JP2008199414A - Photoelectric conversion device, image sensor and imaging system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a photoelectric conversion device capable of detecting a modulation amount of probe light modulated by a THz electromagnetic wave without suppressing a phase bias of the probe light and without saturating a circuit. <P>SOLUTION: The photoelectric conversion device is provided with a photodiode 1, a phtodiode 1' in which cathode terminal is connected to that of the photodiode 1, a capacitor 3 for charge storage inserted between the photodiode 1 and a photodiode 1', a capacitor 3' for charge storage inserted between the capacitor 3 for charge storage and the photodiode 1', and a differential amplifier circuit 6, wherein one input terminal is connected between the photodiode 1 and the capacitor 3 for charge storage, and the other input terminal is connected between the photodiode 1' and the capacitor 3' for charge storage. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a photoelectric conversion device, an image sensor, and an imaging system.

  In recent years, the demand for THz band electromagnetic wave imaging apparatuses is increasing for the purpose of security inspection applications, medical inspection applications, food analysis applications, and environmental monitoring applications, and active development is being promoted (for example, Non-Patent Documents 1 and 2). And Patent Documents 1-3).

  In these techniques, a THz electromagnetic wave in a frequency range of 0.1 THz to 100 THz emitted from a THz electromagnetic wave source is irradiated to the inspection object, and the spatial distribution information of the characteristics of the inspection object is transmitted or reflected to the inspection object. As a modulation amount, and receiving this, the spatial distribution information of the characteristics of the object to be inspected is configured as a two-dimensional image.

  Initially, as described in Non-Patent Document 2, as a method for obtaining two-dimensional information of an inspection object, an irradiated THz electromagnetic wave beam is focused on a part of the inspection object by a lens, and the inspection object is obtained. , The modulated THz electromagnetic wave is sequentially received by a receiver capable of receiving only one-dimensional information at most, and two-dimensional information is formed.

  However, this method requires a long time of several hours in order to collect all data of two-dimensional information, and is impractical as an inspection apparatus that is required to end the inspection in real time.

In order to make up for this drawback, Non-Patent Document 3 discloses an apparatus shown in FIG.
In FIG. 19, ultrashort pulse light having a pulse width of 100 fs is generated from a light source 1701 at a frequency of 1 kHz.

  The pump light 1703 passes through an optical delay line 1705 and is incident on a THz electromagnetic wave emitter 1706 formed by a photoconductive switch having an electrode pair formed on a semi-insulating GaAs wafer with a distance of 10 mm, and a THz electromagnetic wave 1707 is generated. Is done. The THz electromagnetic wave 1707 generated in this way is a highly collimated and wide beam, and is irradiated to a measurement object 1708 having a two-dimensional transmission distribution in a plane perpendicular to the traveling direction of the THz electromagnetic wave 1707. .

  The THz electromagnetic wave 1707 that has passed through the measurement object 1708 becomes a spatially intensity-modulated beam in accordance with the two-dimensional transmission characteristics of the measurement object. This is imaged by a polyethylene lens 1709 in an electro-optic modulator 1713 made of a ZnTe crystal at the subsequent stage.

  The probe light 1704 is rerouted by a mirror 1710, further expanded in beam width by a beam expander 1711, then incident on a silicon mirror 1712 made of a silicon wafer and transmitted through the silicon mirror 1712. 1707 and the optical axis are shared. In other words, they are superimposed.

  The superimposed probe light 1704 and THz electromagnetic wave 1707 are incident on the electro-optic modulator 1713.

  The electro-optic modulator 1713 is made of a ZnTe crystal having a [110] plane arranged perpendicular to the optical axis.

  The electro-optic modulator 1713 is followed by a phase plate 1714, a polarizing plate 1715 that transmits only linearly polarized light having a polarization plane orthogonal to the probe light 1704, and one photo for each pixel that receives the transmitted light from the polarizing plate 1715. A two-dimensional CMOS image sensor 1716 having a diode is arranged in this order.

  The phase plate 1714 has a plane of polarization of the probe light 1704 after passing through the phase plate 1714 under a predetermined reference condition so that the transmitted light of the polarizing plate 1715 has an intensity that does not saturate the CMOS image sensor 1716. It is set so as to form a declination of about 2 to 3 ° from the direction orthogonal to the transmission polarization plane of the polarizing plate 1715.

  This reference condition represents, for example, the case where the THz electromagnetic wave 1707 does not enter the electro-optic modulator 1713 simultaneously with each pulse of the probe light 1704, that is, the case where the THz electromagnetic wave pulse and the probe pulse are asynchronous. Under this reference condition, the probe light 1704 passes through the polarizing plate 1715 with an intensity that is suppressed according to the declination and enters the image sensor 1716.

  In this way, using probe light with linear polarization and controlling the plane of polarization of the probe light to suppress the amount of light transmitted through the polarizing plate is hereinafter referred to as suppressing the phase bias of the probe light.

  At this time, when the THz electromagnetic wave pulse and the probe light pulse enter the electro-optic modulator 1713 at the same time, that is, when both pulses are synchronized, the electro-optic modulator 1713 is compared with the asynchronous case. The deflection angle given to the probe light 1704 varies about 0.02 °.

  In response to the modulation of about 1% with respect to the angle of the polarization plane, the intensity of the probe light 1704 is modulated by the polarizing plate 1715.

  The synchronization circuit 1717 synchronizes with the pulse generation timing of the light source 1701, for example, by permitting or suppressing the generation of THz electromagnetic waves by the THz electromagnetic wave emitter 1706, so that information modulated by the THz electromagnetic waves respectively into two consecutive probe light pulses. And unmodulated information.

The image processing circuit 1718 captures two images obtained by the image sensor 1716 from these two probe light pulses, and obtains a difference between the two images, thereby obtaining an image corresponding to the transmission characteristics of the object to be measured 1708 with respect to the THz electromagnetic wave. Is generated.
Japanese Patent Laid-Open No. 2002-5828 JP 2004-20504 A JP 2005-37213 A Kiomi Sakai ed. "Terahertz Optoelectronics", Springer Verlag, 2005. B. B. Hu and M.H. C. Nuss, Opt. Lett. Vol. 20, p. 1716, (1995). F. Miyamaru, T .; Yonera, M.M. Tani and M.M. Hangyo, Japan Journal of Applied Physics, Vol. 43, p. L489-L491, (2004).

However, in the above-described conventional technology, the following two problems arise.
The first problem is that it is difficult to improve the signal-to-noise ratio (S / N ratio) of the obtained image in order to suppress the phase bias of the probe light in order to avoid saturation of the conventional CMOS image sensor. is there.

  FIG. 20 is a diagram for explaining the characteristics of the polarization filter, and conceptually shows the characteristics of the transmission intensity of the probe light with respect to the angle with respect to the polarization plane of the probe light. By suppressing the phase bias of the probe light, the characteristic of the portion A in FIG. 20 is used. For this reason, a large amount of modulation cannot be obtained, and it becomes difficult to improve the S / N ratio.

  When a high-power short-pulse laser with a reproduction amplifier is used as a THz electromagnetic wave source, the probe light can be modulated by about 10% by the generated THz electromagnetic wave (characteristic of the portion B in FIG. 20). However, in that case, measures to avoid saturation are indispensable. However, for example, simply reducing the intensity of the probe light with an ND filter reduces the modulation amount as well as the absolute value of the signal, and the S / N ratio is not improved.

  In order to obtain an ideal S / N ratio, the probe light incident on the polarization filter when unmodulated is circularly polarized, and equal amounts of polarization components orthogonal to each other have information about the rotation of the polarization plane due to modulation without loss. It is preferable. However, this does not solve the saturation problem described above.

  That is, according to the pixel circuit of the general image sensor shown in FIG. 21, excessive probe light is incident on the photodiode 1801, and the amount of charge generated in the photodiode 1801 is saturated that can be accumulated in the capacitor 1804. The charge amount will be exceeded. Then, after the reset transistor 1807 performs the reset operation according to the control by the reset pulse generation circuit 1806, the charge that has reached the saturation amount due to the excessive light input is passed through the transfer transistor 1802 according to the control by the transfer pulse generation circuit 1803. 1805 is input.

  Under such circumstances, conventionally, a technique for suppressing the phase bias of the probe light has been used, and as a result, the first problem that the S / N ratio is lowered without increasing the signal modulation amount occurs. .

  Furthermore, the polarization filter cannot completely block the polarization component orthogonal to the probe light, and the transmission of light in the μW order is a factor that deteriorates the S / N ratio. As a result, the S / N ratio obtained is extremely low and is only about 1%.

  The second problem is that according to the prior art, in order to take a difference between images obtained from two probe light pulses, it is necessary to temporarily store at least one image in a storage circuit. It is complicated.

  The present invention has been made to solve the above problems, and detects the modulation amount of the probe light modulated by the THz electromagnetic wave without suppressing the phase bias of the probe light and without saturating the circuit. An object is to provide a possible photoelectric conversion device.

  Furthermore, it aims at providing the photoelectric conversion apparatus which can detect the modulation amount of the probe light modulated by the THz electromagnetic wave only by receiving one probe light pulse.

  It is also included in the object of the present invention to provide an image sensor and an imaging system using such a photoelectric conversion device.

