JP2013064726A - Resin identification device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a resin identification device and its method in which little erroneous determination is caused and the accuracy of identification is high, regardless of whether or not a resin contains an additive such as carbon black and filler.SOLUTION: A resin identification device includes: an infrared light analyzer 30 for irradiating resin flakes 1 on a conveyor 20 with infrared light L1 and detecting the intensity of infrared reflection light L2 to perform spectral analysis of the infrared reflection light L2; and a controller 40 for identifying kinds of the resin flakes 1 on the basis of the results of the spectral analysis by the infrared light analyzer 30. The infrared reflectance of a conveyor surface 20a is higher than those of the resin flakes 1, and an identification procedure of the controller 40 is different between a first resin transmitting specific wavelengths of the infrared light L1 and a second resin not transmitting.

Description

  The present invention relates to a resin recycling technique, and more particularly, to a resin identifying apparatus and method for analyzing and identifying a recycled resin for each composition by an optical technique.

  In general, in the recycling of resin in waste home appliances, the portion where the resin can be dismantled by hand is limited. For this reason, small parts and parts having complicated structures need to be mechanically crushed to select metals or resins, and then used as recycled materials.

  In this case, since it is required to separate each material from the pulverized and mixed state, an advanced sorting technique is required. Of these, metals can be sorted by specific gravity or electrical or magnetic force, but resins cannot be sorted by electrical or magnetic force, so classification methods based on specific gravity or electrostatic charge amount, etc. Has been proposed.

  However, since it is difficult to identify similar types of resins by the above classification method, attention is paid to the difference in wavelength (wave number) dependency of resin absorption or reflectance in near-infrared or mid-infrared light. An identification method has been proposed.

  However, when identifying a black resin containing carbon black or the like, it is difficult to identify because the required signal intensity cannot be obtained because the absorption rate in the near infrared band is large. Therefore, it is desirable to use a mid-infrared band that is less affected by the absorption of carbon black for identifying the black resin.

  As a method of identifying individual crushed resin pieces using the mid-infrared band, the sample is sequentially flowed on a conveyor and measured by a diffuse reflection method from above the sample using a Fourier transform infrared spectrophotometer. Technology (infrared spectroscopy) is known (see, for example, Patent Document 1 and Patent Document 2).

However, the conventional resin identification apparatus method (apparatus) has the following problems.
In other words, in order to confirm the resin distinguishability using the mid-infrared band, when analyzing the actual measurement results of various types of resins by the reflection method using the mid-infrared band, A resin (hereinafter referred to as “opaque resin”) to which a material that does not transmit external light (filler, carbon black, etc.) is added and a resin to which the above material is not added (hereinafter referred to as “transparent resin”) It was confirmed that the shapes of the reflection spectra were different.

Specifically, in the case of a transparent resin, the spectrum inherent to the resin shown in Patent Documents 1 and 2 or the peak of a spectrum derived from an additive (hereinafter referred to as “peak” means “peak”). In addition to “valley”, a new peak (hereinafter referred to as “ghost peak”) was observed.
As a result of the investigation, the ghost peak is peculiar in the case of transparent resin (resin that does not contain additives such as carbon black and filler), and the peak shape depends on the type of resin (hereinafter referred to as “resin type”). Turned out to be unique.

  As a result of further investigation, the ghost peak is caused by the fact that infrared light with a wave number that hardly absorbs in the resin passes through the resin, reflects off the sample stage, passes through the resin again, and enters the detector. It was found that the reason is that the reflection intensity of the wave number is apparently stronger. At this time, since the reflection intensity decreases at a wave number that is slightly absorbed, the ghost peak has a complicated shape.

  If the resin used as the measurement sample contains an additive that absorbs infrared light (such as carbon black), a ghost peak appears because infrared light does not pass through the resin used in the measurement sample. There is no.

Such a ghost peak is not found in a general measurement.
In transmission-type infrared absorption measurement generally performed, infrared light is transmitted through a sample having a thickness of about 10 μm and the degree of absorption is measured. However, since the sample resin is thin, most infrared light is Since only the wave number that is transmitted and strongly absorbed inherent to the resin is absorbed, the ghost peak does not appear.

  In addition, in the infrared reflection measurement of a thin film, the degree of absorption of infrared light that is transmitted through a sample having a thickness of about 10 μm, reflected by the lower reflective layer of the sample, and transmitted through the sample again. In this case, a ghost peak does not appear for the same reason as in the transmission infrared absorption measurement.

  In solid infrared reflection measurement, the sample is irradiated with infrared light and the wave number dependence of the infrared light reflected from the outermost surface of the sample is measured. If the sample is opaque, the sample is not reflected. Infrared light transmitted through the sample is not detected, and if the sample is large, the infrared light transmitted through the sample will not be detected in a completely different direction from the detector, so a ghost peak will still appear There is nothing.

  However, if the thickness of the sample is about several millimeters or less, in addition to the infrared light reflected on the sample surface, after passing through the sample, the infrared light reflected on the sample stage and again transmitted through the sample However, a ghost peak appears because it is under a special situation detected by the detector.

