JP2000235636A - Information medium using defect information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent illegal copy and the use and to improve reliability in certification data of an IC card by detecting a crystal defect existing in a speci fied area and using information corresponding to the detected crystal defect as identification information of the article. SOLUTION: A defect measuring area 12 is set in an area where the device pattern of several mm is not formed from the outer periphery of a wafer on a side opposite to a notch 11. Marks 101-104 are formed in the boundary part of the area 12 and a defect is measured with the marks as references. Then, measured crystal defects 14 shown by stars exist. When the defect positions are shown in order from the smallest value of X, they can be shown by (2, 4) (6, 1) (7, 3) (9, 4) (11, 2) (14, 3) (15, 1) (19, 2). Information accompanied by the crystal defects 14 is used as identification information of the article. Thus, illegal copy and the use can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an article identification apparatus and an article identification method.

[0002]

2. Description of the Related Art For example, a card in which money is paid in advance and a corresponding amount of information and functions are coded and recorded, a so-called prepaid card, or an electronic money card which is a kind of IC card capable of transferring money, Cash cards and credit cards are used in which purchases are deducted from their accounts. In order to confirm that the card is a card owned by the person, an ID method is adopted. That is, a method is used in which a password is registered in a card in advance, and when used, the password is confirmed by matching the password. With respect to the IDs of these cards, techniques for preventing duplication and unauthorized use have been developed, and there are the following known examples.

Japanese Patent Laid-Open No. Hei 8-2112 discloses that a credit card, an identification card, and the like can be prevented from being duplicated or counterfeited by using an optical diffraction structure by utilizing an etching technique of semiconductor manufacturing.
No. 15. In this known example, an identification label comprising a diffractive structure formed on a substrate layer based on artificial data and a transparent overcoat layer thereon is disclosed. U.S. Pat. No. 4,661,983 discloses a known invention using a crack as identification information.

As another known invention, as a personal identification method and device of a security system for confirming the identity of a person, there is disclosed a method of identifying a person by using biological information such as a fingerprint, an iris, a fundus blood vessel, and a capillary blood vessel pattern. A method of checking is known. For example, an apparatus using a fingerprint is disclosed in
It is disclosed in JP-A-3-123168. A known invention using a blood vessel fluoroscopic image as personal identification information is disclosed in Japanese Patent Laid-Open No. 7-213.
No. 73.

On the other hand, as a method of managing a substrate in a semiconductor process, Japanese Patent Laid-Open Publication No. Hei 8-3170 discloses a method in which a recognition mark on a wafer is provided with a laser marker or FIB (Focused Ion Beam) to provide letters and numbers due to surface irregularities.
No. 5.

[0006]

At present, when a card is widely used, for example, an I-card of an IC card is used.
It is unacceptable to decrypt and duplicate D for unauthorized use, and various techniques have been developed as described above. On the other hand, it is also true that unauthorized duplication and its use have not been eradicated.

An object of the present invention is to provide an article having identification information which is extremely difficult to be illegally copied and used, and an apparatus and method for identifying an article using the article.

[0008]

SUMMARY OF THE INVENTION The present invention proposes, as a means, to utilize information accompanying a crystal defect which is extremely difficult to artificially configure.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS As one embodiment of the use of information accompanying a crystal defect which is the object of the present invention, a crystal defect itself in a semiconductor can be used. The fact that it is extremely difficult to construct artificially will be described below.

In a semiconductor, crystal defects occur in the crystal during crystal growth. This crystal defect is, for example, a grown-i
Although there are n-defects and oxygen precipitates generated by heat treatment, it is difficult to artificially control the position, depth, size, and the like of these. The reason for this is that the occurrence of these defects is a phenomenon related to the diffusion phenomenon, and it is difficult to control the position and size of each individual defect. These defects have a defect density of the order of 10 ⁶ per 1 cm ^3, for example, in a wafer prepared by the CZ method (Czochralski: pull-up method). This density can be said to be a sufficient number to be used as identification information. Therefore, if attention is paid to the distribution pattern of the locations of the crystal defects, it can be understood that this pattern can be used as information for identifying the substrate. Therefore, the substrate can be identified by generating a code using a random pattern of crystal defects naturally occurring inside the substrate.

Therefore, it is possible to use the defect pattern of the semiconductor substrate as a management code for the substrate. In addition, a chip that has undergone the process is further defective, so that the identification information is further increased, and it can be applied to the management of a chip mounted on an IC card, and a chip having the same code is duplicated. This makes it possible to prevent unauthorized use of the IC card.

Embodiment 1 In this embodiment, an example will be described in which a defect code on a semiconductor substrate is used as a management code for a substrate. In the element manufacturing process, processing on the substrate such as heat treatment, etching, and polishing is performed, so that the mark attached to the substrate surface is deformed or discolored, damaged or disappeared, making it difficult to recognize the mark It may be. Furthermore, in the case where a single company conducts the entire process in a single line from the integrated process, a specialization system in which a specialized company specializing in that process is assigned to each type of process will be adopted. It is expected that a problem will occur in the display of marks to be attached to the substrate and the like in the process management in manufacturing over a period of time.

[0013] If the management code of the substrate is used by utilizing the defect pattern of the semiconductor substrate, the problem of artificially attaching these codes can be solved.

An example of using a crystal defect as a recognition pattern of a semiconductor substrate and detecting the defect will be described below.

As shown in FIG. 1A, a defect measurement region 12 is set in a region where a device pattern of several mm is not formed from the outer periphery of the wafer 1 on the opposite side of the notch 11. FIG.
FIG. 2B schematically shows an example of the region 12 and an example of a crystal defect appearing in the region. The area 12 is surrounded by + -shaped marks 10 _{1 to} 10 ₄ indicating a range for determining this. Select a region on the surface of the wafer 1 where a device pattern is not formed in a region of several mm from the outer periphery opposite to the notch 11 and form a concave or convex mark on the wafer surface with a laser mark or FIB mark at the boundary of the region. Defect measurement is performed by forming 10 ₁ -10 ₄ and using it as a reference for the region 12. A specific example of the defect measurement will be described later.

