Many different types of transitions appear in the far-infrared (FIR) region. FIR spectroscopy is therefore an effective and promising technique for obtaining spectroscopic and structural information on organic and inorganic compounds, polymers, charge–transfer complexes, semiconductors, ionic liquids, and metal clusters.1–9
Atmospheric water vapor exhibits strong spectral absorption corresponding to pure rotational transitions in nearly the entire FIR region.1,10 These water vapor lines may severely obscure the features of the sample. Thus, removal of water vapor disturbance is a key problem with measurement in the FIR. The following three methods have been customarily used to reduce water vapor interferences: 1) employment of a sample shuttle system to alternate background and sample collection under an atmosphere of low and constant humidity, 2) removal of water vapor by purging the instrument with dry N2 or dry air, and 3) evacuating water vapor under vacuum.
The purpose of the purge system or vacuum system is to remove water vapor completely from the spectrometer. A vacuum system prevents atmospheric water interferences completely, but this option is costly and may limit sampling capabilities. Purging the system with N2 or dry air may be a good choice, but is time consuming and sometimes insufficient to prevent water vapor lines from interfering.1 The sample shuttle device works well because the humidity in the spectrometer remains constant during the entire measurement period, and the water vapor quantities in the background and sample measurements are congruent. Therefore, the water vapor peaks are absent. However, the shuttle system is not ideal for some sampling methods, such as attenuated total reflectance (ATR) and diffuse reflectance.10
In general, the sample compartment in the spectrometer has to be open to the atmosphere to allow for sample or background exchange. In this way, the moist air enters the sample compartment and the relative humidity changes with time. In such a case, if an FIR spectrum is recorded immediately after sample introduction, water vapor bands will appear during the recording period.
This paper discusses the method of wet air and dry air spectral titration to effectively remove water vapor interferences. The appearance of water vapor peaks is allowed at the first stage of the collection procedure of the sample single-beam spectrum; at the second stage, water vapor peaks become smaller and smaller by wet or dry air spectral titration. A good FIR spectrum can be obtained by terminating the collection of the spectra when the water vapor bands become unobservable. This is a rapid method for collecting a qualified FIR spectrum under conditions of fluctuant humidity.
Experimental
Materials
L-histidine and D-galactose were purchased from Shanghai Bio Life Science & Technology Co. (Shanghai, China) and were used without further purification.
FIR spectra were measured at 4 cm–1 resolution with a Magna 750 spectrometer (Thermo Fisher Scientific [Nicolet], Madison, WI) equipped with a deuterated L-alanine doped triglycine sulfate (DLaTGS) detector with a polyethylene window, and solid-substrate beamsplitter covering the full FIR spectral region. One hundred scans were accumulated for each background spectrum. The relative humidity and temperature of the laboratory room were around 60% and 20 °C, respectively.
Results and discussion
Measurement methods
Water molecules strongly absorb FIR radiation. If the air humidity level inside the spectrometer is too high, the 100% line becomes very poor, even though the humidity remains constant during the background and sample measurements. The anomalous absorption may account for the poor 100% line at high humidity (greater than 20%~30%).11 In the present study, the relative humidity in the spectrometer was always kept lower than 20% during the measurements. This can be done easily by placing strong desiccants and/or blowing dry air into the spectrometer.
Figure 1 - Diagram of the changes in water vapor content from wet to dry in the spectrometer during the recording period of a sample single-beam spectrum.
The background single-beam spectrum was measured under an ambient fluctuant humidity atmosphere immediately after opening and closing the door of the sample compartment (to simulate real life). Measurement of the sample single-beam spectrum is illustrated in Figure 1. The average of the relative humidity in the spectrometer was always lower than 20%, although the humidity in the sample compartment can sometimes be higher than 20%. The period from the scanning start to scanning number Nb is called the wet air stage or first stage. The next stage from the scanning number Nb to Nx is called the dry air stage. Usually the sample spectrum is collected as soon as the door of the sample compartment is closed. Therefore, the water vapor concentration at the first stage is fluctuant. After the scanning number reaches N1, the humidity level of the first stage can be determined by viewing the live spectrum, which is displayed in real time using the multiscan function of the spectrometer. In Figure 1, the first stage is wet. The second stage should be dry in order to titrate the moist air at the first stage. Thus, dry air or N2 is needed between the two stages. Conversely, if the first stage is dry, then an operation to introduce wet air into the spectrometer is needed to make the second stage wet.
Keeping water vapor concentration constant in the spectrometer is very difficult during a period of rapid measurement. In the present method, the relative humidity is controlled so that it is lower or higher than that of the background vapor level. The humidity control is qualitative but not quantitative. Therefore, this can be done easily by a dry air blow or simply by opening the lid of the sample compartment and introducing room air (as a wet air supply) into the sample compartment.
A large N1 value should be selected to ensure a good signal-to-noise ratio. Figure 1 also shows that the scanning number, N1, at the first stage should be large enough for the necessary titration. As a rule, a large N1 needs a large Nx to match, and a large Nx allows wet/dry titration to proceed gradually. If an appropriate Nx is selected, the water vapor peaks in the FIR spectrum can be removed completely.