JP2008083040A - Liquid chromatograph unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem wherein a response 50 ms is not sufficiently short when calculating the number N of theoretical plates, in a peak having 1.0 s of half-value width w<SB>1/2</SB>, i.e. about 0.4 s of standard deviation δ, and to calculate the accurate number N of theoretical plates. <P>SOLUTION: This liquid chromatograph unit of the present invention has a features of setting a proper detection response in a time of a peak width, and a proper data collection interval. The proper detection response in the time of the peak width, and the proper data collection interval, are required to be set because the number N of theoretical plates gets low both in too large and too small detection responses and data collection intervals. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a liquid chromatograph apparatus, and more particularly to a technique suitable for a high-speed liquid chromatograph apparatus for high-speed analysis.

  In the technical field of high performance liquid chromatography, the particle size of column fillers is being refined. In recent years, the chromatogram analysis time is about 1 minute by using a filler having a particle size of 2 μm or less. In addition, monolithic columns that do not use particle-type fillers are also widespread, and the analysis time is similarly realized in about 1 minute. There exists patent document 1 as a thing relevant to this technique.

As the peak retention time t _R (appearance time) of the chromatogram becomes earlier, the width of the peak becomes narrower in proportion to the retention time. The holding time _{t R} is 30s, the number of theoretical plates N (Japanese Pharmacopoeia, 15th Edition: proportional coefficient 5.54) if 5,000 stages, the half-width _{w 1/2} of the peak is 1.0 s. The theoretical plate number N is widely used as a figure of merit for a column indicating good separation.

  If the peak waveform is ideally in the form of a normal distribution function (Gaussian), the theoretical plate number N is as follows. σ is the standard deviation of Gaussian.

For example, the standard deviation σ described above is 0.42 s. Incidentally, the coefficient 5.54 of Equation 1 is derived from 8 log _e 2 that relates the half-value width w _1/2 of the Gaussian peak and the standard deviation σ.
It is known that the theoretical plate number N becomes worse when affected by the response of the detection signal (corresponding to the time constant τ). In other words, if the peak shape is tailed by superimposing an exponential function in a transient manner of an AC circuit, the half-value width w _1/2 is widened and the theoretical plate number N is reduced.

  In general, the spread of the peak waveform is analyzed by the variance V (secondary moment). The Gaussian on which the exponential function is superimposed becomes an exponentially modified Gaussian EMG (Exponentially Modified Gaussian). The variance V is expressed by the following equation.

The reason why the theoretical plate number N is affected by 1% by the time constant τ can be said to be that the time constant τ is about 10% of the standard deviation σ because Equation 3 is the sum of squares. Describing along the above example, when the peak half width w _1/2 is 1.0 s, the theoretical plate number N is estimated to be affected by about 1% when the time constant τ is 0.04 to 0.05 s. . In other words, if the time constant τ is set to 0.04 to 0.05 s, it is about 1% for a peak having a retention time t _R of 30 s and a true theoretical plate number N of 5,000. However, it is estimated that the number of theoretical plates N is not affected. Furthermore, since the number of theoretical plates N in the Japanese Pharmacopoeia is calculated using the half-value width w _1/2 as represented by Equation 1, the degree of influence by the time constant τ is expressed using Equation 3. Is less than 1% after all. This argument is the same even when the true theoretical plate number N is doubled to 10,000, and if the time constant τ is reduced by 1/2 to 0.3 s, 1% will not be affected. . In general, the number of theoretical plates N is not small in measurement uncertainty, and it is rare to discuss a significant difference of 1%, and a practically small time constant τ is not required so far.

In summary, if the retention time t _R is 30 s or the peak at half width w _1/2 is 1.0 s, the time constant τ is set to 0.05 s, that is, 50 ms. It was thought that it was not particularly necessary to set a smaller time constant τ.