  In order to achieve the above object, a photoelectric conversion device of the present invention includes a first photodiode, a second photodiode having a first polarity terminal connected to a first polarity terminal of the first photodiode, A first capacitor inserted between the first photodiode and the second photodiode; a second capacitor inserted between the first capacitor and the second photodiode; A differential amplifier circuit connected between the first photodiode and the first capacitor and having the other input terminal connected between the second photodiode and the second capacitor. . Here, the first photodiode and the second photodiode have equivalent characteristics, and the first capacitor and the second capacitor have equivalent characteristics, and the first photodiode and the second photodiode The photodiode may be provided with a filter that transmits incident light according to characteristics different from each other with an intensity corresponding to a specific physical quantity of the incident light. The filter may be a polarization filter having different polarization transmission characteristics, and the filter may be a wavelength filter having different wavelength transmission characteristics.

  The photoelectric conversion device is used to detect the modulation amount of the probe light modulated with respect to the direction and wavelength of the polarization plane, and the two filters formed in the first photodiode and the second photodiode have a predetermined reference state, for example, When the modulation amount is 0, that is, when the probe light is not modulated, the light amount of the probe light incident on each photodiode is set to be equal. In order to modulate the direction and wavelength of the polarization plane of the probe light, a conventionally known electro-optic modulation element can be used.

  According to this configuration, the probe light that becomes circularly polarized light is incident on the two photodiodes as the unmodulated light, and an equal amount of current is generated in the photodiodes, so that each of the opposing electrodes of the first capacitor and the second capacitor Suppressing the output by flowing an equal amount of current. At the time of modulation, a difference in received light amount corresponding to the modulation amount is generated for each photodiode due to a difference in filter characteristics, and a current corresponding to this difference is caused to flow in the first capacitor and the second capacitor to accumulate charges. Thus, it is possible to detect a voltage signal corresponding to the difference in the accumulated charge amount. Therefore, when unmodulated light is incident, the current generated by the two photodiodes does not contribute to the signal, and the output signal does not saturate. It becomes possible to detect. In addition, the differential amplifier circuit can amplify a voltage signal corresponding to the difference in the amount of accumulated charges with a predetermined gain, and can realize a high S / N ratio of the obtained image.

  As a result, it is possible to detect the modulation amount of the probe light modulated by the THz electromagnetic wave by only receiving one probe light pulse without suppressing the phase bias of the probe light and without saturating the circuit. A conversion device is obtained.

  Here, the first photodiode includes a first conductivity type first region and a second conductivity type second region which are formed on the surface of the first conductivity type semiconductor substrate and are in contact with each other. The diode is composed of a third region of the first conductivity type and a fourth region of the second conductivity type formed on the surface of the semiconductor substrate and in contact with each other, and the shapes of the first region and the third region are equal, The shapes of the second region and the fourth region are equal, and either the first region or the second region is connected to the first electrode formed on the surface of the semiconductor substrate, and one of the third region and the fourth region. May be connected to a second electrode formed on the surface of the semiconductor substrate.

  According to this configuration, the photoelectric conversion device can be configured with higher symmetry, and the circuit configuration of the photoelectric conversion device can be easily formed on the semiconductor substrate as an integrated circuit.

  The first capacitor and the second capacitor are each a junction capacitor of a PN junction, the first electrode is a second conductivity type electrode of the first capacitor, and the second electrode is the second capacitor. It may be a second conductivity type electrode having a capacitance.

  According to this configuration, the two capacitors, the first capacitor and the second capacitor, are junction capacitors formed by PN junctions between the first electrode and the second electrode and the semiconductor substrate, respectively. Therefore, by reducing the impurity concentrations of the first electrode, the second electrode, and the semiconductor substrate, the size of the two capacitors can be reduced, and a higher S / N ratio of the obtained image can be realized.

  The first capacitor and the second capacitor are each a junction capacitor of a Schottky junction, the first electrode is an electrode of the first capacitor that is Schottky connected to the semiconductor substrate, and the second electrode is The second capacitor electrode may be Schottky connected to the semiconductor substrate.

  According to this configuration, the two capacitors become junction capacitors by Schottky junctions between the first electrode and the second electrode and the semiconductor substrate, respectively. Accordingly, the size of the two capacitors is reduced by reducing the impurity concentration of the first electrode, the second electrode, and the semiconductor substrate, or by increasing the height of the Schottky junction between the first electrode, the second electrode, and the semiconductor substrate. In addition, a higher S / N ratio of the obtained image can be realized.

  The first capacitor and the second capacitor may have a structure in which a dielectric is sandwiched between conductors.

  According to this configuration, it is possible to reliably form a capacitor having a high dynamic range on the semiconductor substrate, which cannot be realized simply by modifying the semiconductor substrate by controlling the film thickness and dielectric constant of the dielectric film. It becomes.

  The film thickness of the first capacitor dielectric may be equal to the film thickness of the second capacitor dielectric.

  According to this configuration, the two capacitors are the same, and when an equal amount of photocurrent is generated by the two photodiodes, an equal amount of charge is reliably accumulated in each capacitor. Therefore, it is possible to reliably realize a desired characteristic of suppressing signal output when unmodulated light is incident.

  The photoelectric conversion device may further include a discharge switch that is inserted between the first capacitor and the second capacitor and discharges the charges of the first capacitor and the second capacitor.

  According to this configuration, it is possible to cancel, that is, perform a reset operation, the accumulated charges of the two capacitors.

  The discharge switch may be a MOSFET, and the MOSFET may include a gate electrode formed on the semiconductor substrate so as to be positioned between the first electrode and the second electrode.

  According to this configuration, it is possible to control the depletion layer thickness under the gate electrode in the semiconductor substrate by applying a voltage to the gate electrode. Therefore, it is possible to control the magnitudes of the two capacitors located on both sides of the gate electrode by the external voltage, and to change the magnitude of the voltage signal in accordance with the accumulated charge amount. Further, a voltage can be applied to the gate electrode to form a channel under the gate electrode in the semiconductor substrate. Accordingly, the two capacitors can be electrically short-circuited, that is, can be reset.

  The discharge switch is a junction FET, and the junction FET is a second conductivity type gate formed on the surface of the semiconductor substrate so as to be positioned between the first electrode and the second electrode. You may have an electrode.

  According to this configuration, it is possible to control the depletion layer thickness under the gate electrode in the semiconductor substrate by applying a voltage to the gate electrode. Therefore, it is possible to control the magnitude of the capacitance located on both sides of the gate electrode by the external voltage and change the magnitude of the voltage signal according to the amount of accumulated charge. Furthermore, a voltage can be applied to the gate electrode, and the depletion layer width can be reduced under the gate electrode in the semiconductor substrate. Accordingly, the two capacitors can be electrically short-circuited, that is, can be reset.

  The photoelectric conversion device is further inserted between a first transfer switch inserted between the first photodiode and the first capacitor, and between the second photodiode and the second capacitor. And a second transfer switch.

  According to this configuration, it is possible to independently control the electrical connection state between the photodiode and the capacitor connected. In particular, when a photocurrent with a large amount of incident light that exceeds the photodiode capacity is generated, the photodiode is electrically short-circuited to the capacitor by the first transfer switch and the second transfer switch, so that charge is always accumulated in the capacitor. It is possible to continue and no signal saturation occurs. On the other hand, when the amount of incident light is small, the photodiode can be electrically cut off from the capacitor, the charge can be accumulated in the photodiode, and transferred to the capacitor by the first transfer switch and the second transfer switch when the charge is read. is there.

  In addition, the present invention includes a plurality of pixel circuits that are two-dimensionally arranged and each have the photoelectric conversion device, and a readout circuit that reads an output signal of the photoelectric conversion device included in each pixel circuit. It can also be set as an image sensor.

  According to this configuration, the difference between the photocurrents generated by the two photodiodes can be extracted as image information.

  The present invention is also a method for driving the photoelectric conversion device, wherein the first voltage is applied to the gate electrode so that a depletion layer is expanded in a region below the gate electrode of the semiconductor substrate. A driving method of the photoelectric conversion device, wherein a voltage corresponding to a difference between charges stored in the capacitor and the second capacitor is read from the differential amplifier circuit, can be used.

  According to this configuration, a voltage signal corresponding to the difference between the accumulated charges of the two capacitors is output in a state where the depletion layer under the gate electrode in the semiconductor substrate is expanded to reduce the capacitance. Therefore, a higher S / N ratio of the obtained image can be realized.

  The present invention is also a method for driving the photoelectric conversion device, wherein the first voltage is applied to the gate electrode so that a depletion layer is narrowed in a region below the gate electrode of the semiconductor substrate. A driving method of the photoelectric conversion device can also be characterized in that the charges accumulated in the capacitor and the second capacitor are discharged.

  According to this configuration, the charge accumulated in the capacitor is discharged by narrowing the depletion layer under the gate electrode in the semiconductor substrate. Accordingly, it is possible to perform a capacitance resetting operation without providing an independent reset gate, and it is possible to suppress an increase in device area.

  The present invention also provides an electromagnetic wave source that generates a THz electromagnetic wave having a frequency in a range of 0.1 THz to 10 THz, a light source that generates probe light, and the THz electromagnetic wave that has passed through or reflected from a subject as the probe light. A superimposing optical element that superimposes, an electrooptic modulation element that receives the superimposed THz electromagnetic wave and probe light and modulates a specific physical quantity of the probe light according to the electric field of the THz electromagnetic wave, and the modulated probe An imaging system including the image sensor that captures light can also be provided.

  According to this configuration, it is possible to realize a THz imaging system that can operate at high speed and has a high S / N ratio.

  According to the photoelectric conversion device of the present invention, the difference between the photocurrents generated in the two photodiodes can be amplified to the output signal, so that the filters that transmit the incident light according to the different characteristics with the intensity according to the specific physical quantity are respectively If provided in the photodiode, it is possible to detect how much the incident light is modulated with respect to a specific physical quantity as a difference between photocurrents generated in the two photodiodes.

