JP2005268384A - Photoelectric transducer film-laminated solid-state imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a photoelectric transducer film-laminated solid-state imaging device capable of efficiently extracting optical charge from a photoelectric transducer film. <P>SOLUTION: In a photoelectric transducer film-laminated solid-state imaging device, a photoelectric transducer film 23 is laminated on a semiconductor substrate with a signal reading circuit formed thereon, and a signal corresponding to the quantity of incident light photoelectric-transduced by the photoelectric transducer film 23 is read to the outside by the signal reading circuit. The photoelectric transducer film 23 includes a photoelectric transducer layer wherein a first quantum dots 43 contributed to photoelectric transduction and a second quantum dots 44 having a band gap greater than that of the first quantum dots 43 are substantially uniformly distributed. Appropriately, InN quantum dots are used as the first quantum dots 43, and GaN quantum dots are used as the second quantum dots 44. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a photoelectric conversion film stacked solid-state imaging device configured by stacking a photoelectric conversion film on a semiconductor substrate on which a signal readout circuit is formed.

  As a prototype element of the photoelectric conversion film laminated solid-state imaging device, for example, there is one described in Patent Document 1 below. In this solid-state imaging device, three photosensitive layers are stacked on a semiconductor substrate, and red (R), green (G), and blue (B) electrical signals detected in each photosensitive layer are transmitted to the surface of the semiconductor substrate. Reading is performed by the MOS circuit formed in the circuit.

  A solid-state imaging device having such a configuration has been proposed in the past. Thereafter, a large number of light receiving portions (photodiodes) are integrated on the surface portion of the semiconductor substrate, and red (R), green (G), and blue are integrated on each light receiving portion. The CCD type image sensor and CMOS type image sensor in which the color filters of each color (B) are laminated have advanced remarkably, and now, an image sensor in which millions of light receiving parts (pixels) are integrated on one chip is used as a digital still camera. It comes to be installed.

  However, the CCD type image sensor and the CMOS type image sensor have progressed to their limits, and the aperture size of one light receiving part is about 2 μm, which is close to the wavelength order of incident light, and the manufacturing yield is poor. Faced with the problem.

  In addition, the upper limit of the amount of photocharge accumulated in one miniaturized light receiving portion is as small as about 3000 electrons, which makes it difficult to express 256 gradations neatly. For this reason, it is difficult to expect an image sensor more than the current type in the CCD type or the CMOS type in terms of image quality and sensitivity.

  Therefore, as a solid-state imaging device that solves these problems, the solid-state imaging device proposed in Patent Document 1 has attracted attention, and image sensors described in Patent Document 2 and Patent Document 3 are newly proposed. It is becoming like this.

  In the image sensor described in Patent Document 2, ultrafine particles of silicon are dispersed in a medium to form a photoelectric conversion layer, and a plurality of photoelectric conversion layers having different ultrafine particle sizes are stacked on a semiconductor substrate. In each photoelectric conversion layer, electrical signals corresponding to the received light amounts of red, green and blue are generated.

  The same applies to the image sensor described in Patent Document 3, in which three nanosilicon layers having different particle diameters are stacked on a semiconductor substrate, and red, green, and blue electrical signals detected by each nanosilicon layer are detected. Is read out to the storage diode formed on the surface portion of the semiconductor substrate.

JP 58-103165 A Japanese Patent No. 3405099 Japanese Patent Laid-Open No. 2002-83946

  When silicon is used as the ultrafine particles used in the photoelectric conversion layer, the electron-hole pairs generated by receiving light are recombined in a short time on the surface of the ultrafine particles, and the charge is extracted before recombination. Therefore, there is a problem that the performance as a solid-state imaging device is not sufficient.

  In order to put the photoelectric conversion film laminated solid-state imaging device into practical use, it is necessary to solve the problem of what material the photoelectric conversion film is formed of.

  An object of the present invention is to provide a photoelectric conversion film stacked solid-state imaging device capable of efficiently taking out photoelectric charges from a photoelectric conversion film.

