CN108169215A - A kind of computational methods of emission spectrometer time of integration upper limit setting - Google Patents

Abstract

A kind of computational methods of emission spectrometer time of integration upper limit setting, including step：Obtain blank spectrogram of the excitation light source under the conditions of N number of different gradient times of integration；Calculate the average blank spectrogram under the conditions of N number of different gradient times of integration；To blank spectral intensity I average at different wave length_bLinear regression fit is done with time of integration T, obtains blank background spectral temporal proportionality coefficient k_dWith background constant b_d；Set ccd detector range upper limit FS；Set the confidence coefficient t of detection limit；Set light source blank spectral ripple horizontal properties value [RSD]_b；It sets the element range upper limit to be detected and estimates the multiple α of detection limit with the element；The time of integration upper limit is calculated according to formula.Due to that can realize computer aided calculation by formula, liberate operation, it only needs to test blank spectrogram of the excitation light source under the conditions of the different gradient times of integration, avoid to carrying out integral condition optimization respectively under the conditions of different elements, different characteristic spectral line, various concentration.

Description

A calculation method for setting upper limit of integration time of emission spectrometer

technical field

The invention relates to the technical field of emission spectrum analysis, in particular to a calculation method for setting the upper limit of the integration time of an emission spectrometer.

Background technique

Emission spectroscopy is an analytical chemical method for precise quantification based on the linear relationship between the concentration of the element to be measured and the intensity of the characteristic spectral line it emits. Generally, the elements in the sample are excited by plasma to emit characteristic spectral lines, which are collected into the spectroscopic detection system through the front optical path to cause dispersion in space, and are photoelectrically converted by the detector to finally obtain the full spectrum picture.

Inductively coupled device CCD is a common photoelectric conversion device, often used as a detector for spectroscopic detection systems, such as micro-fiber spectrometers, echelle grating spectrometers, etc.

When using emission spectrometry to quantify elements, it is first necessary to establish an element concentration-intensity standard curve, so standard samples with gradient concentrations will be introduced during the test. In order to ensure the accuracy of the test results, it is necessary to ensure that the obtained element spectra have a good signal-to-noise ratio and signal-to-background ratio, so the instrument parameters need to be optimized, and the integration time of the CCD is one of them. Generally speaking, when other conditions are fixed, for a standard sample with a fixed concentration of elements, the number of photons entering the CCD per unit time is constant, so if the CCD integration time is small, the net signal of the element after CCD photoelectric conversion is relatively small, It is easy to be submerged in the dark noise of the CCD, that is, the signal-to-noise ratio/signal-to-background ratio of the element signal is low at this time. And when the dividing time is too large, it is easy to cause CCD saturation or exceed the linear range of CCD response, both of which will damage the accuracy of the test results.

In order to obtain better characteristic spectral lines of the elements to be measured, the integration time of the CCD is generally optimized during the test process, and the upper limit value of the integration time is obtained.

There are currently two schemes for setting the integration time of the CCD. One scheme is the trial and error method, that is, for a certain element, the element standard sample with the concentration of the upper limit of the range is first introduced, and the integration time value is continuously adjusted. Line intensity value to determine the final suitable integration time.

Another solution is the preset method, that is, through the summary of the previous test results, the upper limit of the integration time of different characteristic lines of different elements under different concentration conditions is given or the preset integration time is given, and it is compared with the database. The form is written into the software and called when used.

However, the disadvantages of the above two schemes are: 1. Optimization and trial and error are required for different elements, different characteristic lines, and different concentrations, which increases the test time and test cost; 2. The transplantation effect of this scheme is poor, requiring Do lots of repetitive testing.

Contents of the invention

This application provides a calculation method for setting the upper limit of the integration time of the emission spectrometer. The detector of the emission spectrometer is a CCD detector, and the calculation method for setting the upper limit of the integration time of the CCD detector includes steps:

Obtain the blank spectrum of an excitation light source under N different gradient integration time conditions, where N is greater than or equal to 2;

Calculate the average blank spectrum under N different gradient integration time conditions;

Perform linear regression fitting on the average blank spectral intensity I _b at different wavelengths and the integration time T to obtain the blank background spectral time proportional coefficient k _d and the background constant b _d ;

Set the CCD detector range upper limit FS;

Set the confidence coefficient t of the detection limit;

Set the characteristic value of the light source blank spectrum fluctuation level [RSD] _b ;

Set the upper limit of the range of the element to be detected and the multiple α of the estimated detection limit of the element;

According to the formula Calculate the upper limit of the integration time.

