Technical Field
[0001]
The present invention relates to an analysis apparatus and the like used for analysis such as gas
component analysis.
B ackground Art
[0002]
As an analysis method (TDLAS: Tunable Diode Laser Absorption Spectroscopy) adapted to
sweep an oscillation wavelength by modulating the injection current of a semiconductor laser, and
obtain the absorption spectrum of measurement target gas to quantify concentration, a wavelength
modulation method (WMS: Wavelength Modulation Spectroscopy) adapted to apply current
modulation having small amplitude at a frequency sufficiently higher than a frequency used to modulate
the current for the wavelength sweeping, and obtain the spectrum from a signal lock-in detected at a
frequency twice the sufficiently higher frequency to quantify the concentration is often used in order to
increase detection sensitivity.
[0003]
For example, Patent Literature 1 discloses an example of a laser gas analyzer based on the
wavelength modulation method (or also referred to as a frequency modulation method). In this
example, the need for a reference gas cell for scanning the ab sorption wavelength of measurement target
gas, i.e., for scanning the emission wavelength of a laser element is eliminated, and without stabilizing
the emission wavelength to a specific wavelength, or without separately using a meter such as a dust
meter even in a dusty environment, gas concentration can be accurately measured by correcting the
amplitude of a gas absorption waveform using a received light amount correction coefficient.
Citation List
Patent Literature
[0004]
Patent Literature 1: Japanese Unexamined Patent Publication JP-A2009-47677
Summary of Invention
Technical Problem
[0005]
However, in a normal WMS, when the entire signal intensity uniformly varies due to the
variation of laser intensity from that at the time of zero calibration, an indicated value varies, and that
cause drift. In order to suppress the drift, as disclosed in Patent Literature 1 described above, it is
necessary to monitor the laser intensity in some way to make a correction simultaneously.
[0006]
Also, in the WMS, in order to quantify concentration from a spectrum obtained by wavelength
sweeping, a spectrum calculation process (such as obtaining the difference between the maximum and
minimum value within a specific range of the spectrum) is further required.
[0007]
Further, in the WMS, for the lock-in detection, it is necessary to apply modulation at a
frequency sufficiently higher than the modulation frequency for the wavelength sweeping, and in order
to respond to such a high frequency signal, sufficient response speed is required for a detector as well.
For a near-infrared region often used in the WMS, quantum detectors having sufficient response speed
are relatively easily available at low cost; however, for a mid-infrared region preferable for gas analysis,
quantum detectors having high response speed are mostly expensive, and it is difficult to perform the
WMS using an inexpensive but low response speed thermal detector.
[0008]
Still further, when part of a spectrum of a component other than a measurement target
component is superposed on a spectrum obtained by the WMS, in order to remove the interference
effect of it, complicated calculation processes such as spectrum fitting, baseline estimation, and
multivariate analysis are required.
[0009]
The present invention is made in consideration of the problems as described above, and a
desired object thereof is to allow an analysis apparatus utilizing light absorption to accurately measure
the absorbance or concentration of a measurement target component with simple calculation while
eliminating some effects such as the effect of a variation in light source intensity and the effect of
3
interference by another substance.
Solution to Problem
[0010]
That is, the analysis apparatus according to the present invention is one adapted to analyze a
measurement target component contained in a sample, and the analysis apparatus includes: a light source
adapted to emit reference light that is light whose wavelength is modulated at a predetermined
modulation frequency; a light detector adapted to detect the intensity of measurement target light that is
light after the reference light has transmitted through the sample, and the intensity of the reference light;
a first calculation part adapted to calculate an intensity ratio logarithm that is the logarithm of the ratio
between the intensity of the measurement target light and the intensity of the reference light; a frequency
component extraction part adapted to, from the intensity ratio logarithm, extract a frequency component
having a frequency n times (n is an integer equal to or more than 1) the modulation frequency; and a
second calculation part adapted to, on the basis of the result of the frequency component extraction by
the frequency component extraction part, calculate the concentration or absorbance of the measurement
target component.
[0011]
In such a configuration, since the intensity ratio logarithm (hereinafter also referred to as an
absorbance signal) that is the logarithm of the ratio between the intensity of the measurement target light
and the intensity of the reference light is calculated before the extraction of the frequency component, for
example, even when the entire signal intensity uniformly varies due to the variation of laser intensity
from that at the time of zero calibration (at the time of measuring the intensity of the reference light), a
measurement result is not affected. Therefore no drift occurs. Accordingly, the need to monitor the
light source intensity for drift suppression simultaneously is eliminated.
[0012]
Also, a value obtained by the frequency component extraction part is proportional to the value
of the concentration of the measurement target component directly. As a result, the need for a spectrum
calculation process for concentration quantification, which is required forWMS, can be eliminated, and
only a single modulation frequency is required, leading to a simpler system configuration and low cost.
4
[0013]
Further, since the frequency for extracting the component can be set equal to the modulation
frequency or set to at most several times the modulation frequency, analysis accuracy equivalent to that
of WMS can be obtained using a low response speed light detector not applicable to WMS. This effect
5 remarkably appears in the mid-infrared region. This is because a light detector whose response speed
is high in the mid-infrared region is relatively expensive, a nd therefore being able to use an inexpensive
thermal detector is directly linked to being able to enjoy a big benefit in terms of cost.
[0014]
Also, according to the present invention, even when the sample contains an interference
10 component, the unconventional dramatic idea of converting such a problem into a linear problem based
on logarithmic calculation and finally leading to a direct problem of solving simultaneous equations
allows the concentration of the measurement target component to be surely measured. An example of
a specific configuration for this is as follows.