  The photoelectric conversion film stack type solid-state imaging device of the present invention has a photoelectric conversion film stacked on a semiconductor substrate on which a signal readout circuit is formed, and a signal corresponding to the amount of incident light photoelectrically converted by the photoelectric conversion film is the signal readout. In the photoelectric conversion film stacked solid-state imaging device read out to the outside by the circuit, the first quantum dots contributing to photoelectric conversion and the second quantum dots having a larger band gap than the first quantum dots are substantially uniformly dispersed. The photoelectric conversion layer has a photoelectric conversion layer.

  With this configuration, recombination of hole-electron pairs generated when light is incident on the first quantum dots is suppressed, and the extraction of photocharges from the photoelectric conversion film is facilitated.

  The photoelectric conversion film laminated solid-state imaging device of the present invention is characterized in that the photoelectric conversion film is configured by laminating a hole transport layer composed of the second quantum dots on the photoelectric conversion layer.

  With this configuration, the hole can be easily transported, and the photocharge can be easily taken out from the photoelectric conversion film.

  The photoelectric conversion film laminated solid-state imaging device of the present invention is characterized in that a p-type impurity is doped in the second quantum dots of the hole transport layer.

  With this configuration, the hole can be easily transported, and the photocharge can be more easily taken out from the photoelectric conversion film.

  The photoelectric conversion film stacked solid-state imaging device of the present invention is characterized in that an n-type impurity is doped in the second quantum dots of the photoelectric conversion layer.

  With this configuration, electrons can be transported more easily, and photocharges can be more easily taken out from the photoelectric conversion film.

  In the photoelectric conversion film stacked solid-state imaging device of the present invention, the second quantum dot of the hole transport layer is doped with p-type impurities, and the second quantum dot on the hole transport layer side of the photoelectric conversion layer is doped. A p-type impurity is doped.

  With this configuration, the hole can be easily transported, and the photocharge can be more easily taken out from the photoelectric conversion film.

  The photoelectric conversion film laminated solid-state imaging device of the present invention is characterized in that an n-type impurity is doped in the second quantum dots on the opposite side of the photoelectric conversion layer from the hole transport layer.

  With this configuration, electrons can be transported more easily, and photoelectric charges can be more efficiently taken out from the photoelectric conversion film.

  In the photoelectric conversion film stacked solid-state imaging device of the present invention, the first quantum dots are InN, and the second quantum dots are GaN.

  The use of InN, in which recombination is suppressed from silicon, improves the performance as a solid-state imaging device, and the use of GaN facilitates the transport of holes and electrons.

  The photoelectric conversion film laminated solid-state imaging device of the present invention is characterized in that the photoelectric conversion film sandwiched between two transparent electrode films is laminated in three layers via a transparent insulating film.

  With this configuration, a color image can be captured.

  In the photoelectric conversion film laminated solid-state imaging device of the present invention, among the three layers of photoelectric conversion films, the maximum value of light absorption of the photoelectric conversion film of the first layer is a wavelength of 400 to 500 nm, the second layer The photoelectric conversion film is provided in each photoelectric conversion film so that the maximum value of light absorption of the photoelectric conversion film is a wavelength of 500 to 560 nm, and the maximum value of light absorption of the photoelectric conversion film of the third layer is a wavelength of 560 to 640 nm. The average particle diameter of the first quantum dots is determined.

  With this configuration, it is possible to extract image data divided into three primary colors of red (R), green (G), and blue (B).

  The photoelectric conversion film laminated solid-state imaging device of the present invention is characterized in that the photoelectric conversion film sandwiched between two transparent electrode films is stacked in four layers via a transparent insulating film.

  With this configuration, various signal processing is possible, and a color image with good color reproducibility can be captured.

  In the photoelectric conversion film stacked solid-state imaging device of the present invention, the maximum value of light absorption of the first photoelectric conversion film among the four photoelectric conversion films is a wavelength of 420 to 480 nm, the second layer. The maximum value of light absorption of the photoelectric conversion film is 480 to 520 nm, the maximum value of light absorption of the photoelectric conversion film of the third layer is 520 to 560 nm, and the maximum value of the photoelectric conversion film of the fourth layer is The average particle diameter of the first quantum dots provided in each photoelectric conversion film is determined so that the maximum value of light absorption is a wavelength of 560 to 620 nm.

