ASTM D2668 Standard Test Method for 2,6-di-tert-Butyl- p-Cresol and 2,6-di-tert-Butyl Phenol in Electrical Insulating Oil by Infrared Absorption
4. Apparatus
4.1 With equipment description referring to compliance, the equipment shall be in accordance with Section 6 of Practices D2144. Accordingly, the use of Fourier-transform rapid scan infrared (FTIR) spectrophotometers is permitted by reference to that test method.
5. Sampling
5.1 Obtain the sample in accordance with Practices D923.
6. Calibration and Standardization
6.1 When the manufacturer of the oil is known and the base oil is available, use it to prepare the standards. For oils of unknown origin, use base oils which meet the requirements of Specification D3487. Some base oils may provide a better match than others and therefore it is desirable to have several available.
6.2 Prepare standards containing between 0.05 and 0.4 weight percent of 2,6-ditertiary-butyl paracresol or 2,6-ditertiary-butyl phenol dissolved in an uninhibited base oil. Alternatively, the range of prepared standards may be increased to 0.5 weight percent if certain oils to be investigated are believed to contain greater amounts of inhibitor. Obtain a spectrum, at the desired band, of each standard in accordance with Practices D2144. Cells with a standard path length of 0.3 to 1.0 mm are recommended. Other path lengths may be found more suitable for different instruments or particular wave lengths. Other sample path lengths may be used provided the instrument sensitivity can be adjusted to compensate for this change. The dip in the curve for the inhibited oil should provide a distinctive increase in the absorbance at the critical wavelength or frequency (Note 3). Repeat the procedure on each of the standards making at least three scans on each standard. (See Note 2) Record all settings of the spectrophotometer used in obtaining the respective spectra (Note 4).
NOTE 2 - The current method precision is based on manually determined results where exactly three scans were determined for each standard. Newer instruments are capable of automatically performing scans much more rapidly, which can reduce the variability of results determined. In such cases, it is recommended that the number of scans be increased to statistically compensate for any outliers. Laboratories will need to determine the minimum number of scans that should be used in their instrument standardization and test specimen analyses to satisfy their testing needs.
NOTE 3 - Where desired, a chart having a non-linear wavelength scale as the abscissa may be used.
NOTE 4 - In making these tests, transmission-scaled charts may be used, but in this case special rulers and nomographs or logarithmic tables will be necessary for determining the intensity measurements. Alternatively, instrument software capable of recording all settings of the spectrophotometer used in obtaining the respective spectra, may be used.
6.3 The quantitative determination is made from the following equation which is derived from Beer's law:
Absorbance = A - Ao
where:
Ao = absorbance units of base oil, and
A = absorbance units of oil containing 2,6-ditertiary-butyl paracresol or 2,6-ditertiary-butyl phenol.
6.4 Manual Plotting Routine for Generating Calibration Curve - Designate the point of maximum absorbance of the absorbance band as Point A. Draw a tangent to the spectrum curve and a second line through Point A perpendicular to the absorbance lines, as shown in Fig. 1. Designate the intersection of these two lines as Point Ao. Read the values of absorbance at these points on the charts of the three scans made on each test specimen to the nearest 0.001 absorbance unit (with the aid of a reading glass) and subtract the values of A o from those of A. When the average of the three values for each of the specially prepared test specimens is plotted against the concentration, a straight line is obtained. The best straight line through the calibration data points should be drawn or determined by linear regression analysis. This is the calibration curve from which the unknown concentration of the 2,6-ditertiary-butyl paracresol or 2,6-ditertiary-butyl phenol in a test specimen may be determined. One such calibration curve is shown as Fig. 2. Fig. 3 and Fig. 4 illustrate sections of differential scans. Figs. 5-7 show FTIR scans of uninhibited, a similar oil with 0.55 % DBPC and the resulting FTIR differential scan.
6.4.1 Automatic Plotting Routine for Generating Calibration Curve - For instruments capable of automatically generating a standard calibration curve, follow manufacturer's instructions. Perform at least three scans for each standard analyzed across the calibration range of interest (see Note 2). Develop a calibration curve which has a minimum correlation value of 0.99 to ensure the linearity of the calibration curve.
6.5 When frequent determinations are made on a routine basis, periodic checks of one or more standards are recommended, since the characteristics of electronics components in spectrophotometers change with time. If the absorbance of the standards differ from the calibration curve by more than the limits given in 8.2, a new curve should be obtained. Some laboratories have developed tighter limits than these. A new curve should also be obtained whenever there is a change in operating conditions, such as a change in light source, scan speed, and so forth.
NOTE 5 - It is recommended that the solutions of known concentration be stored in amber-colored bottles for a period not to exceed one year in order to facilitate a quick check of the characteristics of the spectrophotometer in relation to the calibration curve.