[0015]
15 That is, the frequency component extraction part is one adapted to, from the intensity ratio
logarithm (absorbance signal) calculated by the first calculation part, extract mutually different
frequency components whose number is equal to or more than the sum of the number of the
measurement target component and the number of interference components, the analysis apparatus
stores independent frequency components that are frequency components of absorbance signals when
20 the measurement target component a nd the respective interference components are independently
present or stores independent absorbance signals (independent intensity ratio logarithms) that are
absorbance signals of the measurement target component a nd the respective interference components
per unit concentration, and the second calculation part is one adapted to calculate the concentration of the
me asurement target component on the basis of the result of the frequency component extraction by the
25 frequency component extraction part a nd the independent frequency components or the independent
absorbance signals.
[0016]
Specifically, it can be cited that the second calculation part is one adapted to calculate the
concentration of the measurement target component by solving simultaneous equations including the
5
result of the extraction of the respective frequency components by the frequency component extraction
part, the independent frequency components or the independent intensity ratio logarithms of the
measurement target component and the respective interference components, and the concentrations of
the measurement target component and the respective interference components. In this configuration,
5 simple a nd certain calculation, i.e., solving the simultaneously equations allows the concentration of the
measurement target gas, from which an interference effect has been removed, to be determined.
[0017]
In order to make it possible to determine the concentration with an error due to measurement
noise being small, it is preferable that the frequency component extraction part is one adapted to extract
10 frequency components whose number is larger than the sum of the number of the measurement target
component and the number of the interference components, and the second calculation part is one
adapted to calculate the concentration of the measurement target component with use of a least squares
method from simultaneous equations where the number of unknowns is larger than the sum of the
number of the measurement target component and the number of the interference components.
15 [0018]
As the frequency component extraction part, one adapted to extract the frequency component
or the frequency components by performing lock-in detection at the frequency n times (n is an integer
equal to or more than 1) the modulation frequency can be cited.
[0019]
20 As a specific method for calculating the intensity ratio logarithm by the first calculation part,
for example, a method adapted to calculate the intensity ratio logarithm by obtaining the ratio between
the intensity of the measurement target light and the intensity of the reference light and obtaining the
logarithm of the ratio, or a method adapted to calculate the intensity ratio logarithm by obtaining the
logarithm of the intensity of the measurement target light as well as obtaining the logarithm of the
25 intensity of the reference light, a nd subtracting the logarithm of the intensity of the reference light from
the logarithm of the intensity of the measurement target light can be cited.
[0020]
The analysis apparatus of the present invention is applicable to analyzing gas or the like.
Specifically, it can be cited that the light source is a semiconductor laser adapted to emit the
6
reference light modulated in a wavelength range including the peak wavelength of the light absorption
spectrum of the measurement target component, the analysis apparatus further includes a ga s cell into
which the sample gas is introduced, the gas cell is irradiated with the reference light emitted from the
semiconductor laser, and the light detector is arranged in the light path of the measurement target light
5 having transmitted through the gas cell.
[0021]
Also, it is preferable that the analysis apparatus further includes a light source control part
adapted to make the light source operate in quasi-continuous-wave (quasi-CW) oscillation, a nd generate
a temperature change based on current modulation to sweep an oscillation wavelength.
10 When a low response speed detector such as a thermopile detects pulse train signals, the
wavelength modulation by the light source control part allows the pulse train signals to be averaged, a nd
therefore the same result can be obtained as that obtained when making the light source operate in
continuous-wave (CW) oscillation and using a high response speed light detector such as an MCT. In
this case, as the pulse width of the quasi-CW oscillation is decreased, wavelength resolution increases,
15 and as the duty ratio is increased, signal intensity increases. Also, as compared with the CW oscillation,
the quasi-CW oscillation makes it possible to decrease the power consumption of the light source,
facilitate waste heat disposal, and further prolong the life of the light source.
[0022]
In addition, it is preferable that the analysis apparatus further includes a signal separation part
20 adapted to, from a light intensity signal obtained by the light detector, separate a signal corresponding to
a part of a pulse from the light source.
By separating a signal corresponding to a part of a pulse from the light source as described, the
pulse width of the pulse does not directly affect wavelength resolution, and therefore a reduction in
wavelength resolution can be suppressed without shortening the pulse width. As a result, as compared
25 with a conventional quasi-CW oscillation method, the wavelength resolution can be significantly
improved. In addition, since shortening the pulse width to prevent the reduction in wavelength
resolution is not required, the technical difficulty required for electronics for driving the light source is
reduced, and correspondingly, cost is also reduced.
[0023]
7
When the semiconductor laser used as the light source operates in pulse oscillation,
temperature changes to thereby change a wavelength. The degree of such a transient temperature
change (wavelength change) becomes smaller toward the latter half of a pulse, and therefore by
separating a signal corresponding to the latter half, the wavelength resolution is improved. For this
5 reason, it is preferable that the signal separation part separates a signal corresponding to the latter half of
a pulse of the light source. This configuration makes it possible to significantly improve the
wavelength resolution as compared with the conventional quasi-CW oscillation method by relatively
widening the pulse width of a pulse (e.g., approximately 100 ns) and setting a sampling position to a
time position within the pulse as far back as possible (e.g., 85 to 95 ns after the rise of the pulse). In
10 addition, the sampling is performed within a predetermined time width.