  With this configuration, it is possible to capture a color image with further excellent color reproducibility.

  In the photoelectric conversion film stacked solid-state imaging device of the present invention, the signal amount detected by the photoelectric conversion film of the second layer is calculated from the signal amount detected by the photoelectric conversion film of the fourth layer. The red signal amount is obtained by subtraction.

  With this configuration, red reproduction according to human visibility is possible.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the photoelectric conversion film laminated | stacked solid-state image sensor which can take out photoelectric charge (signal charge) from a photoelectric conversion film efficiently.

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

  FIG. 1 is a schematic cross-sectional view of one pixel of a photoelectric conversion film stacked solid-state imaging device according to an embodiment of the present invention. In this embodiment, three layers of photoelectric conversion films are stacked to extract an electrical signal corresponding to the three primary colors of red (R), green (G), and blue (B), that is, a configuration for capturing a color image. However, only one layer of the photoelectric conversion film may be provided to capture a monochrome image, for example, a monochrome image.

  In FIG. 1, on the surface portion of a P well layer 1 formed on an n-type silicon substrate 50, a high-concentration impurity region 2 for red signal storage, a MOS circuit 3 for reading red signals, and a green signal storage A high concentration impurity region 4, a green signal readout MOS circuit 5, a blue signal storage high concentration impurity region 6, and a blue signal readout MOS circuit 7 are formed.

  Each MOS circuit 3, 5, 7 is composed of a source and drain impurity region formed on the surface of the semiconductor substrate and a gate electrode formed via a gate insulating film 8. An insulating film 9 is laminated and planarized on the gate insulating film 8 and the gate electrode. In some cases, a light shielding film is formed on the surface of the insulating film 9, but when the light shielding film is formed, an insulating film 10 is further laminated so as to insulate the light shielding film. This is because the light shielding film is often formed of a metal thin film. When the light shielding film is not provided at this location, the illustrated insulating films 9 and 10 may be integrated.

  Signals corresponding to the signal charges accumulated in the high-concentration impurity regions 2, 4, and 6 for color signal accumulation described above are read out by the MOS circuits 3, 5, and 7, and although not shown, Although it is taken out by the formed readout electrode, its configuration is the same as that of a conventional CMOS image sensor.

  In this example, a signal corresponding to the amount of signal charge is read out by a MOS circuit formed on a semiconductor substrate. However, the charge stored in the high-concentration impurity regions 2, 4, and 6 for storing color signals is used as a conventional CCD. Similar to the type image sensor, it may be configured to move along the vertical transfer path and read out along the horizontal transfer path.

  The above configuration is manufactured by a conventional CCD image sensor or CMOS image sensor semiconductor process, and a photoelectric conversion film stacked solid-state imaging device is manufactured by adding the configuration described below.

  A transparent electrode film 11 is formed on the insulating film 10 shown in FIG. The transparent electrode film 11 is electrically connected to the high concentration impurity region 2 for red signal accumulation by the electrode 12. This electrode 12 is electrically insulated from areas other than the transparent electrode film 11 and the high concentration impurity region 2. Then, a red detection photoelectric conversion film 13 is formed on the transparent electrode film 11, and a transparent electrode film 14 is further formed on the photoelectric conversion film 13. That is, the photoelectric conversion film 13 is sandwiched between the pair of transparent electrode films 11 and 14. Note that the electrode film 11 which is the lowermost layer may be made opaque to be used as a light shielding film.

