Stable isotope measurement method and apparatus by spectroscopy

[0002] This is a division of application Ser. No. 09/341,045, filed Jul. 1, 1999, which is incorporated herein by reference, which is a 371 of PCT/JP98/00097, filed Jan. 12, 1998.

[0003] 1. Technical Field of the Invention

[0004] Isotopic analyses are useful for diagnosis of a disease in a medical application, in which metabolic functions of a living body can be determined by measuring a change in the concentration or concentration ratio of an isotope after administration of a drug containing the isotope. In the other fields, the isotopic analyses are used for studies on the photosynthesis and metabolism of plants, and for ecological tracing in a geochemical application.

[0005] The present invention relates to stable isotope measurement methods and apparatus for spectrometrically measuring the concentration or concentration ratio of an isotopic gas on the basis of the light absorption characteristics of the isotope.

[0006] 2. Background Art

[0007] It is generally known that gastric ulcer and gastritis are caused by bacteria called helicobacter pylori(HP) as well as by a stress.

[0008] If the HP is present in the stomach of a patient, an antibiotic or the like should be administered to the patient for bacteria removal treatment. Therefore, it is indispensable to check if the patient has the HP. The HP has a strong urease activity for decomposing urea into carbon dioxide and ammonia.

[0009] Carbon has isotopes having mass numbers of 12, 13 and 14, among which ¹³C having a mass number of 13 is easy to handle because of its non-radioactivity and stability.

[0010] If the concentration of ¹³CO₂as a final metabolic product or the concentration ratio of ¹³CO₂to ¹²CO₂in breath of a patient is successfully measured after urea labeled with the isotope ¹³C is administered to the patient, the presence of the HP can be confirmed.

[0011] However, the concentration ratio of ¹³CO₂to ¹²CO₂in naturally occurring carbon dioxide is 1:100. Therefore, it is difficult to determine the concentration ratio in the breath of the patient with high accuracy.

[0012] There have been known methods for determining the concentration ratio of ¹³CO₂to ¹²CO₂by way of infrared spectroscopy (see Japanese Examined Patent Publications No. 61-42219 (1986) and No. 61-42220 (1986)).

[0013] In the method disclosed in Japanese Examined Patent Publication No. 61-42220, two cells respectively having a long path and a short path are provided, the path lengths of which are adjusted such that the light absorption by ¹³CO₂in one cell is equal to the light absorption by ¹²CO₂in the other cell. Light beams transmitted through the two cells are led to the detectors, in which the light intensities are measured at wavelengths which ensure the maximum sensitivity. In accordance with this method, the light absorption ratio for the concentration ratio of ¹³CO₂to ¹²CO₂in naturally occurring carbon dioxide can be adjusted to 1. If the concentration ratio is changed, the light absorption ratio also changes by the amount of a change in the concentration ratio. Thus, the change in the concentration ratio can be determined by measuring the change in the light absorption ratio.

[0014] A. However, the method for determining the concentration ratio according to the aforesaid literature suffers from the following drawback.

[0015] Calibration curves for determining the concentrations of ¹²CO₂and ¹³CO₂should be prepared by using gas samples each having a known ¹²CO₂concentration and gas samples each having a known ¹³CO₂concentration.

[0016] To prepare the calibration curve for the ¹²CO₂concentration, the ¹²CO₂absorbances are measured for different ¹²CO₂concentrations. The ¹²CO₂concentrations and the ¹²CO₂absorbances are plotted as abscissa and ordinate, respectively, and the calibration curve is determined by the method of least squares.

[0017] The calibration curve for the ¹³CO₂concentration is prepared in the same manner as described above.

[0018] The ¹³CO₂concentration or the ¹³CO₂concentration ratio (which is herein meant by ¹³CO₂concentration/¹²CO₂concentration) in the breath as a test gas sample is typically determined by way of infrared spectroscopy. In this case, since a test sample gas, or breath is exhaled from a living body as a result of the metabolism, the breath contains water vapor in a concentration proximate to saturation.

[0019] In the infrared spectroscopy, the absorption of infrared radiation with a particular wavelength by a test gas sample is utilized for determination of the absorbance for the test gas sample.

[0020] FIG. 5 is a graph obtained by plotting the measured values of the ¹³CO₂concentration ratio changes with respect to the humidities of test gas samples having different humidities ranging from 0% to 100% wherein the ¹³CO₂concentration ratio with respect to a 0%-humidity gas sample is used as a reference gas sample.

[0021] As can be seen from the graph, the measured values of the ¹³CO₂concentration ratio are not the same, but vary depending on the humidity.

[0022] Therefore, if the ¹³CO₂concentration or the ¹³CO₂concentration ratio of a test gas sample containing moisture is measured in ignorance of this fact, the measured value is apparently greater than the true value.