As a result, it was found that the peak used for resin identification (due to the resin) is distorted by the ghost peak, so that it may be difficult to identify the resin. More specifically, when the ghost peak shape of the resin coincides with the peak shape (due to the resin type) used for identification of other resins by chance, the possibility of erroneous identification increases.
Particularly difficult to distinguish is when, for example, PS (Polystyrene) resin and ABS (Acrylonitrile-Butadiene-Styrene) resin are identified.

Although each infrared spectrum of PS resin and ABS resin is very similar, a significant difference is the presence or absence of a peak (hereinafter referred to as “CN peak”) due to CN bonding in the vicinity of wave number 2200 cm ^−1. . That is, it can be identified that the one having a CN peak is an ABS resin, and the one having no CN peak is a PS resin.
However, in the case of the transparent PS resin, it has been experimentally found that a ghost peak having a shape similar to the CN peak exists at the position of the CN peak appearing in the ABS resin and is easily misidentified by the ghost peak.

JP 60-89732 A JP-A-8-300354

The conventional resin identification device and method have a problem that the resin identification peak may be distorted by the ghost peak, which may make the resin identification difficult.
In particular, since a ghost peak exists in the vicinity of the CN peak of the transparent PS resin, it is difficult to determine the presence or absence of the CN peak for distinguishing each infrared spectrum of the PS resin and the ABS resin, and there is a problem that the identification accuracy is lowered. .

  The present invention has been made in order to solve the above-described problems. A resin identification device and method with high identification accuracy with few misjudgments regardless of the presence or absence of additives such as carbon black and filler in the resin. The purpose is to obtain.

  A resin identification device according to the present invention includes an infrared light analysis device that sequentially irradiates a plurality of identification target resins on a sample stage with infrared light, and obtains a spectrum of each reflected infrared light, and an infrared light analysis A controller for identifying a plurality of types of resin to be identified based on an infrared reflection spectrum from the apparatus, wherein the infrared light analyzer is based on the spectrum of the acquired infrared reflected light The infrared light irradiated to the identification resin is transmitted through the identification resin, reflected from the sample stage, and again determined whether there is infrared light in a specific wavelength region that has been transmitted through the identification resin. The type of resin to be identified can be determined by using an identification algorithm in which at least part of the wavelength region of the infrared reflection spectrum for identification is different according to the result of the presence / absence determination of infrared light in a specific wavelength region by an infrared light analyzer. To identify .

  According to the present invention, an additive such as carbon black or filler is added to the identified resin by changing the spectral analysis procedure for the identified resin that generates the ghost peak and the identified resin that does not generate the ghost peak. Regardless of the presence or absence of ghost peaks (whether or not a ghost peak is generated), it is possible to provide a resin identification device and method with high identification accuracy with few erroneous determinations.

It is explanatory drawing which shows the infrared reflection spectrum of the to-be-identified resin (transparent and opaque PP resin board, PS resin board, and ABS resin board) in Embodiment 1 of this invention. It is explanatory drawing which expands and shows the peak vicinity resulting from CH coupling | bonding of the infrared reflection spectrum of FIG. It is explanatory drawing which shows the infrared reflection spectrum of the to-be-identified resin (Opaque PS resin board and opaque ABS resin board) in Embodiment 1 of this invention. It is explanatory drawing which shows the infrared reflection spectrum of the to-be-identified resin (transparent PS resin board and transparent ABS resin board) in Embodiment 1 of this invention. It is explanatory drawing which shows the infrared reflection spectrum of the to-be-identified resin (Transparent PS resin flakes (3 types) and transparent PS resin board) in Embodiment 1 of this invention. It is explanatory drawing which shows the infrared reflection spectrum of the to-be-identified resin (Transparent ABS resin flakes (3 types) and transparent PS resin board) in Embodiment 1 of this invention. It is a block diagram which shows the resin identification device which concerns on Embodiment 1 of this invention. It is a flowchart which shows the resin identification method which concerns on Embodiment 1 of this invention. It is a flowchart which shows the other example of the resin identification method which concerns on Embodiment 1 of this invention. It is explanatory drawing which shows the ghost peak of the to-be-identified resin (PE resin) in Embodiment 1 of this invention.

Embodiment 1 FIG.
First, before explaining the resin identification apparatus and method according to Embodiment 1 of the present invention, the explanatory diagrams of FIGS. 1 to 4 based on the analysis results of Experiment 1 and FIGS. 5 and 5 based on the analysis results of Experiment 2 will be described. The basic technical concept and the ghost peak in the first embodiment of the present invention will be described in detail with reference to FIG.

[Experiment 1]
FIG. 1 is an explanatory diagram showing an infrared reflection spectrum of a resin to be identified (transparent and opaque, PP resin plate, PS resin plate and ABS resin plate) according to Embodiment 1 of the present invention, and the horizontal axis represents the wave number [cm. ⁻¹ ], the vertical axis represents the reflection intensity of infrared light.

In FIG. 1, the reflection spectrum of the opaque resin is indicated by a thin line, and the reflection spectrum of the transparent resin is indicated by a thick line.
Note that the “opaque resin” in FIG. 1 is a resin to which 1% of carbon black is added, and the “transparent resin” is a resin to which no carbon black or filler is added.