In the area 12 surrounded by the marks 10 _{1 to} 10 ₄ , a measured defect 14 exists as shown by an asterisk. The example shown in FIG. 1B schematically shows a state in which the in-plane distribution of defects in a range of a depth of 0.5 μm from the surface is measured and enlarged. Not all of the defects 14 appear to be the same size, and some appear large as shown, while others are barely visible. As shown by the broken line, the position coordinates of the defect in the range of this area are shown at regular intervals in the XY directions, and the position coordinates of the partition defect are shown in two-dimensional coordinates for each defect.
Here, the description using the two-dimensional coordinates is a concept, and it goes without saying that the binary coordinates are not actually displayed on the substrate 1. In the case of FIG. 1B, a two-dimensional X
It is shown by the Y coordinate, X is divided into 22 and Y is divided into 5,
The position of one defect is shown in XY coordinates in ascending order of the value of X. In this case, the defect position is represented by (X, Y) X
(2, 4) (6, 1)
(7, 3) (9, 4) (11, 2) (14, 3) (1
5, 1) (19, 2). Here, (13,
The fact that 1) is neglected means that the signal level is low, and there is a possibility of noise, so that it was deleted in the course of signal processing. When these defect coordinates are displayed in hexadecimal code in order from the left, 24-61-73-94-112-143-
151-192. This is the code generated from the defect pattern in a one-to-one correspondence. This code can be displayed on a defect inspection device and displayed as a code recognizable by an operator.

Thus, for recognition of a semiconductor chip,
By utilizing internal defects such as grown-in defects generated during crystal growth and oxygen precipitates and slip dislocations generated by heat treatment in the crystal, on the surface which is susceptible to artificial changes such as foreign substances adhered to the surface and surface scratches Can be prevented. Since the defect measurement area 12 to be set only needs to be an area which is not affected by the circuit pattern, the defect measurement area 12 is not limited to the opposite side of the surface of the wafer 1 from the notch 11, and the area around the notch 11 where no device pattern is formed is selected. Is also good. Although the example of FIG. 1A has been described using a notch as an example, it goes without saying that this may be an orientation flat.

FIG. 2 is a block diagram showing an example of an apparatus for detecting a crystal defect. Sample table 1 on which substrate 1 to be measured is placed
A light source 140 ₁ , 140 ₂ that emits light of different wavelengths is provided below the light source 40. These light sources
Light of each wavelength is irradiated onto the defect measurement area 12 on the back surface side of the substrate 1 with a time lag. By the photodetector 141,
Scattered light generated from surface or internal defects is detected for each irradiation wavelength. The detected scattered light intensity signal and the defect detection position are digitized and stored in the temporary memory 142.
The data processing unit 143 performs an encoding process according to the distribution of defects based on the stored data as described with reference to FIG. The display 144 displays the coded board ID.

The same result can be obtained by simultaneously irradiating these light sources from both sides. In this case, however, a device for separating scattered light generated from a defect on the sample surface or inside by an irradiation wavelength is used. It is necessary to collect light.

The method of detecting scattered light generated from a surface or internal defect by the photodetector 141 for each irradiation wavelength and evaluating the defect structure is described in "A NEW MEASUREMENT METHOD OF MICRO" related to the presentation by the inventors. DEFECTS NEAR
SURFACE OF Si WAFERS; OPTICAL SHALLOW DEFECT ANALY
ZER (OSDA) "Mat. Res. Soc. Symp. Proc. Vol. 442, 1
This is described in detail in 997. The photodetector 141 may be a two-dimensional detector, and may be, for example, any of a photodiode array, a phototransistor array, and a charge-coupled device (CCD).

In order to adjust the sensitivity of the detection signal from the substrate to be constant, it is also useful to add a sensitivity adjustment mark in the defect measurement area 12. That is, FIG.
(B) Correcting the signal intensity of the defect detection of the photodetector 141 based on the detection signal of the sensitivity adjustment mark when it is said whether or not the signal at the address (13, 1) is regarded as valid. Therefore, the reliability of the reproducibility of the signal strength measurement can be maintained. It is also useful to provide the defect measurement area 12 in a certain area in advance on the opposite side to the surface on which the circuit pattern is formed, similarly to the sensitivity adjustment mark. The marks 10 _{1 to} 10 ₄ in the defect measurement area 12 may also be used as sensitivity adjustment marks. Further, FIG.
The mark 10 ₁ -10 ₄ of the defect measurement area 12 shown in FIG.
Are not necessarily required at the four corners. The reason is that this has only a meaning of defining the defect measurement area 12, and if the area of the defect measurement area is determined in advance, the measurement can be performed if it is known from which direction the measurement should be performed. Therefore, as in the example illustrated the mark 10 _1, + if characters provided marks each which side of the line of the mark Oke promised whether the region 12, is not necessarily the other three marks.

The present embodiment can be applied to a semiconductor substrate made of SOI (silicon on insulator).
More specifically, it has a first semiconductor substrate and a second semiconductor substrate, oxidizes the first semiconductor substrate, and then heats the oxidized first and second substrates together. The present invention can also be applied to semiconductor substrates that have been manufactured. In this substrate, a circuit pattern is to be formed on the bonded surface of the substrate obtained by polishing the bonded second substrate. Therefore, if the outer surface of the first substrate is the surface on which the defect measurement region 12 is formed, Good.

The method of identification without artificially marking the substrate surface as in this embodiment has an effect on wafer management by a maker who creates a wafer. In general, wafer manufacturers do not want to mark products as much as possible before shipping. Therefore, it is effective to be able to manage only by setting a specific position of the wafer as the defect measurement area 12 without artificially adding a mark. In addition, when applied to a semiconductor of an SOI semiconductor substrate, propagation of free carriers due to light applied to the back surface is prevented by the SiO ₂ layer, and
There is an advantage that a malfunction of the circuit pattern based on the propagation of er can be prevented.

Embodiment 2 This embodiment is an example in which a crystal defect of an IC chip built in an IC card is used as an ID of the IC card.

As described above, the surface resolution or reading resolution is 10 μm, and the area identified as identification information is 10 m.
In the case of m × 10 mm and a depth of 1 μm, the number of states in which 10 defects are distributed is 10 ⁶⁰ . Actually, a chip used in an IC card generally put to practical use is 10 chips.
mm × 10 mm and thickness of about 50 μm. If the area used as identification information is 1 mm × 1 mm and the depth is 1 μm, the number of states in which 10 defects are distributed is 10 ⁴⁰ , which is a sufficient amount of information to be used for recognition. An embodiment using an IC card will be specifically described with reference to the drawings.

FIG. 3 is an exploded view of an example of a commonly used IC card. In this example, 31-
Four sheets indicated by reference numeral 34 are stacked, and are shown by being slightly shifted. Reference numeral 33 denotes a substrate sheet on which a wiring pattern 23 is formed and chips necessary for fulfilling the function of an IC card are arranged.
The wiring pattern 23 is a wiring between these chips.

