GB2285505A

GB2285505A - Amplitude modulation of carrier wave by atomic absorption signal

Info

Publication number: GB2285505A
Application number: GB9500122A
Authority: GB
Inventors: Michael Ron Hammer; Colin Andrew Porter
Original assignee: Varian Australia Pty Ltd
Current assignee: Varian Australia Pty Ltd
Priority date: 1994-01-11
Filing date: 1995-01-05
Publication date: 1995-07-12
Anticipated expiration: 2015-01-05
Also published as: AUPM329594A0; DE19500069A1; GB9500122D0; GB2285505B

Description

39 2285505

1 b=OD AND MEANS FOR CARRYING OUT ATOMIC SPECTROSCO Technical Field

This invention relates to a method and means for carrying out atomic spectroscopy such as used for the analytical determination of various elements, and is applicable to atomic absorption spectroscopy (AAS).

Background

In AAS the concentration of a chemical element in a sample is determined by measuring the degree to which atoms of the element absorb light of a wavelength which characterize those atoms. Usually, the light is generated by a discharge lamp and the sample to be analysed is in the form of a solution which is dissociated in any one of several ways to provide a cloud of atoms in the light path. For example, the cloud of atoms may be produced by spraying the sample solution into a flame or by depositing a small amount of the sample solution on a heated filament, rod, or other non flame device. The light passing through the sample region is received by a suitable detector and the degree of light absorption is measured. Absorption is determined by measuring the amount of light received by the detector with the sample present and again with the sample absent (reference) and ratioing the two readings.

AA spectra however suffer from the problems of noise and background signals.

The usual sources of noise components in the measured absorbent spectra include: flame emission noise, lamp noise, nebuliser noise, EHT noise, molecular and particulate (background) noise, A/D conversion noise and photon noise.

Prior attempts to eliminate noise and background

2 signals from AA spectra include the following:

(A) Double Beam Ai)i)roacli Instead of the conventional single light beam systems, a double beam system has been proposed whereby sample and reference beams are measured cyclically and ratioed. Noise sources common to both beams (ie.

correlated) are in theory eliminated. Such noise sources include lamp, EHT and some photon noise. However, uncorrelated noise sources are not eliminated and in fact they increase by a factor of 1.4 over the noise sources of the single beam approach.

39 In practice, the measurements on the two beams are performed alternatively via a time division multiplexed approach. Each beam is measured for 1-2 milliseconds with a repetition rate of 10 milliseconds. The effect of this is to make the noise cancelling effect of the double beam approach less than perfect and has the following effects on the nominally correlated noise sources; The 1-2 millisecond integration of each signal acts as a low pass filter attenuating those noise frequency components above about 500-1000 Hz.

The repetition rate of 10 milliseconds means that noise frequency components below 50 Hz remain correlated and are therefore eliminated.

The noise frequencies between 50Hz and about 1000 Hz become uncorrelated and are increased by 1.4 times over a single beam approach.

In addition, the effect of sampling for 1-2 milliseconds at a 10 millisecond repetition rate means 3 that noise frequencies between 50 Hz and 1000 Hz are aliased (their frequencies are shifted by the sampling rate to lower frequencies) thus increasing the low frequency noise. Subsequent integration of the signal cannot eliminate this effect and output signal to noise is degraded. A double beam approach reduces some low frequency noise sources (in particular lamp drift) but increases the effect of some other noise sources and the aliasing due to a sampled approach further degrades overall system noise performance. It appears that at best the reduction and amplification roughly balance so that double beam approaches have about the same noise performance as the early unsampled single beam systems.

There are two limitations on the degree to which background signals can be eliminated in the double beam system: static error and dynamic error.

Static error results when the physical position or energy distribution of the light beam used for measuring background absorption is different to that used for atomic absorption. Since the sample doing the absorbing is not completely homogeneous, the different spatial energy distribution between the two beams causes an extraneous variation in measured absorbance between the two beams and hence an error.

