SU1497538A1 - Method of determining residual moisture in dry substances by nuclear magnetic resonance technique - Google Patents

Abstract

The invention relates to biotechnology, in particular, to the analysis of dry microbial preparations, enzymes and other proton-containing solids, and can be used in microbiological, pharmaceutical, medical, food and other areas of the national economy. The aim of the invention is to extend the measurement range towards low residual moisture. In the method, the entire visible part of the free induction signal is recorded, the amplitudes of the slow and fast components of the free induction decay at zero time are determined, the transverse relaxation time of the slow component and the humidity are determined according to a definite dependence. By the time of transverse relaxation of the slow component, the quality of the drying of the sample is judged by comparison with the reference value. 3 il.

Description

In the method, the entire visible part of the FID is measured after being exposed to a 90 HF high frequency pulse, the amplitude of the FID components (H and Hj) is determined at time zero (in the middle of the high frequency pulse)

Residual moisture can be determined {) eaten out of a linear relationship.

,(one)

where B is the residual moisture of the biological product in g per 100 g of dry residue; X is the ratio between the amplitudes of the slow and fast component of the FID,%; K - calibration factor

equal to the specific mass content of protons in the dry residue relative to water; D is a calibration factor equal to the value of KX of the dry residue.

Characterizing a biological product by residual moisture, it is important to know not only the amount of moisture, but also how it is distributed in the biological product. The number of moisture distribution on the one side is due to the material composition of the biological product, and on the other, it can be created by the drying process itself.

In the first case, the water in the dry bis preparation is unevenly distributed between the components included in its composition. For example, microbiological biological preparations contain a protective medium (up to 40% of the dry weight of a biological preparation), which is dried almost completely to 0.2-0.4%, while the cells of the microorganisms have a moisture content of 6-7% after in a single biological preparation, these components create an average moisture content. If the composition of a biological preparation can vary, then the average humidity obviously will not reflect the humidity of the microorganism cells entering it. Excessive moisture in the cells of microorganisms can be compensated for by an excess amount of protective environment and the average humidity is preserved. However, the excessive moisture of the cells in gdvt to an increase in Tj, which can be controlled by the NMR method.

In the second case, the uneven distribution of moisture in the biological product

associated with the drying process itself. It occurs when dried in a thick layer due to marginal effects, or as a result of thawing, during lyophilization (loss of vacuum, non-observance of the temperature mode of drying, etc.). All this leads to an increase in the effective value of the slow component.

The magnitude of the transverse relaxation of the slow component in both the above cases, reflecting the state of water in the sample, characterizes the quality of drying.

The method is carried out as follows.

The sample is packaged in a vial and placed in a heating chamber for preheating the sample, for example, up to 38 ° C. The temperature is not critical, its stability is important both for calibration and for the measurement process. This is the preliminary preparation of the sample. When measuring a large number of samples, such preparation only slightly increases the time by one measurement,

A sample of the heating cabinet is then placed in the NMR sensor of the instrument. A 90 ° pulse is applied and the entire FID is measured using an amplitude digital converter (ADP). The result is stored in a microprocessor buffer memory. The sample temperature during the measurement is maintained the same as in the heating cabinet. The amplitudes of the slow (H) and fast (H) component of the FID at the zero time point and the transverse relaxation time of the slow component are determined by approximating the visible parts of the signal with Gaussian functions. From the ratio of these components, we obtain the value of X.

In the range of low residual moisture (2–9%), the K value in the formula (O can be determined by one of three methods: I calculated theoretically if the basic chemical composition of the biological product is known, the number of protons per unit weight of the dry part is determined and divided by value for water

2

which is equal to 73 ii determined ex 1 about

perimeterly from the ratio of the NMR amplitudes per unit weight for the same volume of dry biopsy

parata and water; III by constructing a linear calibration dependence for a series of reference samples using the least squares method.

The D value is defined as the KX value for a completely dry sample. It is most convenient to determine it with samples with a known humidity.

For a residual moisture range of 9%, the values of the calibration coefficients K and D are preserved if the composition of the biopreparation does not include readily water-soluble substances (for example, enzymes, DNA, etc.). In the presence of readily soluble substances in the composition of the biological product, the linear relationship (1) remains, the value of K decreases, and the value of K and D are determined according to method III.