In this embodiment, the authentication chip 22, the control chip 24 and the photodetector 8 are mounted. To mount these chips on a substrate, a known technique may be used. Numeral 32 denotes a spacer sheet, and a notch 3 corresponding to the size of each chip is provided at a position corresponding to each chip.
It has 5, 36 and 37. The thickness of the spacer sheet 32 substantially corresponds to the height of the chip. 31,
Reference numeral 34 denotes a cover sheet, which is arranged so as to sandwich the spacer sheet 32 and the substrate sheet 33 from both sides thereof. A generally used IC is provided on the outer surface of the cover sheet 31.
Similarly to the card, a terminal group 78 for making necessary electrical connection between the authentication chip mounted on the substrate sheet 33 and an external device.
Is provided. The connection between the terminal group 78 and, for example, the wiring pattern 23 or the authentication chip 22 and the control chip 24 will not be described because commonly used techniques can be applied as they are.

In the IC card of this embodiment, this terminal group 7
8, a transparent window 72 is provided at the center of a terminal 79 generally used as a ground terminal. Further, a transparent window 70 for inputting a fingerprint is provided. Although described later in detail, the transparent window 72
Is provided at a position corresponding to the center of the back surface of the authentication chip 22, through which a crystal defect of the authentication chip 22 is detected to evaluate the validity of the ID of the IC card. In addition, transparent window 7
0 is provided at a position corresponding to the photodetector 8, and the fingerprint image attached to the transparent window 70 by the user can be detected by the photodetector 8, and the validity of the IC card user can be evaluated.

FIG. 4 shows a partial cross-sectional view of the embodiment of the IC card shown in FIG. 3 at the part of the authentication chip 22, and FIG. 5 shows the embodiment of the IC card shown in FIG. FIG. 2 shows a partial cross-sectional view of a section of the container 8. In the figure, reference numerals 31 to 34 denote sheets described in FIG. The part shown without reference number between each sheet is an adhesive for bonding each sheet. Sheet 31-
As described later, the adhesive for bonding 32 is preferably substantially transparent because it is necessary to transmit light.

As is apparent from FIG. 4, the wiring pattern 23 is formed on the surface of the substrate sheet 33, and the circuit on the circuit pattern surface 131 of the authentication chip 22 is connected to the bump 1.
38 is connected to the wiring pattern 23. Since the back of the verifying chip 22 corresponds to the position of the transparent window 72, the light 4 from the outside through the transparent window 72 _1, 4 ₂ receiving scattered light from a very shallow layer of internally reflecting or back from the back 4 ₃
Can be taken out. The light 4 ₁ , 4 ₂ is the light of the light source 140 ₁ , 140 ₂ described in FIG. 2, and the reflected light 4 ₃ from the back surface or the scattered light 4 ₃ from the extremely shallow layer inside the back surface is used as the input light of the photodetector 141. Then, a crystal defect of the authentication chip 22 can be detected. The notch indicated by the reference numeral 73 on the back of the authentication chip 22 is shown in FIG.
Corresponds to the + -shaped marks 10 _{1 to} 10 ₄ indicating the range for determining the defect measurement area 12 described in
In this embodiment, this is a linear cut connecting the + -shaped marks 10 ₁ and 10 ₃ . Needless to say, a similar effect can be obtained by using a projection instead of the notch.

As is apparent from FIG. 5, the wiring pattern 23 is formed on the surface of the substrate sheet 33, and the surface 150 of the photodetector 8 on which the accumulated charge-coupled devices (CCD) are arranged. On the back side, the charge-coupled device (CC
The output circuit of D) forms the wiring pattern 23 by the bump 138.
It is connected to the. Here, the accumulated charge-coupled device (CC
The technique of leading out the output circuit of D) to the back side is similar to a commonly used technique of this type, and will not be described. Since the surface of the photodetector 8 on which the charge-coupled device (CCD) is arranged corresponds to the position of the transparent window 70, the transparent window 70
Through which light corresponding to the fingerprint pattern 91 attached to the outer surface of the transparent window 70 is received. If the output of the photodetector 8 is to be compared with a fingerprint pattern registered in advance by signal processing by the control chip 24 to determine a match or non-coincidence, an attempt is made to use an IC card and the transparent window 7 is used.
That is, it is possible to determine whether or not a person who has attached a fingerprint pattern to 0 is appropriate to use the IC card.

In this case, the fingerprint pattern of the valid user of the IC card must be registered in the IC card in advance. For example, the following may be performed. First, when supplying an IC card, a legitimate user is asked to attach a fingerprint pattern on the outer surface of the transparent member of the transparent window 70. In this state, this IC card is set in the data setting terminal of the IC card. By operating the data setting terminal, the image data of the fingerprint pattern read at this time may be held in a memory built in the authentication chip 22 that permits writing only once and thereafter permits reading only. Further, at the time of this operation, a crystal defect is taken in from the back surface of the authentication chip 22 and encoded as described in FIG.
What is necessary is just to register as ID of C card. In this case, since the ID of the IC card is used for authentication every time the card is used, a department that can be commonly accessed via a network, such as a certification authority described later with reference to FIG. Must be registered with However, since the data setting terminal of the IC card generally used at present does not have a function of irradiating the IC card with light, in order to actually operate the IC card of this embodiment, the setting terminal is required. Need to have an optical system provided in a card reader as described in the embodiments described later.

The reason why the image data of the fingerprint pattern is held in the memory built in the authentication chip 22 is that, in the present embodiment, the crystal defect of the authentication chip 22 is focused on, Since the validity is evaluated, the data which should not be falsified is the authentication chip 22.
Is more effective in improving security. Therefore, even if the image data of the fingerprint pattern is stored in the memory built in the control chip 24, there is no problem in realizing the function of authenticating the validity of the user in the IC card.

As in the present embodiment, the authentication chip 22
In addition to evaluating the validity of the IC card by paying attention to the crystal defects of the above, a conventional authentication method by assigning a personal identification number (authentication method by a soft scramble code using a secret key such as RAS) is also used. It is also useful.
This is because when the case where the IC card is used is not so important, if the authentication using only the password is sufficient, the IC card can be used easily.

In the embodiment shown in FIG. 4 and FIG.
2, 70 were formed by cutting out the cover sheet 31 and fitting a transparent member such as quartz glass. This, however, can be varied in many ways, for example,
A layer of a transparent member such as quartz glass may be provided between the cover sheet 31 and the spacer sheet 32, and a thin cover sheet 31 in which a portion corresponding to the transparent window is cut out may be provided thereon.