[0023] One approach to this problem is to remove the moisture contained in the breath sample as the test gas sample through molecular sieving or with the use of a moisture absorbent such as magnesium perchlorate prior to the measurement. However, some problems may be encountered in this approach since the approach requires a large space for housing the moisture absorbent, there is no means for checking if the moisture is completely removed by the moisture absorbent, and the moisture absorbent should periodically be replaced with a new one.

[0024] It is, therefore, an object of the present invention to provide a stable isotope measurement method and apparatus for spectrometrically analyzing an isotopic gas, wherein a test gas sample containing carbon dioxide ¹³CO₂as a component gas is introduced into a cell and the concentration or concentration ratio of the component gas is precisely measured and corrected by measuring moisture content in the test gas sample.

[0025] A stable isotope measurement method for spectrometrically analyzing an isotopic gas in accordance with the present invention comprises: a first step of introducing a test gas sample into a cell and determining the absorbance of light transmitted therethrough at a wavelength suitable for the component gas ¹³CO₂; a second step of determining a concentration of the component gas in the test gas sample on the basis of a calibration curve prepared through measurement on test gas samples each containing the component gas in a known concentration; and a third step of measuring a concentration of water vapor contained in the test gas sample and correcting a concentration of the component gas contained in the test gas sample in accordance with the measured water vapor concentration on the basis of a correction curve prepared through measurement on test gas samples each containing water vapor in a known concentration.

[0026] A stable isotope measurement method for spectrometrically analyzing an isotopic gas in accordance with the present invention comprises: a first step of introducing a test gas sample containing carbon dioxide ¹³CO₂and carbon dioxide ¹³CO₂as component gases into a cell and determining the absorbances of light transmitted therethrough at wavelengths suitable for the respective component gases; a second step of determining a concentration ratio between the component gases in the test gas sample on the basis of a calibration curve prepared through measurement on test gas samples each containing the component gases in known concentrations; and a third step of measuring a concentration of water vapor contained in the test gas sample and correcting a concentration ratio between the component gases contained in the test gas sample in accordance with the measured water vapor concentration on the basis of a correction curve prepared through measurement on test gas samples each containing water vapor in a known concentration.

[0027] When compared with the prior art method, each of the aforesaid methods additionally include the third step in which the concentration ratio of the component gas is corrected in accordance with the measured water vapor concentration on the basis of the correction curve prepared through the measurement on the test gas samples each containing water vapor in a known concentration.

[0028] Although the concentration of the component gas should basically be represented by a single true value, the measured value of the concentration of the component gas varies depending on the water vapor concentration. In view of this fact, the aforesaid methods improve the measurement accuracy of the concentration ratio of the component gas.

[0029] The water vapor concentration may otherwise be determined by means of any of various humidity sensors, or may be calculated from the absorbance determined spectrometrically on the basis of the water molecule spectrum.

[0030] In the method of claim 2, the correction curve in the third step is prepared by determining the light absorbances at the wavelengths suitable for the respective component gases for the plurality of test gas samples containing water vapor in different concentrations, then determining the concentrations of or concentration ratios between the respective component gases in the test gas samples on the basis of the calibration curve, and plotting ratios or differences between the concentrations of or the concentration ratios between the respective component gases in the gas samples thus determined with respect to the water vapor concentrations, and the correction in the third step is achieved by obtaining a concentration correction value or a concentration ratio correction value for the component gases by fitting the water vapor concentration of the test sample gas obtained in the third step to the correction curve, and then dividing the concentrations of or the concentration ratio between the respective component gases in the test gas sample obtained in the second step by the concentration correction value or the concentration ratio correction value obtained on the basis of the correction curve, or subtracting the concentration correction value or the concentration ratio correction value from the concentrations of or the concentration ratio between the respective component gases in the test gas sample.

[0031] A stable isotope measurement apparatus for spectrometrically analyzing an isotopic gas in accordance with the present invention is a measurement apparatus adapted to perform the aforesaid methods for spectrometrically analyzing the isotopic gas and comprises, as data processing means, absorbance calculation means for determining the absorbances of light transmitted through the test gas sample introduced into the cell on the basis of light intensities measured at the wavelengths suitable for the respective component gases, concentration calculation means for determining the concentration ratio of the component gases on the basis of the calibration curve prepared through the measurement on the test gas samples each containing the component gases in known concentrations, water vapor concentration measuring means for measuring the concentration of water vapor contained in the test gas sample, and correction means for correcting the concentration ratio between the component gases in the test gas sample in accordance with the, measured water vapor concentration on the basis of the correction curve prepared through the measurement on the gas samples each containing water vapor in a known concentration.