39 Dynamic error results when measurement of background and atomic absorption occurs at different times. Any change in the sample background absorbance between the two measurements will appear as a background correction error. In practice, interpolation techniques can be used to reduce, but not eliminate, this error.

(B) Zeeman Corrected Instruments Zeeman corrected instruments use only a single light source and use the same light beam for measurement of both 4 sample and reference absorption. They do not have the problem, therefore, of static error, but still are subject to dynamic error.

Zeeman corrected instruments modulate atomic absorption without affecting non-atomic (or background) absorption and thereby provide a means of discriminating between atomic and non-atomic absorption. Modulating the absorbance signal using the Zeeman effect has generally involved either physically rotating a prism polariser or rotating the plane of polarization magneto-optically.

These techniques are effective for simple background correction, however they are physically limited to frequencies of modulation of less than 100 Hz and still rely on the traditional method of sequentially measuring sample and reference signals.

Disclpsure of the Invention It is an object of the present invention to provide a method for carrying out atomic spectroscopy which improves background detection performance. It is a further object of the invention to improve the detection limits of a spectrometer, and particularly an atomic absorption spectrometer.

A method according to the invention is characterised in that the absorbance signal is modulated onto a high f requency carrier - eg., in the range f rom 1 KHz to 100 MHz - so that the absorbance signal appears as an AM modulation on the carrier. That is, 0 amplitude indicates 0 absorption, and increasing amplitude indicates increasing absorption. The signal may be detected by way of a narrow band filter centered at the carrier frequency or by synchronous detection via a demodulating amplifier locked to the frequency of the modulating signal.

39 Such a method enables an improved signal to noise ratio because many of the noise sources which create a problem for background correction have very low noise components at high frequency.

Some of the noise sources referred to above result in additive noise at the output and also cause modulation of the sample absorbance. By way of example, lamp intensity variations directly affect the output even without a sample being present and also effect the size of the absorption signal since halving the light halves the measured absorption signal. In these cases the additive noise should be significantly reduced by use of a method according to the present invention, but the modulation effect on the sample absorption may not be affected.

Since such modulation effects are multiplicative they affect precision but not detection limit. In general, the additive noise effect is much greater than the modulation effect.

The adverse consequences of the additive noise effect can be at least substantially removed by measuring and relating two signals in carrying out the method according to the invention. The first is the signal amplitude at the modulation frequency, which is the AC component of the output and will be referred to as the AC (or RF) signal, and this gives the (ref erence-s ample) /2 level directly. The second signal is the DC output f rom the detector (average value), which is the baseband signal and will be referred to as the DC signal, and this gives (reference+ sample)/2. At very low absorbance where the sample approximates the reference, the ratio of these two signals multiplied by 0.43429 gives the absorbance directly. At higher absorbances the computation is somewhat more complex but readily handled within a microprocessor: AC/2 + DC gives the reference level, and DC - AC/2 gives the sample signal. Alternatively, it 39 6 would be possible to always compute AC/DC and use a lookup table conversion. It is possible to measure the DC and AC outputs from the system simultaneously and continuously without time division multiplexing. This avoids the aliasing effect referred to earlier and ensures that noise sources common to both signals are cancelled correctly.

In carrying out the method of the present invention, the absorbance signal may be modulated onto the carrier by any suitable technique, including use of the Doppler effect, the Stark effect, or the Zeeman effect. In any of those cases, any appropriate means may be adopted to achieve a sufficiently high speed of modulation. By way of example, an electro-optic cell has been found satisfactory for that purpose.

39 The invention also provides atomic absorption spectroscopic apparatus including a light source for generating a beam of light, atomising means for generating a cloud of atoms of a sample to be analysed in the path of said beam of light, and modulating means which is operable to modulate onto a high frequency carrier an absorbance signal which is representative of the degree to which light at a predetermined frequency is absorbed by the sample, whereby the amplitude of the modulated signal is representative of the degree of said absorption.