In FIG. 1, the resin plate in which the presence or absence of each additive is controlled is used in order to obtain a reflection spectrum caused only by each resin type.
In addition, in resin flakes (flakes) that are actually identified, there is a possibility that the peak shape of the reflection spectrum may be distorted or another peak may appear due to the influence of the surface irregularities, size, additives, etc. is there.

Therefore, in Experiment 1, first, the basic function is confirmed using an ideal resin plate whose additive and shape are controlled. Thereafter, in Experiment 2 (described later), it is confirmed whether or not the same is true even in the case of actual resin flakes.
In the actual identification of the resin flakes, the infrared reflection spectrum of the resin flakes is compared with the ideal infrared reflection spectrum (resin plate spectrum) of each resin.

As a specific measurement apparatus in Experiment 1, for example, FT-IR5500 manufactured by JEOL Ltd. was used, and as an infrared light detector, an MCT (HgCdTe) detector was used. Further, as measurement conditions, integration was performed 10 times at a resolution of 4 cm ⁻¹ , a reflection measurement jig was used as the optical system, and the incident angle and the reflection angle were both 10 degrees. Furthermore, a sample stage made of stainless steel was used.

As shown in FIG. 1, the transparent and opaque, PP resin plate, the infrared reflection spectrum of the PS resin plate and ABS resin plate, both in and around 1500cm around ^-1 wavenumber 3000 cm ^-1, due to the CH bond A plurality of CH peaks CHp1 and CHp2 are observed.
Further, 2400 cm ^-1 near the absorption peak P3,1300～1800cm ^-1 due to the CO2 in the atmosphere is also recognized in any small peaks due to moisture in the atmosphere P4, P5 is near 3500～4000Cm ^-1 .

FIG. 2 is an explanatory diagram showing an enlargement of the vicinity of the CH peak CHp1 (3000 cm ⁻¹ ) in FIG.
As is clear from FIG. 2, the peaks in the vicinity of 3000 cm ⁻¹ are the same in both the PP resin plate, the PS resin plate, and the ABS resin plate, which are transparent and opaque. Moreover, the reflection spectrum of the PS resin plate and the ABS resin plate near the CH peak CHp1 (3000 cm ⁻¹ ) is very similar.
Therefore, it can be seen from the peak shape around 3000 cm ⁻¹ whether it is PP resin, PS resin or ABS resin.

Further, as is clear from FIG. 1, the remarkable difference in the infrared reflection spectrum between the PS resin plate and the ABS resin plate is the presence or absence of the CN peak CNp due to the CN bond appearing in the vicinity of 2200 cm ⁻¹ .

FIG. 3 is an enlarged explanatory view showing the vicinity of the CN peak CNp at 2200 cm ⁻¹ in FIG. 1, and shows infrared reflection spectra of the opaque PS resin plate and the opaque ABS resin plate.
In FIG. 3, a CN peak CNp is recognized in the infrared reflection spectrum of the opaque ABS resin plate, whereas a CN peak is not recognized in the infrared reflection spectrum of the opaque PS resin plate.

FIG. 4 is an explanatory view showing the vicinity of the CN peak CNp in the same manner as FIG. 3 and shows the infrared reflection spectra of the transparent PS resin plate and the transparent ABS resin plate.
Also in FIG. 4, the CN peak CNp is recognized in the infrared reflection spectrum of the transparent ABS resin plate.

However, in this case, even in the infrared reflection spectrum of the transparent PS resin plate, a shoulder (see a one-dot chain line frame) that does not originally occur is observed near the wave number (2200 cm ⁻¹ ) where CN coupling appears. In particular, when the resin to be identified is a resin flake, the fluctuation of the base line is often large, so that the shoulder portion may be erroneously determined as a CN peak.

Therefore, as a result of Experiment 1, it has been clarified that there arises a problem that the identification accuracy between the transparent PS resin and the transparent ABS resin is lowered due to the shoulder in FIG.
The shoulder means a shape having a sharp inflection point (namely, a shoulder line portion) although it does not have a sharp tip shape as much as the CN peak CNp.

[Experiment 2]
FIG. 5 is an explanatory diagram showing an infrared reflection spectrum of the identified resin in Embodiment 1 of the present invention, examples of infrared reflection spectra of transparent PS resin flakes (three types), and infrared of the transparent PS resin plate. The reflection spectrum is shown.

In Figure 5, the infrared reflection spectrum of the transparent PS resin flakes (see thin line), the none of the three, in the vicinity of 3500 cm ^-1 and near 2000 cm ^-1, infrared reflection spectrum of the transparent PS resin plate (thick line see ) And ghost peaks Gp1 and Gp2 having a shape similar to that of FIG.
Further, it can be seen that the peak intensities of the ghost peaks Gp1 and Gp2 are larger than the intensity of the CH peak CHp1 in the vicinity of 3000 cm ⁻¹ .