[0032] In the methods or apparatus for spectrometrically analyzing the isotopic gas in accordance with the present invention, when a test gas sample containing carbon dioxide ¹³CO₂as a component gas is introduced into the cell and then spectrometrically analyzed, the concentration ratio of the component gas is corrected in accordance with the water vapor concentration in the test gas sample. Therefore, the concentration ratio of the component gas can be determined with a higher accuracy.

[0033] B. In the infrared spectrometric analysis, the ¹²CO₂concentration in a breath sample obtained before the drug administration is calculated from the measured ¹²CO₂absorbance on the basis of a ¹²CO₂calibration curve, while the ¹³CO₂concentration in the breath sample is calculated from the measured ¹³CO₂absorbance on the basis of a ¹³CO₂calibration curve. The ¹²CO₂and ¹³CO₂concentrations in the breath sample obtained after the drug administration are determined in the same manner.

[0034] If the CO₂concentrations in the two breath samples are substantially the same, it is possible to use narrower ranges of the ¹²CO₂calibration curve and the ¹³CO₂calibration curve. Thus, the measurement accuracy can be improved by using limited ranges of the calibration curves.

[0035] For equalization of the CO₂concentrations in the two breath samples, either one of the breath samples should be diluted. Typically used as a gas for dilution (hereinafter referred to as “diluent gas”) is nitrogen gas which exhibits no absorption in the infrared region of the radiation spectrum (nitrogen gas is used as the diluent gas in the embodiment of the invention disclosed in Japanese Unexamined Patent Publication No. 9-1665546 (1997) which was filed prior to the present invention).

[0036] In this dilution method, however, the diluted breath sample has a different component gas ratio from the undiluted breath sample, because diluent gas contains only nitrogen but breath sample contains oxygen, moisture and etc. as well as nitrogen.

[0037] As a result, the difference in the component gas ratio influences the determination of the ¹³CO₂concentration and the concentration ratios between ¹²CO₂and ¹³CO₂, so that the measured values may be erroneous.

[0038] It is, therefore, another object of the present invention to provide a method for spectrometrically analyzing an isotopic gas, wherein a breath sample as a test gas sample containing a plurality of component gases is introduced into a cell and the concentrations of the component gases are precisely measured through spectrometry by diluting the test gas sample in such a manner that the component gas composition in the test gas sample is not changed.

[0039] To achieve this object, there is provided a stable isotope measurement method for spectrometrically analyzing an isotopic gas, wherein two test gas samples are sampled from a single subject and, if the CO₂concentration of one of the test gas samples is higher than the CO₂concentration of the other test gas sample, the one test gas sample is diluted with air (atmospheric air) to a CO₂concentration level which is equivalent to that of the other test gas sample for measurement of the concentration ratios ¹³CO₂/¹²CO₂in the respective test gas samples (claim 5).

[0040] In this method, the two breath samples are analyzed on condition that the breath samples have the same CO₂concentration level. This makes it possible to use limited ranges of the calibration curves. In addition, the component gas composition in the breath sample is not changed by the dilution because air is used as the diluent gas. As a result, the measurement accuracy can be improved.

[0041] Methods according to present claims 6 and 7 each provide a more specific procedure for the method for spectrometrically analyzing the isotopic gas in accordance with claim 5, and are each based on the precondition that a first test gas sample is first filled in a single cell for measurement of the intensity of light transmitted therethrough and, after the first test gas sample is discharged from the cell, a second test gas sample is filled in the cell for measurement of the intensity of light transmitted therethrough.

[0042] As described above, the CO₂concentrations in the two test gas samples can be generally equalized by diluting either one of the two test gas samples so as not to change the component gas composition of the breath sample. This makes it possible to use limited ranges of the ¹²CO₂and ¹³CO₂calibration curves. The accuracy of the calibration curves is increased as the ranges of the calibration curves to be used are narrowed. Therefore, the measurement accuracy can be improved by limiting the ranges of the calibration curves to be used.

[0043] FIG. 1 is a block diagram illustrating the overall construction of an apparatus for spectrometrically analyzing an isotopic gas.

[0044] FIGS. 2A to 2D are diagrams illustrating gas flow paths in the apparatus for spectrometrically analyzing the isotopic gas. Particularly, FIGS. 2A and 2C are diagrams illustrating gas flow paths to be employed when a cell is cleaned by passing a clean reference gas therethrough. FIG. 2B is a diagram illustrating gas flow path to be employed when a base gas is sucked into a gas injector 21 from a breath sampling bag and then mechanically pushed out into the gas flow path at a constant rate. FIG. 2D is a diagram illustrating a gas flow path to be employed when a sample gas is sucked into the gas injector 21 from a breath sampling bag and then mechanically pushed out into the gas flow path at a constant rate.