FIGS. 6A and 6B show the authentication chip 22.
FIG. 2 is a diagram showing a process flow of an example of a process for manufacturing a wafer and an outline of a change of a wafer in each process. At step 120, devices are fabricated on the wafer surface. After the device is completed, thinned by polishing S ₁ portion from the wafer back with mechanical polishing in step 121. Then, in step 122, the CM
Mirror polishing is performed while etching the S ₂ portion with P. Thereafter, in step 123, the entire back surface of the wafer is subjected to CVD.
The S3 portion is covered with the oxide film 200 by the process. Thereafter, in step 124, a notch 73 is formed on the outer surface of the oxide film 200 using FIB or a laser beam. The notch 73 is for specifying the crystal defect identification region 12 as described above. Thereafter, in step 125, the wafer is cut (scribed) into chips. Thus, the authentication chip 22 having the desired size and thickness h is completed.

Here, the thickness of the authentication chip 22 is set to a desired height h by processing a surface different from the surface on which the circuit pattern 131 is provided by mechanical polishing and then by chemical etching. Specifically, the thickness is reduced to about 50 μm, but it is possible to reduce the thickness to 30 μm. The standard thickness of the entire IC card is generally 0.85 mm, but the thickness of the authentication chip 22 and the cover sheets 31 and 34 and the spacer sheet 32
If the thickness is appropriately selected, even 0.15 mm is possible.

Even if the thickness of the entire IC card is set to such a small value, it is a size sufficient to use a crystal defect of a semiconductor as the ID of the authentication chip. Hereinafter, the authentication chip 22
A crystal defect of a semiconductor used as an ID of a semiconductor will be described.

In the first embodiment, crystal defects in the wafer itself
As described above, the ID of a wafer using this is described. As is well known, the number of crystal defects further increases when the processing of the wafer is added. As described with reference to FIG. 6A, the heat treatment is applied to the authentication chip 22 during the manufacturing process. Due to the heat treatment during the process, SiO ₂ particles are generated inside the silicon wafer. The density at which the SiO ₂ particles are generated may be determined, for example, by the heat treatment of the SiO ₂ nucleation process (800 ° C. for 4 hours) and the SiO ₂ formation process (1000 ° C. for 16 hours), which are commonly performed model annealing of a wafer. It is known that at least 10 ⁸ particles / cm ³ or more are generated inside. Although the depth distribution of this oxygen precipitate occurs to some extent deep from the surface, it is 50
Sufficient precipitate density can be obtained even within μm. For example,
When the volume density is 10 ⁸ pieces / cm ³ , the area density is 10 ² pieces / mm ² within 1 μm from the back surface. These defects are far more than the defects at the time of wafer generation.

That is, when detecting a crystal defect from the back surface of the authentication chip 22, a sufficient density of oxygen precipitate generated by the heat treatment during the process can be detected.
The ID has a sufficient information amount.

In this embodiment, the authentication chip 2
2, the control chip 24 has been described, but it is of course possible to integrate them into one,
It is needless to say that a processor function for performing image processing on the output of the stored charge-coupled device (CCD) may be provided.

FIG. 7 shows a lower surface side of a transparent member in which the mark 73 is fitted in the transparent window 72 instead of forming a mark 73 for specifying the defect measurement area 12 on the back surface of the authentication chip 22. The example provided on the side facing the back surface of the authentication chip 22 is shown. With this configuration, step 124 described in FIG. 6A can be omitted.

Embodiment 3 FIG. 8 is a schematic perspective view illustrating a card reader of the IC card described in FIG. Reference numeral 6 denotes a card reader, and reference numeral 2 denotes an IC card. I
A terminal group 78, a transparent window 72, and a transparent window 70 can be seen on the surface of the C card. As described above, the transparent window 72 is for detecting a crystal defect of the authentication chip 22 and evaluating the validity of the ID of the IC card. The transparent window 70 is for detecting the fingerprint image and evaluating the validity of the card holder. Light sources 62 ₁ and 62 ₂ and a light detector 6 are provided on the inner surface of the card reader 6 facing the front side of the IC card 2, that is, the surface having the transparent window 72 and the transparent window 70.
3 is attached. Further, a light source 10
And a slit 67 are attached. Here, 61 is a concave groove, which is a guide for taking the IC card 2 into the card reader. 66 is a roller for feeding the IC card 2 into the concave groove 61 of the card reader 6. Reference numeral 65 denotes a stop sensor which detects that the IC card 2 has reached a position to be stopped. Reference numeral 74 denotes a switch for confirmation, which is turned on after the IC card 2 is set in the concave groove 61 of the card reader 6. A data processing unit 64 corresponds to the temporary memory 142 and the data processing unit 143 described with reference to FIG. 2 and processes the output of the photodetector 63. Reference numerals 75 and 76 denote displays for displaying the output of the data processing unit 64. Reference numeral 77 denotes a connector, which is used to connect the card reader 6 to a controller and transmit the output to a center via a network, as described later. Here, although not shown, the card reader 6 comes into contact with the two terminal groups 78 of the IC card to read data and the like held in the memory in the card, and, if necessary, from the controller. A contact for sending data into the card is provided.

The reader 6 of the IC card is used as follows. When the IC card 2 is set in the concave groove 61 of the card reader 6 and the confirmation switch 74 is turned on, the rollers 66 provided on both sides of the concave groove 61 are rotated by a drive motor (not shown). Then, the IC card 2 is taken into the concave groove. When the IC card 2 moves to a predetermined position, the stop sensor 65 operates and the IC
Card 2 stops. From the time the IC card 2 is inserted into the concave groove 61 to the time it stops, the front side of the IC card 2
That is, the transparent windows 72 have the light sources 62 ₁ , 62 ₂ and 10
Irradiates light. Light from the light sources 62 ₁ and 62 ₂ illuminates the back surface of the authentication chip 22 through the transparent window 72 as described in FIG. 2 and as described in FIG. 4 or FIG.
While the C card is moved, and the position of the light source 62 _1, 62 ₂ are determined to allow irradiation of the required surface. Similarly,
Light from the light source 10 is applied to the transparent window 70 through the slit 67. The positions of the light source 10 and the slit 67 are determined so that a required surface can be irradiated while the IC card moves.

During the period from when the IC card 2 is inserted into the concave groove 61 to when the IC card 2 stops, the transparent windows 72 and 70 are irradiated with light from respective light sources. Therefore, the photodetector 63 can detect the information of the crystal defect identification area 12 specified by the mark 73 on the back surface of the authentication chip 22, and the output thereof is processed by the data processing unit 64 so that this IC card Can be output to determine whether or not is issued properly. Further, in the inside of the IC card 2, for example, by the control chip 24, the light radiated through the transparent window 70 determines whether or not the fingerprint image attached to the transparent window 70 matches the fingerprint image registered at the time of issuing the IC card. Can be determined. The processing result of the data processing unit 64 and the result of collation of the fingerprint image output from the terminal group 78 through the contact are displayed on the display units 75 and 76, if necessary. Of course, the information may be displayed on a controller connected via the connector 77.