Advantageous Effects of Invention
[0024]
According to the present invention described above, the analysis apparatus utilizing light
15 absorption is capable of accurately measure the absorbance or concentration of a measurement target
component while ensuring a simple and inexpensive configuration and eliminating effects such as the
effect of a variation in light source intensity and the effect of interference by another substance.
Brief Description of Drawings
20 [0025]
FIG. 1 is an overall schematic diagram of an analysis apparatus according to one embodiment
of the present invention;
FIG. 2 is a functional block diagram of an information processor in the same embodiment;
FIG. 3 is a schematic diagram illustrating a method for modulating a laser oscillation
25 wavelength in the same embodiment;
FIG. 4 is a time-series graph illustrating an example of a modulation signal, output signals
from a light detector, and a measurement result in the same embodiment;
FIG. 5 is a flowchart illustrating concentration calculation in a variation;
FIG. 6 is a diagram illustrating drive current (voltage) and a modulation signal in quasi-CW
8
oscillation; and
FIG. 7 is a schematic diagram illustrating a measurement principle based on the quasi-CW
oscillation.
FIG. 8 is an overall schematic diagram of an analysis apparatus according to a variation;
5 FIG. 9 is a functional block diagram of a signal processor in the variation;
FIG. 10 is a schematic diagram illustrating an example of pulse oscillation timings and light
intensity signals of multiple semiconductor lasers in the variation; and
FIG. 11 is a schematic diagram illustrating the configuration of a signal separation part in the
variation.
10
Description of Embodiments
[0026]
In the following, an analysis apparatus 100 according to one embodiment of the present
invention will be described with reference to the drawings.
15 The analysis apparatus 100 is a concentration measurement apparatus adapted to measure the
concentration of a measurement target component (such as CO or CO2 herein) contained in sample gas
such as exhaust gas, and as illustrated in FIG. 1, includes: a cell 1 into which the sample gas is
introduced, a semiconductor laser 2 as a light source adapted to irradiate the cell 1 with laser light; a light
detector 3 adapted to be provided in the light path of laser light having transmitted through the cell 1 and
20 receive the laser light; and an information processor 4 adapted to receive an output signal from the light
detector 3 and on the basis of the value of the output signal, calculate the concentration of the
measurement target component.
[0027]
The respective parts will be described.
25 The cell 1 is one whose light incident and exit ports are formed of a transparent material hardly
absorbing light in the absorption wavelength band of the measurement target component, such as quartz,
calcium fluoride, or barium fluoride. Although not illustrated in FIG. 1, the cell 1 is provided with an
inlet port for introducing gas inside and an outlet port fo r exhausting internal gas, a nd the sample gas is
introduced into the cell 1 through the inlet port and then enclosed.
9
[0028]
The semiconductor laser 2 is a quantum cascade laser (QCL) herein, a type of a
semiconductor laser 2, and oscillates laser light in the mid-infrared region (4 μm to 10 μm). The
semiconductor laser 2 is capable of modulate (change) an oscillation wavelength depending on a given
5 current (or voltage). Note that as long as an oscillation wavelength is variable, another type of laser
may be used, and in order to change an oscillation wavelength, some means may be taken, such as
changing temperature.
[0029]
As the light detector 3, a relatively inexpensive thermal type such as a thermopile is used
10 herein; however, another type, for example, a highly responsive quantum photoelectric element such as
one using HgCdTe, InGaAs, InAsSb, or PbSe may be used.
[0030]
The information processor 4 is one including: an analog electric circuit including a buffer, an
amplifier, and the like; a digital electric circuit including a CPU, a memory, and the like; AD and DA
15 converters acting as interfaces between the analog and digital electric circuits; and the like. In addition,
the CPU and its peripheral devices cooperate in accordance with a predetermined program stored in a
predetermined area of the memory, a nd thereby as illustrated in FIG. 2, the information processor 4
fulfills functions as a light source control part 5 adapted to control the output of the semiconductor laser
2 and a signal processing part 6 adapted to receive an output signal from the light detector 3 and perform
20 a calculation process on the value of the output signal to calculate the concentration of the measurement
target component.
[0031]
The respective parts will be described below in detail.
The light source control part 5 is one adapted to output a current (voltage) control signal and
25 thereby control the current source (or the voltage source) of the semiconductor laser 2, and this allows
the drive current (or the drive voltage) of the semiconductor laser 2 to be changed at a predetermined
frequency, therefore modulating the oscillation wavelength of the laser light outputted from the
semiconductor laser 2 at the predetermined frequency.
[0032]
10
In this embodiment, the light source control part 5 changes the drive current into a sine wave
shape, and modulates the oscillation frequency into a sine wave shape (see a modulation signal in FIG.
4). Also, as illustrated in FIG. 3, the oscillation frequency of the laser light is adapted to be modulated
with the peak of the light absorption spectrum of the measurement target component as a center.
5 [0033]
The signal processing part 6 includes a first calculation part 61, a frequency component
extraction part 62, a second calculation part 63, and the like.
The first calculation part 61 is one adapted to calculate the logarithm of the ratio (hereinafter
also referred to as an intensity ratio logarithm) between the light intensity of the laser light (hereinafter
10 also referred to as measurement target light) having transmitted through the cell 1 in a state where the
sample gas is enclosed and light is absorbed by the measurement target component of the sample gas
and the light intensity of laser light (hereinafter also referred to as reference light) having transmitted
through the cell 1 in a state where light absorption is substantially zero.