[0045] FIGS. 3A to 3E are diagrams illustrating gas flow paths in the apparatus for spectrometrically analyzing the isotopic gas. Particularly, FIGS. 3A and 3D are diagrams illustrating gas flow paths to be employed when a cell is cleaned by passing a clean reference gas therethrough. FIG. 3B-1 is a diagram illustrating a gas flow path to be employed when a predetermined amount of the reference gas is sucked into the gas injector 21. FIG. 3B-2 is a diagram illustrating a gas flow path to be employed when a predetermined amount of air is sucked into the gas injector 21 with a three-way valve V4 opened to the atmospheric air. FIG. 3C is a diagram illustrating a gas flow path to be employed when a base gas is sucked into the gas injector 21 from a breath sampling bag and then mechanically pushed out into the gas flow path at a constant rate. FIG. 3E is a diagram illustrating a gas flow path to be employed when a sample gas is sucked into the gas injector 21 from a breath sampling bag and is mechanically injected into the gas flow path at a constant rate.

[0046] FIG. 4 is a graph prepared in such a manner that sample gases having different humidities and a base gas having a humidity of 0% were prepared by mixing a CO₂gas having a predetermined ¹³CO₂concentration and containing no moisture and a CO₂gas having the predetermined ¹³CO₂concentration and containing moisture, and differences ΔV between an output value for the humidity of the base gas and output values for the humidities of the sample gases detected by a humidity sensor 19 and differences between the ¹³CO₂concentration ratio in the base gas and the ¹³CO₂concentration ratios in the sample gases determined on the basis of a calibration curve were plotted as abscissa and ordinate, respectively.

[0047] FIG. 5 is a graph illustrating a relationship between the humidity and the ¹³CO₂concentration ratio for the sample gas having different humidities.

[0048] With reference to the attached drawings, embodiments of the present invention will hereinafter be described which are adapted for a case where the ¹³CO₂concentration ratio in a breath sample is spectrometrically determined after administration of an urea diagnostic drug labeled with an isotope ¹³C.

[0049] I. Breath sampling test

[0050] Before the urea diagnostic drug is administered to a patient, breath of the patient is collected in a breath sampling bag. The volume of the breath sampling bag is about 250 ml. Then, the urea diagnostic drug is orally administered to the patient and, after a lapse of 10 to 15 minutes, breath of the patient is collected in another breath sampling bag in the same manner as in the previous breath sampling.

[0051] The breath sampling bags obtained before and after the drug administration are respectively attached to predetermined nozzles of an apparatus for spectrometrically analyzing an isotopic gas, and the following automatic control is performed.

[0052] II. Apparatus for spectrometrically analyzing isotopic gas

[0053] FIG. 1 is a block diagram illustrating the overall construction of the apparatus for spectrometrically analyzing the isotopic gas.

[0054] The breath sampling bag containing the breath sample collected after the drug administration (hereinafter referred to as “sample gas”) and the breath sampling bag containing the breath sample collected before the drug administration (hereinafter referred to as “base gas”) are respectively attached to the predetermined nozzles of the apparatus. The breath sampling bag containing the base gas is connected to a valve V3 through a resin or metal pipe (hereinafter referred to simply as “pipe”), while the breath sampling bag containing the sample gas is connected to a valve V2 through a pipe.

[0055] A reference gas (any gas exhibiting no absorption at a wavelength for measurement, e.g., nitrogen gas) is supplied from a gas tank to the apparatus. The reference gas flows through a pressure release valve 31, a valve V0, a regulator 32 and a flow meter 33, and is diverged into a reference cell llcthrough a needle valve 35 and into a first sample cell llafor measuring the ¹²CO₂absorbance through a valve V1 and a check valve 36.

[0056] A gas injector 21 (volume: 70 cm³) for quantitatively injecting the sample gas or the base gas is connected to a flow path between the valve V1 and the first sample cell 11avia a three-way valve V4. The gas injector 21 is a syringe-like device having a piston and a cylinder. The piston is driven by cooperation of a motor M1, a feed screw connected to the motor M1 and a nut fixed to the piston.

[0057] As shown in FIG. 1, a cell chamber 11 has the first sample cell 11ahaving a shorter length for measuring a ¹²CO₂absorbance, a second shorter cell 11bhaving a longer length for measuring a ¹³CO₂absorbance, and the reference cell 11cthrough which the reference gas is passed. The first sample cell 11acommunicates with the second sample cell 11b. Gas is introduced into the first sample cell 11aand then into the second sample cell 11b, and discharged therefrom. The reference gas is introduced into the reference cell 11c. Then, a portion of the reference gas flows into a case 10 housing the cell chamber 11 and discharged therefrom, and the other portion of the reference gas flows into an infrared radiation source device L and discharged therefrom. Specifically, the first and second sample cells 11aand 11bhave lengths of 13 mm and 250 mm, respectively, and the reference cell 11chas a length of 236 mm.