In one embodiment, the average blank spectrogram under N different gradient integration time conditions is calculated, specifically:

Perform M tests on the blank spectrum under each gradient integration time condition, and no saturation value appears in the full spectrum under the highest integration time condition, and M is greater than or equal to 1;

Computes the average blank spectrogram for M tests.

In one embodiment, the upper limit FS of the range of the CCD detector is set as the full scale value, or FS is set according to the linear influence range of the CCD detector.

In one embodiment, the blank background spectral time scaling coefficient k _d is, k _d =k _n +k _p , wherein, k _n is the time scaling coefficient of the dark noise of the CCD detector, and k _p is the time scaling coefficient of the plasma background .

According to the calculation method for setting the upper limit of the integration time in the above-mentioned embodiment, since the computer-aided calculation can be realized through the formula, the workers are freed from operation, and after the experimental conditions are determined, it is only necessary to test the blank spectrum of the excitation light source under different gradient integration time conditions once. It avoids the optimization of integration conditions for different elements, different characteristic lines, and different concentration conditions; the method is based on the objective laws of physics and the definition of detection limit standards, which is accurate and reliable; the method has good portability and scalability, and can Compatible with the detection limit definition method, adapting to different CCD types and their upper limit of measurement range, different excitation light source system stability level, different upper limit of measurement range of analyte elements, different analyte elements and characteristic wavelengths, etc.

Description of drawings

Fig. 1 is the calculation flowchart of CCD integration time upper limit setting;

Fig. 2 is the average blank spectrogram under different gradient integration time conditions;

Fig. 3 is the blank background spectral time scale factor spectrogram obtained by linear fitting at different wavelengths;

Fig. 4 is the blank background spectral constant spectrogram obtained by linear fitting at different wavelengths;

Fig. 5 is linear fitting correlation coefficient spectrogram at different wavelengths;

Figure 6 is the upper limit spectrum of integration time obtained at different wavelengths.

Detailed ways

The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings.

This example provides a calculation method for setting the upper limit of the integration time of the emission spectrometer, especially when the detector of the emission spectrometer is a CCD detector, the calculation method for setting the upper limit of the integration time of the CCD detector is provided. The flow chart is shown in Figure 1 , specifically include the following steps.

S1: Obtain the blank spectrum of an excitation light source under N different gradient integration time conditions, where N is greater than or equal to 2.

S2: Calculate the average blank spectrum under N different gradient integration time conditions.

In this step, M tests are performed on the blank spectrogram under each gradient integration time condition, that is, the test is repeated M times under each gradient integration time condition, and the spectrogram under the highest integration time condition among N gradient integration times There is no saturation value in , and the highest integration time here refers to the maximum value among the N gradient integration times. If there is a saturation value, remove the saturation value so that it does not enter the calculation, or reduce the maximum integration time to avoid saturation of the intensity , M is greater than or equal to 1; then, calculate the average blank spectrum of M tests, and calculate the average blank spectrum under N different gradient integration time conditions by analogy.

S3: Perform linear regression fitting on the average blank spectral intensity I _b at different wavelengths and the integration time T to obtain the blank background spectral time proportional coefficient k _d and background constant b _d .

In this step, the linear regression fitting method is: I _b =k _d T+b _d , after the linear regression fitting, the blank background spectrum time proportional coefficient k _d and the background constant b _d are obtained.

S4: Set the upper limit FS of the range of the CCD detector.

Among them, the upper limit FS of the CCD detector range can be set as its full-scale value, or FS can be set according to the linear influence range of the CCD detector.

S5: Set the confidence coefficient t of the detection limit.

S6: Set the characteristic value [RSD] _b of the fluctuation level of the blank spectrum of the light source.

S7: Set the upper limit of the range of the element to be detected and the multiple α of the estimated detection limit of the element.