[0034]
15 Describing more specifically, both of the former and latter light intensities are measured by the
light detector 3, the resulting pieces of measurement result data are stored in a predetermined area of the
memory, and the first calculation part 61 refers to the pieces of measurement result data to calculate the
intensity ratio logarithm.
[0035]
20 In addition, the former measurement (hereinafter also referred to as sample measurement) is of
course performed on every sample gas. The latter measurement (hereinafter also referred to as
reference measurement) may be performed any of before and after every sample measurement, or may
be performed, for example, once at appropriate timing, and the result of the measurement may be stored
in the memory and used in common for every sample measurement.
25 [0036]
Note that in this embodiment, in order to obtain the state where light absorption is substantially
zero, zero gas whose light absorption is substantially zero in a wavelength band where the measurement
target component absorbs light, for example, N2 gas is enclosed in the cell 1; however, another gas may
be enclosed, or the cell 1 may be evacuated.
11
[0037]
The frequency component extraction part 62 is one adapted to lock-in detect the intensity ratio
logarithm (hereinafter also referred to as an absorbance signal) calculated by the first calculation part 61
with a reference signal having a frequency n times (n is an integer equal to or more than 1) the
5 modulation frequency, and extract a frequency component having the frequency of the reference signal
from the intensity ratio logarithm. Note that the lock-in detection may be performed by digital
calculation or by calculation through an analog circuit. Also, the frequency component may be
extracted using not only the lock-in detection but also a method such as Fourier series expansion.
[0038]
10 The second calculation part 63 is one adapted to, on the basis of the result of the detection by
the frequency component extraction part 62, calculate the concentration of the measurement target
component.
[0039]
Next, an example of the operation of the analysis apparatus 100 will be described while
15 describing the respective parts above in detail.
[0040]
First, as described above, the light source control part 5 controls the semiconductor laser 2 to
modulate the wavelength of the laser light at the modulation frequency with the peak of the absorption
spectrum of the measurement target component as a center.
20 [0041]
Then, when zero gas is enclosed in the cell 1 by an operator or automatically, the first
calculation part 61 having sensed the enclosure performs the reference measurement.
Specifically, the first calculation part 61 receives an output signal from the light detector 3 in a
state where the zero gas is enclosed in the cell 1, and stores the value of the output signal in a
25 measurement result data storage part. When displaying the value of the output signal from the light
detector 3 in the reference measurement, i.e., reference light intensity in a time-series graph, a curve as
illustrated in (a) of FIG. 4 is obtained. That is, only a change in light output caused by the modulation
of the laser drive current (voltage) is displayed as the output signal from the light detector 3.
[0042]
12
Subsequently, when the sample gas is enclosed in the cell 1 by the operator or automatically,
the first calculation part 61 performs the sample measurement. Specifically, the first calculation part 61
receives an output signal from the light detector 3 in a state where the sample gas is enclosed in the cell 1,
and stores the value of the output signal in a predetermined area of the memory. When displaying the
5 value of the output signal from the light detector 3 in the sample measurement, i.e., measurement target
light intensity in the time-series graph, a curve as illustrated in (b) of FIG. 4 is obtained. It turns out that
an absorption peak appears in every half period of the modulation.
[0043]
After that, the first calculation part 61 synchronizes respective pieces of measurement data
10 with the modulation period, and calculates the intensity ratio logarithm between the light intensity of the
measurement target light and the light intensity of the reference light. Specifically, the first calculation
part 61 performs the calculation in accordance with the following expression (Expression 1).
[Expression 1]
15 Here, Dm(t) represents the measurement target light intensity, Dz(t) the reference light intensity,
and A(t) the intensity ratio logarithm (absorbance signal). When displaying the absorbance signal in
the graph with the horizontal axis as time, a curve as illustrated in (c) of FIG. 4 is obtained.
[0044]
The intensity ratio logarithm calculation may be performed by calculating the ratio between
20 the measurement target light intensity a nd the reference light intensity a nd then obtaining the logarithm
of the ratio, or obtaining the logarithm of the measurement target light intensity and the logarithm of the
reference light intensity, and then obtaining the difference between them.
[0045]
Then, the frequency component extraction part 62 lock-in detects the intensity ratio logarithm
25 with the reference signal having the frequency twice the modulation frequency, i.e., extracts the
frequency component having the frequency twice the modulation frequency, and stores the resulting
data (hereinafter also referred to as lock-in data) in a predetermined area of the memory.
[0046]
13
The value of the lock-in data is a value proportional to the concentration of the measurement
target component, and the second calculation part 63 calculates, on the basis of the value of the lock-in
data, a concentration indicated value indicating the concentration of the measurement target component.
[0047]
5 In such a configuration, even when laser light intensity varies due to some cause, a constant
offset is only added to the above-described intensity ratio logarithm but a waveform does not change.
Accordingly, the value of each frequency component calculated by lock-in detecting the intensity ratio
logarithm added with the constant offset does no t change, and a concentration indicated value also does
not change, therefore making it possible to expect accurate measurement.
10 [0048]
Describing the reason for this in detail, it is as follows.
In general, when expanding the absorbance signal A(t) into a Fourier series, it is expressed by
the following expression (Expression 2).
In addition, in the expression (Expression 2), an represents a value proportional to the
15 concentration of the measurement target component, and on the basis of the value an, the second
calculation part 63 calculates the concentration indicated value indicating the concentration of the
measurement target component.
[Expression 2]
20 Here, fm represents the modulation frequency, and n represents a multiple of the modulation
frequency.
[0049]
On the other hand, A(t) is also expressed by the above expression (Expression 1).