[0058] A discharge pipe extending from the second sample cell 11bis provided with an O₂sensor 18 and a humidity sensor 19. Usable as the O₂sensor 18 are commercially available oxygen sensors, for example, a solid electrolyte gas sensor such as a zirconia sensor and an electrochemical gas sensor such as a galvanic cell sensor. Usable as the humidity sensor 19 are commercially available sensors such as utilizing a porous ceramic resistor and a polymer resistor.

[0059] The infrared radiation source device L has two waveguides 23aand 23bfor guiding an infrared beam. The generation of the infrared radiation may be achieved arbitrarily, for example, a ceramic heater (surface temperature: 450° C.) and the like can be used. A rotary chopper 22 for periodically blocking and passing the infrared beams is provided adjacent to the infrared radiation source device L. A light path along which an infrared beam emitted from the infrared radiation source device L is transmitted though the first sample cell 11aand the reference cell 11cis herein referred to as “first light path”, while a light path along which an infrared beam is transmitted through the second sample cell 11bis herein referred to as “second light path”.

[0060] A reference character D denotes an infrared beam detector for detecting the infrared beams transmitted through the cells. The infrared beam detector D has a first interference filter 24aand a first detection element 25adisposed in the first light path, and a second interference filter 24band a second detection element 25bdisposed in the second light path.

[0061] The first interference filter 24a(band width: about 20 nm) transmits infrared radiation having a interference of about 4,280 nm for measurement of the ¹²CO₂absorbance. The second interference filter 24b(band width: about 50 nm) transmits infrared radiation having a wavelength of about 4,412 nm for measurement of the ¹³CO₂absorbance. Usable as the first and second detection elements 25aand 25bare any elements capable of detecting infrared radiation, for example, a semiconductor infrared sensor such as of PbSe.

[0062] The first interference filter 24aand the first detection element 25aare housed in a package 26afilled with an inert gas such as Ar. Similarly, the second interference filter 24band the second detection element 25bare housed in a package 26bfilled with an inert gas.

[0063] The whole infrared beam detector D is maintained at a constant temperature (25° C.) by means of a heater and a Peltier element 27. The detection elements in the packages 26aand 26bare kept at 0° C. by means of a Peltier element.

[0064] The cell chamber 11 is formed of a stainless steel, and vertically or laterally sandwiched between heaters 13.

[0065] The cell chamber 11 has two tiers. The first sample cell 11aand the reference cell 11care disposed in one tier, and the second sample cell 11bis disposed in the other tier. The first light path extends through the first sample cell 11aand the reference cell 11cwhich are disposed in series, and the second light path extends through the second sample cell 11b. Reference characters 15, 16 and 17 denote sapphire transmission windows through which the infrared radiation is transmitted.

[0066] The cell chamber 11 is kept at a constant temperature (40° C.) by controlling the heaters 13.

[0067] III. Measuring procedure

[0068] In the measurement, the CO₂concentrations of the base gas and the sample gas are adjusted to substantially the same level. For this purpose, the CO₂concentrations of the base gas and the sample gas are measured in a preliminary measurement. If the preliminarily measured CO₂concentration of the base gas is higher than the preliminarily measured CO₂concentration of the sample gas, the CO₂concentration of the base gas is measured after the base gas is diluted to a CO₂concentration level equivalent to that of the sample gas, and then the CO₂concentration of the sample gas is measured in a main measurement.

[0069] If, in the main measurement, the preliminarily measured CO₂concentration of the base gas is lower than the preliminarily measured CO₂concentration of the sample gas, the CO₂concentration of the base gas is measured as it is, and the CO₂concentration of the sample gas is measured after the sample gas is diluted to a CO₂concentration level equivalent to that of the base gas.

[0070] The measuring procedure includes reference gas measurement, preliminary base gas measurement, reference gas measurement, preliminary sample gas measurement, reference gas measurement, base gas measurement, reference gas measurement, sample gas measurement and reference gas measurement which are to be performed in this order.

[0071] III-1. Preliminary base gas measurement

[0072] The gas flow path and the cell chamber 11 in the apparatus for spectrometrically analyzing the isotopic gas are cleaned by passing the clean reference gas therethrough, and a reference light intensity is measured.

[0073] More specifically, the reference gas is sucked into the gas injector 21 with the three-way valve V4 opened to the side of the cell chamber 11 and with the valve V1 opened as shown in FIG. 2A, and then mechanically pushed out into the flow path from the gas injector 21 with the valve V1 closed to clean the first sample cell 11aand the second sample cell 11b. The reference gas is constantly passed through the reference cell 11c.