JP-A-7-103959

Conventionally, it was the above-mentioned background art recognition level. However, when the theoretical plate number N is calculated according to the experiment, the half width w _1/2 is 1.0 s, that is, the peak with the standard deviation σ of about 0.4 s is sufficiently short when the response is 50 ms. I have found that there is no. There are several possible causes for this.

  (1). If the data collection interval is not appropriate, the peak height cannot be measured accurately.

  (2). If the data collection interval is not appropriate, the peak half width cannot be measured accurately.

  (3). If the data collection interval is not appropriate, it is affected by detection error (uncertainty).

  (4). If the response is not appropriate, it is affected by detection error (uncertainty).

  A feature of the present invention is that an appropriate detection response and an appropriate data collection interval are set for the peak width time. If the detection response and the data collection interval are too large or too small, the number of theoretical plates is lowered. Therefore, it is necessary to set an appropriate detection response and an appropriate data collection interval for the peak width time.

  According to the present invention, a more accurate theoretical plate number can be calculated.

  A method for speeding up the high-speed liquid chromatograph apparatus will be described, and a mode for setting an appropriate detection response and an appropriate data collection interval will be described.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 shows a system configuration example of the present invention. The liquid chromatograph apparatus 1 includes an analytical instrument module pump 2, an autosampler (injection apparatus) 3, a column oven 4, a detector 5, and a data processing apparatus 10. The data processing apparatus 10 includes a system control unit 6 and a data processing unit 7, and the system control unit 6 includes an analytical instrument control unit 8 and a parameter storage unit 9. At the time of analysis, a command is issued from the system control unit 6 of the data processing apparatus 10, and a series of parameters of the analysis sequence is transmitted (downloaded) to the pump 2, autosampler 3, column oven 4, and detector 5 of each module. . Each time the sample is injected, the data processing unit 7 digitally receives the detection signal from the detector 5, temporarily forms it as a chromatogram waveform, and analyzes the data.

  FIG. 2 shows a flow path configuration example of the present invention. The liquid chromatograph device 1 includes a pump 12a for feeding the eluent 11a and a pump 12b for feeding the eluent 11b as a liquid feeding system. In this example, a so-called high pressure gradient elution method in which two eluents are mixed by a joint 13 and a mixer 14 under high pressure is used. A sample is injected from the autosampler 15 and fed into an analysis column 16 (for example, an inner diameter of 2 mm, a length of 50 mm, and a filler particle size of 2 μm). Each solute in the sample is separated and developed in the analysis column (stationary phase) 16 by the eluent (mobile phase) 11a or the eluent (mobile phase) 11b. Each solute has a different retention time due to the difference in partition ratio present in the mobile phase or in the stationary phase. In order to stabilize the retention time of each solute, the analytical column 16 is kept constant using the column oven 17. Each separated and developed solute reaches the cell 20 of the detector 18 with a difference in retention time, and is discharged into the waste liquid tank 19.

For the description of the present invention, the process from detection to data processing will be described in detail from the viewpoint of time resolution. The detector 5 includes an ultraviolet (UV) detector, a visible (VIS) detector, a diode array detector (DAD), a fluorescence (FL) detector, a differential refractive index (RI) detector, a conductivity detector, and the like. It can be used. In this example, a typical UV detector is used, and a flow cell having an optical path length of 5 mm and a light transmitting volume of 0.98 μL is mounted. In order to calculate the absorbance, each of the reference light amount and the sample light amount is detected using a detection photocell. The primary detection signal is analog, but each AD is quickly converted. Once the absorbance is calculated and the signal processing is repeated, the previous digital value and the current digital value are repeatedly weighted and weighted to obtain a response equivalent to the time constant τ (characteristic response time). A detection signal having the same can be generated. In this system example, the response is 0.01, 0.02,
A selection of 0.05 is possible. A time interval between detection data points for sending the detection signal to the data processing apparatus 10 is defined as a data collection interval. In this system example, it is possible to select data collection intervals of 10, 20, 50, and 100 ms.