[0050]
25 Next, an absorbance signal A’(t) obtained when the laser light intensity varies α times due to
some cause during the measurement is expressed as the following expression (Expression 3).
[Expression 3]
14
[0051]
As is clear from this expression (Expression 3), A’(t) is equal to the absorption signal A(t),
which is a signal in the absence of a variation in the laser light intensity, only added with –ln(α) as a
5 constant value, and therefore it turns out that even when the laser light intensity changes, the value an of
each frequency component does not change.
[0052]
Accordingly, the concentration indicated value determined on the basis of the value of the
frequency component having the frequency twice the modulation frequency is not affected.
10 The above is the example of the operation of the analysis apparatus 100 at the time when the
sample gas does not contain an interference component other than the measurement target component.
[0053]
Next, an example of the operation of the present analysis apparatus 100 at the time when at
least one interference component (e.g., H2O) absorbing light at the peak light absorption wavelength of
15 the measurement target component is contained in the sample gas will be described.
[0054]
First, a principle will be described.
Since the light absorption spectrum of the measurement target component a nd the light
absorption spectrum of the interference component are different in shape, absorbance signals obtained
20 when the respective components are independently present are different in waveform, and therefore the
ratios of respective frequency components are different (linearly independent). By utilizing this and
solving simultaneous equation using the relationship between the values of respective frequency
components of measured absorption signals a nd preliminarily obtained respective frequency
components of absorbance signals of the measurement target component and the interference
25 component, the concentration of the measurement target component, from which the interference effect
has been corrected, can be obtained.
[0055]
Given that absorbance signals per unit density obtained when the measurement target
15
component and the interference component are independently present are respectively represented by
Am(t) a nd Ai(t), and respective frequency components of the absorbance signals are represented by anm
and ani, the following expressions (Expressions 4 and 5) hold.
[Expression 4]
[Expression 5]
[0056]
An absorbance signal value A(t) obtained when the measurement target component and the
10 interference component are present at concentrations of Cm a nd Ci, respectively is expressed by the
following expression (Expression 6) on the basis of the linearities of the respective absorbances.
[Expression 6]
[0057]
15 Given here that frequency components of A(t) at fm and 2fm are respectively represented by a1
and a2, the following simultaneous equations (Expression 7) hold from the above expression
(Expression 6).
[Expression 7]
20 [0058]
The frequency components anm and ani (n is a natural number a nd n = 1, 2 herein) respectively
obtained when the measurement target component and the interference component are independently
16
present can be preliminarily obtained by flowing corresponding span gases, and therefore simple and
certain calculation, i.e., solving the above simultaneous equations (Expression 7) allows the
concentration Cm of the measurement target gas, from which the interference effect has been eliminated,
to be determined.
5 [0059]
The analysis apparatus 100 operates on the basis of the above-described principle.
That is, the analysis apparatus 100 in this case stores, in a predetermined area of the memory,
the frequency components a1m, a2m, a1i, and a2i (independent frequency components in claims) of the
absorbance signals obtained by some means such as preliminarily make measurements while flowing
10 corresponding span gases when the measurement target component and the interference component are
independently present. Specifically, as in the above example, fo r each of the measurement target
component and the interference component, the measurement target light intensity and the reference
light intensity are measured to calculate the intensity ratio logarithm (absorbance signal) between them,
and from the intensity ratio logarithm, the frequency components a1m, a2m, a1i, and a2i are obtained by the
15 lock-in detection or another means and then stored. Note that it may be adapted not to store the
frequency components but to store the absorbance signals per unit concentration Am(t) and Ai(t)
(independent logarithmic intensity ratios in claims), and calculate the frequency components a1m, a2m, a1i,
and a2i from the above expression (Expression 4).
[0060]
20 Then, the analysis apparatus 100 specifies the measurement target component a nd the
interference component by inputs from an operator or the like.
[0061]
Subsequently, the first calculation part 61 calculates the intensity ratio logarithm A(t) in
accordance with the above expression (Expression 1).
25 After that, the frequency component extraction part 62 lock-in detects the intensity ratio
logarithm with reference signals respectively having the modulation frequency fm and the frequency 2fm
twice the modulation frequency fm to extract the respective frequency components a1 and a2 (pieces of
lock-in data), and stores them in the predetermined area of the memory.
[0062]
17
Then, the second calculation part 63 substitutes the values of the pieces of lock-in data a1 and
a2 and the values of the frequency components a1m, a2m, a1i, and a2i stored in the memory into the aboveexpression
(Expression 7), or performs equivalent calculation to calculate concentration (or a
concentration indicated value) Cm indicating the concentration of the measurement target gas, from
5 which the interference effect has been removed. At this time, the concentration (concentration
indicated value) Ci of each interference component may be calculated.
[0063]
Note that even when two or more interference components can be assumed to be present, by
adding higher order frequency components whose number corresponds to the number of the
10 interference components and solving simultaneous equations where the number of unknowns is equal to
the number of components, the concentration of the measurement target component, from which the
interference effect has been removed, can be determined.
[0064]
That is, in general, when the total number of a measurement target component and
15 interference components existing is n, given that the frequency component of the k-th gas component at
i x fm is represented by aik, and the concentration of the k-th gas component is represented by Ck, the
following expression (Expression 8) holds.
[Expression 8]
20 [0065]
By solving the simultaneous equations with n unknowns expressed by this expression
(Expression 8), the concentrations of the measurement target component a nd the interference
components can be determined.
