Many biomolecules are only active in aqueous solution, and after dehydration, not only their biological activity changes greatly, but also their structure undergoes certain changes. Therefore, solvent water is irreplaceable in the study of active biomolecules. Water is a strongly polar molecule, and there are wide and strong infrared absorption bands in the 3400 and 1640 cm-1 regions, and the absorption bands of water often overlap or partially overlap with the absorption bands of the biomolecule of interest, which makes the infrared spectroscopy of biomolecules complicated and highly difficult. In early infrared textbooks, it was often said that "infrared spectroscopy is not suitable for the analysis of aqueous solutions or aqueous substances", which illustrates the severity of water peak interference. In many practices, in order to obtain a high-quality infrared spectrum of a substance of interest, the infrared absorption peaks (interfering peaks) of water must be subtracted.
One way to remove the water absorption bands is to replace water (H2O) with deuterated water (D2O) as a solvent. Due to isotopic effects, the IR band positions of D2O and H2O are significantly different. However, active hydrogen (—Cooh, —NH2, etc.) is present in many biomolecules, and the exchange reaction between active hydrogen (H) and deuterium (D) in D2O can occur. In general, to ensure that the exchange reaction is complete, the D2O solution must be left for a long enough time or undergo multiple dissolution processes. Hydrogen deuterium exchange can cause some infrared absorption bands to move and even cause local conformational changes in biomolecules. More seriously, the study of the structure of proteins in aqueous (H2O) solution and their physiological activity will also be of little practical significance, as the structure of proteins in D2O may differ from their structure in H2O.
In addition to the use of D2O solvents, spectral subtraction is also widely used to subtract disturbances from water peaks. In this technique, the spectral difference between the infrared absorbance spectrum AS of the solution and the absorbance spectrum AW of the solvent water (or aqueous reference solution) is subtracted to obtain a spectrum that subtracts the interference of water absorption peaks. Although this research has been carried out for many years and has become the main means to obtain infrared spectra of aqueous solutions, there is still room for improvement in the reproducibility of spectra and the uncertainty of deviation.
When measuring infrared spectra, the single beam spectrum of the sample to be measured (sample single beam spectrum) and the single spectrum spectrum (background single beam spectrum) of the reference sample are usually obtained, respectively, and the ratio of the two is the transmittance spectrum of the sample to be measured. According to the Lambert-Beale law, if the water is exactly the same in the background spectrum and in the sample spectrum, the absorption peak of the water will not appear on the final infrared spectrum. However, it is difficult to control the consistency of the amount of water in the reference sample and the sample to be measured. The Attenuated Total Reflection (ATR) accessory is currently the conventional method for obtaining infrared spectra of aqueous solutions. However, until now, it has been nearly impossible to achieve a consensus on the total number of effective molecules of water in the reference sample and the sample to be measured using ATR measurement techniques.
To measure infrared spectroscopy, the sample single beam spectrum of the solution (K2CO3 solution or BSA solution) is measured first, followed by the background single beam spectrum of the reference sample. Unlike traditional measurement methods, a double background sample is used to collect the background spectrum. First, scan the empty ATR crystal n times (n should be large enough), pause, fill the ATR crystal with reference solution (15% NaCl solution or pure water), continue scanning, observe the infrared spectrum map, you can see the whole process of the absorption peak of water disappearing from strength to weakness, and stop scanning when the infrared spectrum without the interference of anhydrous absorption peaks is obtained. The block diagram of the experimental route is shown in Figure 1.
The infrared spectrum of pure water may be somewhat different from that of water in solution, including the position of the bands, the shape of the absorption peaks, etc. Therefore, when preparing the reference sample, it is required that the state of the water in the reference sample be as consistent as possible with the state of the water in the sample to be measured. The infrared spectra of pure water and water in 10% BSA solution are basically the same, so pure water is used as the reference sample in this case. But the infrared spectrum of water in a 10% potassium carbonate solution is different from that of pure water. According to the report, the infrared spectrum of water in potassium carbonate sodium chloride mixed salt solution is very similar to that of water in sodium chloride salt solution, therefore, in the case of 10% potassium carbonate solution, 15% sodium chloride solution is selected as the reference sample.