  FIG. 6 is an explanatory diagram showing the infrared reflection spectrum of the resin to be identified in Embodiment 1 of the present invention. Examples (three types) of the infrared reflection spectrum of the transparent ABS resin flakes and the infrared of the transparent ABS resin plate are shown. The reflection spectrum is shown.

In FIG. 6, in the case of the infrared reflection spectrum (see thin line) of the transparent ABS resin flake, similar to the infrared reflection spectrum (see thick line) of the transparent ABS resin plate in the vicinity of 2000 cm ⁻¹ , as in FIG. A ghost peak Gp2 is observed.
Also in FIG. 6, the peak intensity of the ghost peak Gp2 is the CH peak CHp1 near 3000 cm ⁻¹ and the CN peak CNp near 2200 cm ⁻¹ (it is too small to be confirmed on the scale of the vertical axis in FIG. 6). It can be seen that it is greater than the strength.

As is apparent from FIGS. 5 and 6, as a result of Experiment 2, the ghost peak shape of the transparent PS resin flake and the transparent ABS resin flake is similar to the peak shape of the transparent PS resin plate and the transparent ABS resin plate. I understood.
From the above, the transparent PS resin flakes and the transparent ABS resin flakes can distinguish PS and ABS based on the peak shape of the ghost peak.

As the measurement apparatus in Experiment 2 (FIGS. 5 and 6), as in Experiment 1 described above (FIGS. 1 to 4), for example, FT-IR5500 manufactured by JEOL Ltd. is used, and MCT ( A HgCdTe) detector was used.
In addition, a reflection measuring jig was used as the optical system, and the incident angle and the reflection angle were both 10 degrees. In addition, a sample stage made of stainless steel (semi-gloss) was used.

In this case, it has been confirmed that even if a semi-gloss rolled aluminum plate (material A5052) is used instead of the stainless steel sample stage, substantially the same infrared reflection spectrum including a ghost peak can be obtained.
It has also been confirmed that a stronger ghost peak can be obtained by changing the sample stage to an aluminum mirror. Conversely, it has been confirmed that when a material having a low infrared reflectance (for example, alumina) is used as the sample stage, the ghost peak becomes small.

From the above, infrared light of a specific wave number that can be transmitted through the transparent resin plate (or transparent resin flakes) is transmitted through the transparent resin plate (or transparent resin flakes) and reflected on the surface of the sample stage. It turns out that it passes through a transparent resin plate (or transparent resin flakes) again and is detected as ghost peaks Gp1 and Gp2 by being caught by a detector.
Therefore, the shapes of the ghost peaks Gp1 and Gp2 are considered to be unique to the resin type, and it is clear that the resin type can be identified using the ghost peaks Gp1 and Gp2.

Hereinafter, the resin identification apparatus according to Embodiment 1 of the present invention will be described with reference to FIG. 7 in consideration of the results of Experiments 1 and 2 described above.
FIG. 7 is a block diagram schematically showing the entirety of the resin identifying apparatus according to Embodiment 1 of the present invention.

  In FIG. 7, the resin identification device includes a supply device 10 that supplies resin flakes 1 (identified resin) as indicated by broken line arrows, a conveyor 20 that conveys the resin flakes 1 as indicated by dashed-dotted arrows, and a conveyor 20 Infrared light analysis device 30 for irradiating resin flake 1 with infrared light L1 and controller 40 for identifying the type of resin flake 1 based on the analysis result of infrared light analysis device 30.

  Here, a Fourier transform infrared spectrometer (FT-IR) is used as the infrared light analyzer. Although not shown in FIG. 7, the infrared light analysis device 30 includes an emitting unit that emits infrared light L <b> 1, and a light receiving unit (detector) that receives infrared reflected light L <b> 2 from the resin flake 1. It has. Moreover, the surface 20a of the conveyor 20 which mounts the resin flakes 1 functions as a sample stage.

First, the supply device 10 sequentially supplies a plurality of resin flakes 1 (identified resins) onto the conveyor 20.
At this time, the supply of the resin flakes 1 onto the conveyor 20 is performed so that the individual resin flakes 1 are arranged in a straight line with no predetermined gap therebetween. Adjacent resin flakes 1 need only be spaced apart from each other, and the spacing may not be constant.

  The conveyor 20 conveys the resin flakes 1 directly below the infrared light analyzer 30, and the infrared light analyzer 30 irradiates the resin flakes 1 with the infrared light L1 and analyzes the intensity of the infrared reflected light L2. To obtain a spectrum.

  Finally, the controller 40 compares the spectrum of the infrared reflected light acquired by the infrared light analyzer 30 with the standard spectrum of the infrared reflected light L2 of each resin type acquired in advance, thereby The resin type of flake 1 is determined.

  Although not specifically shown here, it is also possible to sort out the resin by ejecting air or the like based on the identification result by the controller 40 and eliminating the resin flakes 1 other than the target species from the track of the conveyor 20. It is.

  In addition, when the resin mixture ratio is confirmed based on the determination result of the controller 40, only the basic configuration shown in FIG. 7 is sufficient. For example, 1000 resin flakes 1 can be extracted from the population and the mixing ratio can be examined.

