WO1997033156A1

WO1997033156A1 - Apparatus for distinguishing kinds of organic vapors

Info

Publication number: WO1997033156A1
Application number: PCT/EP1997/001189
Authority: WO
Inventors: Kenji Motosugi; Yusuke Takano
Original assignee: Hoechst Research & Technology Deutschland Gmbh & Co.Kg
Priority date: 1996-03-08
Filing date: 1997-03-10
Publication date: 1997-09-12
Also published as: JPH09243630A

Abstract

A compact apparatus for safely distinguishing kinds of organic vapors is provided. A predetermined number m of detector units are located along a vapor flow path through which one of m kinds (m≥2) of fuel vapors is passed. A light source in each detector unit emits light to associated sensor member exhibiting a unique reflection characteristic to one of the m kinds of different fuel vapors through a first optical fiber. Light reflected from the sensor member is received by a second optical fiber, and a signal having a magnitude corresponding to the received light intensity is outputted from a detector. A signal processing unit compares these signals from each other in order to distinguish the fuel vapor introduced into the fuel vapor flow path.

Description

Apparatus for distinguishing kinds of organic vapors

The present invention relates to an apparatus for distinguishing kinds of organic vapors, for example, fuel vapors.

There have been proposed apparatuses for preventing a driver or a fuel supplier from erroneously supplying a different kind of fuel to his automobile in automobile fuel supply facilities such as gas stations. Japanese Patent Publication No. 4-64958 discloses an apparatus for preventing such erroneous supply of fuel utilizing a catalytic gas sensor. Specifically, this apparatus absorbs a fuel vapor in the fuel tank of an automobile, and measures the concentration of the fuel vapor with the gas sensor to determine the kind of fuel used in the automobile, thus preventing the erroneous fuel supply. The apparatus determines that the fuel under measurement is gasoline when the concentration of the fuel vapor is equal to or more than a preset value, and light oil when the concentration is less than the preset value.

However, since a fuel supply port of a fuel tank arranged in an automobile is open to the atmosphere while fuel is being charged therein, the concentration of the fuel vapor absorbed for determining the kind of fuel may largely depend on external factors such as a weather condition. Therefore, even with an automobile using gasoline, a measured concentration of the vapor may be lower than an actual value to take a gasoline powered automobile for a light-oil powered automobile, thereby leading to possible erroneous supply of fuel. The disadvantage of erroneous determination cannot be avoided as long as the determination relies on the measurement of the concentration of a fuel vapor.

Japanese Patent Laid-open Application No. 2-85200 discloses a method of determining the kind of oil relying on the difference in exhaust sound generated when an absorbed fuel vapor is exhausted through a whistle-like mechanism. This method requires that a fuel vapor under measurement has a concentration at a certain level or more since the difference in exhaust sound is caused by the density of the fuel vapor. This method therefore has a disadvantage that the determination is not reliably made when the concentration of a fuel vapor is extremely low.

To solve the disadvantages mentioned above, the assignee of the subject application has proposed in Japanese Patent Application No. 6-315025 (filed on December 19, 1994) an apparatus for distinguishing kinds of organic vapors utilizing a plurality of sensor members each exhibiting an inherent reflection characteristic to one of different organic vapors. The proposed distinguishing apparatus is extremely effective in distinguishing kinds of organic vapors. In one of the embodiments disclosed in the foregoing patent application, an apparatus for distinguishing kinds of organic vapors irradiates sensor members with light from a light source through a collimator, and converges light reflected by the sensor members onto detectors through a collimator. This apparatus advantageously increases the degree of freedom in arrangement of the light source and the detectors and is therefore applicable to a configuration in which the light source and the detector cannot be arranged near a flow path of a fuel vapor. However, since the collimator used in the apparatus is rather expensive and can only be reduced in size to a certain limit, the apparatus for discriminating kinds of organic vapors has limits of a reduction in size and cost.

In view of the problems mentioned above, it is an object of the present invention to provide an apparatus for distinguishing kinds of organic vapors which is capable of eliminating erroneous determination of kinds of organic vapors, more reliably and safely discriminating kinds of organic vapors, and achieving a reduction in cost and size.

To achieve the above object, the present invention provides an apparatus for distinguishing kinds of organic vapors comprising:

m sensor members each exhibiting a unique reflection characteristic to m kinds of different organic vapors, where m is a positive integer equal to or more than two; m light sources respectively associated with the m sensor members for emitting light for irradiating the sensor members associated therewith; first m optical fibers respectively associated with the m sensor members for introducing light emitted from the m light sources to the sensor members associated therewith; second m optical fibers respectively associated with the m light sources and the first m optical fibers for introducing light reflected by the sensor members; m detectors associated with one of the m light sources and the first m optical fibers for receiving the light introduced by the second m optical fibers associated therewith, each of the m detectors outputting a signal indicative of the intensity of received light; and a processing circuit for comparing magnitude of the signals generated by the m detectors for each of the m organic vapors with each other to derive differences between the respective signals and for distinguishing kinds of organic vapors based on signs of the differences, whereby the processing circuit processes the signals generated by the m detectors when the m sensor members are positioned in an organic vapor under measurement in order to determine which of the m organic vapors the organic vapor under measurement is. The organic vapor is, for example, a fuel vapor.

Preferably, the apparatus further comprises a housing where light emitting ends of the first optical fibers and light receiving ends of the second optical fibers are arranged opposite to the sensor members associated therewith with respect to a flow path of the organic vapor under measurement. The flow path may be straight, curved, or bent from an inlet to an outlet of an organic vapor. By bending the flow path to allow the center of an organic vapor flow to pass near the surfaces of polymer thin films of the sensor members, a response speed can be improved.

Each of the sensor members includes a substrate for reflection and a polymer thin film formed on the substrate, and the polymer thin film of the sensor member exhibits a unique different reflection characteristic to the m organic vapors. The reflection characteristic of the sensor member is defined by the product of the thickness of the polymer thin film and a refractive index of the polymer thin film.

The light sources may comprise light emitting means including light emitting diodes and semiconductor lasers, and the m detectors may comprise light detecting means including photodiodes and phototransistors.

The m sensor members exhibit unique and different reflection characteristics to m different kinds of organic vapors. When the m sensor members are positioned in an organic vapor under measurement and irradiated with light from the light sources, the light from the light sources is reflected by the m sensor members with a magnitude corresponding to the kind and the concentration of the organic vapor. The reflected light is received through the second optical fibers by the m detectors which in turn compare the magnitudes of signals generated for m organic vapors to derive differences between the respective signals. Signs of the derived differences can be used to determine an organic vapor under measurement from the m organic vapors.