When measured with a single ATR accessory, if the empty germanium crystal is used as the reference sample and the 10% potassium carbonate solution is used as the sample to be measured, the absorption peak intensity of water at 1640 cm-1 is approximately 0075 (absorbance). If 15% sodium chloride solution is used as the reference sample, and 10% potassium carbonate solution is also used as the test solution, a negative but small water absorption peak occurs at 1640 cm-1. This is because a 10% potassium carbonate solution contains less water as compared to a 15% sodium chloride solution. The reason for the small absorption peak is that the water content of the two solutions, although there is a difference, is very small.
With different reference samples (empty germanium crystals or sodium chloride solution), the absorption peaks of water (potassium carbonate solution) were positive and negative. So, when measuring the background spectrum, first use a blank germanium crystal as a reference sample, scan it many times, pause, and then fill the 15% sodium chloride solution on the germanium crystal, and then continue scanning, as the number of scans increases, what will happen?
Figure 2 shows the experimental results. 1643 cm-1 is the absorption peak of water, and 1400 cm-1 is the absorption peak of carbonate (CO23+). In the experiment, the sample single beam spectrum of the 10% K2CO3 aqueous solution of the sample to be tested was first measured, and then the blank germanium crystal was used as the reference sample according to the above description, and the background spectrum (blank germanium crystal) was scanned 20 times, and then paused. Figure 2a is obtained, where the absorption peak of water (1643 cm-1) is positive. At 1,2 ,...The intensity of the absorption peaks of carbonate and water remained unchanged over the 20 scan accumulations. After the pause, the reference sample was replaced with a 15% sodium chloride aqueous solution, and the background spectrum was scanned again. With the addition of new scans, the absorption intensity of the carbonate continues to remain the same (since the 15% sodium chloride solution does not contain carbonate). However, with each additional scan, the water absorption decreases slightly (15% sodium chloride is greater than 10% potassium carbonate). When the sodium chloride solution was scanned 48 times, the water peak at 1643 cm-1 had become significantly smaller, as shown in Figure 2b. With each additional scan of the aqueous sodium chloride solution, the water absorption peak becomes smaller. By the time the number of scans reaches 300, the absorption peak of the water has completely disappeared, at which point stopping the scan yields a satisfactory infrared spectrum free of interference from the absorption peak, see Figure 2c. The experiment was not operated using differential spectroscopy. Therefore, when actually measuring, the ordinate can be in units such as absorbance, reflectance or transmittance.
Figure 2c shows the direct acquisition of IR spectra without water-interfering peaks using ATR techniques. It is important to emphasize that the scanning sequence of the two reference samples has an important influence on the quality of the infrared spectrum obtained. The water content of blank germanium crystals (water 0) and 10% potassium carbonate solution is quite different, while the water content of 15% sodium chloride solution and 10% potassium carbonate solution is very small. Therefore, the blank germanium crystal was scanned as the first reference sample, so that the IR peak of the water showed a large absorption peak, as shown in Figure 2a. It is easy to understand that in the second stage, the strong water absorption peak gradually decreases to obtain the desired deduction effect. If the 15% sodium chloride solution is the first reference sample, the water absorption peak (negative) is small but not negligible. In the second stage, the process of disappearance of the weak absorption peak (short time) is not conducive to observation. The number of scans (n) of the first reference sample (blank germanium crystal) should not be too small, the larger the number of scans n, the better the deduction effect, but also the more accumulation time is required. In order to obtain a satisfactory signal-to-noise ratio, the number of scans of the first reference sample should be controlled between 20 and 32 times. The actual number of scans of the second reference sample, m, must be selected in real time based on the change in intensity of the water peak being deducted. The principle is that the water disturbance peak signal is negligible and stops the scan.
In order to evaluate the effect of the new method on subtracting water peaks, we compared the results with the traditional differential spectroscopy technique. The infrared spectra of 10% potassium carbonate and 15% sodium chloride were measured using blank germanium crystals as reference samples. Select the appropriate subtractive factor and subtract the infrared spectra of the two solutions to obtain Figure 3a.