[0066]
25 Also, it may be adapted to add a higher harmonic component of an order higher than n to
18
prepare simultaneous equations where the number of equations is larger than the number of gas
components, and determine the concentrations of the respective components using the least squares
method. In doing so, the concentrations can be determined with errors due to measurement noise
being small.
5 [0067]
Note that when the total number of a measurement target component and interference
components is n, the concentrations of the respective components are calculated, and there are some
components whose concentrations are equal to or less than a predetermined threshold value, it is
conceivable to recalculate the concentrations of components other than the components having
10 concentrations equal to or less than the threshold value.
[0068]
Specifically, as illustrated in FIG. 5, the second calculation part 63 solves the simultaneous
equations with n unknowns expressed by the above expression (Expression 8) to calculate the
concentrations of the n components (S1). Then, a determination part provided in the signal processing
15 part 6 determines whether or not there is any component whose component is equal to or less than the
predetermined threshold value (S2). When there are j components whose concentrations are equal to
or less than the threshold value, the second calculation part 63 recalculates the concentrations of (n – j)
components other than the j components whose concentrations are equal to or less than the threshold
value using simultaneous equations with (n – j) unknowns expressed on the basis of the same idea as for
20 the above expression (Expression 8) (S3). This makes it possible to accurately calculate the
concentrations of the existing gas components. The above series of calculation is repeated until no
component whose concentration is equal to or less than the threshold value is detected, or the
concentration of the measurement target component is repeatedly calculated a predetermined number of
times.
25 [0069]
In addition, as operation after it has been determined that no component whose concentration
is equal to or less than the threshold value is left, for example, an embodiment adapted to determine
whether or not any of calculated concentrations has an abnormal value can be cited (S4). In S4, when
there is an abnormal value, the second calculation part 63 checks previously calculated concentrations
19
(S5) to determine whether or not any of the previously calculated concentration has an abnormal value.
On the other hand, in S4, when there is no abnormal value, the concentrations no t exhibiting any
abnormal value are outputted (S6).
[0070]
5 Note that the present invention is not limited to the above-described embodiment.
For example, the light source control part 5 in the above-described embodiment is o ne adapted
to make the semiconductor laser operate in CW oscillation, but as illustrated in FIG. 6, may be one
adapted to make the semiconductor laser operate in quasi-CW oscillation. In this case, the light source
control part 5 outputs the current (or voltage) control signal to thereby control the current source (or the
10 voltage source) fo r the semiconductor laser 2, and sets the drive current (the drive voltage) from the
current source (or the voltage source) to a predetermined threshold value or more for pulse oscillation.
Specifically, the light source control part 5 is one adapted to make the semiconductor laser operate in the
quasi-CW oscillation as the pulse oscillation having a predetermined pulse width (e.g., 10 to 50 ns, duty
ratio of 5 %) and repeated with a predetermined period (e.g., 1 to 5 MHz). In addition, the light source
15 control part 5 is one adapted to change the drive current (drive voltage) from the current source (or the
voltage source) at a predetermined frequency with the drive current (drive voltage) set to a value that is
for the wavelength sweeping and lower than the threshold value fo r the pulse oscillation, and thereby
change temperature to sweep the oscillation wavelength of the laser light. A modulation signal for
modulating the drive current changes in a triangular wave shape, sawtooth wave shape, or sine wave
20 shape, and its frequency is, for example, 1 to 100 Hz.
[0071]
A light intensity signal obtained by the light detector as a result of making the semiconductor
laser operate in quasi-CW oscillation is as illustrated in FIG. 7. As illustrated, an absorption spectrum
can be obtained as the entire pulse train. In this case, as the light detector, the use of a high response
25 speed detector such as an MCT is not required, but the use of a lo w response speed thermal detector
such as a thermopile is possible. In addition, as compared with the CW oscillation, the quasi-CW
oscillation makes it possible to decrease the power consumption of the light source, facilitate waste heat
disposal, and further prolong the life of the light source.
[0072]
20
Also, as illustrated in FIG. 8, the analysis apparatus 100 may be one including multiple
semiconductor lasers 2 as light sources adapted to irradiate the measurement cell 1 with laser lights. In
this case, as illustrated in FIG. 9, a signal processer 4 in the analysis apparatus 100 fulfills functions as a
light source control part 5 adapted to control the outputs of the semiconductor lasers 2, a signal
5 separation part 7 adapted to separate signals corresponding to the respective semiconductor lasers 2 from
a light intensity signal obtained by the light detector 3, a signal processing part 6 adapted to receive the
signals resulting from the separation by the signal separation part 7 and corresponding to the respective
semiconductor lasers 2, and perform a calculation process on the values of the signals to calculate the
concentrations of measurement target components, and the like.
10 [0073]
The light source control part 5 is one adapted to make the respective multiple semiconductor
lasers 2 operate in pulse oscillations, and modulate the oscillation wavelengths of the laser lights at a
predetermined frequency. Also, the light source control part 5 is adapted to control the multiple
semiconductor lasers 2 such that the multiple semiconductor lasers 2 have the oscillation wavelengths
15 corresponding to the respective different measurement target components, and the respective multiple
semiconductor lasers 2 operate in the pulse oscillations with mutually different oscillation periods at
mutually different oscillation timings.
[0074]
Specifically, the light source control part 5 outputs current (or voltage) control signals to
20 thereby control the current sources (or the voltage sources) for the respective semiconductor lasers 2.
As illustrated in FIG. 3, the light source control part 5 in the present embodiment is o ne adapted to make
the respective semiconductor lasers 2 operate in quasi-CW oscillations as pulse oscillations having a
predetermined pulse width (e.g., 10 to 100 nm, duty ratio 5 %) and repeated with a predetermined period
(e.g., 1 to 5 MHz).