That is, if the resin flakes 1 (resin flake group) obtained by sorting the resin types are used as a population, the purity (the number of resin flakes 1 that become impurities) can be known.
In addition, if the resin flake 1 (resin flake group) mixed after crushing is used as a population, the mixing ratio for each resin type can be known, and optimization of the subsequent separation process conditions and each resin type of the final product You can know the amount in advance.

In Embodiment 1 of the present invention, the conveyor surface 20a corresponding to the sample stage is made of at least a material having high reflectivity of infrared light having a wave number with a ghost peak (specifically, a metal such as stainless steel or aluminum). is there. Here, “the reflectance is high” means that the reflectance is higher than that of the resin flakes 1 that are identification objects.
This makes it possible to amplify and detect the intensity of the ghost peak.

Next, a spectrum analysis procedure (resin identification method) of the resin identification apparatus according to Embodiment 1 of the present invention shown in FIG. 7 will be described with reference to FIG.
FIG. 8 is a flowchart showing the resin identification method according to Embodiment 1 of the present invention, and shows a spectrum analysis procedure by the controller 40 for identifying the resin type from the infrared reflection spectrum of the resin flake 1 (identified resin). ing.

  In FIG. 8, as a specific example, when three types of resin flakes of PP resin, PS resin, and ABS resin are mixed, the case where the resin type is determined based on the infrared reflection spectrum of each resin flake 1 is shown. ing.

In this case, as described above, in the transparent or opaque PP resin, the transparent or opaque PS resin, or the transparent or opaque ABS resin, the CH peak CHp1 due to CH appears in the vicinity of the wave number of 2700 to 3000 cm ⁻¹ , and the transparent or opaque ABS resin , A CN peak CNp due to CN appears in the vicinity of a wave number of 2200 cm ⁻¹ .
Further, in the vicinity of a wave number 3200～3700Cm ^-1, ghost peaks Gp1 appeared by transparent PS resin, in the vicinity of a wave number 1700～2200Cm ^-1, transparent PP resin, ghost peaks Gp2 appears by the transparent PS resin and a transparent ABS resin.

In FIG. 8, the controller 40 first determines whether or not there is a ghost peak Gp2 (near 2000 cm ⁻¹ ) in the infrared reflection spectrum from the resin flake 1 (identified resin) (step S1). If it is determined that there is no (that is, the resin flake 1 is opaque), then the CH peak CHp1 (near 2700 to 3000 cm ⁻¹ ) is compared (step S6).

  On the other hand, when it is determined in step S1 that the ghost peak Gp2 is present (resin flake 1 is transparent), first, the shape of the ghost peak Gp2 is changed to the standard spectrum of the transparent PP resin (for example, the transparent PP resin in FIG. 1). (Spectrum of the plate) and standard PS or standard ABS (standard spectrum of transparent PS resin or transparent PPABS resin) and the similarity between them is determined (step S2).

In step S2, if the ghost peak Gp2 matches the standard PP (standard spectrum of transparent PP resin), it is determined that the resin flake 1 is a transparent PP resin (step S11).
If it is determined that the ghost peak Gp2 coincides with the standard PS or standard ABS (transparent PS resin or transparent ABS resin standard spectrum), then the ghost peak Gp1 (near 3500 cm ⁻¹ ) is compared (step S3). ).

In step S3, the ghost peak Gp1 is shaped into a standard spectrum of transparent PS resin (for example, the spectrum of the transparent PS resin plate in FIG. 1) and a standard spectrum of transparent ABS resin (for example, the transparent ABS resin plate in FIG. 1). If the ghost peak Gp1 matches the standard PS, it is determined as PS (step S4).
If the ghost peak Gp1 matches the standard ABS, the ABS is determined (step S5).

  That is, as described above, since the standard spectral shapes of the ghost peak Gp2 of the transparent PS resin and the transparent ABS resin are similar to each other, the controller 40 determines that the infrared reflection spectrum from the resin flake 1 is transparent in step S2. It is divided into the case of matching with the PP resin and the case of matching with the transparent PS resin or the transparent ABS resin. Further, in step 3, the infrared reflection spectrum from the resin flake 1 matches with the transparent PS resin or is transparent. It is determined whether it matches the ABS resin.

For comparison of similarities, the covariance value is determined within the range of the ghost peak Gp2 for the infrared reflection spectrum of the resin flake 1 and each standard infrared reflection spectrum, and the peak is determined from the magnitude of each value. A method of quantifying and comparing similarities is applied.
At this time, an appropriate threshold value is provided, and the most similar ones exceeding the threshold value are regarded as matching, and the case where both are below the threshold value can be regarded as “no ghost peak Gp2”. The method for quantifying the similarity is not limited to covariance, and standard deviation may be used.

As a specific classification method, in step S1, when the covariance of the transparent PP with the standard spectrum exceeds a certain threshold, it can be determined that the transparent PP matches the transparent PP.
Further, when the covariance of transparent PS or the standard spectrum of transparent PS exceeds a certain threshold value, it can be determined that it matches the PS resin or ABS resin.