  For example, if the specific physical quantity is the direction of the polarization plane, the filter is a polarization filter, and if the specific physical quantity is the wavelength, the filter is a wavelength filter.

  At this time, since the output signal is based on the difference in the amount of incident light, the circuit is not saturated even if the amount of incident light itself is large. Therefore, it is possible to detect the modulation amount of the incident light with a high S / N ratio without stopping the conventional phase bias of the incident light in order to prevent circuit saturation. Moreover, the detection can be performed by receiving the modulated incident light only once.

  This effect is particularly remarkable when the photoelectric conversion device of the present invention is applied to an image sensor and an imaging system. Since the characteristics of the object reflected in the amount of modulation of incident light can be imaged with a single shooting, the apparatus can be simplified and real-time imaging can be realized.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a circuit diagram illustrating an example of the configuration of the photoelectric conversion apparatus according to the first embodiment.

  In this photoelectric conversion device, the same polarity (P side) terminals of two photodiodes 1 and 1 'having the same characteristics are grounded, and the other terminals are transfer switches such as MOSFETs (Metal Oxide Semiconductor-Field Effect Transistors). The charge storage capacitors 3 and 3 ′ having the same characteristics are connected via the terminals 2 and 2 ′. That is, the two photodiodes 1 and 1 'are connected in series with their polarities reversed, and the cathode terminal of the photodiode 1 and the cathode terminal of the photodiode 1' are connected. The charge storage capacitor 3 is inserted between the two photodiodes 1 and 1 ′, and the charge storage capacitor 3 ′ is inserted between the charge storage capacitor 3 and the photodiode 1 ′. The transfer switches 2 and 2 'follow the control circuit 4, and the control circuit 4 generates a control signal for transferring the charges accumulated in the photodiodes 1 and 1' to the charge storage capacitors 3 and 3 ', respectively. A capacitance terminal 5 on the photodiode side of the charge storage capacitor 3 between the photodiode 1 and the charge storage capacitor 3 is connected to one terminal of the differential amplifier circuit 6. A capacitor terminal 5 ′ on the photodiode side of the charge storage capacitor 3 ′ between the photodiode 1 ′ and the charge storage capacitor 3 ′ is connected to the other terminal of the differential amplifier circuit 6. The differential amplifier circuit 6 amplifies the voltage between both terminals of the capacitor terminals 5 and 5 ′. The switches 7 and 7 ′ such as MOS FETs are switches that discharge the charge storage capacitors 3 and 3 ′ according to the output signal of the reset signal generation circuit 8.

  In this photoelectric conversion device, when light of equal intensity is incident on the photodiodes 1 and 1 ′, an equal amount of photocurrent is generated, and the charge storage capacitors 3 and 3 ′ are charged with opposite polarities. Since the stored charge amounts of the charge storage capacitors 3 and 3 ′ are equal, no voltage is generated between the capacitor terminals 5 and 5 ′. On the other hand, when the light intensity incident on the photodiodes 1, 1 'is different, the photoelectric flow rates generated in the photodiodes 1, 1' are different. Accordingly, the accumulated charge amounts of the charge storage capacitors 3 and 3 ′ are different, and the charge storage capacitors 3 and 3 ′ are charged with the opposite polarity. A corresponding voltage is generated. This voltage is amplified by the differential amplifier circuit 6, and a signal proportional to the difference in the amount of charges accumulated in the charge storage capacitors 3, 3 ′, that is, the difference in incident light intensity to the photodiodes 1, 1 ′ is obtained. Is output from.

  Here, the photodiodes 1 and 1 ′, the transfer switches 2 and 2 ′, the switches 7 and 7 ′, and the charge storage capacitors 3 and 3 ′ are collectively referred to as a photoelectric conversion unit 11. The charge storage capacitors 3, 3 'are examples of the first capacitor and the second capacitor of the present invention, respectively. The transfer switches 2 and 2 'are examples of the first transfer switch and the second transfer switch of the present invention, respectively.

  FIG. 2 is a timing chart showing main signals generated or given to each part of the above-described photoelectric conversion device.

  A 1 kHz probe light pulse is incident on the photodiodes 1, 1 ′, and a pulse current flows according to the amount of light. The period A is when the same amount of light is incident on the photodiodes 1 and 1 ′ and the same photocurrent flows, and the period B is when the amount of incident light on the photodiode 1 is greater than the amount of incident light on the photodiode 1 ′. . In the period B, a voltage corresponding to the difference in the accumulated charge amount of the charge storage capacitors 3 and 3 ′ is generated as a voltage between the capacitor terminals 5 and 5 ′.

  FIG. 3 schematically shows a cross section passing through the photodiode 1 ′ of the photoelectric conversion unit 11 in the photoelectric conversion device according to the present embodiment.

A PN junction composed of a P layer 301 and an N layer 302 formed in contact with the lower layer is formed on the outermost surface of the semiconductor substrate 300. This PN junction constitutes the photodiode 1. Similarly, the PN junction composed of the P layer 301 ′ and the N layer 302 ′ constitutes the photodiode 1 ′. The N ⁺ electrodes 304 and 304 ′ constitute the drains of the transfer switches 2 and 2 ′, and form PN junctions with the semiconductor substrate 300, respectively, to form charge storage capacitors 3 and 3 ′ that are junction capacitances of the PN junctions. To do. The N ⁺ electrode 304 is formed on the surface of the semiconductor substrate 300 so as to be positioned between the P layer 301 and the N layer 302 and the P layer 301 ′ and the N layer 302 ′. The N ⁺ electrode 304 ′ is formed on the surface of the semiconductor substrate 300 so as to be positioned between the N ⁺ electrode 304 and the P layer 301 ′ and the N layer 302 ′. The transfer gates 303 and 303 ′ constitute the gates of the transfer switches 2 and 2 ′. The photodiodes 1 and 1 ′ are formed in regions adjacent to each other with the transfer gates 303 and 303 ′ and the N ⁺ electrodes 304 and 304 ′ interposed therebetween.

Here, the P layer 301 and the P layer 301 ′, the N layer 302 and the N layer 302 ′, the N ⁺ electrode 304 and the N ⁺ electrode 304 ′, and the transfer gate 303 and the transfer gate 303 ′ are simultaneously formed by the same manufacturing process. Therefore, the shape is the same.

The P layer 301 and the N layer 302 are examples of the first region and the second region of the present invention, respectively. The P layer 301 ′ and the N layer 302 ′ are examples of the third region and the fourth region of the present invention, respectively. Further, the N ⁺ electrode 304 is an example of the first electrode of the present invention, and the N ⁺ electrode 304 ′ is an example of the second electrode of the present invention.

  The entire device is covered with a protective layer 305 made of a silicon nitride film, and a light shielding metal 306 is disposed in the protective layer 305 between the two photodiodes 1 and 1 ′.

  On-chip lenses 307 and 307 'for increasing the amount of collected light are formed above the photodiodes 1 and 1'.

  Further, a second protective layer 308 is formed on the on-chip lenses 307 and 307 '. On the second protective layer 308, a polarizing filter 309 that is located above the photodiode 1 and transmits only polarized light having an oscillating electric field perpendicular to the paper surface, and a polarizing filter 309 that is located above the photodiode 1 ′. A polarizing filter 309 ′ that transmits only polarized light that is orthogonal to the transmitted polarized light is formed. That is, polarization filters 309 and 309 ′ that transmit incident light according to different characteristics from each other with an intensity corresponding to a specific physical quantity of the incident light are formed above the photodiodes 1 and 1 ′. For example, a polarizing filter 309 that transmits only P-polarized light is formed above the photodiode 1, and a polarizing filter 309 ′ that transmits only S-polarized light is formed above the photodiode 1 ′.

  Both the polarizing filters 309 and 309 'are formed by forming a grating in which copper, which is a wiring metal, is arranged for 10 periods with a width of 0.25 μm and an interval of 0.25 μm. At this time, the directions of the gratings are orthogonal between the polarizing filters 309 and 309 ′, and have transmission characteristics with respect to the orthogonally polarized light.

  In the photoelectric conversion device, the photodiodes 1 and 1 ′ have the same characteristics, and the photoelectric conversion device has a configuration that is symmetric with respect to the light shielding metal 306. Accordingly, the circuit configuration of the photoelectric conversion device can be easily formed as an integrated circuit on the semiconductor substrate 300, and the manufacturing process is simplified, so that the cost of the photoelectric conversion device can be reduced.

In addition, the charge storage capacitors 3 and 3 ′ are junction capacitors by PN junctions between the N ⁺ electrodes 304 and 304 ′ and the semiconductor substrate 300, respectively. Therefore, by reducing the impurity concentration of the N ⁺ electrodes 304 and 304 ′ and the semiconductor substrate 300, the size of the charge storage capacitors 3 and 3 ′ can be reduced, and a higher S / N ratio of the obtained image can be realized. Is possible.

  Further, it is possible to independently control the electrical connection state between the charge storage capacitors 3 and 3 ′ connected to the photodiodes 1 and 1 ′ by the transfer switches 2 and 2 ′. In particular, when a large amount of incident light is generated and a photocurrent exceeding the capacity of the photodiodes 1 and 1 ′ is generated, the transfer switches 2 and 2 ′ electrically connect the photodiodes 1 and 1 ′ to the charge storage capacitors 3 and 3 ′. By short-circuiting, it becomes possible to keep accumulating charges in the charge accumulating capacitors 3, 3 'at all times, and no signal saturation occurs. On the other hand, when the amount of incident light is small, the photodiodes 1, 1 ′ are electrically cut off from the charge storage capacitors 3, 3 ′, charges are accumulated in the photodiodes 1, 1 ′, and the transfer switch 2 is read out when reading the charges. 2 'can be transferred to the charge storage capacitors 3, 3'.