25 [0075]
Also, as illustrated in FIG. 6, the light source control part 5 is o ne adapted to change drive
currents (drive voltages) from the current sources (or the voltage sources) at a predetermined frequency,
and thereby change temperatures to sweep the oscillation wavelengths of the laser lights. Further, as
illustrated in FIG. 3, the oscillation wavelengths of the laser lights are modulated with the peaks of the
21
absorption spectra of the measurement target components as centers. Modulation signals for changing
the drive currents are signals that change in a triangular wave shape, sawtooth wave shape, or sine wave
shape, and have a frequency of, fo r example, 1 to 100 Hz. Note that FIG. 6 illustrates an example
where a modulation signal changes in a triangular wave shape.
5 [0076]
The light intensity signal obtained by the light detector 3 as a result of making one of the
semiconductor lasers 2 operate in the quasi-CW oscillation in the above manner is as illustrated in FIG. 7.
As illustrated, an absorption spectrum can be obtained as the entire pulse train.
[0077]
10 Also, the light source control part 5 makes the multiple semiconductor lasers 2 operate in the
pulse oscillations at the mutually different timings. Specifically, as illustrated in FIG. 10, the multiple
semiconductor lasers 2 sequentially operate in the pulse oscillations, and within one period of pulse
oscillation produced by one of the semiconductor lasers 2, one pulses from the other semiconductor
lasers 2 are included. That is, between mutually adjacent pulses from o ne of the semiconductor lasers
15 2, one pulses from the other semiconductor lasers 2 are included. In this case, pulses from the multiple
semiconductor lasers 2 are oscillated so as to prevent the overlap of the pulses.
[0078]
The signal separation part 7 is one adapted to separate the signals corresponding to the
respective multiple semiconductor lasers 2 from the light intensity signal obtained by the light detector 3.
20 As illustrated in FIG. 11, the signal separation part 7 in the present embodiment includes: multiple
sample-and-hold circuits 71 provided corresponding to the respective multiple semiconductor lasers 2;
and AD converters 72 adapted to perform digital conversion of light intensity signals resulting from
separation by the sample-and-hold circuits 71.
[0079]
25 Each of the sample-and-hold circuits 71 uses a sampling signal synchronized with a current (or
a voltage) control signal for a corresponding semiconductor laser 2 to, at timing synchronized with
timing at which the semiconductor laser 2 operates in pulse oscillation, separate a signal corresponding
to the semiconductor laser 2 from the light intensity signal obtained by the light detector 3, and holds the
signal. Note that the sample-and-hold circuit 71 is configured to separate and hold the latter half of the
22
signal, which corresponds to a part of a pulse from the semiconductor laser 2. Since a signal
corresponding to the part of a pulse from the semiconductor laser 2 is separated, the pulse width of the
pulse does no t directly affect wavelength resolution, and therefore a reduction in wavelength resolution
can be suppressed without shortening the pulse width. As a result, as compared with a conventional
5 quasi-CW oscillation method, the wavelength resolution can be significantly improved. In addition,
since shortening the pulse width to prevent the reduction in wavelength resolution is not required, the
technical difficulty required for electronics for driving the semiconductor lasers 2 is reduced, and
correspondingly, cost is also reduced.
[0080]
10 When a semiconductor laser operates in pulse oscillation, temperature changes to thereby
change a wavelength. The degree of such a transient temperature change (wavelength change)
becomes smaller toward the latter half of a pulse, and therefore by separating a signal corresponding to
the latter half, the wavelength resolution is improved. For this reason, it is preferable that from the light
intensity signal obtained by the light detector 3, the signal separation part separates a signal
15 corresponding to the latter half of a pulse from each light source. This configuration makes it possible
to significantly improve the wavelength resolution as compared with the conventional quasi-CW
oscillation method by relatively widening the pulse width of a pulse (e.g., approximately 100 ns) and
setting a sampling position to a time position within the pulse as far back as possible (e.g., 85 to 95 ns
after the rise of the pulse). In addition, the sampling is performed within a predetermined time width.

WE CLAIM:
1. An analysis apparatus adapted to analyze a measurement target component contained in a
sample, the analysis apparatus comprising:
a light source adapted to emit reference light that is light whose wavelength is modulated at a
5 predetermined modulation frequency;
a light detector adapted to detect intensity of measurement target light that is light after the
reference light has transmitted through the sample, and intensity of the reference light;
a first calculation part adapted to calculate an intensity ratio logarithm that is a logarithm of a
ratio between the intensity of the measurement target light and the intensity of the reference light;
10 a frequency component extraction part adapted to, from the intensity ratio logarithm, extract a
frequency component having a frequency n times (n is an integer equal to or more than 1) the
modulation frequency; and
a second calculation part adapted to, on a basis of a result of the frequency component
extraction by the frequency component extraction part, calculate concentration or absorbance of the
15 measurement target component.
2. The analysis apparatus as claimed in claim 1, adapted to analyze the measurement target
component in a sample containing one or more interference components, wherein
the frequency component extraction part is one adapted to, from the intensity ratio logarithm
20 calculated by the first calculation part, extract mutually different frequency components whose number
is equal to or more than a sum of the number of the measurement target component and the number of
the interference components,
the analysis apparatus storing independent frequency components obtained by extracting
frequency components from intensity ratio logarithms when the measurement target component and the
25 respective interference components are independently present or storing independent intensity ratio
logarithms that are intensity ratio logarithms of the measurement target component and the respective
interference components per unit concentration, wherein
the second calculation part is one adapted to calculate the concentration of the measurement
target component on a basis of a result of the extraction of the respective frequency components by the
25
frequency component extraction part a nd the independent frequency components or the independent
intensity ratio logarithms.