S8: Calculate the upper limit of the integration time according to the formula.

The formula in this step is: The derivation of the formula is as follows:

The characteristic spectrum emitted by the element/sample to be measured is converted into an electrical signal by the spectroscopic detection system after being excited by the plasma. If a CCD detector is selected as the detector of the spectroscopic detection system, the intensity of the pixel corresponding to the characteristic spectrum is I. The light intensity is specifically the sum of the dark noise I _n of the CCD detector, the plasma background spectrum I _p , and the net signal intensity I _s of the signal characteristic spectrum, namely

I=I _n +I _p +I _s (1);

From the working principle of the CCD detector, it can be seen that the spectral intensity of the CCD response has a linear relationship with the setting of the integration time, that is, for the CCD dark noise I _n and the plasma background spectrum I _p :

I _n =k _n T+b _d (2)

I _p =k _p T (3)

Among them, T is the integration time of CCD, k _n and b _d are the time scale coefficient and dark noise constant of CCD dark noise respectively; k _p is the time scale coefficient of plasma background.

For the net signal intensity I _s of the characteristic spectrum of an element, in addition to having a linear relationship with the integration time, according to the basic principle of the emission spectrometry, it is also proportional to the concentration of the element to be measured, so:

I _s =k _s CT (4)

Among them, k _s is the time scale coefficient of the signal, and the concentration of the element to be measured is C;

Let k _d =k _n +k _p , and define I _b as the blank background spectral intensity obtained by CCD when the element to be measured is not introduced, then I _b =I _n +I _p =k _d T+b _d , k _d is the corresponding blank background spectrum time scale coefficient, after substituting formulas 2 to 3 into formula 1, we get:

I=k _s CT+k _d T+b _d (5)

For the detection limit, it is generally defined as when the instrument is in normal working condition, inject a series of standard solutions, make a working curve, measure the blank solution n times continuously, and take the concentration corresponding to the standard deviation t times of the blank value for n times as the detection limit. Therefore, t is the confidence coefficient in the detection limit calculation method, which is related to the number of tests n;

Assuming that the standard deviation of the n times blank value is [SD] _b , the slope of the standard curve is k _s T according to formula 5, then according to the definition of the detection limit _DL , its general expression can be obtained:

Assuming that the concentration C of the element to be detected is α times the detection limit DL of the element, that is

C＝α·DL (7)

Obtained by formulas 6 and 7:

k _s CT = αt[SD] _b (8)

Substitute Equation 8 into Equation 5 and let the relative standard deviation of the blank value n times be and end up with:

In the above formula, if the CCD intensity I is equal to the upper limit FS of the range allowed by the CCD, then the upper limit of the CCD integration time setting is obtained as follows:

For ease of understanding, first take the Ocean Optics HR4000 series spectrometer as an example, and use the computer to calculate the upper limit of the integration time at a certain characteristic wavelength of a certain element through the above formula (10). The excitation source is argon microwave plasma (ie ArMPT);

First, perform operation steps S1-S3, and perform linear regression fitting according to the average blank spectral intensity and integration time to obtain b _d =650, k _d =30;

Ocean Optics HR4000 spectrometer has 14 AD conversion digits, so its CCD full-scale value FS=2 ¹⁴ -1=16383;

According to the "JJG768-2005 Emission Spectrometer Verification Regulations", take the confidence coefficient t=3;

For the stable ArMPT, its plasma stability is relatively high, and the characteristic value of the light source background spectrum fluctuation level [RSD] _b = 0.3%;

Take the upper limit of the measurement range of the element to be measured as 100 times the estimated detection limit of the element, that is, α=100;

Substituting the above parameters into Formula 10, T _max =267ms is obtained, that is, the upper limit of the CCD integration time is obtained.

It should be noted that the CCD range upper limit FS allowed in step 4 can also take into account the linear response range of the CCD, such as FS is 95% of its full scale, then FS=95%×(1 ¹⁴ -1)=15563.85, at this time _Tmax = 253.65ms.