  FIG. 4 shows a layout of the photoelectric conversion device 11 according to the present embodiment as viewed from above the photoelectric conversion unit 11 including the photodiodes 1, 1 ′, transfer switches 2, 2 ′, and charge storage capacitors 3, 3 ′. It is shown schematically.

The N layers 302 and 302 ′ of the photodiodes 1 and 1 ′ are entirely covered with P layers 301 and 301 ′, respectively. The charges generated in the photodiodes 1 and 1 ′ are transferred to the N ⁺ electrodes 304 and 304 ′ constituting the charge storage capacitors 3 and 3 ′, respectively, according to the outputs from the control circuit 4 by the transfer gates 303 and 303 ′. Further, the voltage between the terminals of the N ⁺ electrodes 304 and 304 ′ is input to the differential amplifier circuit 6 through the wirings 405 and 405 ′, respectively. Further, reset gates 407 and 407 ′ constituting the gates of the switches 7 and 7 ′ are arranged adjacent to the N ⁺ electrodes 304 and 304 ′, and the accumulated charges are transferred according to the output of the reset signal generation circuit 8 according to the output of the differential amplifier circuit 6. It discharges after reading.

  FIG. 5 is a diagram schematically illustrating a configuration example of a THz imaging system using the above-described photoelectric conversion device.

  A light source 501 generates an ultrashort pulse train having a pulse width of 100 fs at a frequency of 1 kHz. The polarization beam splitter 502 separates P-polarized light as pump light 503 and S-polarized light as probe light 504. The pump light 503 is incident on the ZnTe crystal [110] plane constituting the THz electromagnetic wave emitter 506 perpendicularly through the optical delay line 505, and a THz electromagnetic wave 507 having a frequency in the region of 0.1 THz to 10 THz is generated.

  After the traveling direction of the probe light 504 is adjusted by the mirror 508, the probe light 504 is incident on the silicon mirror 509 made of a silicon wafer, shares the optical axis with the THz electromagnetic wave 507 transmitted through the silicon mirror 509, and the [110] plane is The light enters the electro-optic modulator 510 made of ZnTe crystal arranged perpendicular to the optical axis. A quarter wave plate 511 and an image sensor 512 are arranged in this order at the subsequent stage of the electro-optic modulator 510.

  As will be described in detail later, the image sensor 512 includes a plurality of pixels that are two-dimensionally arranged and each include the above-described photoelectric conversion device, and a readout circuit that reads an output signal of the photoelectric conversion device of each pixel. Prepare.

  In the subsequent stage, the quarter-wave plate 511 emits the probe light 504 when the THz electromagnetic wave 507 does not enter the electro-optic modulator 510 simultaneously with each pulse of the probe light 504, that is, when the THz electromagnetic wave pulse and the probe pulse are asynchronous. Set to full circular polarization state. At this time, since circularly polarized light is incident on the image sensor 512, the P-polarized component and the S-polarized component are equal. Therefore, an equal amount of current is generated in each of the internal photodiodes 1, 1 ′, and the charge storage capacitor 3, 3 'no charge is accumulated.

  On the other hand, when the THz electromagnetic wave pulse and the probe light pulse are simultaneously incident on the electro-optic modulator 510, that is, when both pulses are synchronized, a specific physical quantity of the probe light 504 according to the electric field of the THz electromagnetic wave 507. Is modulated, and the polarization state of the probe light 504 after passing through the electro-optic modulator 510 is rotated as compared with the case where the probe light 504 is asynchronous with the THz electromagnetic wave 507 and further becomes elliptically polarized light. The polarization state after transmission through 511 is not completely circularly polarized but is elliptically polarized.

  Accordingly, the intensity of light incident on the two photodiodes 1 and 1 ′ inside the image sensor 512 that captures the probe light 504 is different, and different amounts of photocurrents are generated. Therefore, the charge storage capacitors 3 and 3 ′ The charge is accumulated, and a voltage signal, that is, a THz electromagnetic wave detection signal is output. As a result, it is possible to detect THz electromagnetic waves without saturating the photodiodes 1 and 1 ′ that are problems. Furthermore, the configuration of the photoelectric conversion device of this embodiment does not require a bulk polarization beam splitter or a polarization prism, which is necessary in the configuration of the conventional photoelectric conversion device, and the photoelectric conversion device has two photodiodes 1. Since 1 ′ is integrated, there is an advantage that the optical system is simplified.

  FIG. 6 is a block diagram illustrating an example of a functional configuration of the image sensor 512. The image sensor 512 is realized by two-dimensionally arranging a plurality of pixel circuits each including the above-described photoelectric conversion device and arranging a vertical scanning circuit 601 and a horizontal scanning circuit 602.

  The vertical scanning circuit 601 includes the control circuit 4 and the reset signal generation circuit 8 shown in FIG. 1. In addition to the control signals output from these circuits, a row selection signal is sent via the row data selection line 603. , And supply to the pixel circuit for each row.

  Each pixel circuit includes a readout circuit that reads out an output signal of the pixel circuit in addition to the photoelectric conversion unit 11 and the differential amplifier circuit 6 of the photoelectric conversion device illustrated in FIG. 1. That is, it includes a switch 610 and terminals 607, 609, and 606 for receiving the control signals from the control circuit 4 and the reset signal generation circuit 8 and the row selection signal.

  By closing the switch 610 according to the row selection signal, the output signal of the pixel circuit is output to the column data readout line 604.

  The horizontal scanning circuit 602 sequentially reads out the output signal from the pixel circuit in the selected row for each pixel via the column data readout line 604.

  Note that on the premise that this image sensor is applied to a THz imaging system, the above-described on-chip lenses 307 and 307 ′ and the polarization filter 309 are provided on the photodiodes 1 and 1 ′ included in the photoelectric conversion unit 11 of each pixel. 309 ′ is provided.

  As described above, in the image sensor 512 of the present embodiment, there is a certain signal charge accumulation time corresponding to one screen data read time. The operation described in the timing chart of FIG. 2 is performed in each pixel circuit within this charge accumulation time, whereby charges corresponding to the difference in the amount of light incident on the photodiodes 1, 1 ′ are charged. 'Accumulated and read out.

  Accordingly, the difference in the amount of incident light on the two photodiodes 1, 1 'in each pixel circuit is detected on each screen, and is reproduced as a two-dimensional image by an external image reproduction circuit.

FIG. 7 is a diagram schematically showing a modification of the THz imaging system described above.
An ultrashort pulse train having a pulse width of 100 fs is generated from the light source 701 at a frequency of 1 kHz. The polarization beam splitter 702 separates P-polarized light as pump light 703 and S-polarized light as probe light 704. The pump light 703 passes through an optical delay line 705 and is incident on a THz electromagnetic wave emitter 706 configured by a photoconductive switch having an electrode pair formed on a semi-insulating GaAs wafer with a distance of 10 mm, and a THz electromagnetic wave 707 is generated. Is done.

  The THz electromagnetic wave 707 generated in this way is a highly collimated beam, and is irradiated to the object to be measured 708 having a two-dimensional transmission characteristic distribution with respect to the THz electromagnetic wave 707. The THz electromagnetic wave 707 that has passed through the device under test 708 becomes a spatially intensity-modulated beam along with the two-dimensional transmission characteristic distribution of the device under test 708. This is imaged by a polyethylene lens 709 in an electro-optic modulator 713 made of a subsequent ZnTe crystal.

  The probe light 704 is rerouted by a mirror 710, further expanded in beam width by a beam expander 711, and then incident on a silicon mirror 712 composed of a silicon wafer and transmitted through the silicon mirror 712. The optical axis is shared with 707, and the light enters the electro-optic modulator 713 made of a ZnTe crystal whose [110] plane is arranged perpendicular to the optical axis.

  A quarter-wave plate 714 and an image sensor 715 are arranged in this order at the subsequent stage of the electro-optic modulator 713. The image sensor 715 is equivalent to the image sensor 512 described above.

  In the subsequent stage, the ¼ wavelength plate 714 completely transmits the probe light 704 when the THz electromagnetic wave 707 does not enter the electro-optic modulator 713 simultaneously with each pulse of the probe light 704, that is, when the THz electromagnetic wave pulse and the probe pulse are asynchronous. Set to a circular polarization state. At this time, since circularly polarized light is incident on the image sensor 715, the P-polarized component and the S-polarized component are equal. Therefore, an equal amount of photocurrent is generated in the photodiodes 1, 1 'of each internal pixel, No charge is stored in the charge storage capacitors 3 and 3 ′.

  On the other hand, when the THz electromagnetic wave pulse and the probe light pulse are simultaneously incident on the electro-optic modulator 713, that is, when the two pulses are synchronized, the polarization state of the probe light 704 after being transmitted to the electro-optic modulator 713 Compared with the case of asynchronous with the THz electromagnetic wave 707, the polarization axis rotates and becomes elliptically polarized light. Therefore, the polarization state after transmission through the quarter-wave plate 714 does not become completely circularly polarized light. Become.