[0074] In turn, the base gas is sucked into the gas injector 21 from the breath sampling bag with the valve V3 opened as shown in FIG. 2B, and then mechanically pushed out into the flow path from the gas injector 21 at a constant flow rate. At this time, the intensity of light transmitted through the base gas is measured by means of the detection elements 25aand 25b, and the CO₂concentration of the base gas is determined from its absorbance on the basis of a calibration curve.

[0075] III-2. Preliminary sample gas measurement

[0076] The gas flow path and the cell chamber 11 in the apparatus for spectrometrically analyzing the isotopic gas are cleaned by passing the clean reference gas therethrough, and a reference light intensity is measured.

[0077] More specifically, the reference gas is sucked into the gas injector 21 with the valve V1 opened as shown in FIG. 2C, and then pushed out into the flow path from the gas injector 21 with the valve V1 closed to clean the first sample cell 11aand the second sample cell 11b.

[0078] In turn, the sample gas is sucked into the gas injector 21 from the breath sampling bag with the valve V2 opened as shown in FIG. 2D, and then mechanically pushed out into the flow path from the gas injector 21 at a constant flow rate. At this time, the intensity of light transmitted through the sample gas is measured by means of the detection elements 25aand 25b, and the CO₂concentration of the sample gas is determined from its absorbance on the basis of the calibration curve.

[0079] III-3. Reference measurement

[0080] The gas flow path is changed, and then the reference gas is passed therethrough to clean the gas flow path and the cell chamber 11. After a lapse of about 30 seconds, light intensities are measured by means of each of the detection elements 25aand 25b.

[0081] More specifically, the reference gas is sucked into the gas injector 21 with the valve V1 opened as shown in FIG. 3A, and then pushed out into the flow path from the gas injector 21 with the valve V1 closed to clean the first sample cell 11aand the second sample cell 11b. At this time, the intensities of light transmitted through the reference gas are measured by means of the detection element 25aand the detection element 25b. The light intensities thus obtained by the first and second detection elements 25aand 25bare represented by ¹²R₁and ¹³R₁, respectively.

[0082] III-4. Base gas measurement

[0083] The CO₂concentration of the base gas obtained by the first detection element 25ain “III-1. Preliminary base gas measurement” is compared with the CO₂concentration of the sample gas obtained by the first detection element 25ain “III-2. Preliminary sample gas measurement”. If the CO₂concentration of the base gas is higher than the CO₂concentration of the sample gas, the base gas is diluted with the air or reference gas in the gas injector 21 to a CO₂concentration level equivalent to that of the sample gas, and then the light intensity measurement is performed on the base gas thus diluted.

[0084] More specifically, a predetermined amount of the reference gas is sucked into the gas injector 21 with the valve V1 opened as shown in FIG. 3B-1. In turn, the base gas is sucked into the gas injector 21 with the valve V3 opened as shown in FIG. 3C, and mixed with the reference gas. Since the CO₂concentrations of the two breath samples are adjusted to substantially the same level by thus diluting the base gas with the reference gas, the ranges of the ¹²CO₂and ¹³CO₂calibration curves to be used can be narrowed.

[0085] Alternatively, a predetermined amount of air may be sucked into the gas injector 21 with the three-way valve V4 opened to the atmospheric air as shown in FIG. 3B-2. In turn, the base gas is sucked into the gas injector 21 with the three-way valve V4 opened to the cell chamber and with the valve V3 opened as shown in FIG. 3C, and then mixed with the air.

[0086] Since the CO₂concentrations of the two breath samples are adjusted to substantially the same level by thus diluting the base gas with the air, the ranges of the ¹²CO₂and ¹³CO₂calibration curves to be used can be narrowed.

[0087] It should be noted that the measuring procedure employing the dilution method shown in FIG. 3B-2 is characterized in that the CO₂concentrations of the two breath samples are adjusted to substantially the same level, and does not necessarily require to employ a step of constantly maintaining the CO₂concentration at a constant level as described in Japanese Examined Patent Publication No. 4-124141 (1992). The use of limited ranges of the calibration curves can be achieved simply by adjusting the CO₂concentrations of the base gas and the sample gas to substantially the same level. Since the CO₂concentrations of the base gas and the sample gas may vary within a range of 1% to 6% in actual measurement, it is very troublesome to always maintain the CO₂concentrations at a constant level.

[0088] If the CO₂concentration of the base gas is lower than the CO₂concentration of the sample gas, the base gas is not diluted, but the base gas is subjected to the measurement as it is.

