(1997; 248 pages) [French]
General recommendations for the preparation and use of infrared spectra in pharmaceutical analysis1
1WHO Expert Committee on Specifications for Pharmaceutical Preparations. Thirty-fourth Report. Geneva, World Health Organization, 1996 (WHO Technical Report Series, No. 863).
In pharmaceutical analysis the region of the electromagnetic spectrum used is 4000-600cm-1 (wavelength 2.5-16.7 µm), i.e. the mid-infrared. Spectrophotometric measurements in this region are mainly used for identification purposes. Except for enantiomers, which have identical spectra in solution, the infrared spectrum of any given substance is unique. Polymorphism and other factors, such as variations in crystal size and orientation, the grinding procedure, and the possible formation of hydrates may, however, be responsible for minor, and occasionally substantial, variations in the infrared spectrum of a substance in the solid state. The infrared spectrum is not usually greatly affected by the presence of small quantities of impurities in the substance tested. For identification purposes, the spectrum may be compared with that of a reference substance, concomitantly prepared, or with a reference spectrum.
The terms absorbance, transmittance, absorptivity and absorption spectrum are defined in The international pharmacopoeia. 3rd ed., Vol. 1, pp. 33-34, in the chapter “Spectrophotometry in the visible and ultraviolet regions”.
Conventional infrared spectrometers disperse the infrared radiation by means of either gratings or prisms. The development of computerized laboratory equipment provides the additional option of using an interferometer coupled to a computer for the reduction of the data, by performing a Fourier transformation of the interferogram, to generate an infrared spectrum. These instruments are called Fourier transform infrared spectrometers (FTIRs). Apart from small differences in the low-frequency cut-off, all of the above types of infrared instruments generate comparable data and can generally be used interchangeably for qualitative analyses. However, each instrument will possess specific signal-to-noise and resolution characteristics.
Spectrophotometers suitable for use for identification tests should normally operate in the range 4000-600 cm-1 (2.5-16.7 µm) or in some cases up to 250 cm-1 (40 µm). If the attenuated total reflectance technique is to be used, the instrument must be equipped with a suitable attachment consisting of a single or multireflecting element. The attachment and a suitable mounting should permit its alignment in the spectrophotometer for maximum transmission.
3. Method of verification of frequency scale and resolution
The spectrum of a polystyrene film of suitable thickness, normally between 0.03 mm and 0.05 mm, is recorded. This includes maxima at the following frequencies, expressed as wavenumbers in cm-1: 3027, 2851, 2924, 1944, 1871, 1802, 1601, 1583, 1181, 1154, 1069, 1028, 907, 699. Acceptable tolerances are ± 8 cm-1 for the range 4000-2000 cm-1 and ± 4 cm-1 for the range 2000-600 cm-1.
The difference between the percentage transmittance of the absorption minimum at 2870 cm-1 and that of the absorption maximum at 2851 cm-1 should be greater than 18 and the difference between the percentage transmittance of the absorption minimum at 1589 cm-1 and that of the absorption maximum at 1583 cm-1 should be greater than 12.
Precautions should be taken to minimize exposure to atmospheric moisture during sample preparation. It is advisable to store the halide salts, the sodium chloride or other similar plates, and all necessary accessories in a desiccator at room temperature over silica gel, and to prepare the samples in an area of controlled temperature and humidity; alternatively, all manipulations should be carried out under an infrared lamp.
5. Use of solvents
The solvent used in infrared spectrophotometry must not affect the cell, which usually consists of a halide salt such as sodium chloride or potassium bromide. Where possible, spectral grade solvents should be used.
No solvent is completely transparent throughout the entire infrared spectrum. Carbon tetrachloride R1 is practically transparent (up to 1 mm of thickness) over the range 4000-1700 cm-1 (2.5-5.9 µm). Dichloromethane R and dibromomethane R are useful solvents. Carbon disulfide IR2 (up to 1 mm in thickness) is suitable as a solvent up to 250 cm-1 (40 µm) except in the 2400-2000 cm-1 (4.2-5 µm) and the 1800-1300 cm-1 (5.6-7.7 µm) regions, where it has strong absorption. Its weak absorption in the 875-845 cm-1 (11.4-11.8 µm) region should be noted. Other solvents have relatively narrow regions of transparency.