When the authentication of the IC card 2 and the necessary input processing by the controller are completed, for example, the switch 74 for confirmation is turned off, the roller 66 is reversed, and the IC card is ejected from the concave groove 61 of the card reader 6. .

As is clear from the above description, in the present embodiment, the fingerprint image is only held in the IC card when the IC card is issued. Therefore, every time the IC card is used, the fingerprint image can be prevented from flowing as a data to an organization or an organization such as a bank or a company via a network, thereby leaking personal privacy information and the danger of unauthorized use. Can be reduced.

Embodiment 4 FIG. 9 is a diagram showing an overall system configuration for explaining an example of a method of using an IC card and a card reader according to this embodiment. FIG. 10 is a diagram showing signal processing in the card reader. It is a flowchart which shows an example. Reference numeral 800 denotes a card reader controller, which is used to supply necessary information to the card reader 6 and the system, and to extract a signal read by the card reader 6. Reference numeral 801 denotes an input unit, and 802 denotes a transmission / reception unit. The card reader 6 includes the light sources 62 ₁ , 6 described above.
A reading unit 811 mainly including 2 ₂ , 10 or the photodetector 63 and a display unit 8 mainly including the display units 75 and 76.
12. Comparison processing unit 813 mainly including data processing unit 64
And an output unit 814.

Reference numeral 820 denotes a network, and the controller 800 is connected to the network 820.

Reference numeral 830 denotes a certificate authority,
0 to perform a function of authenticating whether or not the code in the defect measurement area 12 detected and transmitted from the IC card matches the ID code registered when the IC card was issued. It is. 831 is a transmission / reception unit, and 832 is an ID collation unit.

Reference numeral 840 denotes a bank B, and the network 82
0, and performs processing in accordance with the deposit / withdrawal information transmitted from the controller 800 and the information transmitted from the certificate authority. 841 is a transmitting / receiving unit, 842 is a processing unit, and 844 and 845 are customer account data memories.

Now, a customer K purchases a product C at a store A,
It is assumed that payment is made by debiting the bank B personal account using an IC card. In store A, customer K
Then, after having the transparent window 70 have a fingerprint, the IC card is received. This is applied to the IC card reader 6 described in FIG. As a result, in the card reader 6, detection of a fingerprint image attached to the transparent window 70, I
The ID code of the verification chip 22 is detected by the defect measurement area 12 read from the back surface of the C card, and the fingerprint is compared, that is, the detected fingerprint image is compared with a previously registered fingerprint image. By this fingerprint collation, the suitability of the card holder is determined by the comparison processing unit 81.
3 is displayed on the display unit 812. At this time, if the fingerprint collation becomes “match”, that is, if the card holder is appropriate, the ID code of the verification chip 22 is also output to the output unit 81.
4 to the controller 800. If the fingerprint collation is "mismatch", that is, if the card holder is incorrect, the subsequent processing is prohibited. FIG. 10 shows the flow of this signal processing. If the cardholder has a proper display, the store A inputs the price of the product C via the input unit 801. In response to this, the price of the product C, the read ID code, and the data such as the settlement account registered at the time of issuing the card are transmitted to the network 820 via the transmission unit 802. The certificate authority 830 determines that the read I
Upon receiving the D code, it determines whether the ID code was registered at the time of issuing the card, and sends the result to the network 820. When the bank B receives the settlement account data, it waits for the authentication result from the certificate authority 830, and if it is determined to be valid, subtracts the price from the principal account data 843 and stores it in the A store account data 844. Add the price. When the settlement process by Bank B is completed,
This result is transmitted to the controller 800 of the store A, and the store A ejects the card from the card reader and returns the card to the customer K. Here, if the A store account does not exist in the bank B, it goes without saying that the deposit processing is performed in another bank having the A store account connected to the network. That is, according to the present embodiment, the validity of the card holder is evaluated at the stage of fingerprint collation, and the validity of the IC card is evaluated by the ID of the authentication chip 22. A check will be applied. Moreover, as described above, if the fingerprint image data is stored in the authentication chip 22, even if the authentication chip is exchanged improperly, it is impossible to pass this double check. It can be said.

On the other hand, when this system is operated only in a specific bank affiliate, this certificate authority
The certificate authority 830 only needs to exist as a part of the bank since it exists only as a part of the function of the bank.

As described above, the validity of the IC card is determined in addition to the ID code of the verification chip 22 by the defect measurement area 12 read from the back of the IC card.
If the sales amount is smaller than a predetermined price in this embodiment so that it can be evaluated using a password generally used at present, only the process using the password may be performed. When doing this,
Since the burden on the network can be reduced, there is an advantage that the responsiveness can be made sufficiently large even when the system becomes large. Embodiment 5 In the above embodiment, a circuit pattern is formed. A defect measurement area 12 is set on the back surface of the substrate, and a code obtained from a crystal defect in this area is used as a card ID. Here, an embodiment will be described in which a defective bit pattern of a memory caused by a crystal defect is used for recognition instead of the crystal defect itself. When a memory device is manufactured, a defective bit is generated. However, by manufacturing the memory device on a CZ wafer having a controlled defect density, the number of defective bits per chip can be suppressed to a substantially constant range. CZ crystal defects are known to adversely affect device electrical characteristics and cause defects (eg, “The Analysis of the Defecti”).
ve Cells Induced by copin a 0.3-micron-tecnology N
ode DRAM "by M.MURANAKA et. al, JJAP37 (1998) 1240-1
243). It is possible to detect a defect based on the defective electrical characteristics of the memory device. In other words, a defective bit map of a memory device due to a defective electrical characteristic of the memory device is representative of a crystal defect (defect due to damage caused by a process), and the same bit map cannot be produced artificially. Similarly to the embodiment described above, the present invention can be used for a code for recognizing a semiconductor chip. Since there are various types of memory devices and different defective bit densities, memory devices (DRAM, SRAM, flash memory, FRAM, magnetic memory, or the like) having different integration levels may be used depending on the security level of authentication. However, part of the address space information may be used.

In this embodiment, since it is necessary to measure the defective bit pattern of the memory device in the IC card, it is not necessary to have an optical system as described with reference to FIG. It is necessary to read a code corresponding to the defective bit pattern and register it with a certificate authority. As a defective bit pattern, a method of digitally measuring a bit unit characteristic with 0 or 1 of good or bad and recording it together with an address, such as an IC tester, and an analog electric characteristic of an element corresponding to one address are used. A method of recording with multiple values is possible.