[0089] The base gas is mechanically pushed out into the flow path from the gas injector 21 at a constant flow rate and, at this time, light intensity measurement is performed by means of the detection elements 25aand 25b.

[0090] The light intensities thus obtained by the first and second detection elements 25aand 25bare represented by ¹²B and ¹³B, respectively.

[0091] III-5. Reference measurement

[0092] The cleaning of the gas flow path and the cells and the light intensity measurement on the reference gas are performed again by employing the flow path shown in FIG. 3D.

[0093] The light intensities thus obtained by the first and second detection elements 25aand 25bare represented by ¹²R₂and ¹³R₂, respectively.

[0094] III-6. Sample gas measurement

[0095] If the base gas is diluted in “III-4. Base gas measurement”, the sample gas is sucked into the gas injector 21 from the breath sampling bag as shown in FIG. 3E, and then mechanically pushed out into the flow path from the gas injector 21 at a constant flow rate. At this time, light intensities are measured by the detection elements 25aand 25b.

[0096] If the base gas is not diluted in “III-4. Base gas measurement”, the sample gas is diluted with the reference gas or air to a CO₂concentration level equivalent to that of the base gas in the gas injector 21, and then the intensities of light transmitted through the sample gas is measured by means of the detection elements 25aand 25b.

[0097] The light intensities thus obtained by the first and second detection elements 25aand 25bare represented by ¹²S and ¹³S, respectively.

[0098] III-7. Reference gas measurement

[0099] The cleaning of the gas flow path and the cells and the light intensity measurement on the reference gas are performed again.

[0100] The light intensities thus obtained by the first and second detection elements 25aand 25bare represented by ¹²R₃and ¹³R₃, respectively.

[0101] IV. Data processing

[0102] IV-1. Calculation of absorbances of base gas

[0103] The ¹²CO₂absorbance ¹²Abs(B) and the ¹³CO₂absorbance ¹³Abs(B) of the base gas are calculated on the basis of the transmitted light intensities ¹²R₁and ¹³R₁for the reference gas, the transmitted light intensities ¹²B and ¹³B for the base gas and the transmitted light intensities ¹²R₂and ¹³R₂for the reference gas obtained in accordance with the aforesaid measuring procedure.

[0104] The ¹²CO₂absorbance ¹²Abs(B) is calculated from the following equation:

[0105] ¹²Abs(B)=−log[2×¹²B/(¹²R₁+¹²R₂)]

[0106] The ¹³CO₂absorbance ¹³Abs(B) is calculated from the following equation:

[0107] ¹³Abs(B)=−log[2×¹³B/(¹³R₁+¹³R₂)]

[0108] Since the calculation of the absorbances is based on the light intensities obtained in the base gas measurement and the averages (R₁+R₂)/2 of the light intensities obtained in the reference measurements performed before and after the base gas measurement, the influence of a drift (a time-related influence on the measurement) can be eliminated. Therefore, when the apparatus is turned on, there is no need for waiting until the apparatus reaches a complete thermal equilibrium (it usually takes several hours), so that the measurement can be started immediately after the turn-on of the apparatus.

[0109] IV-2. Calculation of absorbances of sample gas

[0110] The ¹²CO₂absorbance ¹²Abs(S) and the ¹³CO₂absorbance ¹³Abs(S) of the sample gas are calculated on the basis of the transmitted light intensities ¹²R₂and ¹³R₂for the reference gas, the transmitted light intensities ¹²S and ¹³S for the sample gas and the transmitted light intensities ¹²R₃and ¹³R₃for the reference gas obtained in accordance with the aforesaid measuring procedure.

[0111] The ¹²CO₂absorbance ¹²Abs(S) is calculated from the following equation:

[0112] ¹²Abs(S)=−log[2×¹²S/(¹²R₂+¹²R₃)]

[0113] The ¹³CO₂absorbance ¹³Abs(S) is calculated from the following equation:

[0114] ¹³Abs(S)=−log[2×¹³S/(¹³R₂+¹³R₃)]

[0115] Since the calculation of the absorbances is based on the light intensities obtained in the sample gas measurement and the averages of the light intensities obtained in the reference measurements performed before and after the sample gas measurement, the influence of a drift can be eliminated.

[0116] IV-3. Calculation of concentrations

[0117] The ¹²CO₂concentration and the ¹³CO₂concentration are calculated by using calibration curves.

[0118] The calibration curves are prepared on the basis of measurement performed by using test gas samples of known ¹²CO₂concentrations and test gas samples of known ¹³CO₂concentrations.