In terms of the transverse relaxation time of the slow component, the quality of sample drying is compared by comparison with the reference value.

Example 1. Consider the measurement of residual moisture in Baker's yeast. To obtain samples with different humidity, the lyophilized present yeast cells were placed for a time in a desiccator in an atmosphere of 100% relative humidity.

After the desiccator, the sample was placed in an ampoule and kept in a heating cabinet at.

The obtained samples were initially measured on an SxP-100 NMR spectrometer, then the same sample was dried (105-1 hours). The measurement results are shown in FIG. 2,

The free induction signal was measured at a frequency of 90 MHz after 90 ° c.h. pulse 2.5 µs. The launch period is equal to ls - 3s, the number of savings is 10,

Consequently, the measurement itself took 30 seconds. FID was measured every 0.5 ps. All amplitudes minus the baseline were recorded in 1024 microprocessor channels.

The sample weight was 200 mg. The slow and fast component of the FID was approximated by Gaussian functions, the approximation zones were chosen at Ou 50–70 μs, 0.2 10–20 μs (Fig. 1). The time for mathematical processing was Ci i O c,

The value of K was calculated theoretically and was equal, 11

0

five

0

five

The experimental value is obtained from the graph

, 27iO, 14, .D (19, li2,6)%.

The indicated correspondence of K values when measuring a wide range of objects greatly simplifies and shortens the calibration process. For most microbial biologics, biopolymers, and other protic substances, the value is therefore a determination of the value of this coefficient is eliminated.

The NMR error of measuring the residual moisture in absolute units was ± 0.5%.

Measurement of the residual moisture in these samples according to the prototype method turned out to be practically impossible due to the large measurement error, which was ± 3%.

For a residual moisture range of 9%, the NMR link of the parameter X and B remains linear with the coefficients

, 27 ± 0,016,, 91 ± 0,49%.

The coefficients are obtained by the method of least squares according to the graph in FIG. 2

The time T. of the slow component was 150 µs. With non-uniform drying, partial thawing during lyophilization, the T value increased to 400 μs or more.

Example2. As an object of nonbiological nature, we take a proton-containing substance that adsorbs water and does not have substances soluble in water. An example of this is ion exchange resins, for example

five

45

measures, KB-4P2 in sodium form.

The difference between the present measurements and Example 1 is that zone 0 (Fig. 1) was from 100 µs to 200 µs. The results of NMR measurements and drying (105 ° - 1 h) are shown in FIG. 3. In the range of residual humidity of 0-130%, a linear-Q dependence is observed.

The parameters of the gauge equation (3), obtained by the least squares method, are

, 42 ± 0.02,.

For comparison, the value of V was calculated theoretically and the chemical formula KB-4P2 in the sodium form: .42.

This shows that knowledge of the chemical composition can completely eliminate the problem of calibration.

The error in measuring the residual moisture for KB-41T2 did not exceed ± 0.4% in absolute units.

Analysis of various samples of ion exchange resin showed that the transverse relaxation time T slow Components for the same residual moisture can vary from batch to batch. The value of T / in some samples increased to two times. Moreover, this difference was preserved throughout the whole range of humidity.

The ion exchange resin, which has a shorter relaxation time T of the slow component, more strongly binds water. Therefore, in this case T allows sorting the resin by adsorption properties.

Claims (1)

Thus, the proposed method allows to measure the residual moisture of biological products starting with 2-3% with an error not exceeding 0.5%, and also to control the quality of drying of the samples. Invention Formula Method for determination of residual moisture in dry substances impulse nuclear magnetic resonance method, based on measuring the free induction signal (FID) after exposure to a dry substance containing protons, with a 90 ° high-frequency pulse and determining amplitude values of the slow and fast FID component, in order to the measurement range towards low residual moisture, record the entire visible part of the FID, find amplitudes of the component signals at zero time by extrapolation according to Gaussian law, and determine the residual Humidity linear dependence , where B is residual moisture; X is the ratio between the amplitudes of the slow and fast components of the FID; K is a calibration factor, equal to the specific mass content of protons in the dry residue relative to water; D is the KH value of the dry residue. V.ch. pulse Phage.G AT,% 12.5-SHO7 .5sA 25 About Oh. 100 76.0 yy.o 28.0 ao l.% Fig, g Century% ShchO120 .0– 90.0 shcho30, 0