From the comparison in Figure 3, it can be concluded that the new two-reference sample method (Fig. 3b) has a better effect and a significant improvement in the signal-to-noise ratio, especially in the 1640 cm-1 range. Another advantage of the double reference sample is that it is easy to operate and can be monitored in real time as needed. In the single ATR attachment, the double reference sample method successfully subtracted water interference with good results.
Double reference sample (blank germanium crystal + water) with background spectra containing information about water and air. The required background spectrum is information about the appropriate thickness of the water layer. The double reference method can only achieve good results if the single beam spectrum of the double reference sample is highly similar to the single beam spectrum of water. The limitations of this approach are discussed below.
Figure 4 shows the infrared spectrum obtained using the two-sample method. The reference samples were all air, and the samples to be tested were either polystyrene (PS) (Figure 4a) or double (polystyrene + air). Figure 4a shows the infrared spectrum of the PS film, and Figure 4b, c, d, and e show the infrared spectrum of the double sample (polystyrene + air). The spectra in Figure 4b, c, d, and e are all significantly distorted compared to Figure 4a. i.e., the spectra in Figure 4b, c, d, and e are not representative of the infrared spectrum of PS. Under what conditions can the infrared spectra of air and polystyrene samples be highly similar to those of pure polystyrene?
Enlarge the 2000 1640cm-1 area in Figure 4 to obtain Figure 5. Comparing the peaks in Figure 5b, c, d, and e with their corresponding peaks in a, the intensities are different, but the peak shape distortion is not obvious. There is no strong absorption peak in the 2000 1640cm-1 region, and the absorption of infrared light is small. Therefore, the absorbance of the infrared peak is an important factor affecting the distortion. The weaker the absorption peak, i.e., the smaller the absorbance a, the smaller the degree of distortion. The stronger the absorption, as shown in Figure 4 with an asterisk absorption band, the more severe the distortion.
As you can imagine, another factor influencing distortion is the proportion of the real matter (PS) signal in the final spectrum in the double sample. M scans for PS and n scans for air, for a total of (n+m) scans. When n is equal to zero, it is equal to the spectrum of ps, without distortion. When M is equal to zero, the spectrum of air is obtained, and the PS spectrum is used as the judgment standard, and the distortion is the most serious. Distortion can be judged by m (n+m), and it is not distorted when the value is equal to the maximum value of 1. The larger the value of m (n+m) (closer to 1), the less distortion is obtained from the cumulative spectrum.
Figure 6 shows the ATR spectrum for water, a mixed ATR spectrum for water and air. All the reference samples in Figure 6 were blank GE crystals, i.e., no samples were added to the surface of the germanium crystals, and the background spectrum was scanned 32 times. Figure 6a shows the ATR spectrum of water, with 64 water samples taken. From Figure 6a, it is found that the absorbance at the strongest peak of water at 3400 cm-1 is about 0 in a single ATR accessory measurement3. The absorbance at 1640cm-1 is about 0075。In Figure 6b, C and D, the samples were collected by scanning the water sample several times and scanning the empty ATR crystal several times, so the ATR spectra in Figure 6b, C and D are mixed double sample (water + air) spectra.
The peak patterns of the mixed spectra 6b and 6a (pure water ATR spectrum) are highly similar, as illustrated by the difference between them (Figure 6F). m (n+m) = 48 (16+48) = 0. in Figure 6b75, that is, for a single ATR accessory measurement, as long as m (n+m) is greater than 075, the resulting mixed spectra are not significantly distorted, and can be used to obtain high-quality infrared spectroscopy measurements with water interference removed. Mixed spectrum 6d, m (n+m) = 025 Compared with the water spectrum 6a, there is a significant distortion, so the difference between Figure 6a and Figure 6d is far from the ideal straight line, see Figure 6h. In practice, as long as the water content of the reference solution (e.g., 15% NaCl solution) is slightly higher than that of the solution to be measured (10% potassium carbonate solution) for a single ATR attachment, the calculations show that high-quality infrared spectra without water peak interference can be obtained by this new method.