  FIG. 3 is a plot of an appropriate response τ (s) against the peak width σ (s). Using a log-log graph, the value estimated to be optimum is represented by a center line (empirical value), and a line is drawn with the upper and lower limits multiplied by √10 and 1 / √10 times as a guide.

  Similarly, FIG. 4 is a plot of an appropriate data collection interval ΔD (s) against the peak width σ (s). Similarly, the upper and lower limits were similarly drawn with a line of √10 times and 1 / √10 times as a guideline.

  In FIG. 3 and FIG. 4 of the graph, the peak width σ is shown from 0.001 s for the following reason. Going back in the history of liquid chromatography, the column packing particle size has been reduced by an order of magnitude from 20 μm to 2 μm, so that it has become possible to perform analysis at about two orders of magnitude from tens of minutes to tens of seconds. Can do. Currently, the peak width σ is about 0.1 to 1.0 s. If it is assumed that the speed increase will be promoted by similar technical innovation in the future, the peak width σ is 0.001 to 0.01 s. It is based on the prediction that there is a possibility of targeting the peak of the degree.

  FIG. 5 shows an example of a chromatogram when the response and the data collection interval are both set to 50 ms and 10 ms, and Table 1 shows a comparison of theoretical plate numbers at the time of the chromatogram of FIG.

  In the case of the theoretical plate number N where the fast peak width σ is 0.26, 0.30, and 0.37 s, if the setting is changed from 50 ms to 10 ms for each of the response and the data collection interval, there is an effect of about 10%. Recognize.

In the present invention, from the viewpoint of speeding up liquid chromatography, the particle size of the column packing material is reduced, the column length L (column volume when multiplied by the column cross-sectional area) is shortened, and the linear velocity u (column cross-sectional area is reduced). This is effective when adopting a method of increasing the flow rate when multiplied. In short, the speeding up is to minimize the (elution) time t ₀ of the non-retaining peak. When the time t ₀ is shortened, the peak width σ (s) is proportional to and shortened. Actually, since the flow rate is mainly used at 1.0 ml / min, not only the parameter t ₀ related to time and the peak width σ (s) but also the column volume, the flow cell volume, etc. are proportionally small for speeding up. There is a need to. Otherwise, the volume of the flow cell cannot be ignored, and the volume occupied by the solute is widened and a tailing peak is generated as in the discussion of the time constant for the peak width.

  For example, in the previous embodiment, since the peak width σ is about 0.3 s, when the flow rate is 1.0 ml / min, the peak width σ is 5 μL in terms of volume, and the flow cell volume 0.98 μL has an effect. Therefore, a flow cell having a small volume (for example, 0.3 μL) is required.

The internal volume of the piping system also affects the spread outside the column and cannot be ignored. It is known that the spread outside the column (dispersion σext ² ) is proportional to the fourth power of the inner radius r of the pipe.

Here, L is the pipe length, F is the flow rate, and DE is the diffusion coefficient. In general, in this example, it is necessary to suppress the dispersion σext ² to about several tens of μL ^2, and it is desirable to use a tube having an inner diameter of 0.1 mm and a radius of r = 0.05 mm or less (eg, benzene DE = 10 ^{− 9} m ² s ⁻¹ ).

  In addition, it is known that not only the diffusion in the pipe but also the pipe connection portion and the structure in which the shape in the radial direction of the flow is enlarged or reduced, the diffusion is extremely large. For this reason, it is necessary to minimize the dimensional change in the radial direction.

  In addition, the discussion on the time and volume requires exactly the same response even in the case of the monolith column. A monolith column generally has a high porosity and a low load resistance with respect to a flow rate, so that the pressure loss is relatively small. Eventually, the flow rate is increased by increasing the flow rate or increasing the linear velocity than the packing column.

  The liquid chromatograph apparatus of the present invention is intended for high-speed analysis, and increases the linear velocity to a column having a certain length to increase the speed. The length of the column is roughly found by the required number of theoretical plates. Generally, the linear velocity is determined by the pressure resistance (upper limit) of the liquid chromatograph apparatus, and the upper limit of the high speed (minimum value of analysis time) is determined.