  A transparent insulating film 15 is formed on the transparent electrode film 14, and a transparent electrode film 16 is formed on the transparent insulating film 15. The transparent electrode film 16 is electrically connected to the high-concentration impurity region 4 for storing the green signal by the electrode 17. This electrode 17 is electrically insulated from areas other than the transparent electrode film 16 and the high concentration impurity region 4. A green color detection photoelectric conversion film 18 is formed on the transparent electrode film 16, and a transparent electrode film 19 is formed on the photoelectric conversion film 18. That is, the photoelectric conversion film 18 is sandwiched between the pair of transparent electrode films 16 and 19.

  A transparent insulating film 20 is formed on the transparent electrode film 19, and a transparent electrode film 21 is formed on the transparent insulating film 20. The transparent electrode film 21 is electrically connected to the high concentration impurity region 6 for storing the blue signal by the electrode 22. This electrode 22 is electrically insulated from areas other than the transparent electrode film 21 and the high concentration impurity region 6. A blue color photoelectric conversion film 23 is formed on the transparent electrode film 21, and a transparent electrode film 24 is formed on the photoelectric conversion film 23. That is, the photoelectric conversion film 23 is sandwiched between the pair of transparent electrode films 21 and 24.

A transparent insulating film 25 is provided as the uppermost layer. In this embodiment, a light shielding film 26 for limiting the incident range of incident light to the pixel is provided in the transparent insulating film 25. The reason why the light shielding film 26 is provided in the uppermost layer in this embodiment is to further reduce the color mixture between pixels. As the homogeneous transparent electrode film, tin oxide (SnO ₂ ), titanium oxide (TiO ₂ ), indium oxide (InO ₂ ), and indium oxide-tin (ITO) thin film are used, but the present invention is not limited thereto. Examples of the formation method include a laser ablation method and a sputtering method.

  The structures of the photoelectric conversion films 23, 18, and 13 are basically the same, but the particle diameters of InN quantum dots provided in the films are different. The InN quantum dots in the blue color photoelectric conversion film 23 have the smallest particle size, then the InN quantum dots in the green color detection photoelectric conversion film 18 have an intermediate particle size, and the red color detection photoelectric conversion. The particle size of InN quantum dots in the film 13 is the largest, and in any case, the particle size is on the order of nanometers.

  These particle sizes are selected so that light absorption at the corresponding wavelength is increased and a large number of electron-hole pairs are generated. That is, the red detection is performed so that the maximum value of light absorption is 400 to 500 nm in the photoelectric conversion film 23 for blue detection, and the maximum value of light absorption is 500 to 560 nm in the photoelectric conversion film 18 for green detection. In the photoelectric conversion film 13 for use, the particle size is selected to be 560 to 640 nm.

  The embodiment shown in FIG. 1 is an example of a photoelectric conversion film stacked solid-state imaging device that detects three primary colors of red, green, and blue. However, a configuration that can detect four colors is also possible. FIG. 2 is a schematic cross-sectional view of one pixel of a photoelectric conversion film stacked solid-state imaging device that detects four colors, and is an intermediate color (GB: emerald) of green (G) and blue (B) with respect to the configuration of FIG. A layer in which a photoelectric conversion film 31 for detecting color) is sandwiched between transparent electrodes 32 and 33 is provided between a green detection layer and a blue detection layer. That is, the photoelectric conversion films 23, 31, 18, and 13 are provided in order from the top in ascending order of the wavelength of light to be detected.

  In this example, the maximum value of light absorption in the photoelectric conversion film 23 is 420 to 480 nm, the maximum value of light absorption in the photoelectric conversion film 31 is 480 to 520 nm, the maximum value of light absorption in the photoelectric conversion film 18 is 520 to 560 nm, The particle size of the quantum dots is determined so that the maximum value of light absorption in the conversion film 13 is 580 to 620 nm.

  Then, a high-concentration impurity region 36 for storing signal charges of intermediate colors is formed in the semiconductor substrate, and an electrode 35 electrically connected to the high-concentration impurity region 36 and the transparent electrode 32 and electrically insulated from other components is provided. A MOS circuit 37 for reading signal charges in the high concentration impurity region 36 is provided on the semiconductor substrate. Naturally, the transparent insulating film 34 is provided between the transparent electrode film 31 and the upper transparent electrode film 21.