DescriDtion of Drawings It will be convenient to hereinafter describe the invention in greater detail by reference to a Zeeman-type instrument, but it is to be understood that the invention is not restricted to use of the Zeeman effect. It will be further convenient to hereinafter describe the invention by reference to the accompanying drawings, and it is to be understood that the particularity of those drawings and of the associated description do not

7 supercede the generality of the preceding description of the invention.

Figure 1 shows in diagrammatic form an example instrument which has been configured in accordance with the present invention.

Figure 2 shows in diagrammatic form an alternative embodiment of the invention.

Figure 3 shows in diagrammatic form the polarization rotation at the output of the electro-optic cell which is used in both the Figure 1 and Figure 2 embodiments.

Detailed Descriptign of Embodiments In the instrument diagrammatically illustrated in Figure 1, a beam I of unpolarized light from a light source such as a hollow cathode lamp 2, is passed through a polarizer 3. The resulting beam 4 of linearly polarized light is directed through an electro-optic cell 5 which is operable to rotate the plane of polarization in direct proporation to an applied electric field 6. That is, the plane of polarization changes in response to changes in the voltage applied to the cell 5. The output of the cell 5 passes through an analyte cloud produced by an atomizer 7, which may be of any suitable type such as a flame atomizer or a graphite tube furnace. A DC magnet 8 is located at the atomizer 7 and is operable to cause Zeeman splitting of the analyte cloud.

The beam of light leaving the atomizer 7 is therefore modulated at the frequency at which the plane of polarization rotates under the influence of the cell 5, and the amplitude of that modulation is proportional to the absorbance of the analyte. This modulated light then enters a monochromator 9 where the desired analytical 39 8 wavelength is selected, and that wavelength is subsequently detected by a photo-multiplier 10 or other suitable means. In the particular arrangement shown, the electrical output of the photo-multiplier 10, which is preferably amplified, is split into two signal paths 11 and 12 which respectively represent the AC (or RF) and DC signals previously referred to. For that purpose, the signal paths 11 and 12 will be subjected to synchronous detection and low pass filtering respectively. The ratio of the two signals can then be used to determine the true absorbance of the sample.

Since the background signal is not subject to Zeeman splitting, it is not modulated and therefore does not contribute to the AC signal. It merely reduces the available light level and this is a common mode signal eliminated when the ratio of the AC and DC signals is computed.

39 The embodiment shown in Figure 2 is substantially the same as the embodiment of Figure 1, with the exception that the relative order of the polarizer 3, the electro-optic cell 5 and the atomizer 7, has been reversed.

BY utilizing an electro-optic cell to rotate the plane of polarization, it is possible to obtain very high modulation frequencies. The electro- optic cell may be of any suitable type, including the following, each of the which is well known:

Kerr cell Pockel's cell Photo-elastic cell Nematic liquid crystal device All four types operate on the principle of induced birefringence in which an external influence causes the velocity of one plane of polarized light to be retarded 9 more than the other. The effect of this is to rotate the primary plane of polarization and alter its state from linear to elliptical and back to linear after 900 rotation (see Figure 3). This 900 rotation is termed the "half-wave retardation". The difference between the four cells lies essentially in the mechanism by which birefringence is induced.

Liquid crystal devices operate on the basis that their structure alters in response to application of a small electric field. The birefringence of photo-elastic cells is due to mechanically induced stress typically provided by a piezo-electric transducer. Both Kerr and Pockel's cells function along generally the same lines.

That is, an applied electric field directly alters the optical anisotropy of their crystalline structure. The difference between them is that the Kerr effect varies with the square of the electric field whereas the Pockel's effect varies linearly.

39 Some standard Kerr cells may not be suitable for use in an instrument according to the present invention because they have a lower wavelength limit of 250 nanometers. Use of lithium borate, or beta barium borate, in such a cell however, extends the performance of such cells and makes them suitable for use in the present invention. Other materials might also make such cells suitable.