Fig. 1 is a block diagram schematically illustrating the configuration of a fuel vapor distinguishing apparatus to which the present invention is applied;

Fig. 2 is a schematic diagram illustrating the configuration of a detector unit appearing in Fig. 1 ;

Fig. 3 is a graph for explaining an interference enhanced reflection method which is the operation principle of the present invention;

Fig. 4 is a graph illustrating exemplary relationships between outputs of two detector units having different sensitivities to a gasoline vapor and to a light oil vapor and the concentrations of these vapors; Fig. 5 is a graph illustrating other exemplary relationships between outputs of the two detector units having different sensitivities to a gasoline vapor and to a light oil vapor and the concentrations of these vapors;

Fig. 6 is a graph illustrating further exemplary relationships between outputs of the two detector units having different sensitivities to a gasoline vapor and to a light oil vapor and the concentrations of these vapors;

Fig. 7(a) is a vertical sectional view illustrating a fuel vapor distinguishing apparatus according to the present invention which comprises two of the detector units appearing in Fig. 1 ;

Fig. 7(b) is a cross-sectional view taken along a line b-b of the fuel vapor distinguishing apparatus illustrated in Fig. 7(a);

Fig. 8 is a schematic circuit diagram illustrating an example of a signal processing unit shown in Fig. 1 for distinguishing a gasoline vapor from a light oil vapor;

Fig. 9 is a block circuit diagram illustrating another example of the signal processing unit shown in Fig. 1 ;

Fig. 10 is a graph illustrating the relationships between the relative concentration of a gasoline vapor and outputs of three different detector units of the present invention;

Fig. 11 is a graph illustrating the relationships between the relative concentration of a light oil vapor and outputs of the detector units used in the measurements of Fig. 10;

Fig. 12 is a graph illustrating the relationships between the relative concentration of a methanol vapor and outputs of the detector units used in the measurements of Fig. 10;

Fig. 13 is a schematic circuit diagram illustrating an example of the signal processing unit appearing in Fig. 1 modified to discriminate between a gasoline vapor, a light oil vapor, and a methanol vapor; and

Fig. 14 is a block diagram generally illustrating a system for supplying a fuel vapor to the fuel vapor discriminating apparatus of the present invention.

An embodiment of an apparatus for distinguishing kinds of organic vapors according to the present invention will hereinafter be described with reference to the accompanying drawings in conjunction with an example in which the present invention is applied to discrimination of fuel vapors. Fig. 1 generally illustrates the configuration of a fuel vapor distinguishing apparatus to which the present invention is applied. The illustrated fuel vapor distinguishing apparatus allows m (m is an integer equal to or larger than two) kinds of fuel vapors to be determined, wherein one of m kinds of fuel vapors is introduced into a vapor flow path 1 formed in a housing HS from one end 11 and exhausted from the other end 12. First - m-th detector units 21 - 2m are located along the vapor flow path 1.

The detector units have the same configuration comprising light source units 21 _E,

22_E, 2_mE; first optical fibers 21 _F1, 22_F1, .... 2m_F1 for transmitting light outputted from associated light sources; sensor members 21 _s, 22_s 2m_s located in the vapor flow path 1 so as to be directly irradiated with light emitted from light emitting ends of the first optical fibers and to reflect the light; second optical fibers 21 _F2, 22_F2, .... 2m_F2 for receiving the light reflected by the sensor members at their light receiving ends; and detectors 21 _D, 22_D, ..., 2m_D for detecting the intensities of light transmitted through the second optical fibers to output signals proportional to the detected intensities of the light. The signals outputted from the detectors 21 _D - 2m_D are supplied to a signal processing unit 3 which in turn performs processing such as comparisons in magnitude on the output signals to discriminate a fuel vapor 7 introduced into the vapor flow path 1.

The sensor members 21 _s - 2m_s have different characteristics from each other. In other words, a sensor member 2k_s of an arbitrary detector unit 2k (k is a positive integer equal to or larger than two and equal to or smaller than m) exhibits a set of reflection characteristics to m kinds of fuel vapors which is different from a set of reflection characteristics of a sensor member 2j_s of an arbitrary detector unit 2j (j is a positive integer different from k, equal to or larger than two and equal to or smaller than m) to the m kinds of fuel vapors.

While Fig. 1 illustrates that each of the first detector unit 21 - the m-th detector unit 2m has a light source, light from a single light source may be distributed to m lines of first optical fibers 21 _F1, 22_F1, .... 2m_F1. When a light source is provided in each detector unit, an open end of the first optical fiber and an open end of the second optical fiber may be coupled to a third optical fiber through an optical coupler/ distributor such that the sensor member is irradiated with light emitted from the third optical fiber and reflected light from the sensor member is transmitted to the detector. In essence, any configuration may be utilized as long as the light source and the detector are positioned away from the housing HS; the respective sensor members are irradiated with light transmitting through the first optical fibers; and light reflected by the sensor members is transmitted to the detectors through the second optical fibers. In addition, from a viewpoint of safety, the vapor flow path 1 is desirably sealed to the external space.

Next, the first detector unit 21 will be exemplarily described with reference to Fig. 2 in terms of its specific configuration, as the representative of the m detector units 21 - 2m. Referring specifically to Fig. 2, a light emitting end of the light source 21 _E is coupled to one end of the first optical fiber 21 _F1, the other end of which protrudes into the vapor flow path 1 to irradiate the sensor member 21 _s with light transmitted therethrough. A light receiving end of the detector 21 _D is coupled to one end of the second optical fiber 21 _F2, the other end of which protrudes into the vapor flow path 1 to receive light reflected by the sensor member 21 _s.

As illustrated in Figs. 1 and 2, since the present invention eliminates a collimator for converging light from the first optical fiber 21 _F1 to irradiate the sensor member 21 _s with the converged light or for converging light reflected from the sensor member 21 _s to transmit the converged light to the second optical fiber 21 _F2, it is preferable that optical fibers having large core diameters be used for the first optical fiber 21 _F1 and the second optical fiber 21 _F2, and that the distance between the light emitting end of the optical fiber 21 _F1 and the surface of the sensor member 21 _s as well as the distance between the light receiving end of the optical fiber 21 _F2 and the surface of the sensor member 21 _s be as small as possible. For example, the core diameter is preferably 0.5 millimeters (mm) or more, and 1 mm or more is particularly preferable. The distance between the light emitting end of the optical fiber 21 _F1 and the surface of the sensor member 21 _s as well as the distance between the light receiving end of the optical fiber 21 _F2 and the surface of the sensor member 21 _s is preferably 3 mm or less. The light emitting end and the light receiving end are oriented such that light emitted from the light emitting end and light received by the light receiving end are at an angle of 60 degrees or less to the normal of the sensor member 21 _s, and 45 or 30 degrees is particularly preferable in consideration of the ease of machining. While ordinary optical fibers may be used for the optical fibers 21 _F1, 21 _F2, plastic fibers (POF) are appropriate in practice. Furthermore, the light emitting end and the light receiving end may be secured in metal cylinders arranged in the housing HS, as required, to ensure a mechanical accuracy for fixedly mounting these ends.