  The advantage of detecting an intermediate color having a wavelength of 480 to 520 nm is to correct red according to human visibility. As shown in FIG. 3 as α, β, and γ, human visual sensitivity is negative in green (G), red (R), and blue (B). For this reason, even if color reproduction is performed by detecting only positive sensitivities of R, G, and B with a solid-state imaging device, an image seen by humans cannot be reproduced. Therefore, β having the greatest negative sensitivity, that is, negative red sensitivity is detected by the photoelectric conversion film 31, and the sensitivity to human red is obtained by subtracting this negative sensitivity from the red sensitivity detected by the photoelectric conversion film 13. Can be reproduced.

  FIG. 4 is a circuit diagram of the MOS circuits 3, 5, and 7 shown in FIG. This MOS circuit is composed of three FET elements for each of R, G, and B, and the circuit configuration is the same as the circuit used in a conventional CMOS image sensor. In the solid-state imaging device of FIG. 2, only three FET elements for intermediate colors (GB) are added for one pixel.

  In the conventional CMOS type image sensor, it is necessary to provide a “light-receiving part” on the semiconductor surface. Therefore, when manufacturing these MOS circuits on the semiconductor surface, the light-receiving part must be manufactured in a narrow place because of the widening of the light-receiving part. It was. However, in the photoelectric conversion film laminated solid-state imaging device of the present embodiment, it is not necessary to provide a “light receiving portion” on the semiconductor surface, so that the MOS circuit can be easily manufactured. In addition, since there is room in the wiring space, in FIG. 4, one of R, G, and B is sequentially read while being selected by a select signal, but the wiring connection is such that R, G, and B can be read together. It becomes easy. This is the same even if the readout circuit is not a MOS circuit but a type in which a charge transfer path is provided like a CCD image sensor.

  The structure shown in FIGS. 1 and 2 is for one pixel, but these pixels are provided in an array in the vertical and horizontal directions on the surface side of the semiconductor substrate. However, it is not necessary to divide and stack the photoelectric conversion films according to each pixel, and the photoelectric conversion films can be stacked in a single structure on the entire surface of the semiconductor substrate. One pixel and one pixel can be separated by forming one of the pair of transparent electrodes sandwiching each photoelectric conversion film into one pixel and one pixel.

  When light from a subject is incident on the photoelectric conversion film stacked solid-state imaging device of FIG. 1 or 2 described above, blue light of the incident light is absorbed by the photoelectric conversion film 23 and green light is converted into a photoelectric conversion film. The red light is absorbed by the photoelectric conversion film 13. In the case of FIG. 2, emerald light that is an intermediate color (GB) of blue and green is absorbed by the photoelectric conversion film 31.

  The quantum dots (ultrafine particles) constituting the photoelectric conversion film 23 absorb incident light and generate electron-hole pairs. This electron-hole pair recombines with time and emits blue light. However, when a voltage is applied between the transparent electrodes 24 and 21, before the electrons are recombined, It flows from the transparent electrode 21 through the electrode 22 to the high concentration impurity region 6.

Similarly, electrons corresponding to the green incident light amount generated in the photoelectric conversion film 18 flow into the high concentration impurity region 4, and electrons corresponding to the red incident light amount generated in the photoelectric conversion film 13 are in the high concentration impurity region. 2, electrons corresponding to the amount of incident light of the emerald color flow to the high concentration impurity region 36 (FIG. 2). Then, the MOS circuits 3, 5, 7, and 37 read out the electrons having the charge amount corresponding to each color to the outside.

  FIG. 5 is a schematic cross-sectional view of the photoelectric conversion films 23, 18, 13, and 31. The photoelectric conversion film 23 provided between the transparent electrode films 24 and 21 (18, 13, and 31 is the same, and only the particle size of the quantum dots is different) is the photoelectric conversion layer 41 in the illustrated example. And a hole transport layer 42.

  The photoelectric conversion layer 41 includes InN quantum dots 43 made of ultrafine particles and a material having a larger band gap than InN, in this example, GaN quantum dots 44, and both 43 and 44 are uniformly dispersed. It is configured.