  Accordingly, the light intensity incident on the photodiodes 1 and 1 'pair of the respective pixels inside the image sensor 715 that captures the probe light 704 is different, and different amounts of photocurrents are generated. Therefore, the charge storage capacitors 3 and 3' Is accumulated, and a voltage signal, that is, a THz electromagnetic wave detection signal is output.

  The two-dimensional signal detected and output based on the above principle is output to the monitor as video information by the image reproduction device 716. Therefore, THz electromagnetic wave imaging can be performed without suppressing the phase bias of the probe light, which is a problem, without saturating the circuit, and only by receiving one probe light pulse without saturating the photodiode. Is possible.

(Second Embodiment)
The photoelectric conversion device having a function equivalent to that of the photoelectric conversion device (see FIG. 1) described in the first embodiment can also be realized by another configuration.

  FIG. 8 schematically shows a cross section passing through the photodiodes 1, 1 ′ of the photoelectric conversion unit 11 of the photoelectric conversion device according to the present embodiment.

  A PN junction composed of a P layer 301 and an N layer 302 formed therebelow is formed on the outermost surface of the semiconductor substrate 300. This PN junction constitutes the photodiode 1. Similarly, a PN junction composed of a P layer 301 ′ and an N layer 302 ′ constitutes a photodiode 1 ′. The nickel silicide electrodes 804 and 804 'constitute the drains of the transfer switches 2 and 2', and are connected to the semiconductor substrate 300 by Schottky to form charge storage capacitors 3 and 3 'that are junction capacitors of Schottky junctions. The nickel silicide electrode 804 is formed on the surface of the semiconductor substrate 300 so as to be positioned between the P layer 301 and the N layer 302 and the P layer 301 ′ and the N layer 302 ′. The nickel silicide electrode 804 ′ is formed on the surface of the semiconductor substrate 300 so as to be located between the nickel silicide electrode 804 and the P layer 301 ′ and the N layer 302 ′. The transfer gates 303 and 303 'constitute the gates of the transfer switches 2 and 2'. The photodiodes 1 and 1 'are formed in regions adjacent to each other with the transfer gates 303 and 303' and the nickel silicide electrodes 804 and 804 'interposed therebetween. Here, the nickel silicide electrodes 804 and 804 'are formed by depositing Ni of 200 angstroms in the electrode region and annealing at 350 ° C.

  Here, since the nickel silicide electrode 804 and the nickel silicide electrode 804 'are simultaneously formed by the same manufacturing process, their shapes are equal. The nickel silicide electrode 804 is an example of the first electrode of the present invention, and the nickel silicide electrode 804 'is an example of the second electrode of the present invention.

  The entire device is covered with a protective layer 305 made of a silicon nitride film, and a light shielding metal 306 is disposed in the protective layer 305 between the two photodiodes 1 and 1 ′.

  On-chip lenses 307 and 307 'for increasing the amount of collected light are formed above the photodiodes 1 and 1'.

  Further, a second protective layer 308 is formed on the on-chip lenses 307 and 307 '. On the second protective layer 308, a polarizing filter 309 that is located above the photodiode 1 and transmits only polarized light having an oscillating electric field perpendicular to the paper surface, and a polarizing filter 309 that is located above the photodiode 1 ′. A polarizing filter 309 ′ that transmits only polarized light that is orthogonal to the transmitted polarized light is formed. For example, a polarizing filter 309 that transmits only P-polarized light is formed above the photodiode 1, and a polarizing filter 309 ′ that transmits only S-polarized light is formed above the photodiode 1 ′.

  Both the polarizing filters 309 and 309 'are formed by forming a grating in which copper, which is a wiring metal, is arranged for 10 periods with a width of 0.25 μm and an interval of 0.25 μm. At this time, the directions of the gratings are orthogonal between the polarizing filters 309 and 309 ′, and have transmission characteristics with respect to the orthogonally polarized light.

  In the photoelectric conversion device, the photodiodes 1 and 1 ′ have the same characteristics, and the photoelectric conversion device has a configuration that is symmetric with respect to the light shielding metal 306. Accordingly, the circuit configuration of the photoelectric conversion device can be easily formed as an integrated circuit on the semiconductor substrate 300, and the manufacturing process is simplified, so that the cost of the photoelectric conversion device can be reduced.

  The charge storage capacitors 3 and 3 ′ are junction capacitors by Schottky junctions between the nickel silicide electrodes 804 and 804 ′ and the semiconductor substrate 300, respectively. Accordingly, by reducing the impurity concentration of the nickel silicide electrodes 804 and 804 ′ and the semiconductor substrate 300, or by increasing the Schottky junction height between the nickel silicide electrodes 804 and 804 ′ and the semiconductor substrate 300, It is possible to reduce the size of 3 ′ and realize a higher S / N ratio of the obtained image.

(Third embodiment)
The photoelectric conversion device having a function equivalent to that of the photoelectric conversion device (see FIG. 1) described in the first embodiment can also be realized by another configuration.

  FIG. 9 schematically shows a cross section passing through the photodiodes 1, 1 ′ of the photoelectric conversion unit 11 of the photoelectric conversion device according to the present embodiment.

  A PN junction composed of a P layer 301 and an N layer 302 formed thereunder is formed on the outermost surface of the semiconductor substrate 300. This PN junction constitutes the photodiode 1. Similarly, a PN junction composed of a P layer 301 ′ and an N layer 302 ′ constitutes a photodiode 1 ′. On the nickel silicide electrodes 901 and 901 ', high dielectric constant films 902 and 902' made of bismuth, strontium and titanium oxide are deposited to a thickness of 100 nm, and the surfaces of the high dielectric constant films 902 and 902 'are made of an upper electrode 903 made of TiN. Covered with. The upper electrode 903, the high dielectric constant films 902 and 902 ', and the nickel silicide electrodes 901 and 901' constitute a charge storage capacitor 3, 3 'having a structure in which a dielectric is sandwiched between conductors. The nickel silicide electrodes 901 and 901 'constitute the drains of the transfer switches 2 and 2'. The transfer gates 303 and 303 'constitute the gates of the transfer switches 2 and 2'. The photodiodes 1 and 1 'are formed in regions adjacent to each other with the transfer gates 303 and 303', the upper electrode 903, and the high dielectric constant films 902 and 902 'interposed therebetween.

  Here, since the nickel silicide electrode 901 and the nickel silicide electrode 901 ', the high dielectric constant film 902 and the high dielectric constant film 902' are simultaneously formed by the same manufacturing process, their shapes are equal. For example, the film thickness of the high dielectric constant film 902 is equal to the film thickness of the high dielectric constant film 902 '.

  The entire device is covered with a protective layer 305 made of a silicon nitride film, and a light shielding metal 306 is disposed in the protective layer 305 between the two photodiodes 1 and 1 ′.

  On-chip lenses 307 and 307 'for increasing the amount of collected light are formed above the photodiodes 1 and 1'.

  Further, a second protective layer 308 is formed on the on-chip lenses 307 and 307 '. On the second protective layer 308, a polarizing filter 309 that is located above the photodiode 1 and transmits only polarized light having an oscillating electric field perpendicular to the paper surface, and a polarizing filter 309 that is located above the photodiode 1 ′. A polarizing filter 309 ′ that transmits only polarized light that is orthogonal to the transmitted polarized light is formed. For example, a polarizing filter 309 that transmits only P-polarized light is formed above the photodiode 1, and a polarizing filter 309 ′ that transmits only S-polarized light is formed above the photodiode 1 ′.

  Both the polarizing filters 309 and 309 'are formed by forming a grating in which copper, which is a wiring metal, is arranged for 100 periods with a width of 0.25 μm and an interval of 0.25 μm. At this time, the directions of the gratings are orthogonal between the polarizing filters 309 and 309 ′, and have transmission characteristics with respect to the orthogonally polarized light.

  In this photoelectric conversion device, the charge storage capacitor 3 composed of the upper electrode 903, the high dielectric constant film 902 and the nickel silicide electrode 901, and the charge storage capacitor 3 composed of the upper electrode 903, the high dielectric constant film 902 ′ and the nickel silicide electrode 901 ′. 'Means that the area and the dielectric film thickness are equal, and the photodiodes 1 and 1' connected to these have the same characteristics, so that the photoelectric conversion device is electrically symmetric with respect to the upper electrode 903. . Therefore, when an equal amount of photocurrent flows through the two photodiodes 1, 1 ', no voltage is generated between the capacitor terminals 5, 5'. On the other hand, when the photocurrents flowing through the two photodiodes 1 and 1 ′ are different, a net charge is accumulated between the nickel silicide electrodes 901 and 901 ′, and a voltage is generated between the capacitor terminals 5 and 5 ′.

  Further, bismuth / strontium / titanium oxide, which is a high dielectric material, is used for the charge storage capacitors 3 and 3 ′. Therefore, a large amount of charge can be stored in the charge storage capacitors 3 and 3 'at a low voltage, which is particularly advantageous when the amount of background light is high.

  Further, the photodiodes 1 and 1 ′ have the same characteristics, and the photoelectric conversion device has a symmetric configuration with respect to the upper electrode 903. Accordingly, the circuit configuration of the photoelectric conversion device can be easily formed as an integrated circuit on the semiconductor substrate 300, and the manufacturing process is simplified, so that the cost of the photoelectric conversion device can be reduced.

  The charge storage capacitors 3 and 3 'include an upper electrode 903, high dielectric constant films 902 and 902', and nickel silicide electrodes 901 and 901 '. Therefore, the charge storage capacitors 3 and 3 ′ having a high dynamic range that cannot be realized simply by modifying the semiconductor substrate 300 by controlling the film thickness and the dielectric constant of the high dielectric constant films 902 and 902 ′ are ensured. It can be formed on the semiconductor substrate 300.