[0119] For preparation of the calibration curve for ¹²CO₂, the ¹²CO₂absorbances for different ¹²CO₂concentrations ranging from about 0.5% to about 6% are measured. The ¹²CO₂concentrations and the ¹²CO₂absorbances are plotted as abscissa and ordinate, respectively, and an approximate curve is determined by the method of least squares. An approximate quadratic curve, which includes relatively small errors, is employed as the calibration curve in this embodiment.

[0120] For preparation of the calibration curve for ¹³CO₂, the ¹³CO₂absorbances for different ¹³CO₂concentrations ranging from about 0.006% to about 0.07% are measured. The ¹³CO₂concentrations and the ¹³CO₂absorbances are plotted as abscissa and ordinate, respectively, and an approximate curve is determined by the method of least squares. An approximate quadratic curve, which includes relatively small errors, is employed as the calibration curve in this embodiment.

[0121] Strictly speaking, the ¹³CO₂absorbance determined by individually measuring a gas sample containing ¹²CO₂and a gas sample containing ¹³CO₂may be different from the ¹³CO₂absorbance determined by measuring a gas sample containing both ¹²CO₂and ¹³CO₂. This is because the interference filters each have a certain bandwidth and the ¹²CO₂absorption spectrum partially overlaps the ¹³CO₂absorption spectrum. Since gas samples containing both ¹²CO₂and ¹³CO₂are to be analyzed in this measurement method, the overlap of these spectra should be corrected in determination of the calibration curves. The calibration curves to be employed in this measurement are corrected for the overlap of the absorption spectra.

[0122] The ¹²CO₂concentration and ¹³CO₂concentration of the base gas and the ¹²CO₂concentration and ¹³CO₂concentration of the sample gas determined by using the aforesaid calibration curves are represented by ¹²Conc(B), ¹³Conc(B), ¹²Conc(S) and ¹³Conc(S), respectively.

[0123] IV-4. Calculation of concentration ratios

[0124] The concentration ratio of ¹³CO₂to ¹²CO₂is determined. The concentration ratios in the base gas and in the sample gas are expressed as ¹³Conc(B)/¹²Conc(B) and ¹³Conc(S)/¹²Conc(S), respectively.

[0125] Alternatively, the concentration ratios in the base gas and in the sample gas may be defined as ¹³Conc(B)/[¹²Conc(B)+¹³Conc(B)] and ¹³Conc(S)/[¹²Conc(S)+¹³Conc(S)], respectively. Since the ¹²CO₂concentration is much higher than the ¹³CO₂concentration, the concentration ratios expressed in the former way and in the latter way are substantially the same.

[0126] IV-5. Determination of ¹³C change

[0127] A ¹³C difference between the sample gas and the base gas is calculated from the following equation:

[0128] Δ¹³C=[Concentration ratio of sample gas−Concentration ratio of base gas]×10³/Concentration ratio of base gas (Unit: per mill (per thousand))

[0129] IV-6. Correction of ¹³C change

[0130] The difference Δ¹³C in the ¹³CO₂concentration ratio between the base gas and the sample gas is subjected to a correction for water vapor concentration (correction for humidity) according to the present invention.

[0131] For this purpose, the difference Δ¹³C in the ¹³CO₂concentration ratio is corrected with the use of a graph prepared by plotting difference Δ¹³C in the ¹³CO₂concentration ratios with respect to outputs of the humidity sensor 19.

[0132] More specifically, the preparation of the graph is achieved in the following manner. A 3% CO₂/N₂balance gas having a humidity of 0% is filled in two gas sampling bags, and water vapor is charged to saturation into one of the gas sampling bag for preparation of a 3% CO₂/N₂balance gas having a humidity of 100%. By mixing these two gases, five sample gases having different humidities ranging from 0% to 100% and a base gas having a humidity of 0% is prepared. An output of the humidity sensor 19 indicative of the humidity of the base gas and outputs of the humidity sensor 19 indicative of the humidities of the sample gases are obtained. The differences ΔV between the output for the base gas and the outputs for the sample gases are plotted as abscissa. Since the humidity of the base gas is 0%, the differences ΔV in the output correspond to values indicative of the humidities of the sample gases. Then, the differences in the ¹³CO₂concentration between the base gas and the sample gases are plotted as ordinate. Thus, the preparation of the graph is completed.

[0133] Experimentally obtained values are shown in Table 1.

[0134] Although the outputs of the sensor indicative of the humidity of the base gas should basically be the same level, the measured output values varied with a drift. This is because the response speed of the humidity sensor 19 was problematic and the measurement was performed before the humidity sensor 19 did not reach complete equilibrium. The values in Table 1 are plotted as shown in a graph of FIG. 4.

[0135] The differences Δ¹³C in the ¹³CO₂concentration ratio between the base gas and the sample gases are corrected on the basis of the graph and the differences in the output of the humidity sensor 19 between the base gas and the sample gases.