1 R: of reagent-grade quality.
2 IR: of suitable purity for use in spectrophotometry in the infrared region.
6. Preparation of the substance to be examined
To obtain a suitable infrared absorption spectrum, it is necessary to follow the instructions given below for the preparation of the substance. Substances in liquid form may be tested directly or in a suitable solution. The usual methods of preparation for solid substances include dispersing the finely ground solid specimen in mineral oil, incorporating it in a transparent disc or pellet obtained by mixing it thoroughly with previously dried potassium halide and compressing the mixture in a die, or preparing a solution in a suitable solvent. Preparation of the substance for the attenuated total reflectance technique is described separately.
6.1 Method 1
The solid substance should be triturated with dry, finely powdered potassium halide (normally potassium bromide). When hydrochlorides are being examined, potassium chloride should be employed to avoid the risk of halide exchange.
The ratio of substance to halide salt should be about 1 to 200-300, e.g. 1.5 mg in 300 mg of the halide salt in the case of prism instruments, or about 1.0 mg in 300 mg of the halide salt for grating or Fourier transform instruments. The mixture should be carefully ground by means of an agate mortar and pestle for 1 minute. In exceptional cases, the use of a ball mill may be indicated, but the resulting risk of producing polymorphic changes generally outweighs any improvement in resolution. The triturate should then be uniformly spread in a suitable die and compressed, under vacuum, at a pressure of about 800 MPa. As an alternative, potassium halide discs can be prepared by means of a hand-held minipress. The disc thus produced is mounted in a suitable holder.
Several factors, e.g. inadequate or excessive grinding or moisture or other impurities in the halide carrier, may give rise to unsatisfactory discs. Unless its preparation presents particular difficulties, a disc should be rejected if visual inspection shows lack of uniformity or if the transmittance at about 2000 cm-1 (5 µm), in the absence of a specific absorption band, is less than 75% without compensation.
The quality of a spectrum is often improved by placing a blank disc of the appropriate potassium halide, of similar thickness to that of the sample disc, in the reference beam.
6.2 Method 2
A small quantity’ of the finely ground substance should be triturated with the minimum amount of a suitable mineral oil (e.g. Nujol) or other suitable liquid to give a smooth creamy paste; 10 mg of the substance to be examined combined with 1 - 2 drops of mineral oil is often sufficient to prepare a satisfactory mull. The prepared mull should appear opaque.
A portion of the mull is then compressed between two flat sodium chloride or other suitable halide-salt plates.
If the spectrum of the mineral oil used interferes with regions of interest, an additional dispersion of the substance in a medium such as a suitable fluorinated hydrocarbon oil or hexachlorobutadiene R is prepared, and the spectrum recorded in those regions where the mineral oil shows strong absorption.
6.3 Method 3
A capillary film of the liquid held between two sodium chloride plates or a filled cell of suitable thickness is used.
6.4 Method 4
A solution in a suitable solvent is prepared and a concentration and cell thickness are chosen to give a satisfactory spectrum over a sufficiently wide wave number range. Generally, good spectra are obtained with concentrations of 1-10% w/v for a cell thickness of 0.1-0.5 mm. To compensate for the absorption of the solvent, a cell of matched path-length containing the solvent used is placed in the reference beam or a spectrum of the solvent is obtained so as to permit differentiation between solvent and sample absorptions. Alternatively, the solvent absorbance spectrum versus air may be subtracted from the solution spectrum versus air to obtain the absorbance spectrum of the solute. (When an FTIR instrument is used, the spectrum of the solvent recorded under identical conditions can be subtracted digitally.)
6.5 Method 5
Gases are examined in a cell with windows transparent to infrared radiation and having an optical path-length of about 100 mm. The cell is evacuated and filled to the desired pressure through a stopcock or needle valve by means of a suitable gas-transfer line between the cell and the container of the substance to be examined. If necessary, the pressure in the cell is adjusted to atmospheric pressure with a gas transparent to infrared radiation (e.g. nitrogen R or argon R). To avoid absorption interferences due to water, carbon dioxide or other atmospheric gases, an identical cell that is either evacuated or filled with the gas transparent to infrared radiation is placed in the reference beam.