The sensor member 21 _s includes a polymer thin film 21 _p formed on a planar substrate such as a silicon wafer to face the light emitting end and the light receiving end. The polymer thin film 21 _p has properties of changing its thickness and refractive index due to a reaction with a first fuel vapor or due to absorption or insorption of the first fuel vapor. The polymer thin film 21 _p may be formed on the substrate, for example, by a spin coating method. The light source 21 _E may be any arbitrary one which emits visual light or near infrared light, and may comprise, for example, a semiconductor laser or a light emitting diode. The light emitting diode is advantageous over the semiconductor laser in size and cost. The detector 21 _D may be implemented, for example, by a photodiode or a phototransistor.

Polymer formed as the polymer thin film for the sensor members 21 _s - 2m_s in the detector units 21 - 2m is preferably a homopolymer or copoiymer having a recurring unit represented by the following chemical formula (I):

CH-

X-C-R

where

X represents -H, -F, -Cl, -Br, -CH₃ -CF₃, -CN, or -CH₂-CH₃;

R¹ represents -R² or -Z-R²;

Z represents -O-, -S-, -NH-, -NR^2'-, -(C=Y)-, -(C=Y)-Y-, -Y-(C=Y)-, -(SO₂)-,

-r-(so₂)-_t -(so₂)-r-, -r-(so₂)-r-, -NH-(C=O)-, -(C=O)-NH-, -(C=O)-NR^2'-

-Y-(C=Y)-Y-, or -O-(C=O)-(CH₂)_n-(C=O)-O-;

Y represents the same or different O or S;

Y represents the same or different O or NH; n represents an integer ranging from 0 to 20; and

R² and R² represent the same or different hydrogen, a linear alkyl group, a branched alkyl group, a cycloalkyi group, an un-saturated hydrocarbon group, an aryl group, a saturated or un-saturated hetero ring, or substitutes thereof. It should be noted that R¹ does not represent hydrogen, a linear alkyl group, or a branched alkyl group.

In the formula, X is preferably H or CH₃; R¹ is preferably a substituted or non- substituted aryl group or -Z-R²; Z is preferably -O-, -(C=O)-O-, or -O-(C=O)-; R² is preferably a linear alkyl group, a branched alkyl group, a cycloalkyi group, an un- saturated hydrocarbon group, an aryl group, a saturated or un-saturated hetero ring, or substitutes thereof.

A polymer used as the polymer thin film for the present invention may be a polymer consisting of a single recurring unit (I), a copoiymer consisting of another recurring unit and the above-mentioned recurring unit (I), or a copoiymer consisting of two or more species of the recurring unit (I). The recurring units in the copoiymer may be arranged in any order, and a random copoiymer, an alternate copoiymer, a block copoiymer or a graft copoiymer may be used by way of example. Particularly, the polymer thin film is preferably made from polymethacrylic acid esters or polyacrylic acid esters. The side-chain group of the ester is preferably a linear or branched alkyl group, or a cycloalkyi group with the number of carbon molecules ranging preferably from 4 to 22.

Polymers particularly preferred for the polymer thin film are listed as follows: poly(dodecyl methacrylate); poly(isodecyl methacrylate); poly(2-ethylhexyl methacrylate); poly(2-ethylhexyl methacrylate-co-methyl methacrylate); poly(2-ethylhexyl methacrylate-co-styrene); poly(methyl methacrylate-co-2-ethylhexyl acrylate); poly(methyl methacrylate-co-2-ethylhexyl methacrylate); poly(isobutyl methacrylate-co-glycidyl methacrylate); poly(cyclohexyl methacrylate); poly(octadecyl methacrylate); poly(octadecyl methacrylate-co-styrene); poly(vinyl propionate); poly(dodecyl methacrylate-co-styrene); poly(dodecyl methacrylate-co-glycidyl methacrylate); poly(butyl methacrylate); poly(butyl methacrylate-co-methyl methacrylate); poly(butyl methacrylate-co-glycidyl methacrylate); poly(2-ethylhexyl methacrylate-co-glycidyl methacrylate); poly(cyclohexyl methacrylate-co-glycidyl methacrylate); poly(cyclohexyl methacrylate-co-methyl methacrylate); poly(benzyl methacrylate-co-2-ethylhexyl methacrylate); poly(2-ethylhexyl methacrylate-co-diacetoneacrylamide); poly(2-ethylhexyl methacrylate-co-benzyl methacrylate-co-glycidyl methacrylate); poly(2-ethylhexyl methacrylate-co-methyl methacrylate-co-glycidyl methacrylate); poly(vinyl cinnamate) poly(butyl methacrylate-co-methacrylate); poly(vinyl cinnamate-co-dodecyl methacrylate); poly(tetrahydrofurfuryl methacrylate); poly(hexadecyl methacrylate); poly(2-ethylbutyl methacrylate); poly(2-hydroxyethyl methacrylate); poly(cyclohexyl methacrylate-co-isobutyl methacrylate); poly(cyclohexyl methacrylate-co-2-ethylhexyl methacrylate); poly(butyl methacrylate-co-2-ethylhexyl methacrylate); poly(butyl methacrylate-co-isobutyl methacrylate); poly(cyclohexyl methacrylate-co-butyl methacrylate); poly(cyclohexyl methacrylate-co-dodecyl methacrylate); poly(butyl methacrylate-co-ethyl methacrylate); poly(butyl methacrylate-co-octadecyl methacrylate); poly(butyl methacrylate-co-styrene); poly(4-methyl styrene); poly(cyclohexyl methacrylate-co-benzyl methacrylate); poly(dodecyl methacrylate-co-benzyl methacrylate); poly(octadecyl methacrylate-co-benzyl methacrylate); poly(benzyl methacrylate-co-tetrahydrofurfuryl methacrylate); poly(benzyl methacrylate-co-hexadecyl methacrylate); poly(dodecyl methacrylate-co-methyl methacrylate); poly(dodecyl methacrylate-co-ethyl methacrylate); poly(2-ethylhexyl methacrylate-co-dodecyl methacrylate); poly(2-ethylhexyl methacrylate-co-octadecyl methacrylate); poly(2-ethylbutyl methacrylate-co-benzyl methacrylate); poly(tetrahydrofurfuryl methacrylate-co-glycidyl methacrylate); poly(styrene-co-octadecyl acrylate); poly(octadecyl methacrylate-co-glycidyl methacrylate); poly(4-methoxystyrene); poly(2-ethylbutyl methacrylate-co-glycidyl methacrylate); poly(styrene-co-tetrahydrofurfuryI methacrylate); poly(2-ethylhexyl methacrylate-co-propyl methacrylate); poly(octadecyl methacrylate-co-isopropyl methacrylate); poly(3-methyl-4-hydroxystyrene-co-4-hydroxystyrene); poly(styrene-co-2-ethylhexyl methacrylate-co-glycidyl methacrylate);