In apparatus as shown in Figure 1, polarizer 3 is preferably a crystal polarizer and electro-optic cell 5 is a photo-elastic modulator (manufactured by Hinds International, U.S.A.) operated at 50 KHz, which gives a modulation frequency of 100 KHz (ie. the frequency of the carrier is 100 KHz). Atomizer 7 may be a standard Varian atomizer and DC magnet 8 is such to provide a field strength of about 1 Tesla. Other components of the apparatus are standard.

It will be apparent from the foregoing general and detailed descriptions of the invention that the invention provides a relatively simple and effective means for improving the background correction performance of a spectrometer.

39 11

Claims

CLAIMS:

A method of conducting atomic absorption spectroscopy including the steps of, passing a beam of light through a sample to be analysed, generating an absorbance signal which is representative of the degree to which light at a predetermined frequency is absorbed by the sample, and modulating the absorbance signal onto a high frequency carrier so that absorbance of the predetermined frequency is represented by the amplitude of the modulated signal.
2. A method according to claim 1, wherein said modulation is effected by modulating said light beam prior to passing that beam through said sample.
3. A method according to claim 1, wherein said modulation is effected by modulating the residue of said light beam which emerges from said sample.

39
4. A method according to any one of claims 1 to 3, wherein the frequency of the high frequency carrier is in the range from 1 KHz to 100 MHz.
5. A method according to claim 4, wherein the frequency of the high frequency carrier is about 100 KHz.
6. A method according to any one of claims 1 to 5, wherein the Zeeman effect is employed in modulating the absorbance signal onto the carrier.
7. A method according to any one of claims 1 to 6, wherein the amplitude of the modulated signal is detected synchronously with the frequency of the carrier.
8. A method according to claim 7 wherein an average output signal is detected and a ratio of the amplitude signal and the average signal is used to determine the 12 absorbance by the sample of light at the predetermined frequency.
9. Atomic absorption spectroscopic apparatus including a light source for generating a beam of light, atomising means for generating a cloud of atoms of a sample to be analysed in the path of said beam of light, and modulating means which is bperable to modulate onto a high frequency carrier an absorbance signal which is representative of the degree to which light at a predetermined frequency is absorbed by the sample, whereby the amplitude of the modulated signal is representative of the degree of said absorption.
10. Apparatus according to claim 9, wherein said modulating means is operable to modulate said beam of light prior to passage of that beam of light through said sample.
11. Apparatus according to claim 9, wherein said modulating means is operable to modulate the residue of said beam of light which emerges from said sample.
12. Apparatus according to any one of claims 9 to 11, wherein the said modulating means is operable to modulate the absorbance signal onto a carrier at a frequency in the range from 1 KHz to 100 MHz.
13. Apparatus according to claim 12, wherein the said modulating means is operable as modulate the absorbance signal onto a carrier at a frequency of about 100 KHz.
14. Apparatus according to any one of claims 9 to 13, wherein the said modulating means includes an electro-optic cell for rotating the plane of polarization of polarized light that passes therethrough.

39 13
15. Apparatus according to any one of claims 9 to 14, wherein the apparatus includes a magnet for causing the Zeeman effect.
16. Apparatus according to any one of claims 9 to 15, including detection means for detecting and outputting two signals, said signals being the amplitude of the modulated signal and an average of the modulated signal.
17. Apparatus according to claim 16, wherein the said detection means is operable to detect the said amplitude signal synchronously with the frequency of the carrier, and simultaneously to detect said average output signal.
18. A method of conducting atomic absorption spectroscopy substantially as hereinbefore described with reference to Figure 1 or Figure 2 of the drawings.
19. Atomic absorption spectroscopic apparatus substantially as hereinbefore described with reference to Figure 1 or Figure 2 of the drawings.

39