It should be noted that the method in this example does not limit the elements to be measured and the characteristic wavelengths. This method is valid for all wavelengths in the spectral range captured by the spectroscopic detection system, that is, the parameters in the above steps S4-S7 can be used for different wavelengths. Different values are used, that is, different characteristic lines of different elements are processed separately, which will be described in conjunction with specific examples below.

Obtain the blank spectrum of an excitation light source under 11 different gradient integration time conditions, and the number of tests under each gradient integration time condition is 11, and ensure that no saturation value appears in the full spectrum under the highest integration time condition.

Calculate the average blank spectrum under different gradient integration time conditions, and the average blank spectrum for 11 calculations can be obtained, as shown in Figure 2. In Figure 2, T=4, 6, 8, 12, 16, 24, The spectrograms under different gradient integration time conditions of 32, 48, 64, 96, and 128ms, N=11, and the maximum integration time is 128ms; it should be noted that the so-called gradient integration time is not necessarily at equal intervals, as long as it is increasing Any time.

Perform linear regression fitting on the average blank spectral intensity I _b and the integration time T at different wavelengths to obtain the blank background spectral time proportional coefficient k _d and the background constant b _d , as shown in Figure 3 and Figure 4 respectively. In this example, different The number of fitting points at the wavelength is 11. In order to further verify the rationality of the linear regression, Fig. 5 shows the correlation coefficient R ² spectrum of the linear fitting at different wavelengths, R ² ∈ (0.992,1].

Set the allowable CCD range upper limit FS, in this example, FS=2 ¹⁴ -1=16383 for all wavelengths.

Set the confidence factor t in the calculation method of the detection limit. According to the "JJG768-2005 Emission Spectrometer Verification Regulations", take the confidence factor t=3.

Set the characteristic value [RSD] _b of the fluctuation level of the light source blank spectrum. For a stable ArMPT, its plasma stability is relatively high, and the characteristic value of the fluctuation level of the light source background spectrum [RSD] _b = 0.3%.

Set the upper limit of the measurement range of the analyte and the multiple α of the estimated detection limit of the element, and take the upper limit of the measurement range of the analyte as 100 times the estimated detection limit of the element, that is, α=100.

The upper limit of the integration time T _max is calculated according to the formula (10), and the spectrum of the upper limit of the integration time at different wavelengths is obtained, as shown in FIG. 6 .

The above uses specific examples to illustrate the present invention, which is only used to help understand the present invention, and is not intended to limit the present invention. For those skilled in the technical field to which the present invention belongs, some simple deduction, deformation or replacement can also be made according to the idea of the present invention.

Claims (4)

1. a kind of computational methods of emission spectrometer time of integration upper limit setting, the detector of the emission spectrometer are visited for CCD Survey device, which is characterized in that the computational methods of the time of integration upper limit setting of the ccd detector include step：

Blank spectrogram of the excitation light source under the conditions of N number of different gradient times of integration is obtained, N is more than or equal to 2；

Calculate the average blank spectrogram under the conditions of N number of different gradient times of integration；

To blank spectral intensity I average at different wave length_bLinear regression fit is done with time of integration T, when obtaining blank background spectrum Between proportionality coefficient k_dWith background constant b_d；

Set ccd detector range upper limit FS；

Set the confidence coefficient t of detection limit；

Set light source blank spectral ripple horizontal properties value [RSD]_b；

It sets the element range upper limit to be detected and estimates the multiple α of detection limit with the element；

According to formulaCalculate the time of integration upper limit.

2. computational methods as described in claim 1, which is characterized in that under the conditions of calculating N number of different gradient times of integration Average blank spectrogram, specially：

M test, and the highest in N number of gradient time of integration are carried out to the blank spectrogram under the conditions of each gradient time of integration Do not occur saturation value in spectrogram under the conditions of the time of integration, M is more than or equal to 1；

Calculate the average blank spectrogram of M test.

3. computational methods as claimed in claim 2, which is characterized in that the ccd detector range upper limit FS is set as full amount Journey value or the linear effect range set FS according to ccd detector.

4. computational methods as described in claim 1, which is characterized in that the blank background spectral temporal proportionality coefficient k_dFor k_d =k_n+k_p, wherein, k_nFor the time scale coefficient of ccd detector dark noise, k_pTime scale coefficient for plasma background.