When the ghost peak Gp2 of the infrared reflection spectrum from the resin flake 1 coincides with the standard PP resin, it is determined as PP resin (step S11).
If the ghost peak Gp2 matches the standard PS resin or the standard ABS resin, the process proceeds to step S3, where the ghost peak Gp1 is covariance with the standard spectrum of the transparent PS resin and the standard spectrum of the transparent ABS resin. From the comparison of these covariance values, the more similar one is determined as the resin type.

That is, if the ghost peak Gp1 (near 3500 cm ⁻¹ ) in the infrared reflection spectrum of the resin flake 1 matches the standard PS in step S3, the controller 40 determines that the resin flake 1 is a PS resin. (Step S4) If it matches the standard ABS, it is determined that the resin flake 1 is an ABS resin (Step S5).

If the determination result for the ghost peak Gp2 (near 2000 cm ⁻¹ ) does not match any of the PP resin, PS resin, and ABS resin (resin flake 1 is opaque) in step S1, the process proceeds to step S6, and the resin The CH peak CHp1 (near 2700 to 3000 cm ⁻¹ ) in the infrared reflection spectrum of flake 1 is compared with the similarity with PP resin, PS resin or ABS resin.

If it is determined in step S6 that the CH peak CHp1 of the resin flake 1 matches the standard PP resin, it is determined that the resin flake 1 is a PP resin (step S7).
In Step S6, when it is determined that the CH peak CHp1 of the resin flake 1 matches the standard PS resin or the standard ABS resin, the CN peak CNp (near 2200 cm ⁻¹ ) is subsequently compared (Step S6). S8).

  In step S8, the controller 40 compares the CN peak CNp of the resin flake 1 with the similarity between the standard PS resin and the standard ABS resin, and determines that the more similar resin is the resin type of the resin flake 1. .

  That is, if it is determined in step S8 that the CN peak CNp of the resin flake 1 matches the standard PS, it is determined that the resin flake 1 is a PS resin (step S9), and if it matches the standard ABS, the resin flake 1 Is determined to be an ABS resin (step S10).

As a result, whether or not the resin flake 1 is transparent or opaque, the presence / absence determination (step S1) of the ghost peak Gp2 (near 2000 cm ⁻¹ ) is compared with the ghost peak Gp1 (near 3500 cm ⁻¹ ) (step S3). By comparing the CH peak CHp1 (near 3000 cm ⁻¹ ) (step S 6) and the CN peak CNp (near 2200 cm ⁻¹ ) (step S 8), the resin type of the resin flake 1 can be identified with high accuracy. it can.

The processing routine shown in FIG. 8 is merely an example of a spectrum analysis procedure (resin identification method) of the resin identification device according to Embodiment 1 of the present invention. For example, the processing routine shown in FIG. 9 is applied. In addition, PP resin, PS resin and ABS resin can be identified.
In FIG. 9, the same processes as those described above (see FIG. 8) are denoted by the same reference numerals as described above.

In FIG. 9, first, the infrared reflection spectrum of resin flake 1 is compared with the infrared reflection standard spectra of PP resin, PS resin, and ABS resin for CH peak CHp1 (near 3000 cm ⁻¹ ) (step S6). If it matches the standard PP resin, it is determined that the resin flake 1 is a PP resin (step S7).

On the other hand, when the infrared reflection spectrum of the resin flake 1 at the CH peak CHp1 coincides with the standard PS resin or the standard ABS resin in Step S6, the presence or absence of the ghost peak Gp2 (near 2000 cm ⁻¹ ) is subsequently determined. The determination is made based on the similarity to the ghost peak Gp2 (step S1).

If it is determined in step S1 that the ghost peak Gp2 is present (resin flake 1 is transparent), a comparison determination based on the similarity of the ghost peak Gp1 (near 3500 cm ⁻¹ ) is subsequently performed (step S3).

  In step S3, if the ghost peak Gp1 in the infrared reflection spectrum of the resin flake 1 matches the standard PS resin, it is determined that the resin flake 1 is a PS resin (step S4), and if it matches the standard ABS, It is determined that the resin flake 1 is ABS resin (step S5).

On the other hand, if it is determined in step S1 that the ghost peak Gp2 is not present in the infrared reflection spectrum of the resin flake 1 (the resin flake 1 is opaque), then the similarity of the CN peak CNp (near 2200 cm ⁻¹ ). Based on the comparison, a determination is made (step S8).

That is, if the CN peak CNp in the infrared reflection spectrum of the resin flake 1 matches the standard PS resin, it is determined that the resin flake 1 is a PS resin (step S9), and if it matches the standard ABS, the resin flakes It is determined that 1 is an ABS resin (step S10).
Thus, as described above, based on the similarity to the standard resin spectrum, whether the resin flake 1 is a PS resin or an ABS resin can be identified with high accuracy.

  Note that the identification procedure according to the first embodiment of the present invention is not limited to the processing routine of FIG. 8 or FIG. 9, in order to improve the identification accuracy of the resin type, in addition to the comparison determination at the wave number, The similarity between the infrared reflection spectrum and the standard resin may be determined including other wave numbers.