It should be noted that in the methacrylate ester polymers or copolymers listed above, acrylate may be substituted for methacrylate. The polymers may be crosslinked on their own, or they may be crosslinked by introducing a compound that has crosslinking reactive groups. Suitable crosslinking reactive groups include, for example, an amino group, a hydroxyl group, a carboxyl group, an epoxy group, a carbonyl group, a urethane group, and derivatives thereof. Other examples include maleic acid, fumaric acid, sorbic acid, itaconic acid, cinnamic acid, and derivatives thereof. Materials having chemical structures capable of forming carbene or nitrene by irradiation of visible light, ultraviolet light, or high energy radiation may also be used as crosslinking agents. Since a film formed from crosslinking polymer is insoluble, the polymer forming the polymer thin film of the sensor may be crosslinked to increase the stability of the sensor. The crosslinking method is not particularly limited, and methods utilizing irradiation of light or radioactive rays may be used in addition to known crosslinking methods, for example, a heating method. 13

As illustrated in Fig. 2, the light emitting end of the first optical fiber 21 _F1 is slightly displaced from the light receiving end of the second optical fiber 21 _F2 in the lengthwise direction of the vapor flow path 1 such that these ends face the sensor member 21 _s. The present invention, however, is not limited to this particular positioning of the optical fibers, and the plane formed by the light emitting end of the first optical fiber 21 _F1, the light receiving end of the second optical fiber 21 _F2, and the normal of the sensor member 21 _s may form any angle with respect to the lengthwise direction of the vapor flow path 1.

The determination of fuel vapors made by the fuel vapor distinguishing apparatus illustrated in Fig. 1 relies on interference enhanced reflection which is now described in detail. As a fuel vapor passes through the vapor flow path 1 of the fuel vapor distinguishing apparatus, the product of a thickness and a refractive index of the polymer thin film changes in each of the sensor members 21 _s - 2m_s. This change is detected by the associated detector as a change in intensity of reflected light from the polymer thin film, i.e., a change in reflection characteristic thereof, and converted into an electrical signal for output. Actually, since a change in refractive index is thought to be small, it is the thickness of the polymer thin film that mainly exhibits a change due to the existence of a fuel vapor. It should be noted that such a change in thickness due to a certain fuel vapor is not similar in all the sensor members, and each sensor member exhibits a different thickness changing profile when a different fuel vapor is introduced into the vapor flow path 1. More specifically, a polymer thin film of a sensor member among the sensor members 21 _s - 2m_s exhibits a larger change in thickness than the remaining sensor members in response to a fuel vapor introduced into the vapor flow path 1. The former is a sensor member having a higher sensitivity to the fuel vapor, while the remaining ones have a lower sensitivity to the same fuel vapor. Thus, the m detector units 21 - 2m produce a set of different reflection characteristics in response to each fuel vapor introduced into the vapor flow path 1.

Fig. 3 illustrates the relationship between a change in thickness and the reflectivity of a polymer thin film having a higher sensitivity to a certain kind of fuel vapor when light having a wavelength of 940 nanometers (nm) is irradiated to the polymer thin film at an incidence angle of 23°. As can be understood from the graph of Fig. 3, the reflectivity repeatedly increases and decreases as the thickness of the polymer thin film increases. Thus, if the thickness of the polymer thin film is set at a value slightly larger than the thickness corresponding to a minimum value of the reflectivity, the reflectivity monotonously increases with respect to an increase in thickness up to a fixed level. Conversely, if the thickness of the polymer thin film is set at a value slightly larger than the thickness corresponding to a maximum value of the reflectivity, the reflectivity monotonously decreases with respect to an increase in thickness down to a fixed level. Since the thickness of the polymer thin film is thought to be a function which monotonously increases as the concentration of a fuel vapor goes up, the selection of the thickness of the polymer thin film at the maximum value or minimum value, as described above, will result in a monotonously increasing or monotonously decreasing functional relationship established between the concentration of the fuel vapor and the reflectivity (i.e., the output of the detector). The output of the detector here refers to a value calculated by subtracting an output of the detector generated when air exists from an output of the detector generated when the fuel vapor exists. Since the relationship between a change in thickness of the polymer thin film and a change in reflectivity of same also depends on an incidence angle of light to the polymer thin film and a reflection angle of the light on the polymer thin film, the sensitivity of the detector can be improved when the light emitting end of the first optical fiber, the light receiving end of the second optical fiber, and the sensor member are positioned to reduce the incidence angle of light to the polymer thin film and the reflection angle of the light on the polymer thin film, if the light equally includes a P-polarized light component and an S-polarized light component.

Incidentally, the relationship between the concentration of a fuel vapor and a change in thickness of a polymer thin film (hence the relationship between the concentration of a fuel vapor and the reflectivity of the polymer thin film) is uniquely determined by the kind of fuel and the kind of polymer used for a sensor member. Thus, a reflection characteristic or a sensitivity inherent to a particular fuel vapor can be realized in principle by appropriately selecting the kind of polymer used for a sensor member. Ideally, a sensor preferably has a high sensitivity only to a particular fuel vapor such as gasoline, light oil, and so on and a zero sensitivity to other fuel vapors. However, it is actually difficult to reduce the sensitivity to undesired fuel vapors to zero whatever polymer is selected. To cope with this problem, the present invention provides m detector units using m kinds of polymers having different sensitivities to m kinds of fuel vapors, as illustrated in Fig. 1, such that the kind of fuel vapor or the kind of fuel is predicted from a combination of comparisons between magnitudes of respective signals outputted from the detector units. For example, assume that two kinds of fuel vapors, i.e., gasoline vapor and light oil vapor are to be discriminated, and that a detector unit A including a sensor member made of polymer y and a detector unit B including a sensor member made of different polymer δ are used for discriminating the two kinds of fuel vapors. Assume also that the relationships between outputs of the detector unit A and the detector unit B and the concentrations of gasoline vapor and light oil vapor are defined as illustrated in Fig. 4. Specifically, an output S(A)₁ of the detector unit A is always larger than an output S(B)₁ of the detector unit B when a fuel vapor under measurement is a gasoline vapor, while an output S(A)₂ of the detector unit A is always smaller than an output S(B)₂ of the detector unit B when a fuel vapor under measurement is a light oil vapor. Then, a gasoline vapor or a light oil vapor is introduced into the fuel vapor flow path with the detector units A, B located along the fuel vapor flow path. When S(A)₁ > S(B)_{1 t} the fuel vapor under measurement can be predicted to be a gasoline vapor irrespective of the concentration thereof. On the other hand, when S(A)₂ < S(B)₂, the fuel vapor under measurement can be predicted to be a light oil vapor irrespective of the concentration thereof. Assume in the alternative that the sensitivities of the detector units A, B are set such that the output S(A)₁ of the detector unit A is always smaller than the output S(B)₁ of the detector unit B when a fuel vapor under measurement is a gasoline vapor, while the output S(A)₂ of the detector unit A is always larger than the output S(B)₂ of the detector unit B when the fuel vapor under measurement is a light oil vapor, as illustrated in Fig. 5. When S(A)₂ > S(B)₂, the fuel vapor under measurement can be predicted to be a light oil vapor irrespective of the concentration thereof. On the other hand, when S(A) < SfB)^ the fuel vapor under measurement can be predicted to be a gasoline vapor irrespective of the concentration thereof. Furthermore, the relationships illustrated in Figs. 4, 5 need not be satisfied directly by the outputs of the detector units. Alternatively, for example as illustrated in Fig. 6, even if S(A)₁ > S(B)_t and S(A)₂ > S(B)₂ are satisfied when a fuel vapor under measurement is a gasoline vapor or a light oil vapor, the output of any detector unit may be amplified or attenuated with a proper gain to be converted such that the outputs of the detector units satisfy the relationships illustrated in Fig. 4, provided that S(A)₁/S(B)₁ for a gasoline vapor is larger than S(A)₂/S(B)₂ for a light oil vapor.