  The film thickness of the high dielectric constant film 902 is equal to the film thickness of the high dielectric constant film 902 '. Accordingly, the two charge storage capacitors 3 and 3 ′ are the same, and when the two photodiodes 1 and 1 ′ generate an equal amount of photocurrent, the equal amount of charge is surely equal to each other. Accumulated in capacities 3, 3 ′. Therefore, it is possible to reliably realize a desired characteristic of suppressing signal output when unmodulated light is incident.

(Fourth embodiment)
FIG. 10 is a circuit diagram illustrating an example of a structure of the photoelectric conversion device in this embodiment.

  In this photoelectric conversion device, the same polarity (P side) terminals of two photodiodes 1, 1 ′ having the same characteristics are grounded, and the other terminal is a charge storage capacitor 3 via transfer switches 2, 2 ′. 3 '. The transfer switches 2 and 2 'follow the control circuit 4, and the control circuit 4 generates a control signal for transferring the charges accumulated in the photodiodes 1 and 1' to the charge storage capacitors 3 and 3 ', respectively. Capacitance terminals 5 and 5 'on the photodiode side of the charge storage capacitors 3 and 3' are connected to a differential amplifier circuit 6 that amplifies the voltage between both terminals. Also, a switch 17 such as a junction FET (Field Effect Transistor) and a MOSFET is inserted between the charge storage capacitors 3 and 3 ′, and the charge storage capacitors 3 and 3 ′ are stored according to the output signal of the reset control circuit 19. It is a switch that discharges electric charges simultaneously.

  Here, the photodiodes 1 and 1 ′, the transfer switches 2 and 2 ′, the switch 17, and the charge storage capacitors 3 and 3 ′ are collectively referred to as a photoelectric conversion unit 11. The switch 17 is an example of a discharge switch according to the present invention.

  FIG. 11 is a timing chart showing main signals generated or given to each part of the above-described photoelectric conversion device.

  The period A is when the same amount of light is incident on the photodiodes 1, 1 'and the same photocurrent flows, and the period B is when the different amounts of light are incident on the photodiodes 1, 1' and different photocurrents flow. In the period B, the photocurrents generated in the two photodiodes 1 and 1 ′ are different, and accordingly, the accumulated charges in the charge accumulating capacitors 3 and 3 ′ are also different, resulting in a difference in accumulated charges between the capacitor terminals 5 and 5 ′. The principle that the corresponding voltage is generated is the same as that of the photoelectric conversion device of the first embodiment. In the final stage of both periods A and B in which the voltage corresponding to the difference between the stored charges in the charge storage capacitors 3 and 3 ′ is read from the differential amplifier circuit 6, the output of the reset control circuit 19 is composed of a single transistor. When the switch 17 is turned on, the stored charges in the charge storage capacitors 3 and 3 ′ are simultaneously read, and the operation is completed.

  FIG. 12 schematically shows a cross section passing through the photodiodes 1, 1 ′ of the photoelectric conversion unit 11 of the photoelectric conversion device according to the present embodiment.

A PN junction composed of a P layer 301 and an N layer 302 formed therebelow is formed on the outermost surface of the semiconductor substrate 300. This PN junction constitutes the photodiode 1. Similarly, the PN junction composed of the P layer 301 ′ and the N layer 302 ′ constitutes the photodiode 1 ′. The N ⁺ electrodes 304 and 304 ′ form a PN junction with the semiconductor substrate 300 to form charge storage capacitors 3 and 3 ′. The transfer gates 303 and 303 ′ constitute the gates of the transfer switches 2 and 2 ′. The control gate electrode 1001 constitutes the gate of the switch 17. The photodiodes 1 and 1 ′ are formed in regions adjacent to each other with the control gate electrode 1001, the transfer gates 303 and 303 ′, and the N ⁺ electrodes 304 and 304 ′ interposed therebetween.

A control gate electrode 1001 constituting a MOS insulated gate made of an insulating film formed on the semiconductor substrate 300 is disposed between the two N ⁺ electrodes 304 and 304 ′.

  The entire device is covered with a protective layer 305 made of a silicon nitride film, and a light shielding metal 306 is disposed in the protective layer 305 between the two photodiodes 1 and 1 ′.

  On-chip lenses 307 and 307 'for increasing the amount of collected light are formed above the photodiodes 1 and 1'.

  Further, a second protective layer 308 is formed on the on-chip lenses 307 and 307 '. On the second protective layer 308, a polarizing filter 309 that is located above the photodiode 1 and transmits only polarized light having an oscillating electric field perpendicular to the paper surface, and a polarizing filter 309 that is located above the photodiode 1 ′. A polarizing filter 309 ′ that transmits only polarized light that is orthogonal to the transmitted polarized light is formed.

  Both the polarizing filters 309 and 309 'are formed by forming a grating in which copper, which is a wiring metal, is arranged for 10 periods with a width of 0.25 μm and an interval of 0.25 μm. At this time, the directions of the gratings are orthogonal between the polarizing filters 309 and 309 ′, and have transmission characteristics with respect to the orthogonally polarized light.

  In the photoelectric conversion device, the photodiodes 1 and 1 ′ have the same characteristics, and the photoelectric conversion device has a configuration that is symmetric with respect to the control gate electrode 1001. Accordingly, the circuit configuration of the photoelectric conversion device can be easily formed as an integrated circuit on the semiconductor substrate 300, and the manufacturing process is simplified, so that the cost of the photoelectric conversion device can be reduced.

  The photoelectric conversion device described in this embodiment can perform charge-voltage conversion with higher sensitivity than the photoelectric conversion device of the first embodiment. The reason is shown below.

FIG. 13 is a diagram showing the control gate voltage (V _G ) dependency of the control gate electrode 1001 on the capacitance (C _G ) between the N ⁺ electrodes 304 and 304 ′. This characteristic is obtained when N ⁺ polysilicon is used as the control gate electrode 1001, the impurity concentration of the semiconductor substrate 300 is 1 × 10 ¹⁵ cm ^−3, and the thickness of the oxide film is set to 5 to 10 nm. is there.

When the control gate voltage becomes lower than the flat band voltage (V _FB = −0.4 V), the depletion layer under the control gate electrode 1001 of the semiconductor substrate 300 spreads, and the capacitance between the N ⁺ electrodes 304 and 304 ′ is extremely small. Become. For example, if V _G = −2 V, when a flat band voltage is applied, which is a capacity substantially equivalent to the charge storage capacitors 3 and 3 ′ in the photoelectric conversion device of the first embodiment that does not have the control gate electrode 1001. Capacitance C _G = about 1/10 (C _G = 20 fF) of 200 fF. Accordingly, V _G = −2 V is applied in advance to the control gate electrode 1001, and the voltage corresponding to the difference between the accumulated charges of the charge storage capacitors 3 and 3 ′ is read out from the differential amplifier circuit 6, whereby each photodiode When the amount of charge generated at 1 and 1 ′ is different, the voltage between the capacitor terminals 5 and 5 ′ is 10 times as compared with the case where the gate bias is V _G = V _FB . That is, 10 times higher sensitivity can be realized. Note that the flat band voltage varies depending on the material of the control gate electrode 1001, the impurity concentration of the semiconductor substrate 300, and the thickness of the oxide film between the control gate electrode 1001 and the semiconductor substrate 300.

Further, by setting the control gate voltage to a voltage equal to or higher than the flat band voltage (V _G ≧ V _FB ), an inversion layer below the control gate electrode 1001 of the semiconductor substrate 300 is formed, the depletion layer is narrowed, and the N ⁺ electrode The capacitor between 304 and 304 ′ is short-circuited, and the accumulated charge is discharged. That is, by controlling only one control gate voltage, it is possible not only to increase the signal sensitivity as variable, but also to realize a reset operation, so that the element area can be greatly reduced.

(Fifth embodiment)
A photoelectric conversion device having a function equivalent to that of the photoelectric conversion device (see FIG. 10) shown in the fourth embodiment can be realized by another configuration using a PN junction gate instead of a MOS gate.

  FIG. 14 schematically shows a cross section passing through the photodiodes 1, 1 ′ of the photoelectric conversion unit 11 of the photoelectric conversion device according to the present embodiment.

A PN junction composed of a P layer 301 and an N layer 302 formed therebelow is formed on the outermost surface of the semiconductor substrate 300. This PN junction constitutes the photodiode 1. Similarly, the PN junction composed of the P layer 301 ′ and the N layer 302 ′ constitutes the photodiode 1 ′. The N ⁺ electrodes 304 and 304 ′ form a PN junction with the semiconductor substrate 300 to form charge storage capacitors 3 and 3 ′. The transfer gates 303 and 303 ′ constitute the gates of the transfer switches 2 and 2 ′. The control gate electrode 1201 constitutes the gate of the switch 17. The photodiodes 1 and 1 ′ are formed in regions adjacent to each other with the control gate electrode 1201, the transfer gates 303 and 303 ′, and the N ⁺ electrodes 304 and 304 ′ interposed therebetween.

On the surface of the semiconductor substrate 300 between the two N ⁺ electrodes 304 and 304 ′, a control gate electrode 1201 constituting a PN junction gate is disposed.

  The entire device is covered with a protective layer 305 made of a silicon nitride film, and a light shielding metal 306 is disposed in the protective layer 305 between the two photodiodes 1 and 1 ′.