In this case, the codes of the generated IC card are (1) the case where the address of the defective bit is coded, and (2) the address of the defective bit and the degree of the defect of the bit (in the case of multi-level recording). However, the latter code is safer for copy protection. This is briefly described as follows. Considering the possibility of duplication of a chip whose defective bit address is known, a memory is manufactured by manufacturing a semiconductor mask so that the bit of the defective bit address becomes defective. However, even if a device is manufactured based on this semiconductor mask, it is impossible to reproduce the characteristic of the defective bit, although the array of defective bits cannot be reproduced.

Hereinafter, a method of using an electrical defect random pattern will be described more specifically with reference to FIGS. 11A, 11B and 11C. ), FIG. 11 (b) and FIG. 11 (c)
FIG. 3 is a schematic diagram illustrating a memory element group in which a defective bit has occurred, such as a RAM, an SRAM, and a flash memory.
In the figure, a two-dimensional area divided into cells is an area to be used as a memory, and each cell is one area of the memory.
Means a bit. At the stage when this memory is manufactured, defective bits are dispersed and generated throughout the area. That is, as shown in the squares or hatched squares, the area ID of the left one-third of the memory area shown in FIG.
Defective bits occur with a statistical probability in the entire area within the area ID. On the other hand, a 2/3 area M-ARY indicated by a dashed-dotted line on the right side of the memory area shown in FIG. Make it not exist. That is, in order to perform the function as the original memory, the normal memory bits existing outside this memory area are replaced with the defective bits in this memory area M-ARY by software processing, and the apparent memory bits are used. That is, there is no upper defective bit.

Here, FIG. 11A shows the distribution of defective bits in the memory for binary data, and means that the bits indicated by black in the area ID are bits that do not store normally. FIG. 11B shows that the memory data itself is 2
An example is shown in which the threshold value for determining the defective bit in the data memory is two levels. Here also, the bits shown in black in the area ID are bits that do not store normally, while the bits that are shown with hatching are bits that do not store normally, but have a certain size. Indicates the stored output. As apparent from a comparison between FIG. 11A and FIG. 11B, FIG. 11B includes a defective bit having a different character from the defective bit in FIG. 11A. By discriminating between them, more rigorous verification can be performed. FIG.
FIG. 11C is a schematic diagram showing the same memory as that of FIG. 11A, except that the threshold value for determining a defective bit is different. As is clear from the comparison between the two, even in the same memory, the defective bit patterns are different.

As described above, a defective pattern obtained corresponding to a certain electric characteristic (leak current value, threshold voltage, etc.) for an address can be used as authentication. The address space used for authentication does not need to use the entire memory, but may be a part. It is also desirable to change the address space by security. For authentication, it is desirable that a memory for recording important data and an authentication pattern that cannot be duplicated be on the same chip. From this point of view, a non-volatile memory is convenient as a memory, and it is preferable to use an area having a memory address for authentication and another area for reading and writing data.

As described above, in the case where a defect generated in the process of manufacturing a memory is used for authentication, more reliable authentication can be achieved by giving consideration to unauthorized tampering.

If there is a case where an IC card is illegally created by artificially creating a defective bit, it is conceivable to create a memory using a mask such that an element corresponding to the address of the defective bit becomes defective. . That is, it is conceivable that the element at the address is formed incompletely or not at all. However, in this method, even if a defective bit can be formed, it is practically impossible to reproduce its electrical characteristics. Therefore, FIG.
It becomes more difficult to reproduce the pattern as shown in FIG. As another method, instead of using a mask as described above, a method of replacing a defective bit address portion to be created with another bit can be considered.
This technique is still used as a bit rescue method, and an operation of actually replacing a defective bit address with a non-defective bit address is performed by software.
If this replacement is permitted, various defective bit patterns will be generated by software, so it is effective to substantially prohibit this replacement. One way to do this is to add a time constraint to calling the bad bit pattern. As a result, there is no time margin for software processing required for bit replacement, and illegal use can be prevented.

On the other hand, from the aspect of detecting the electrical characteristics, the fraud prevention is performed by converting the magnitude of the leak current value under a constant applied intensity of the gate insulating film of the transistor due to a defect in the oxide film into an A / D conversion for each address. A method of obtaining a pattern for each address of the value is performed. In a flash memory, a write bias voltage is used in some combination, a voltage (Vt) of a memory cell after a predetermined time after writing is AD-converted, and a pattern of the value for each address is obtained. Is possible. Also,
As in the examples of FIGS. 11A and 11C, in the flash memory, the write voltage is changed to change the binary defective bit pattern of the memory cell (from the pattern of FIG. 11A to FIG. 11C). The change to the pattern in ()) can be used for the authentication pattern. In this method, specifically, a defective bit pattern may be obtained by changing a bias. It is desirable to specify what bias is to be applied by the certificate authority. In this case, it is needless to say that the bias condition and the defective bit pattern must be registered in the certificate authority. That is, when issuing an IC card, these conditions are specified by the certificate authority,
The bad bitmap is registered. In a state where the IC card is used, the certificate authority checks the registered pattern against the pattern sent from the reader in the same manner as when using the crystal defect itself for authentication.

Other Embodiments In the above embodiment, the defect measurement area 12 is set. However, the circuit pattern itself can be used as data for setting the defect measurement area 12.
That is, from the surface of the substrate on which the circuit pattern is formed,
A code obtained from a crystal defect in a specific area based on a circuit pattern is used as a card ID.
As described above, in an epiwafer having an epitaxial layer provided on a wafer prepared by the CZ method, the heat-treated C
There is no crystal defect in the epitaxial film as much as the Z wafer. The density is of the order of 10 ⁴ per 1 cm ³ , and when detecting from the back surface of the wafer, there is no problem because high-density defects similar to those of a CZ wafer can be used. C
In the case of a Z wafer, since the γ defect density is high in the surface region,
Since the defect measurement area 12 can be set on the surface,
The number of steps for producing the mark 73 can be omitted.

To detect a defect measurement from the surface of the substrate,
Instead of using the circuit pattern itself as data for setting the defect measurement region 12, the defect measurement region 12 may be set in a region on the surface of the substrate different from the region where the circuit pattern is formed. . In this case,
It is needless to say that it is necessary to add a mark indicating the area 12 as in the case of using the back surface.

In the above embodiment, in addition to the card ID, the IC card is provided with the transparent window 70 so that the card holder can be authenticated. However, in order to authenticate the IC card itself,
It may not have this, in which case the IC
The card can be simplified, and the card reader 6 can be simplified accordingly.