Figs. 7(a) and 7(b) generally illustrate the configuration of a distinguishing apparatus comprising a detector unit A having a higher sensitivity to a gasoline vapor than to a light oil vapor and a detector unit B having a higher sensitivity to a light oil vapor than to a gasoline vapor in order to discriminate the two kinds of fuel vapors as explained above. Fig. 7(a) is a plan view, and Fig. 7(b) is a cross- sectional view taken along a line b-b in Fig. 7(a). A vapor flow path 1 is formed within a housing HS for introducing a fuel vapor under measurement into the housing HS and discharging the same from the housing HS. A sensor member A_s of the detector unit A and a sensor member B_s of the detector unit B are located along the vapor flow path 1 in a chamber 1' defined in the middle of the vapor flow path !

Specifically, as illustrated in Figs. 7(a) and 7(b), a light emitting end of a first optical fiber A_F1 having one end coupled to a light source A_E extends through the housing HS and to the vicinity of the sensor member A_s for irradiating the sensor member A_s of the detector unit A with light from the light source A_E. A second optical fiber A_F2 having one end coupled to a detector A_D also extends through the housing HS, and its light receiving end is located near the sensor member A_s to receive light reflected by the sensor member A_s. The light emitting end of the first optical fiber A_F1 and the light receiving end of the second optical fiber A_F2 may only have to exist in a plane including a normal on the surface of the sensor member A_s, and more preferably they are included in a plane perpendicular to the lengthwise direction of the vapor flow path 1. While in Figs. 7(a) and 7(b), the light emitting end and the light receiving end are oriented such that light emitted from the light emitting end and light received by the light receiving end are at 45° to the normal of the sensor member A_s, the present invention is not limited to this particular configuration.

The detector unit B has the configuration similar to that of the detector unit A. A light emitting end of a first optical fiber B_F1 having one end coupled to a light source B_E extends through the housing HS and to the vicinity of the sensor member B_s for irradiating the sensor member B_s of the detector unit B with light from the light source B_E. A second optical fiber B_F2 having one end coupled to a detector B_D also extends through the housing HS, and its light receiving end is located near the sensor member B_s to receive light reflected by the sensor member B_s. The light emitting end of the first optical fiber B_F1 and the light receiving end of the second optical fiber B_F2 may only have to exist in a plane including a normal on the surface of the sensor member B_s, and more preferably they are included in a plane perpendicular to the lengthwise direction of the vapor flow path 1. While in Figs. 7(a) and 7(b), the light emitting end and the light receiving end are oriented such that light emitted from the light emitting end and light received by the light receiving end are at 45° to the normal of the sensor member B_s, the present invention is not limited to this particular configuration.

Each of the light sources A_E, B_E may be implemented by a light emitting diode, for example, a light emitting diode having a center wavelength of 670 nm, while each of the detectors A_D, B_D may be implemented by a silicon photodiode. The first optical fibers A_F1 , B_F1 and the second optical fibers A_F2, B_F2 have a core diameter of, for example, 1 mm. The housing HS may be made of a metal or a plastic material. Preferably, the distance between the center of a flow of a fuel vapor under measurement introduced into the vapor flow path 1 and the surface of the polymer thin film is small, and preferably 1 mm or less. The vapor flow path 1 may be straight, curved, or bent.

The first optical fibers A_F1, B_F1 and the second optical fibers A_F2, B_F2 extend to the chamber I^* through associated throughholes formed through the housing HS, and the light emitting ends of the first optical fibers A_F1, B_F1 and the light receiving ends of the second optical fibers A_F2, B_F2 are secured in the respective throughholes. These light emitting ends and light receiving ends are preferably secured using metal cylinders (for example, ferrules) M in order to improve a mechanical strength and a mechanical accuracy of the structure. When optical fibers having a core diameter of 1 mm are used for the first optical fibers A_F1 , B_F1 and the second optical fibers A_F2, B_F2, ferrules of 3.15 mm in diameter are preferably attached at the respective light emitting ends and light receiving ends of the optical fibers, so that the ferrules are bonded to the housing HS. Alternatively, connectors may be used to secure the optical fibers to the housing HS. As a polymer thin film constituting each sensor member, a film of poly(benzyl methacrylate-co-2-ethylhexyl methacrylate) having a thickness of 150 nm may be used for the sensor member A_s, and a film of poly(octadecyl methacrylate-co-glycidyl methacrylate) having a thickness of 150 nm for the sensor member B_s.

Now, the configuration and operation of the signal processing unit 3 illustrated in Fig. 1 will be described in connection with an example in which the foregoing two detector units A, B illustrated in Figs. 7(a) and 7(b) are used for discriminating gasoline and light oil. The signal processing unit 3 may be implemented by a combination of analog and digital circuits or by a circuit using a microcomputer. Fig. 8 is a block diagram illustrating an example of a circuit implemented by a combination of analog and digital circuits. The illustrated circuit is composed of an output processing unit having two processing lines, a difference processing unit, and a determining unit. The fuel vapor distinguishing apparatus of Fig. 1 operates in synchronism with an external control system which usually supplies the apparatus with air and supplies the same with gasoline or light oil from a certain point of time.