  In addition, liquid chromatographs currently used for general purposes are typically used with a flow rate of about 1 ml / min, and can be used at a lower level from the viewpoint of the amount of solvent (mobile phase) used. desirable. The requirement for a flow rate of 1 ml / min is also a very important input for defining various parameters, and contributes not only to optimizing the volume in the flow path but also to setting the column inner diameter.

  When requesting a theoretical plate number of 5,000 required as the standard separation performance, taking the holding time of 30 s as an example, the standard deviation is 0.42 s from Equation 1. The peak time interval (from the start point to the end point) is 3 to 4 s when the standard deviation is 8 to 10 times, and corresponds to a mobile phase volume of about 100 μL at a flow rate of 1 ml / min. When calculating the peak area, etc., it is said that 30 points from the start point to the end point are required as the number of peak formation points, and not only the time resolution (detection response and data collection interval) but also the flow cell. It is considered that the volumetric resolution of 1/30 times the time interval of the peak is also required for the characteristic size. Also, from the viewpoint of spreading outside the column (diffusion), it can be said that the optical path length volume of the detector flow cell is required to be 3 μL or less for the peak of this scale. In the case of a peak with a standard deviation of 0.42 s at a flow rate of 1 ml / min, 0.01 μL (10 nL) or more would be a practical level because there is an adverse effect such as insufficient light quantity even if it is too small.

  In connection with the aforementioned expansion outside the column, attention should be paid to the inner diameter of the flow path piping. Even if it is not desired to generate the spread outside the column as much as possible, the flow path piping needs several hundred mm and cannot be made zero. When the characteristic volume size of the peak is about 50 μL, it is considered that 1/10 or more times the expansion outside the column, that is, about 5 μL or more is not desirable. In general-purpose liquid chromatographs, a tube having an inner diameter of 0.25 mm was frequently used. In the case of laminar flow, since the diffusion volume is proportional to the tube cross-sectional area, it is necessary to use a tube having a pipe inner diameter of 0.1 mm or less that can obtain a diffusion performance of about 1/6 times. In addition, since excessive pressure loss occurs if the piping is too thin, a tube having an inner diameter smaller than 0.05 mm is not used in the case of a peak with a standard deviation of 0.42 s at a flow rate of 1 ml / min.

  Under the restriction of the required number of theoretical plates and the system pressure that are the subject of the present invention, it is necessary to ensure a typical linear velocity of about 10 mm / s. In order to secure a linear velocity of about 10 mm / s at a flow rate of about 1 ml / min, the inner diameter of the column needs to be about 2 mm or less. Here, it is used that the porosity of the particle packed column is 50 to 60%. On the other hand, feeding a flow rate of about 1 ml / min to an inner diameter column of 0.05 mm or less will cause a pressure loss that is generally difficult to use, although it depends on the particle diameter of the filler and the column length. .

  As a rule of thumb, the upper limit of sample injection volume depends on the column inner diameter. In particular, the non-retention peak in the case of isocratic elution is remarkable, but the upper limit of the sample injection amount is proportional to the column cross-sectional area. This would be to receive the sample solute once at the area of the column inlet. If the upper limit of the sample injection volume is 15 μL for a 4.6 mm inner diameter column that is used for general purposes, it will be about 3 μL for a 2.0 mm inner diameter column. Of course, if the amount of sample is extremely small, not only will it be difficult to obtain a detection signal, but handling will also be difficult and the variation in the amount of suction will be significantly large. For this reason, if it is less than 0.01 μL (10 nL), the practicality is low.

  When the column volume is reduced, the gradient elution delay (duel volume) cannot be ignored. This is because the time program for gradient elution is shortened in proportion to the volume in the column. If the analysis (retention) time is set to 30 s at a flow rate of about 1 ml / min, the duel volume must not exceed the equivalent volume, so 500 μL or less is required. However, in reality, the piping system volume cannot be reduced to zero, so that a duel volume of several μL remains at the minimum.