  On-chip lenses 307 and 307 'for increasing the amount of collected light are formed above the photodiodes 1 and 1'.

  Further, a second protective layer 308 is formed on the on-chip lenses 307 and 307 '. On the second protective layer 308, a polarizing filter 309 that is located above the photodiode 1 and transmits only polarized light having an oscillating electric field perpendicular to the paper surface, and a polarizing filter 309 that is located above the photodiode 1 ′. A polarizing filter 309 ′ that transmits only polarized light that is orthogonal to the transmitted polarized light is formed.

  Both the polarizing filters 309 and 309 'are formed by forming a grating in which copper, which is a wiring metal, is arranged for 10 periods with a width of 0.25 μm and an interval of 0.25 μm. At this time, the directions of the gratings are orthogonal between the polarizing filters 309 and 309 ′, and have transmission characteristics with respect to the orthogonally polarized light.

  In the photoelectric conversion device, the photodiodes 1 and 1 ′ have the same characteristics, and the photoelectric conversion device has a configuration that is symmetric with respect to the control gate electrode 1201. Accordingly, the circuit configuration of the photoelectric conversion device can be easily formed as an integrated circuit on the semiconductor substrate 300, and the manufacturing process is simplified, so that the cost of the photoelectric conversion device can be reduced.

  The photoelectric conversion device described in this embodiment can perform charge-voltage conversion with higher sensitivity than the photoelectric conversion device of the first embodiment. The reason is shown below.

FIG. 15 is a diagram showing the control gate voltage (V _G ) dependency of the control gate electrode 1201 on the capacitance (C _G ) between the N ⁺ electrodes 304 and 304 ′.

When the control gate voltage becomes negative, the depletion layer under the control gate electrode 1201 spreads, and the capacitance between the N ⁺ electrodes 304 and 304 ′ becomes a very small value. For example, if V _G = −2 V, the capacitance when no voltage is applied, which is approximately the same as the charge storage capacitance 3, 3 ′ in the photoelectric conversion device of the first embodiment that does not have the control gate electrode 1201. approximately 1/10 (C _G = 20fF) of C _G = 200 fF. Accordingly, V _G = −2 V is applied in advance to the control gate electrode 1201, and a voltage corresponding to the difference between the accumulated charges of the charge storage capacitors 3 and 3 ′ is read out from the differential amplifier circuit 6, whereby each photodiode When the amount of charge generated at 1 and 1 ′ is different, the voltage between the capacitor terminals 5 and 5 ′ is 10 times that in the case where the gate bias is V _G = V _FB . That is, 10 times higher sensitivity can be realized.

Further, by setting the control gate voltage to a value of 0 V or higher (V _G ≧ 0 V), a high concentration majority carrier layer is formed under the control gate electrode 1201 of the semiconductor substrate 300, the depletion layer is narrowed, and the N ⁺ electrode The capacitor between 304 and 304 ′ is short-circuited, and the accumulated charge is discharged. That is, by controlling only one control gate voltage, it is possible not only to increase the signal sensitivity as a variable, but also to realize a reset operation, thereby achieving a significant reduction in element area.

(Sixth embodiment)
The photoelectric conversion device having a function equivalent to that of the photoelectric conversion device (see FIG. 1) described in the first embodiment can also be realized by another configuration.

  FIG. 16 schematically shows a cross section passing through the photodiodes 1, 1 ′ of the photoelectric conversion unit 11 of the photoelectric conversion device according to the present embodiment.

A PN junction composed of a P layer 301 and an N layer 302 formed therebelow is formed on the outermost surface of the semiconductor substrate 300. This PN junction constitutes the photodiode 1. Similarly, the PN junction composed of the P layer 301 ′ and the N layer 302 ′ constitutes the photodiode 1 ′. The N ⁺ electrodes 304 and 304 ′ form a PN junction with the semiconductor substrate 300 to form charge storage capacitors 3 and 3 ′. The transfer gates 303 and 303 ′ constitute the gates of the transfer switches 2 and 2 ′. The photodiodes 1 and 1 ′ are formed in regions adjacent to each other with the transfer gates 303 and 303 ′ and the N ⁺ electrodes 304 and 304 ′ interposed therebetween.

  The entire device is covered with a protective layer 305 made of a silicon nitride film, and a light shielding metal 306 is disposed in the protective layer 305 between the two photodiodes 1 and 1 ′.

  On-chip lenses 307 and 307 'for increasing the amount of collected light are formed above the photodiodes 1 and 1'.

  Further, a second protective layer 308 is formed on the on-chip lenses 307 and 307 '. On the second protective layer 308, located above the photodiode 1, a narrowband filter 1409 having a peak at a wavelength of 790 nm and a transmission band having a half-value width of 20 nm, and located above the photodiode 1 ′ and peaking at a wavelength of 810 nm And a narrow band filter 1409 ′ having a transmission band with a half width of 20 nm. That is, narrow-band filters 1409 and 1409 ′ that transmit incident light according to different characteristics (wavelength transmission characteristics) with an intensity corresponding to a specific physical quantity of the incident light above the photodiodes 1 and 1 ′, respectively. Is formed.

  In the photoelectric conversion device, the photodiodes 1 and 1 ′ have the same characteristics, and the photoelectric conversion device has a configuration that is symmetric with respect to the light shielding metal 306. Accordingly, the circuit configuration of the photoelectric conversion device can be easily formed as an integrated circuit on the semiconductor substrate 300, and the manufacturing process is simplified, so that the cost of the photoelectric conversion device can be reduced.

  FIGS. 17A and 17B are diagrams showing the transmission characteristics of the narrowband filters 1409 and 1409 ′, respectively.

  Narrowband filters 1409 and 1409 'both have a maximum transmittance of 0.6 and a half width of 20 nm. Such narrow band filters 1409, 1409 'are described in, for example, the document E.I. Hecht, “Optics”, 4th ed. , P. 425-430, Addison Wesley, San Francisco (2002). An interference filter using a multilayer dielectric thin film as shown in FIG. 2 can be separately produced for each photodiode 1, 1 'by photolithography.

  When light having a wavelength of 800 nm is incident on the photoelectric conversion device of this embodiment, the transmission characteristics of the narrowband filters 1409 and 1409 ′ are symmetric with respect to the transmission peak. The generated photocurrents are also equal, so that no net charge is stored in the charge storage capacitors 3, 3 ′, and no signal output is generated.

  On the other hand, when the wavelength of the incident light is shifted by, for example, 5 nm to the low wavelength side, the amount of light transmitted through the narrowband filter 1409 increases and the amount of light transmitted through the narrowband filter 1409 ′ decreases, and the photodiode 1 accordingly. The current generated at 1 ′ increases or decreases, and the charge corresponding to the difference is accumulated as a net charge in the charge storage capacitors 3, 3 ′. By amplifying the difference, a signal output can be obtained. Therefore, the photoelectric conversion device of this embodiment functions as a wavelength shift monitor that detects a modulation amount related to the wavelength of incident light.

  FIG. 18 is a diagram schematically illustrating a configuration example of a THz imaging system using the photoelectric conversion device of the present embodiment.

  An ultrashort pulsed light train having a pulse width of 100 fs is generated from the light source 1601 at a frequency of 1 kHz, and is separated into pump light 1603 and probe light 1604 by a polarizing beam splitter 1602. The pump light 1603 is incident on the ZnTe crystal [110] plane constituting the THz electromagnetic wave emitter 1606 through the optical delay line 1605, and a THz electromagnetic wave 1607 is generated.

  After the traveling direction of the probe light 1604 is adjusted by the mirror 1608, the probe light 1604 enters the silicon mirror 1609 made of a silicon wafer and shares the optical axis with the THz electromagnetic wave 1607 transmitted through the silicon mirror 1609. The light enters the electro-optic modulator 1610 made of a superlattice in which arsenic thin films are alternately stacked.

  The electro-optic modulator 1610 is provided with the photoelectric conversion device described in this embodiment and an image sensor 1611 that images the probe light 1604 is disposed. At this time, when the THz electromagnetic wave 1607 does not enter the electro-optic modulator 1610 at the same time as the probe light 1604, the wavelength of the transmitted light does not change, so that an equal amount of light is transmitted to each of the internal photodiodes 1, 1 ′. Since the incident and generated photocurrents are equal, no signal is detected.

  On the other hand, when the THz electromagnetic wave pulse and the probe light pulse are simultaneously incident on the electro-optic modulator 1610, that is, when the two pulses are synchronized, a specific physical quantity of the probe light 1604 is determined according to the electric field of the THz electromagnetic wave 1607. , And the wavelength of the probe light 1604 after passing through the electro-optic modulator 1610 is shifted by 5 nm to the longer wavelength side compared to the case where the wavelength is asynchronous with the THz electromagnetic wave 1607, and the amount of light incident on the photodiode 1 ′ increases. Since the amount of light incident on the diode 1 is reduced, different photocurrents are generated in the photodiodes 1 and 1 ′, charges are accumulated in the charge storage capacitors 3 and 3 ′, and a voltage signal, that is, a THz electromagnetic wave detection signal is output. Is done.

  Therefore, it is possible to detect THz electromagnetic waves without saturating the photodiode that is the problem.

  In addition, this configuration in which different wavelength filters are provided in the respective photodiodes 1 and 1 ′ directly detects the color difference information from the incident light separately from the use for detecting the modulation degree related to the wavelength of the incident light as described above. It can also be used to

  Furthermore, in the configuration in which neither the polarization filter nor the wavelength filter is provided for each photodiode 1, 1 ′, the difference in the amount of light received by the two photodiodes 1, 1 ′ can be directly detected, so that it can be applied to edge detection in an image. It is also possible to do.