The output processing unit have two processing systems: a first processing system including a sample and hold circuit 11 for receiving an output of the detector unit A having a higher sensitivity to a gasoline vapor than to a light oil vapor through a buffer amplifier 10 and an inverting amplifier 12; and a second processing system including a sample and hold circuit 15 for receiving an output of the detector unit B having a higher sensitivity to a light oil vapor than to a gasoline vapor through a buffer amplifier 14 and an inverting amplifier 16. The two processing systems are set in a sample mode when the control system is supplying air, during which values indicative of output signals from the detector units A, B are accumulated in capacitors 13, 17. Next, at the instance the control system stops supplying air and switches to supply a fuel vapor under measurement, a start pulse SP is applied to the sample and hold circuits 11 , 15 to switch the two processing systems to a hold mode. With this switching operation, the values accumulated in the capacitors 13, 17 are applied to non-inverting inputs of the inverting amplifiers 12, 16, while output signals associated with the fuel vapor under measurement from the detector units A, B are applied to inverting inputs of the inverting amplifiers 12, 16, so that the inverting amplifiers 12, 16 output inverted signals each having a magnitude equal to the product of a magnitude corresponding to the difference between a value for air and a value for the fuel vapor under measurement and a constant gain. In other words, assuming that values calculated by subtracting the output for air from the output for the fuel vapor under measurement are S(A), S(B), and the gains of the inverting amplifiers 12, 16 are Ga, Gb, respectively, the output processing unit generates an output represented by -Ga*S(A) and -Gb*S(B), respectively.

An element which presents a change in resistance in response to a temperature change (for example, a thermistor) may be provided near the detector units A, B such that the element is utilized to adjust the gains of the inverting amplifiers 12, 16 in an appropriate form. In this way, it is possible to correct the dependency of the outputs of the two detector units A, B on the fuel vapor concentration due to temperature.

The difference processing unit comprises a differential amplifier 18 which has an inverting input coupled to the output of the first processing system of the output processing unit and a non-inverting input coupled to the output of the second processing system of the output processing unit, so that the differential amplifier 18 provides the difference between outputs delivered from the two processing systems. In other words, the output Vd of the differential amplifier 18 is equal to Ga*S(A)- Gb*S(B).

Next, the determining unit determines which of Ga*S(A) and Gb*S(B) is larger by checking the sign of the output Vd. To realize this determining function, the determining unit has two comparators 19, 20. The comparators 19, 20 are supplied with the output Vd of the differential amplifier 18 at their respective non-inverting inputs and with threshold levels +V_TH1 and -V_TH1 at their respective inverting inputs. The comparators 19, 20 provide outputs CMP1, CMP2 indicative of the results of comparisons performed thereby between the output Vd and the threshold levels +V_TH1 and -V_TH1 , respectively. Specifically, the comparators 19, 20 output signals at high level when the output Vd is higher than the threshold levels +V_TH1 and -V-η., and signals at low level when the output Vd is lower than the threshold levels +V_TH1 and -V_TH1. The outputs of the comparators 19, 20 are both supplied to an AND circuit 21 and a NOR circuit 22, respectively. Thus, when the outputs CMP1 , CMP2 are both at high level, the AND circuit 21 delivers an output at high level while the NOR circuit 22 delivers an output at low level. Conversely, when the outputs CMP1 , CMP2 are both at low level, the AND circuit 21 delivers an output at low level while the NOR circuit 22 delivers an output at high level. Then, a flip-flop 23 generates a Q-output OUT, at high level and a flip-flop 24 generates a Q-output OUT₂ at low level when the output of the AND circuit 21 is at high level and the output of the NOR circuit 22 is at low level. Conversely, the flip-flop 24 generates the Q-output OUT₂ at high level and the flip-flop 23 generates the Q-output OUT- at low level when the output of the AND circuit 21 is at low level and the output of the NOR circuit 22 is at high level. A timer TM, started by a start pulse SP, applies a clock pulse to the flip-flops 23, 24 at the time a predetermined time has elapsed from the supply of the start pulse SP. The clock pulse causes the outputs of the flip- flops 23, 24 to be latched.

Based on the operations described above, a fuel vapor under measurement can be determined. Specifically, when the Q-output OUT of the flip-flop 23 is at high level, Vd is positive, thus determining that the fuel vapor under measurement is a gasoline vapor. When the Q-output OUT₂ of the flip-flop 24 is at high level, Vd is negative, thus determining that the fuel vapor under measurement is a light oil vapor. The determination result thus produced can be retrieved as a digital signal or an analog signal.

When one of the outputs CMP1, CMP2 of the comparators 19, 20 is positive and the other is negative, Vd lies between the two threshold levels, in which case the determination as to the sign of Vd, i.e., the determination of the kind of a fuel vapor under measurement cannot be made. As the absolute values of the two threshold levels are set closer to zero, failures in determining the sign of Vd would occur less frequently, but extremely low threshold levels would cause an increased possibility of erroneous determination due to noise. It is therefore necessary to appropriately determine the threshold values, while making compromise with a noise level. It should be noted that the sign of Vd cannot be determined also when the concentration of a fuel vapor is extremely low.

The determining unit in Fig. 8 may be implemented by a circuit which integrates the output of the differential amplifier for a predetermined time period to determine the sign of the result of the integration. Such an integration based circuit can produce a larger difference as well as improve an S/N ratio since noise components in the output of the differential amplifier is averaged by the integration.

Next, the signal processing unit 3 implemented using a microcomputer will be described with reference to Fig. 9. Outputs of the two detector units A, B, after passing through buffer amplifiers 30, 31, are converted to digital data by A/D convertors 33, 34, respectively, and supplied to a microcomputer 36 through a data bus. The microcomputer 36 is connected to a memory 37 which stores a program for executing the same operation as the circuit of Fig. 8, and applies determination results to an output OUT as digital data, a digital signal, or an analog signal. Also, as illustrated in Fig. 9, an appropriate temperature sensor TS may be arranged near the detector units A, B such that an output of the temperature sensor TS is applied to an A/D convertor 35 through a buffer amplifier 32 to be converted to a digital signal which is then fetched by the microprocessor 36 to enable the operations performed by the circuit of Fig. 8 for temperature compensation to be executed by software.

Table 1 below lists determination results for various concentrations of a gasoline vapor at a temperature of 20°C when the gain of the inverting amplifier 18 is chosen to be two, and the threshold levels +V_TH1, -V_TH2 are both set at 5 mV. Table 2 lists determination results for a light oil vapor under the same condition as Table 1. The flip-flops 23, 24 are operated to latch the outputs approximately 0.25 seconds after a fuel vapor under measurement is introduced. In the tables below, "1" indicates a high level, and "0" indicates a low level.

Table 2

Relative Light Oil Gasoline Determination Concentration Determination Output Output

0.10 1 0

0.15 1 0

0.20 1 0

0.30 1 0

0.40 1 0

0.50 1 0

0.60 1 0

0.80 1 0

0.98 0

It can be understood from the tables above that the kind of vapor can be determined as long as the concentration of a gasoline vapor or a light oil vapor is above a certain level.