7. Identification by reference substance
Both the substance to be examined and the reference substance are prepared by means of the same method and the spectrum of each from about 4000 to 600 cm-1 (2.5-16.7 µm) is recorded. The concentration of the substance should be such that the strongest peak attributable to it corresponds to a transmittance of about 10%.
If the positions and relative intensities of the absorbance maxima in the spectrum of the substance to be examined are not concordant with those of the spectrum of the reference substance when spectra are obtained by methods 1 or 2, this may be the consequence of differences in crystalline form. To avoid this difficulty, one of the procedures described below may be used for both the substance to be examined and the reference substance:
• Solutions of the reference substance and of the sample, of a suitable concentration, are prepared as described in method 4.
• A small amount (2 or 3 drops) of a concentrated solution in a volatile organic solvent is placed on a blank disc of potassium halide and evaporated to dryness in an oven at 105 °C.
• A small amount (2 or 3 drops) of concentrated solution in a volatile organic solvent is mixed with 300 mg of potassium halide and evaporated to dryness in an oven at 105 °C. Both the reference substance and the substance to be examined are treated in the same manner and then prepared as described in method 1.
• Both the reference substance and the substance to be examined are recrystallized from a suitable solvent.
8. Identification by reference spectrum
The substance to be examined is prepared exactly as described in the note accompanying the International Infrared Reference Spectrum and the spectrum from about 4000 to 600 cm-1 (2.5-16.7 µm) recorded by means of an instrument that is checked frequently to ensure that it meets the standards of performance required. The reference maxima of a polystyrene film should be superimposed on the spectrum of the substance to be examined at about 2851 cm-1 (3.5 µm), 1601 cm-1 (6.25 µm) and 1028 cm-1 (9.73 µm). Other suitable polystyrene bands can be superimposed if interference occurs with the bands of the substance. If these polystyrene maxima are taken into account, the identification is considered to be positive if the principal absorbance maxima in the spectrum of the substance to be examined are concordant with the corresponding maxima in the relevant International Infrared Reference Spectrum. When the two spectra are compared, care should be taken to allow for the possibility of differences in resolving power between the instrument on which the International Infrared Reference Spectrum was prepared and that being used to examine the substance. An International Infrared Reference Spectrum of polystyrene recorded on the same instrument as the collection of the reference spectra should be used for assessing these differences. The greatest variation due to differences in resolving power is likely to occur in the region between 4000 and 2000 cm-1 (2.5 and 5 µm). However, if the positions and relative intensities of the absorbance maxima in the spectrum of the substance to be examined are not concordant with those of the reference spectrum when methods 1 or 2 are used, this may be due to differences in crystalline form. Another procedure as described in section 7, will then be indicated in the note accompanying the reference spectrum.
9. Reflectance techniques
9.1 Attenuated total reflectance technique
The attenuated total reflectance (ATR) technique is best adapted to smooth, flexible surfaces, such as various plastics, or to strongly absorbing liquids and solutions, but can also be employed to determine the infrared absorption spectra of solid substances. It is usually necessary to reduce the solid substance to a fine powder, which is then packed directly against the reflecting element of the attachment. Alternatively, an adhesive tape can be used to facilitate the contact, the powdered substance being spread on the adhesive side of the tape to form an almost translucent layer, after which the powdered side of the tape is pressed on to the reflecting element. The backing plate is then attached, or moderate pressure applied by means of a suitable clamp for 1-2 minutes. Finally, the reflecting element is placed in the holder. The tape used in the procedure should preferably contain a natural rubber adhesive. Some plastic materials may be placed directly on to the reflecting element.
Reflective elements are usually made of zinc selenide (refractive index = 2.3) or germanium (refractive index = 4.0). The correct alignment of the attachment in the apparatus should be carefully checked.
9.2 Diffuse reflectance
In this technique, the surface of a sample reflects light in many different directions. The solid substance is reduced to a fine powder with a non-absorbing matrix (potassium bromide or chloride is suitable for this purpose). The mixture is placed directly in the sample cup holder of the diffuse reflectance instrument. The spectrum of the matrix recorded under identical conditions should be subtracted digitally. Some plastic materials can be placed directly in the sample cup holder of the diffuse reflectance accessory.