In addition, the case where the resin flake 1 (identified resin) is three types of PP resin, PS resin, or ABS resin has been described, but the present invention is not limited to these and can be applied to identification of other resin types. Needless to say.
The gist of the present invention is that in the resin identification step, the ghost peak Gp2 (near 2000 cm ⁻¹ ) is used to determine the transparency or opacity of the resin to be identified, and in each case, the optimum identification method (similarity determination process) ) Is applied.

  In addition to Gp2, a ghost peak is used to determine the transparency or opacity of the resin to be identified, and in each case, an optimum identification method (similarity determination processing) is applied.

  For example, the occurrence of a ghost peak in a transparent resin is confirmed not only in PP resin, PS resin and ABS resin but also in transparent PE (polyethylene) resin, transparent PC (polycarbonate) resin, and many other types of resins. Has been.

FIG. 10 is an explanatory diagram showing the reflection intensity characteristic in the case of PE resin, where the thick line characteristic indicates the reflection spectrum of the opaque PE resin and the thin line characteristic indicates the reflection spectrum of the transparent PS resin.
In FIG. 10, the remarkable ghost peak has generate | occur | produced in 2000 cm < ^-1 > vicinity in the reflection spectrum of transparent PS resin.

  Thus, even when the resin to be identified is a PE resin and includes a transparent resin or an opaque resin and the ghost peak shown in FIG. 10 occurs, the identification procedure at an appropriate wave number is set as described above. By applying the similarity comparison process with the standard resin, the resin type can be identified.

  In addition, as described above, the resin identification using the ghost peak was confirmed by using a stainless steel sample stage, and almost the same effect was obtained even when a rolled aluminum alloy (A5052) was used for the sample stage. It is confirmed that

  Further, when an aluminum-deposited glass on a mirror glass is used as a sample stage, a ghost peak appearing in the reflection spectrum of the transparent resin flake 1 is more emphasized.

  As other specific examples, a method of sticking a metal plate having a narrow width in the traveling direction (one-dot chain line arrow) on the surface 20a of the conveyor 20 in FIG. 7, a method of using a metal mesh as a material of the conveyor 20, and the like There is.

  Further, in FIG. 7, the conveyor 20 that flows in one direction (one-dot chain line arrow) is shown as an example, but instead of the conveyor 20, a metal disk (not shown) that rotates around the central axis is used. May be.

  In this case, the only difference from FIG. 7 is that the conveyor 20 has a longitudinal structure in which the shafts at both ends are turned, whereas the metal disk has a lateral structure. However, in the case of a metal disk, since the resin flake 1 that has made one round returns repeatedly, it is necessary to remove all the resin flakes by air blow or the like when the identification of the resin flake 1 is completed.

  As described above, the resin identification apparatus according to Embodiment 1 (FIG. 7) of the present invention irradiates the plurality of resin flakes 1 (identification resin) on the conveyor surface 20a (sample stage) with infrared light L1. The infrared light analyzing device 30 that detects the intensity of the infrared reflected light L2 from the plurality of resin flakes 1 and performs spectral analysis of the infrared reflected light, and a plurality of the results based on the spectral analysis results from the infrared light analyzing device 30 And a controller 40 for identifying the type of the resin flake 1.

  The resin identification procedure (FIGS. 8 and 9) in the controller 40 is the same as that of the first resin that transmits infrared light having a specific wavelength, even though the plurality of resin flakes 1 are the same type. When both the second resin that does not transmit infrared light is included, the identification procedure for the first resin and the identification procedure for the second resin are different from each other.

The conveyor surface 20a (sample stage) is made of stainless steel or an aluminum alloy so that the infrared reflectance of the surface is higher than the infrared reflectance of the resin flakes 1.
The controller 40 identifies the type of resin to be identified based on the spectrum analysis result from the infrared light analysis device 30.

  In the resin identification method according to the first embodiment (FIGS. 7 to 9) of the present invention, the infrared light L1 is irradiated to the plurality of resin flakes 1 on the conveyor surface 20a, and the infrared reflection from the plurality of resin flakes 1 is performed. A resin identification method for identifying the types of a plurality of resin flakes 1 by detecting the intensity of light L2 and performing spectrum analysis of infrared reflected light, wherein the plurality of resin flakes 1 are of the same type Regardless, when both the first resin that transmits infrared light of a specific wavelength and the second resin that does not transmit infrared light of a specific wavelength are included, the identification procedure of the first resin and the second resin The resin identification procedure is different from each other.

  The first and second resin identification procedures (FIGS. 8 and 9) are a ghost peak determination step (step) for determining the presence or absence of the ghost peak Gp2 based on the infrared reflection spectrum of the resin flake 1 included in the spectrum analysis result. S1).

In the first resin identification procedure, when it is determined in step S1 that the ghost peak Gp2 is not present in the infrared reflection spectrum, the type of the resin flake 1 is selected based on the intrinsic peak (CN peak CNp) of the first resin. A first identification step (step S8) for identifying is provided.
The “peak unique to the resin” means a peak unique to the resin type observed in the spectrum of the infrared reflected light L2, regardless of the presence or absence of additives such as carbon black and filler. It does not include the ghost peak that is said.