As a modification to the apparatus for distinguishing two kinds of fuel vapors described hereto, a fuel vapor distinguishing apparatus for distinguishing three kinds of fuel vapors, for example, a gasoline vapor, a light oil vapor and a methanol vapor, may be realized in a similar manner. This fuel vapor distinguishing apparatus has three detector units A, B, C disposed along a vapor flow path for the purpose of distinguishing three kinds of fuel vapors, similarly to the distinguishing apparatus illustrated in Figs. 7(a) and 7(b). Poly(2-ethylhexyl methacrylate-co- glycidyl methacrylate) may be used for polymer constituting a sensor member of the detector unit A; poly(octadecyl methacrylate-co-glycidyl methacrylate) for a sensor member of the detector unit B; and poly(3-methyl-4-hydroxystyrene-co-4- hydroxystyrene) for a sensor member of the detector unit C. However, polymer materials available to the sensor members are not limited to those mentioned above. These polymer materials are formed on silicon substrates as thin films having a thickness of approximately 400 nm by a spin coat method.

Fig. 10 illustrates outputs of the detector units A, B, C for a gasoline vapor; Fig. 11 illustrates outputs of the detector units A, B, C for a light oil vapor; and Fig. 12 outputs of the detector units A, B, C for a methanol vapor. It can be seen from the graphs of Figs. 10 - 12 that the detector unit A has a higher sensitivity to the gasoline vapor than to the light oil vapor and a very low sensitivity to the methanol vapor, that the detector unit B has a higher sensitivity to the gasoline vapor than to the light oil vapor and a very low sensitivity to the methanol vapor, and that the detector unit C has a high sensitivity to the methanol vapor and a very low sensitivity to the gasoline vapor and the light oil vapor.

Fig. 13 illustrates a circuit for distinguishing a gasoline vapor, a light oil vapor and a methanol vapor using the foregoing detector units A, B, C in view of the sensitivity characteristics thereof. The illustrated circuit, which is an extended version of the circuit of Fig. 8 for distinguishing two kinds of vapors, also comprises an output processing unit, a difference processing unit, and a determining unit. The output processing unit has three separate processing systems corresponding to the three detector units A, B, C, the configuration and operation of which are the same as those of the detector units A, B described in connection with Fig. 8. In the circuit of Fig. 13, outputs of the output processing unit are labelled -Ga*S(A), -Gb^*S(B) and -Gc^*S(C).

The difference processing unit comprises three differential amplifiers 61 , 62, 63. The first differential amplifier 61 is supplied with -Ga*S(A) at an inverting input and with -Gb*S(B) at a non-inverting input, so that an output V1 of the differential amplifier 61 is equal to Ga*S(A)-Gb*S(B). The second differential amplifier 62 is supplied with -Gb*S(B) at an inverting input and with -Gc*S(C) at a non-inverting input, so that an output V2 of the differential amplifier 62 is equal to Gb^*S(B)- Gc^*S(C). The third differential amplifier 63 is supplied with -Gc*S(C) at an inverting input and with -Ga^*S(A) at a non-inverting input, so that an output V3 of the differential amplifier 63 is equal to Gc*S(C)-Ga*S(A).

The determining unit determines the kind of fuel vapor based on whether the outputs V1, V2, V3 of the differential amplifiers 61 , 62, 63 are positive or negative. Similar to the determination described with reference to Figs. 10 - 12, the following method may be employed for determination. Within possible relationships among the magnitudes of the three outputs, a vapor under measurement is determined to be a gasoline vapor when Ga*S(A) > Gb*S(B) > Gc^*S(C) and Ga*S(A) > Gc^*S(C) > Gb*S(B), to be a light oil vapor when Gb*S(B) > Ga*S(A) > Gc*S(C) and Gb*S(B) > Gc^*S(C) > Ga^*S(A) and to be a methanol vapor when Gc*S(C) > Gb*S(B) > Ga*S(A) and Gc^*S(C) > Ga^*S(A) > Gb*S(B).

Expressing the above relationships by signs (i.e., negative or positive) of V1, V2, V3, a vapor under measurement is determined to be a gasoline vapor when V1 >0, V2>0, V3<0 or V1>0, V2<0, V3<0, i.e., V1>0 and V3<0, to be a light oil vapor when V1<0, V2>0, V3<0 or V1<0, V2>0, V3>0, i.e., V2>0 and V1<0 and to be a methanol vapor when V1 >0, V2<0, V3>0 or V1 <0, V2<0, V3>0, i.e., V3>0 and V2<0.

For the purpose of performing the foregoing determining operation, the determining unit comprises six comparators 64 - 69, three AND circuits 70 -72 and three flip- flops 73 - 75. An output V1 of the differential amplifier 61 is supplied to a non- inverting input of the comparator 64 and to an inverting input of the comparator 65. The comparator 64 receives a positive threshold voltage +V_TH at an inverting input and the comparator 65 receives a negative threshold voltage -V_TH at a non-inverting input. As a result, an output V11 of the comparator 64 is at a high level only when V1 >+V_TH, and an output V12 of the comparator 65 is at high level only when V1<-V_TH. Similarly, an output V2 of the differential amplifier 62 is supplied to the comparators 66, 67, while an output V3 of the differential amplifier 63 is supplied to the comparators 68, 69, respectively. Thus, an output V21 of the comparator 66 is at high level when V2>+V_TH; an output V22 of the comparator 67 is at high level when V2<-V_TH; an output V31 of the comparator 68 is at high level when V3>+V_TH; and an output V32 of the comparator 69 is at high level when V3<-V_TH. The output V11 of the comparator 64 and the output V32 of the comparator 69 are applied to the AND circuit 70; the output V12 of the comparator 65 and the output V21 of the comparator 66 to the AND circuit 71 ; and the output V22 of the comparator 67 and the output V31 of the comparator 68 to the AND circuit 72.

As a result, an output A1 of the AND circuit 70 is at high level when V1>+V_TH and V3<-V_TH, i.e., when a fuel vapor under measurement is a gasoline vapor; an output A2 of the AND circuit 71 is at high level when V2>+V_TH and V1 <-V_TH, i.e., when a fuel vapor under measurement is a light oil vapor; and an output A3 of the AND circuit 72 is at high level when V3>+V_TH and V2<-V_TH, i.e., when a fuel vapor under measurement is a methanol vapor. The outputs A1 - A3 of the AND circuits 70 - 72 are latched by their associated flip-flops 73 - 75 when a predetermined time has elapsed after the fuel vapor is supplied, as is the case of the embodiment described with reference to Fig. 8. Then, one of outputs OUT1, OUT2, OUT3 of the flip-flops 73 - 75 goes to high level depending on the kind of the fuel vapor under measurement, thus representing the result of determination.