  As described above, the photoelectric conversion device of the present invention has been described based on the embodiment, but the present invention is not limited to this embodiment. The present invention includes various modifications made by those skilled in the art without departing from the scope of the present invention.

  The photoelectric conversion device of the present invention can be used for security inspection devices, food inspection devices, atmospheric sensors, medical diagnosis devices, and the like.

It is a circuit diagram which shows an example of a structure of the photoelectric conversion apparatus which concerns on the 1st Embodiment of this invention. 6 is a timing chart of main signals related to the operation of the photoelectric conversion device of the embodiment. It is typical sectional drawing of the photoelectric conversion part of the photoelectric conversion apparatus of the embodiment. It is the layout figure which looked at the photoelectric conversion part of the photoelectric conversion apparatus of the embodiment from the upper part of the apparatus. It is a figure which shows typically the structural example of the THz imaging system using the photoelectric conversion apparatus of the embodiment. It is a block diagram which shows an example of a functional structure of the image sensor using the photoelectric conversion apparatus of the embodiment. It is a figure which shows typically the modification of the THz imaging system using the photoelectric conversion apparatus of the embodiment. It is typical sectional drawing of the photoelectric conversion part of the photoelectric conversion apparatus of the 2nd Embodiment of this invention. It is typical sectional drawing of the photoelectric conversion part of the photoelectric conversion apparatus of the 3rd Embodiment of this invention. It is a circuit diagram which shows an example of a structure of the photoelectric conversion apparatus which concerns on the 4th Embodiment of this invention. 6 is a timing chart of main signals related to the operation of the photoelectric conversion device of the embodiment. It is typical sectional drawing of the photoelectric conversion part of the photoelectric conversion apparatus of the embodiment. It is a diagram showing a control gate voltage (V _G) dependence of the control gate electrode in the capacitance (C _G) between the N ⁺ electrode of the photoelectric conversion device of the same embodiment. It is typical sectional drawing of the photoelectric conversion part of the photoelectric conversion apparatus of the 5th Embodiment of this invention. It is a diagram showing a control gate voltage (V _G) dependence of the control gate electrode in the capacitance (C _G) between the N ⁺ electrode of the photoelectric conversion device of the same embodiment. It is typical sectional drawing of the photoelectric conversion part of the photoelectric conversion apparatus of the 6th Embodiment of this invention. (A) It is a figure which shows the transmission characteristic of a narrow-band filter. (B) It is a figure which shows the transmission characteristic of a narrow-band filter. It is a figure which shows typically the structural example of the THz imaging system using the photoelectric conversion apparatus of the embodiment. It is a schematic diagram which shows the structural example of the THz imaging system which concerns on a prior art. It is a figure explaining the characteristic of a polarizing filter. It is a circuit diagram which shows an example of the pixel circuit of the general image sensor which concerns on a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 1 ', 1801 Photodiode 2, 2' Transfer switch 3, 3 'Charge storage capacity 4 Control circuit 5, 5' Capacity terminal 6 Differential amplifier circuit 7, 7 ', 17, 610 Switch 8 Reset signal generation circuit 19 reset control circuit 300 semiconductor substrate 301, 301 ′ P layer 302, 302 ′ N layer 303, 303 ′ transfer gate 304, 304 ′ N ⁺ electrode 305 protective layer 306 light shielding metal 307, 307 ′ on-chip lens 308 second protection Layer 309, 309 ′ Polarizing filter 405, 405 ′ Wiring 407, 407 ′ Reset gate 501, 701, 1601, 1701 Light source 502, 702, 1602, 1702 Polarizing beam splitter 503, 703, 1603, 1703 Pump light 504, 704, 1604 , 1704 Probe light 505, 705, 160 , 1705 Optical delay line 506, 706, 1606, 1706 THz electromagnetic wave emitter 507, 707, 1607, 1707 THz electromagnetic wave 508, 710, 1608, 1710 Mirror 509, 712, 1609, 1712 Silicon mirror 510, 713, 1610, 1713 Electro-optical Modulator 511, 714 1/4 wavelength plate 512, 715, 1611, 1716 Image sensor 601 Vertical scanning circuit 602 Horizontal scanning circuit 603 Row data selection line 604 Column data readout line 606, 607, 609 Terminal 708, 1708 Device under test 709 , 1709 Polyethylene lens 804, 804 ', 901, 901' Nickel silicide electrode 902, 902 'High dielectric constant film 903 Upper electrode 1001, 1201 Control gate electrode 1409, 1409' Bandpass filter 711,1711 beam expander 1714 phase plate 1715 polarizer 1717 synchronizing circuit 1718 image processing circuit 1802 transfer transistor 1803 forward pulse generating circuit 1804 volume 1805 amplifier circuit 1806 reset pulse generator circuit 1807 reset transistor

Claims (17)

A first photodiode;
A second photodiode having a first polarity terminal connected to a first polarity terminal of the first photodiode;
A first capacitor inserted between the first photodiode and the second photodiode;
A second capacitor inserted between the first capacitor and the second photodiode;
A differential amplifier circuit having one input terminal connected between the first photodiode and the first capacitor, and the other input terminal connected between the second photodiode and the second capacitor; A photoelectric conversion device comprising: The first photodiode includes a first conductivity type first region and a second conductivity type second region which are formed on the surface of a first conductivity type semiconductor substrate and are in contact with each other.
The second photodiode includes a third region of a first conductivity type and a fourth region of a second conductivity type that are formed on the surface of the semiconductor substrate and are in contact with each other.
The shapes of the first region and the third region are equal,
The shapes of the second region and the fourth region are equal,
Either the first region or the second region is connected to a first electrode formed on the surface of the semiconductor substrate;
2. The photoelectric conversion device according to claim 1, wherein one of the third region and the fourth region is connected to a second electrode formed on a surface of the semiconductor substrate. The first capacitor and the second capacitor are each a junction capacitor of a PN junction,
The first electrode is a second conductivity type electrode of the first capacitance;
The photoelectric conversion device according to claim 2, wherein the second electrode is a second conductivity-type electrode of the second capacitor. Each of the first capacitor and the second capacitor is a Schottky junction capacitance,
The first electrode is an electrode of the first capacitor that is Schottky connected to the semiconductor substrate;
The photoelectric conversion device according to claim 2, wherein the second electrode is an electrode of the second capacitor that is Schottky connected to the semiconductor substrate. The photoelectric conversion device according to claim 2, wherein each of the first capacitor and the second capacitor has a structure in which a dielectric is sandwiched between conductors. 6. The photoelectric conversion device according to claim 5, wherein a film thickness of the first capacitor dielectric is equal to a film thickness of the second capacitor dielectric. The photoelectric conversion device further includes a discharge switch that is inserted between the first capacitor and the second capacitor and discharges charges of the first capacitor and the second capacitor. The photoelectric conversion device described. The discharge switch is a MOSFET;
The photoelectric conversion device according to claim 7, wherein the MOSFET includes a gate electrode formed on the semiconductor substrate so as to be positioned between the first electrode and the second electrode. The discharge switch is a junction FET,
The said junction type FET has the gate electrode of the 2nd conductivity type formed in the said semiconductor substrate surface so that it may be located between the said 1st electrode and the said 2nd electrode. Photoelectric conversion device. The photoelectric conversion device further includes a first transfer switch inserted between the first photodiode and the first capacitor, and a second transfer switch inserted between the second photodiode and the second capacitor. The photoelectric conversion apparatus according to claim 1, further comprising a transfer switch. The first photodiode and the second photodiode have equivalent characteristics;
The first capacitor and the second capacitor have equivalent characteristics,
The first photodiode and the second photodiode are provided with filters that transmit incident light according to different characteristics with an intensity corresponding to a specific physical quantity of the incident light. Item 11. The photoelectric conversion device according to any one of Items 1 to 10. The photoelectric conversion device according to claim 11, wherein the filter is a polarization filter having different polarization transmission characteristics. The photoelectric conversion device according to claim 11, wherein the filter is a wavelength filter having different wavelength transmission characteristics. A plurality of pixel circuits arranged two-dimensionally, each having the photoelectric conversion device according to any one of claims 1 to 13,
An image sensor comprising: a readout circuit that reads out an output signal of the photoelectric conversion device included in each pixel circuit. A method for driving a photoelectric conversion device according to claim 8 or 9, wherein
In a state where a bias voltage is applied to the gate electrode so that a depletion layer is expanded in a region below the gate electrode of the semiconductor substrate, a voltage corresponding to a difference between charges accumulated in the first capacitor and the second capacitor is applied. Reading from the differential amplifier circuit. A method for driving a photoelectric conversion device. A method for driving a photoelectric conversion device according to claim 8 or 9, wherein
The charge accumulated in the first capacitor and the second capacitor is discharged in a state where a bias voltage is applied to the gate electrode so that a depletion layer is narrowed in a region below the gate electrode of the semiconductor substrate. To drive the photoelectric conversion device. An electromagnetic wave source that generates a THz electromagnetic wave having a frequency in a range of 0.1 THz to 10 THz;
A light source that generates probe light;
A superimposing optical element that superimposes the THz electromagnetic wave after being transmitted or reflected on the subject with the probe light;
An electro-optic modulation element that receives the superimposed THz electromagnetic wave and probe light and modulates a specific physical quantity of the probe light according to the electric field of the THz electromagnetic wave;
An imaging system comprising: the image sensor according to claim 14 that images the modulated probe light.

Images