test substance in the carrier gas, mg.

5.1.2 If the concentration of the test substance in the carrier gas, corresponding to a detector signal is known, the sensitivity is given by the following relationship:
S = E/Cd
where:
E = peak height, mV, and
Cd = concentration of the test substance in the carrier gas at the detector, mg/mL.

5.2 Test Conditions:
5.2.1 Normal butane is the preferred standard test substance.

5.2.2 The measurement must be made within the linear range of the detector.

5.2.3 The measurement must be made at a signal level at least 100 times greater than the minimum detectability (200 times greater than the noise level) at the same conditions.

5.2.4 The rate of drift of the detector at the same conditions must be stated.

5.2.5 The test substance and the conditions under which the detector sensitivity is measured must be stated. This will include but not necessarily be limited to the following:
5.2.5.1 Type of detector (for example, platinum-tungsten filament type),

5.2.5.2 Detector geometry (for example, flow-type, diffusion-type),

5.2.5.3 Internal volume of the detector,

5.2.5.4 Carrier gas,

5.2.5.5 Carrier gas flow rate (corrected to detector temperature),

5.2.5.6 Detector temperature,

5.2.5.7 Detector current,

5.2.5.8 Method of measurement, and

5.2.5.9 Type of power supply (for example, constant voltage, constant current).

5.2.5.10 For capillary detectors, the make-up gas, carrier, and reference flows should be stated.

5.3 Methods of Measurement:
5.3.1 Sensitivity may be measured by any of three methods:
5.3.1.1 Experimental decay with exponential dilution flask (8, 9) (see 5.4),

5.3.1.2 Utilizing the permeation tube (10), under steady-state conditions (see 5.5),

5.3.1.3 Utilizing Young's apparatus (11), under dynamic conditions (see 5.6).

5.3.2 Calculation of TCD sensitivity by utilizing actual chromatograms is not recommended because in such a case the amount of test substance corresponding to the peak cannot be established with sufficient accuracy.

5.4 Exponential Decay Method:
5.4.1 A mixing vessel of known volume fitted with a magnetically driven stirrer is purged with the carrier gas at a known rate. The effluent from the flask is delivered directly to the detector. A measured quantity of the test substance is introduced into the flask, to give an initial concentration, Co, of the test substance in the carrier gas, and a timer is started simultaneously.

5.4.2 The concentration of the test substance in the carrier
gas at the outlet of the flask, at any time is given as follows:
Ct = Co exp [-Fc t/Vf]
where:
Ct = concentration of the test substance at time t after introduction into the flask, mg/mL,
Co = initial concentration of test compound introduced in the flask, mg/mL,
Fc = carrier gas flow rate, corrected to flask temperature 4 mL/min,
t = time, min, and
Vf = volume of flask, mL.

5.4.3 To determine the concentration of the test substance at the detector, Cd, it is necessary to apply the following temperature correction:
Cd = Ct [T f/Td]
where:
Cd = concentration of the test substance at the detector, mg/mL,
Tf = flask temperature, K, and
Td = detector temperature, K.

5.4.4 The sensitivity of the detector at any concentration can be calculated by:
S = E/Cd
where:
S = sensitivity, mV•mL/mg,
E = detector, signal, mV, and
Cd = concentration of the test substance at the detector, mg/mL.

NOTE 1 - This method is subject to errors due to inaccuracies in measuring the flow rate and flask volume. An error of 1 % in the measurement of either variable will propagate to 2 % over two decades in concentration and to 6 % over six decades. Therefore, this method should not be used for concentration ranges of more than two decades over a single run.

NOTE 2 - A temperature difference of 1°C between flask and flow measuring apparatus will, if uncompensated, introduce an error of 1/3 % into the flow rate.

NOTE 3 - Extreme care should be taken to avoid unswept volumes between the flask and the detector, as these will introduce additional errors into the calculations.

NOTE 4 - Flask volumes between 100 and 500 mL have been found the most convenient. Larger volumes should be avoided due to difficulties in obtaining efficient mixing and likelihood of temperature gradients.

5.5 Method Utilizing Permeation Tubes:
5.5.1 Permeation tubes consist of a volatile liquid enclosed in a section of plastic tubing. They provide low concentrations of vapor by diffusion of the vapor through the walls of the tubing. The rate of diffusion for a given permeation tube is dependent only on the temperature. As the weight loss over a period of time can be easily and accurately measured gravimetrically, the rate of diffusion can be accurately determined. Hence, these devices have been proposed as primary standards.

5.5.2 Accurately known concentrations can be prepared by passing a gas over the previously calibrated permeation tube at constant temperature. The concentration of the test substance in the gas can then be easily calculated according to the following relationship:
C = RT/Fc
where:
C = concentration of the test substance in the gas, mg/mL,
RT = permeation rate of the test substance at the temperature of the permeation tube, mg/min, and
Fc = flow rate of the gas over the tube at the temperature of the tube, mL/min.

NOTE 5 - If the flow rate of the gas is measured at a temperature different from the tube temperature, correction must be made, as described in Appendix X1.

5.5.3 When using a permeation tube for the testing of a TCD, the carrier gas is passing over a previously calibrated permeation tube containing the test substance at constant temperature and introduced immediately into the detector, kept at the desired temperature. Knowing the concentration of the test substance in the carrier gas leaving the permeation tube at the temperature of the tube, the concentration at detector temperature can be calculated directly, by applying the correction specified in 5.4.2. Knowing this value and the detector signal, the sensitivity of the detector can be obtained according to the equation given in 5.4.4.

NOTE 6 - Permeation tubes are suitable only for preparing relatively low concentrations in the part-per-million range. Hence for detectors of relatively low sensitivity or of higher noise levels, this method may not satisfy the criteria given in 5.2.3, which requires that the signal be at least 100 times greater than the noise level.

5.6 Dynamic Method:
5.6.1 In this method a known quantity of test substance is injected into the flowing carrier gas stream. A length of empty tubing between the sample injection point and the detector permits the band to spread and be detected as a Gaussian band. The detector signal is then integrated by any suitable method. This method has the advantage that no special equipment or devices are required other than conventional chromatographic hardware. For detectors optimized for capillary column flow rates, uncoated, deactivated, fused silica tubing should be used.

5.6.2 The sensitivity of the detector is calculated from the peak area according to 5.1.1.

NOTE 7 - Care should be taken that the peak obtained is sufficiently wide so the accuracy of the integration is not limited by the response time of the detector or of the recording device.

NOTE 8 - Peak areas obtained by integration (Ai) or by multiplying peak height by peak width at half height (Ac ) differ by 6 % for a Gaussian peak:
Ac = 0.94 Ai