Tables 3 - 5 below list the levels (i.e., high level HI or low level LO) of the outputs V11 , V12, V21 , V22, V31 , V32 of the comparators 64 - 69 and the outputs OUT1 , OUT2, OUT3 of the flip-flops 73 - 75 when a fuel vapor is a gasoline vapor, a light oil vapor or a methanol vapor. The threshold voltages ± V_TH are set to ± 5 mV.

Table 5

Relative

Concentration V11 V12 V21 V22 V31 V32 OUT1 OUT2 OUT3

0.1 LO LO LO HI HI LO LO LO LO

0.2 LO LO LO HI HI LO LO LO LO

0.3 LO HI LO HI HI LO LO LO HI

0.4 LO HI LO HI HI LO LO LO HI

0.5 LO HI LO HI HI LO LO LO HI

0.6 LO HI LO HI HI LO LO LO HI

It can be understood from these tables that one of the outputs OUT1 , OUT2, OUT3 of the flip-flops 73 - 75 goes to high level depending on whether a vapor under measurement is a gasoline vapor, a light oil vapor or a methanol vapor as long as a relative concentration of the vapor is above a certain level, thus making it possible to determine a fuel vapor under measurement.

In this embodiment, a method similar to a so-called shunt flow method for generating an arbitrary humidity is employed so that a gasoline vapor, a light oil vapor or a methanol vapor at an arbitrary concentration can be supplied to the fuel vapor distinguishing apparatus in order to evaluate the performance of the fuel vapor distinguishing apparatus. A system configuration for implementing this method is illustrated in Fig. 14. In Fig. 14, a flow of carrier gas (nitrogen gas) is branched into two, one of which is introduced into a bubbler 78 containing a fuel (gasoline, light oil and the like) through a mass flow controller 77. The carrier gas is substantially saturated with vapors of the fuel in the bubbler 78 and then led to a confluence 79. The other branch of the carrier gas is delivered directly to the confluence 79 after passing through a mass flow controller 80, and the two branched flows are combined at the confluence 79. In this way, a fuel vapor having a concentration determined by the ratio of the mass flow rates of the respective flows is produced. A buffer 81 is connected to a fuel vapor introducing port of the fuel vapor distinguishing apparatus 83 of the present invention through a valve 82. A fuel vapor discharge port of the fuel vapor discriminating apparatus 83 is connected to a suction device (not shown). The fuel vapor distinguishing apparatus 83 is also connected to an air supply through a valve 84. The bubbler 78, the buffer 81 , the valves 82, 84 and the fuel vapor distinguishing apparatus 83 are accommodated in a thermostatic chamber (indicated by a chain line). In an actual application to a fuel supply machine, air is normally sucked when the valve 82 closes and the valve 84 opens, and, when the fuel vapor is to be distinguished, a fuel vapor from the buffer

81 is introduced into the fuel vapor distinguishing apparatus 83 by opening the valve

82 and closing the valve 84.

Assume that a pure liquid is contained in the bubbler 78, that the carrier gas is completely saturated with the fuel vapor in the bubbler 82, that a saturated vapor pressure at a current temperature is Ps, that a flow rate of the carrier gas in the bubbler 78 is α F, and that a flow rate of the carrier gas in the other branch is (1-α)F. Then, a vapor pressure of a generated vapor of the liquid is equal to α Ps. The ratio P/Ps of the vapor pressure of the generated liquid vapor to the saturated vapor pressure is referred to as a relative concentration, and is a relative concentration of the liquid vapor. The relative concentration takes a value ranging from zero to one at any temperature. While a relative concentration cannot be strictly defined for a mixture such as gasoline and light oil, α may be used to be an index for approximately representing the concentration of such a mixture. It is therefore possible to consider α as a relative concentration for a gasoline vapor or a light oil vapor.

As will be apparent from the foregoing description of the present invention made in connection with an embodiment, the present invention has such particular advantages as follows:

(1 ) Since the fuel vapor distinguishing apparatus of the present invention can eliminate erroneous determination of the kind of an organic vapor under measurement and provide reliable determination to the kind of the organic vapor under measurement, erroneous fuel supply can be prevented, for example, at gas stations when the organic vapor is a fuel vapor.

(2) A compact organic vapor distinguishing apparatus can be provided at a low cost.

(3) Since the light source, the detector unit, and the signal processing unit can be disposed at arbitrary positions away from the sensor members, no electrical circuits need be positioned in a region in which an organic vapor may exist. Since an extremely high safety is thereby provided, the fuel vapor distinguishing apparatus of the present invention can be installed even in a place where an explosive atmosphere can occur.

While the preferred form of the present invention has been described, it is understood that modifications will be apparent to those skilled in the art without departing from the spirit of the invention. The scope of the invention, therefore, is to be determined solely by the following claims.

The entire disclosure of Japanese Patent Application No. 8-52023 will be incorporated herein by reference.

Claims

Patent claims

1. An apparatus for distinguishing kinds of organic vapors comprising: m sensor members each exhibiting a unique reflection characteristic to m kinds of different organic vapors, where m is a positive integer equal to or more than two; m light sources respectively associated with said m sensor members for emitting light for irradiating said sensor members associated therewith; first m optical fibers respectively associated with said m sensor members for introducing light emitted from said m light sources to said sensor members associated therewith; second m optical fibers respectively associated with said m light sources and said first m optical fibers for introducing light reflected by said sensor members; m detectors associated with one of said m light sources and said first m optical fibers for receiving the light introduced by said second m optical fibers associated therewith, each of said m detectors outputting a signal indicative of the intensity of received light; and a processing circuit for comparing magnitudes of the signals generated by said m detectors for each of said m organic vapors with each other to derive differences between said respective signals and for distinguishing kinds of organic vapors based on signs of the differences, whereby said processing circuit processes the signals generated by said m detectors when said m sensor members are positioned in an organic vapor under measurement in order to determine which of said m organic vapors said organic vapor under measurement is.

2. The apparatus according to claim 1 , further comprising a housing where light emitting ends of said first optical fibers and light receiving ends of said second optical fibers are arranged opposite to said sensor members associated therewith with respect to a flow path of said organic vapor under measurement.

3. The apparatus according to claim 1 or 2, wherein each of said sensor members includes a substrate for reflection and a polymer thin film formed on said substrate, each of said polymer thin film exhibiting a unique different reflection characteristic to said m organic vapors from each other.

4. The apparatus according to claim 3, wherein said reflection characteristic of said sensor member is defined by the product of the thickness of said polymer thin film and a refractive index of said polymer thin film.

5. The apparatus according to any of claims 1-3, wherein each said light source comprise light emitting means including a light emitting diode and a semiconductor laser, and each said m detectors comprise light detecting means including a photodiode and a phototransistor.

6. The apparatus according to any of claims 1-4, wherein said organic vapors are fuel vapors.