Here, the authentication chip 2 is stored in the IC card.
A description will be given of some implementation formats in the case where 2 is implemented. Note that this mounting format can be similarly applied to the control chip 24 or the photodetector 8.

FIGS. 12A to 13B are diagrams illustrating an example of a method of mounting an authentication chip on an IC card.

FIG. 12A is a cross-sectional view of the case where the flip chip 100 for functioning as the authentication chip 22 is mounted on the wiring pattern 23 provided on the substrate sheet 33 described with reference to FIG. The wiring pattern 131 for functioning as the authentication chip 22 is connected to the wiring pattern 23 by a bump 138. In this example, a transparent film 136 for protection is added to and mounted on the defect recognition surface on the back surface of the surface on which the wiring pattern 131 is provided. As described above with reference to FIGS. 6A and 6B, it is useful to protect the defect recognition surface with a SiO ₂ film.
The attachment surface of the foreign matter attached to the O ₂ film surface can be kept away from the internal defect detection area. Therefore, it is possible to reduce noise for defect recognition due to adhesion of foreign matter.

FIG. 12B is a cross-sectional view showing a mounting example in which the wiring pattern 131 itself for functioning as the authentication chip 22 is used as data for setting the defect measurement area 12. It is configured as a chip carrier, and after adding a wiring 23 ′ for connecting to the wiring pattern 23 to the wiring pattern 131, the wiring 23 ′ is connected to the wiring pattern 23.

FIG. 13A is a cross-sectional view showing another example of a mounting example in which the wiring pattern 131 itself for functioning as the authentication chip 22 is used as data for setting the defect measurement area 12. is there. In this example, the authentication chip 22 is embedded in the notch 33 ′ of the substrate 33. As in the case of FIG. 12B, a chip carrier is configured, and the wiring pattern 131 on the surface of the authentication chip 22 is formed.
Is added to the wiring pattern 23, and then the wiring 23 'is connected to the wiring pattern 23.

FIG. 13B is a cross-sectional view showing an example in which the authentication chip 22 is mounted on the board 33 in the form of a chip-on-board. Reference numeral 201 denotes a molding material, which holds the authentication chip 22 on the substrate 33. A mark 73 is provided on the surface on which the wiring pattern 131 for functioning as the authentication chip 22 is provided as a defect recognition surface. The authentication chip 22 is mounted on the board 139 such that the defect recognition surface can be seen from the cutout 145 of the board 139. Also, the board 139 is connected to the substrate 3 via bumps 138 '.
A wiring pattern 23 ′ for connecting to the wiring 23 on 3 is provided. Wiring pattern 131 of authentication chip 22
And the wiring pattern 23 'on the board 139
38 and the wiring 200. In this example, as in FIG. 13A, the authentication chip 22 is embedded in the notch 33 ′ of the substrate 33. This embodiment has an advantage that the spacer sheet 32 can be made thin.

Although all of the above-described embodiments are based on the crystal defects of the semiconductor substrate, the present invention is not limited to the semiconductor substrate, and all the materials using the materials having the crystal defects as a result are used. Applicable to things. For example, there is a case where an amorphous layer is formed on a substrate and a circuit pattern is formed therein.

As described above, by using a crystal defect which is extremely difficult to manufacture artificially (a non-crystal as well as a non-crystal can be handled in the same way) as a recognition pattern, an illegal operation due to code decoding and duplication is performed. Use can be prevented.

Although all of the above embodiments have been described with reference to a contact type IC card as an example, the present invention relates to a non-contact type IC card.
Applicable to cards. That is, if a card reader having only an optical system for detecting a fingerprint image or detecting an ID code of an authentication chip is provided, a signal that can be electrically handled is a non-contact type signal which is currently proposed. I
It can be processed in the same way as a C card.

[0075]

According to the present invention, the reliability of the card system can be remarkably improved.

[Brief description of the drawings]

FIG. 1A is a diagram illustrating an embodiment in which the present invention is applied to the identification of a semiconductor substrate, and FIG. 1B is an example of a crystal defect appearing in a defect measurement region set in the semiconductor substrate. The figure which shows typically.

FIG. 2 is a block diagram showing an example of an apparatus for detecting a crystal defect.

FIG. 3 is a diagram schematically illustrating the structure of a card according to an embodiment in which the present invention is applied to identification of an IC card.

FIG. 4 is a partial sectional view of the embodiment of the IC card shown in FIG.

FIG. 5 shows an embodiment of the IC card shown in FIG.
FIG.

FIGS. 6A and 6B are views showing an example of a process flow for manufacturing an authentication chip 22 and an outline of a change of a wafer in each process.

FIG. 7 is a diagram showing an example in which a position where a mark of a defect measurement area is provided is changed.

FIG. 8 is a schematic perspective view illustrating a reader of the IC card described in FIG. 3;

FIG. 9 is a view showing an entire system configuration for explaining an example of a method of using the IC card and the card reader according to the embodiment.

FIG. 10 is an exemplary flowchart illustrating an example of signal processing in the card reader.

11 (a), (b) and (c) show DRAM, S
FIG. 3 is a schematic diagram illustrating a memory element group in which a defective bit has occurred, such as a RAM and a flash memory.

12A and 12B are diagrams illustrating an example of a method of mounting an authentication chip on an IC card.

13A and 13B are diagrams illustrating an example of a method of mounting an authentication chip on an IC card.

[Description of Reference Numerals] 1: wafer, 4 _1, 4 ₂ and 4 _3: Light, 6: Card reader: 8 photodetector, 10: light source, 10 ₁ -10 _4: +
Character mark, 11: notch, 12: defect measurement area, 2
2: Authentication chip, 23: Wiring pattern, 24: Control chip, 31: Cover sheet ,: Sheet, 32: Spacer sheet, 33: Board sheet, 34: Cover sheet, 35,
36 and 37: notch, 61: concave groove, 62 ₁ , 6
2 _2: light source, 63: optical detector, 64: data processing section, 6
5: stop sensor, 66: roller, 67: slit, 7
0: transparent window, 72: transparent window, 73: notch, 74:
Switch to confirm, 77: connector, 78: terminal group, 7
9: terminal, 91: fingerprint pattern, 131: circuit pattern surface, 138: bump, 140: sample stage, 140 ₁ , 14
0 _2: light source, 141: optical detector, 142: temporary memory,
143: data processing unit, 150: surface, 200: oxide film,
800: card reader controller, 801: input unit,
802: transmission / reception unit, 811: reading unit, 812: display unit, 813: comparison processing unit, 814: output unit, 820: network, 830: certificate authority, 831: transmission / reception unit, 83
2: ID collation unit, 840: Bank B, 841: Transmission / reception unit,
842: processing unit, 844 and 845: customer account data memory, ID: recognition area, M-ARY: effective memory area.