  Along with this, the capacity of the pump mixer is also limited. Usually, since the duel volume of the piping system is about 200 μL, a mixer capacity of 300 μL or less is required. Although the mixer capacity correlates with the gradient performance, the above-mentioned duel volume exists even if the mixer is not incorporated in the flow path, and the effect of mixing exists even though it is minute.

  As described above, there is a demand for the number of theoretical plates in the column, and the column length must be secured. On the other hand, if the linear velocity is increased with respect to the column having the column length for the purpose of speeding up, the pressure loss increases. For the performance aspects of both the separation performance and the pressure loss, a filler that provides one balance plan is a filler having a particle diameter of less than 3 μm.

  When using a column with a particle size of less than 3 μm, a theoretical plate height of 8 μm or less can be obtained with an optimum linear velocity, and a column length of about 50 mm is necessary to obtain a theoretical plate number of 5,000 or more. It becomes.

  When the flow rate is about 1 ml / min, the linear velocity is about 10 mm / s, and when a viscosity mobile phase of 25 ° C. and 100% methanol is fed, the pressure loss is as low as 30 MPa or less.

  Another balance plan between theoretical plate number and pressure drop is a monolithic column. That is, a pressure loss that is a fraction of that of a particle-type filler can be obtained while obtaining a theoretical plate height when using a filler having a particle diameter of 2 to 3 μm. That is, under the conditions of the above-described embodiment, a pressure loss of 10 MPa or less is obtained.

  The present invention is particularly effective for a high-performance liquid chromatograph apparatus, but is not limited to the above-described embodiments, and various modifications can be made within the scope of the technical idea.

1 is a system configuration diagram showing an embodiment of the present invention. It is a flow-path block diagram which shows the Example of this invention. The figure which shows the optimal response which obtains the exact theoretical plate number according to the peak width for demonstrating the Example of this invention. The figure which shows the optimal data collection interval which obtains the exact number of theoretical plates according to the peak width for demonstrating the Example of this invention. The figure which shows the example of a chromatogram for demonstrating the Example of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... High performance liquid chromatograph, 2 ... Pump, 3 ... Autosampler, 4 ... Column oven, 5 ... Detector, 6 ... System control part, 7 ... Data processing part, 8 ... Analytical instrument control part, 9 ... Parameter storage part DESCRIPTION OF SYMBOLS 10 ... Data processor, 11 ... Eluent, 12 ... Pump, 13 ... Joint, 14 ... Mixer, 15 ... Autosampler, 16 ... Analysis column, 17 ... Column oven, 18 ... Detector, 19 ... Waste liquid tank, 20 …cell.

Claims (12)

  A liquid chromatograph apparatus having at least a column, a detector, and a data processing apparatus, wherein the signal response of the detector is 10 ms or less.   A liquid chromatograph device having a pump, an injection device, a column, a detector and a data processing device, wherein the signal response of the detector is 10 ms or less.   A liquid chromatograph apparatus having at least a column, a detector, and a data processing apparatus, wherein a signal response of the detector is 10 ms or less and a data processing collection time interval is 10 ms or less. .   In a liquid chromatograph apparatus having a pump, an injection device, a column, a detector and a data processing device, the signal response of the detector is 10 ms or less, and the interval of data processing collection time is 10 ms or less. Liquid chromatograph device. The liquid chromatograph apparatus according to claim 1,
An optical path length volume of a detector flow cell is 3 μL or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph characterized by having an inner diameter of a pipe of 0.1 mm or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph characterized by having a sample injection volume of 3 μL or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph having a column inner diameter of 2 mm or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph apparatus having a pump mixer volume of 300 μL or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph having a column filler particle diameter of 3 μm or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph having a duel volume of 500 μL or less. The liquid chromatograph apparatus according to claim 1,
A liquid chromatograph apparatus, wherein the column is a monolith type.

Images