  The second resin identification procedure is a second procedure for identifying the type of the resin flake 1 using another ghost peak Gp1 when it is determined in step S1 that the ghost peak Gp2 is present in the infrared reflection spectrum. Identification step (step S3).

The ghost peak Gp2 corresponds to a region near the wave number 2000 cm ⁻¹ of the infrared reflection spectrum, and the other ghost peak Gp1 corresponds to a region near the wave number 3500 cm ⁻¹ of the infrared reflection spectrum.

  As described above, the transparency or opacity of the resin flake 1 is determined based on the presence or absence of a ghost peak in the infrared reflection spectrum of the resin flake 1, and an optimum identification procedure (algorithm) is determined for each transparent or opaque determination result. ) Is used.

  In other words, when there is no ghost peak Gp2, the resin type is identified by a peak unique to the resin, and when there is a ghost peak Gp2, an erroneous determination is made by identifying the resin type using at least the ghost peak Gp1. By avoiding, the identification accuracy can be improved.

  Therefore, according to the first embodiment of the present invention, by changing the spectrum analysis procedure for the identified resin that generates the ghost peak Gp2 and the identified resin that does not generate the ghost peak Gp2, respectively, Regardless of the presence or absence of additives such as carbon black or filler (whether there is a ghost peak), it is possible to provide a resin identification apparatus and method with high identification accuracy with few erroneous determinations.

  DESCRIPTION OF SYMBOLS 1 Resin flake (identification resin), 10 supply apparatus, 20 conveyor, 20a Conveyor surface (sample stand), 30 Infrared light analyzer, 40 controller, CHp1 CH peak, CNp CN peak, Gp1 another ghost peak, Gp2 ghost Peak, L1 infrared light, L2 infrared reflected light, S1 ghost peak determination step, S3 second identification step, S8 first identification step.

Claims (7)

Infrared light analyzer that sequentially irradiates a plurality of identified resins on a sample stage with infrared light and acquires the spectrum of each reflected infrared light;
Based on the infrared reflection spectrum from the infrared light analyzer, a controller for identifying the type of the plurality of identified resins,
A resin identification device comprising:
The infrared light analyzer is
Based on the spectrum of the acquired infrared reflected light, out of the infrared light irradiated to the identified resin, the specific wavelength region is transmitted through the identified resin, reflected by the sample stage, and again transmitted through the identified resin. Determine the presence or absence of infrared light,
The controller is
By using an identification algorithm in which at least a part of the wavelength region of the infrared reflection spectrum for identification is different according to the result of the presence / absence determination of infrared light in the specific wavelength region by the infrared light analyzer, the identified resin A resin identification device characterized by identifying the type of the resin. The infrared light analyzer is
The specific wavelength region;
A wavelength region for identification that is used when it is determined that there is infrared light of a specific wavelength region that is transmitted through the resin to be identified, reflected by the sample stage, and again transmitted through the resin to be identified;
A wavelength region for identification that is used when it is determined that there is no infrared light in the specific wavelength region that is transmitted through the resin to be identified, reflected by the sample stage, and again transmitted through the resin to be identified;
The resin identifying device according to claim 1, wherein the plurality of resins to be identified are irradiated with infrared light.   The resin identification apparatus according to claim 1, wherein an infrared reflectance of the sample stage is higher than an infrared reflectance of the identified resin.   The resin identifying apparatus according to claim 3, wherein the surface of the sample stage is made of stainless steel or an aluminum alloy. A resin identification method for sequentially irradiating a plurality of identified resins on a sample stage with infrared light, and identifying the types of the plurality of identified resins based on the spectrum of each reflected infrared light,
A step of determining whether there is a peak in an infrared light spectrum of a specific wavelength region that is transmitted through the identified resin, reflected by the sample stage, and again transmitted through the identified resin;
A resin identification method characterized in that all or part of the wavelength region of the infrared spectrum for identification differs according to the determination result of the presence or absence of a peak in the determination step. The infrared light for identification irradiated to the plurality of identification resins is
The specific wavelength region;
A wavelength region for identification that is used when it is determined that there is infrared light of a specific wavelength region that is transmitted through the resin to be identified, reflected by the sample stage, and again transmitted through the resin to be identified;
A wavelength region for identification that is used when it is determined that there is no infrared light in the specific wavelength region that is transmitted through the resin to be identified, reflected by the sample stage, and again transmitted through the resin to be identified;
The resin identification method according to claim 5, wherein: The identification resin includes at least ABS resin or PS resin,
The specific wavelength region includes at least a part of a region having a wave number of 2000 to 2300 cm ⁻¹ in order to distinguish between ABS and PS,
In the specific wavelength region, when it is determined that there is infrared light that is transmitted through the identified resin, reflected by the sample stage, and again transmitted through the identified resin,
The infrared reflection spectrum shape of the resin to be identified in at least a part of a region having a wave number of 3200 to 3600 cm ^-1 is identified by comparing with a standard spectrum shape of PS or ABS prepared in advance. 5. The resin identification method according to 5.

Images