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (reference) G06K 17/00 G06K 17/00 T 19/00 H01L 21/02 A H01L 21/02 B 21/66 Z 21/66 G06K 19/00 Q

Claims (24)

[Claims] 1. A method for identifying an article, comprising: attaching a specific symbol to a surface of the article to designate a specific area; detecting a crystal defect existing in the specific area; A method for identifying an article using information corresponding to the identified crystal defect as identification information of the article. 2. An article having a specific symbol on its surface for designating a specific area, wherein information corresponding to a crystal defect existing in the specific area is used as identification information of the article. 3. The semiconductor substrate according to claim 2, wherein said article is a semiconductor substrate, and said specific symbol is attached to one surface of said substrate. 4. The article is a semiconductor chip, a circuit pattern for performing a predetermined function is formed on a front surface of the semiconductor chip, and the specific symbol is attached to a back surface of the semiconductor chip. The semiconductor chip according to claim 2. 5. The semiconductor substrate according to claim 4, wherein said specific symbol is a notch or a projection provided on a back surface of said semiconductor chip. 6. The semiconductor chip according to claim 4, wherein said semiconductor chip is formed of an SOI (silicon-on-insulator) semiconductor.
The semiconductor chip as described in the above. 7. The article is a semiconductor chip, a circuit pattern for performing a predetermined function is formed on a surface of the semiconductor chip, and the specific pattern is formed in a region different from a region where the circuit pattern is formed. 3. The semiconductor chip according to claim 2, wherein a symbol is attached. 8. The article is a semiconductor chip, and a circuit pattern for performing a predetermined function is formed on a surface of the semiconductor chip, and a specific area of the circuit pattern is specified by the specific symbol. The semiconductor chip according to claim 2, wherein the semiconductor chip is treated as: 9. An IC card containing a semiconductor chip therein, wherein light is guided from outside to a specific area of the semiconductor chip, and reflected light and scattered light from a light-reachable surface of the semiconductor chip are transmitted therefrom. IC with a transparent window that can lead outside
card. 10. A circuit pattern for performing a predetermined function as an IC card is formed on a front surface of the semiconductor chip, and a specific symbol indicating the specific region is attached to a back surface of the semiconductor chip. The IC card according to claim 9, further comprising: guiding the light to an area specified by the specific symbol. 11. The IC card according to claim 9, wherein said specific symbol is a notch or a projection provided on a back surface of said semiconductor chip. 12. The semiconductor chip according to claim 1, wherein said semiconductor chip is SOI (silicon.
10. The IC card according to claim 9, wherein the IC card is formed of an (insulator) semiconductor. 13. The semiconductor chip according to claim 1, wherein a circuit pattern for performing a predetermined function is formed on a surface of the semiconductor chip, and the specific symbol is attached to a region different from a region where the circuit pattern is formed. The IC card according to claim 9. 14. A circuit pattern for performing a predetermined function is formed on a surface of the semiconductor chip, and a specific area of the circuit pattern is treated as an area specified by the specific symbol. 9. The IC card according to 9. 15. An IC card in which first and second two semiconductor chips are housed, wherein light is guided from outside to a specific portion of the first semiconductor chip. A first transparent window that can guide reflected light and scattered light from a surface that can reach the outside, and a second transparent window that can guide a predetermined image from outside to a specific portion of the second semiconductor chip. An IC card having a window and wherein the second semiconductor chip is capable of converting the image into digital data. 16. A means for storing an image converted into digital data by the second semiconductor chip, and a means for comparing an image separately converted into digital data from an image separately input to the second semiconductor chip. The IC card according to claim 15, comprising: 17. A card reader comprising: a semiconductor chip housed therein; and guides light from the outside to a specific portion of the semiconductor chip, and reflects reflected light and scattered light from a surface of the semiconductor chip which light reaches. A groove for introducing an IC card having a transparent window that can be guided to the outside; a light source attached to a wall of the groove facing the IC card; and a movement for moving the IC card relative to the light source. Means for irradiating the transparent window of the IC card with light from the light source to receive scattered light of the semiconductor chip; and generating a digital code corresponding to the IC card from an optical signal received by the light receiving means. Means to do. 18. The IC card includes a second semiconductor chip and a second transparent window through which a predetermined image can be guided to a specific portion of the semiconductor chip from outside.
The second semiconductor chip is capable of converting the image into digital data, and further includes means for storing the image converted into digital data by the second semiconductor chip; Means for comparing an image which has been separately converted into digital data from an input image, wherein a second light source for irradiating the second transparent window with light is attached to a wall of the groove. Item 18. The card reader according to Item 17. 19. The IC card has, on a surface thereof, a group of terminals connected to a semiconductor chip in the IC card, and the group of terminals is formed in the groove of the IC card when the IC card is introduced into the groove. 18. The card reader according to claim 17, wherein the card reader is electrically connected to a group of contacts mounted on a wall facing the device. 20. The IC card has, on a surface thereof, a terminal group connected to a semiconductor chip in the IC card, and the terminal group is provided in the groove of the IC card when the IC card is introduced into the groove. 20. The card reader according to claim 19, wherein the card reader is electrically connected to a group of contacts attached to a wall facing the first semiconductor chip and electrically outputs a comparison result of images separately input to the second semiconductor chip. 21. An IC card authentication method executed by the following procedure: transmitting a code signal corresponding to a crystal defect detected from a semiconductor chip stored in the IC card; Receiving and storing the code signal by an organization; upon receiving the code signal, the organization determines identity by comparing with the previously stored code signal and transmits the result; 22. The IC card authentication method according to claim 21, wherein a defective pattern of a memory is used in place of the code signal corresponding to the crystal defect. 23. A transaction method using an IC card executed according to the following procedure: a transaction financial institution and a transaction amount together with a first code signal corresponding to a crystal defect detected from a semiconductor chip stored in the IC card. Transmitting a second code signal signal corresponding to the following; receiving and storing the first code signal by a predetermined organization when an IC card is issued; Determining the identity by comparing the first code signal with the first code signal and transmitting the result, and the transaction financial institution receives the determination result of the first code signal and determines the second code signal only when it is determined to be appropriate. Accepting the second code signal. 24. The IC card transaction method according to claim 1, wherein when the transaction amount is lower than a predetermined value, instead of authentication by the predetermined organization, an artificial code set at the time of issuing the IC card is used. 24. The transaction method using an IC card according to claim 23, wherein the transaction method is executable.