3. The analysis apparatus as claimed in claim 2, wherein
5 the second calculation part is one adapted to calculate the concentration of the measurement
target component by solving simultaneous equations including the result of the extraction of the
respective frequency components by the frequency component extraction part, the independent
frequency components or the independent intensity ratio logarithms of the measurement target
component and the respective interference components, and concentrations of the measurement target
10 component and the respective interference components.
4. The analysis apparatus as claimed in claim 3, wherein
the frequency component extraction part is o ne adapted to extract frequency components
whose number is larger than the sum of the number of the measurement target component and the
number of the interference components, and
the second calculation part is one adapted to calculate the concentration of the measurement
target component with use of a least squares method from simultaneous equations where the number of
unknowns is larger than the sum of the number of the measurement target component and the number of
the interference components.
5. The analysis apparatus as claimed in claims 2, wherein
the interference components are a largest number of gas components assumed to be contained
in the sample.
25 6. The analysis apparatus as claimed in any one of claims 1, wherein
the frequency component extraction part is one adapted to extract the frequency component or
the frequency components by performing lock-in detection at the frequency n times (n is an integer
equal to or more than 1) the modulation frequency.
26
7. The analysis apparatus as claimed in any of claim s 1, wherein
the first calculation part is one adapted to calculate the intensity ratio logarithm by obtaining
the ratio between the intensity of the measurement target light and the intensity of the reference light and
obtaining a logarithm of the ratio.
8. The analysis apparatus as claimed in any of claims 1, wherein
the first calculation part is one adapted to calculate the intensity ratio logarithm by obtaining a
logarithm of the intensity of the measurement target light as well as obtaining a logarithm of the intensity
of the reference light, and subtracting the logarithm of the intensity of the reference light from the
logarithm of the intensity of the measurement target light.
9. The analysis apparatus as claimed in any one of claims 1, wherein
the measurement target component is one contained in sample gas such as exhaust gas, and
the light source is a semiconductor laser adapted to emit the reference light modulated in a
wavelength range including a peak wavelength of a light absorption spectrum of the measurement target
component,
the analysis apparatus further comprising a cell into which the sample gas is introduced,
wherein
the cell is irradiated with the reference light emitted from the semiconductor laser, and the light
detector is arranged in a light path of the measurement target light having transmitted through the cell.
10. The analysis apparatus as claimed in any one of claims 1, further comprising
a light source control part adapted to make the light source operate in quasi-continuous-wave
oscillation, and generate a temperature change based on current modulation to sweep an oscillation
wavelength.
11. The analysis apparatus as claimed in claim 10, further comprising
a signal separation part adapted to, from a light intensity signal obtained by the light detector,
separate a signal corresponding to a part of a pulse from the light source.
12. The analysis apparatus as claimed in claim 11, wherein
the signal separation part is one adapted to separate a signal corresponding to a latter half of the
pulse ofthe light source.
13. Anon-transitory computer-readable medium storing a program for an analysis apparatus,
the program being applied to an analysis apparatus that in order to analyze a measurement target
component contained in a sample, comprises: a light source adapted to emit reference light that is light
whose wavelength is modulated at a predetermined modulation frequency; and a light detector adapted
to detect intensity of measurement target light that is light after the reference light has transmitted
through the sample, and intensity ofthe reference light,
the program instructing the analysis apparatus to fulfill functions as:
a first calculation part adapted to calculate an intensity ratio logarithm that is a logarithm of a
ratio between the intensity ofthe measurement target light and the intensity ofthe reference light;
a frequency component extraction part adapted to, from the intensity ratio logarithm, extract a
frequency component having a frequency n times (n is an integer equal to or more than 1) the
modulation frequency; and
a second calculation part adapted to, on a basis of a result of the frequency component
extraction by the frequency component extraction part, calculate concentration or absorbance of the
measurement target component.
14. An analysis method adapted to analyze a measurement target component contained in a
sample, the analysis method
emitting reference light that is light whose wavelength is modulated at a predetermined
modulation frequency;
detecting intensity of measurement target light that is light after the reference light has
transmitted through the sample, and intensity ofthe reference light;
calculating an intensity ratio logarithm that is a logarithm of a ratio between the intensity ofthe
measurement target light and the intensity ofthe reference light;
from the intensity ratio logarithm, extracting a frequency component having a frequency n
times (n is an integer equal to or more than 1) the modulation frequency; and
on a basis of a result of the extraction of the frequency component, calculating concentration
or absorbance of the measurement target component.
15. The analysis method as claimed in claim 14, adapted to analyze a measurement target
component in a sample containing one or more interference components, the analysis method
from an intensity ratio logarithm obtained from the sample, extracting mutually different
frequency components whose number is equal to or more than a sum of the number of the interference
components and one,
extracting independent frequency components that are frequency components from intensity
ratio logarithms when the measurement target component and the respective interference components
are independently present, or measuring independent intensity ratio logarithms that are intensity ratio
logarithms of the measurement target component and the respective interference components per unit
concentration, and
calculating the concentration of the measurement target component on a basis of a result of the
extraction of the respective frequency components obtained from the sample and the independent
frequency components or the independent intensity ratio logarithms.