  In the example of FIG. 5, the hole transport layer 42 is provided, but the photoelectric conversion film 23 may be configured only by the photoelectric conversion layer 41. However, it is preferable to provide the hole transport layer 42 because the hole transport efficiency is improved.

  FIG. 6 is a schematic cross-sectional view of another embodiment of the photoelectric conversion film 23 (the same applies to other photoelectric conversion films). The example of FIG. 6 is the same in that InN quantum dots 43 and GaN quantum dots 44 are used, as in the embodiment of FIG. 5, but p-type impurities are used as the GaN quantum dots 44 p of the hole transport layer 42. The difference is that a material doped with n-type impurities is used as the GaN quantum dot 44n of the photoelectric conversion layer 41.

  With this configuration, the hole transport efficiency by the GaN quantum dots 44p is improved and the electron transport efficiency by the GaN quantum dots 44n is also improved. Note that the GaN quantum dots 44 in the photoelectric conversion layer 41 are not necessarily doped with n-type impurities, and GaN quantum dots that are not doped with impurities may be used.

  FIG. 7 is a schematic cross-sectional view of still another embodiment of the photoelectric conversion film 23 (the same applies to other photoelectric conversion films). The example of FIG. 7 is similar to the embodiment of FIG. 5 in that InN quantum dots 43 and GaN quantum dots 44 are used, but p-type impurities are used as the GaN quantum dots 44p of the hole transport layer 42. Further, the GaN quantum dots 44p on the hole transport layer 42 side of the photoelectric conversion layer 41 are also doped with p-type impurities, and the GaN on the anti-hole transport layer side of the photoelectric conversion layer 41 is used. The difference is that a quantum dot 44n is doped with an n-type impurity.

  This configuration also improves the hole transport efficiency due to the GaN quantum dots 44p and the electron transport efficiency due to the GaN quantum dots 44n. The GaN quantum dots 44 on the anti-hole transport layer side, that is, the transparent electrode 21 side in the photoelectric conversion layer 41 do not necessarily need to be doped with n-type impurities, and GaN quantum dots that are not doped with impurities should be used. But you can.

  FIG. 8 is an explanatory diagram of a manufacturing apparatus for manufacturing the photoelectric conversion film shown in FIGS. This manufacturing apparatus includes three vacuum chambers 51, 52, and 53, and a semiconductor substrate 50 on which a photoelectric conversion film is stacked is placed in the vacuum chamber 51. The vacuum chamber 51 is connected to a vacuum pump (not shown) through an exhaust path 51a.

  The vacuum chambers 52 and 53 have the same structure, and both the vacuum chambers 52 and 53 have N2 gas introduction paths 52a and 53a, flow rate adjusting devices 52b and 53b provided in the middle of the introduction paths 52a and 53a, and N2 gas. Exhaust paths 52c and 53c for exhausting the gas, target holders 52d and 53d placed in the chambers 52 and 53, and laser light sources 52e and 53e for irradiating the targets held by the target holders 52d and 53d with pulsed laser light. Prepare.

  The vacuum chamber 51 and the vacuum chambers 52 and 53 are communicated with each other by communication passages 52f and 53f. An orifice is provided in the middle of each communication passage 52f and 53f, and N2 gas is introduced into the chambers 52 and 53. Also, the degree of vacuum in the chamber 51 is maintained at a predetermined degree of vacuum.

  In addition, a mass separation device (not shown) is provided in the communication passages 52f and 53f, and the ultrafine particles emitted from the targets in the chambers 52 and 53 by the pulse laser beam irradiation are first ionized and then accelerated. Finally, when bent by the deflection electrode, only ultrafine particles having a particle diameter within a predetermined range are incident on the semiconductor substrate 50 in the chamber 51.

In such a manufacturing apparatus, after the degree of vacuum in the vacuum chambers 51, 52, 53 is evacuated to 1.0 × 10 ⁻⁹ Torr, N 2 gas is introduced into the chambers 52, 53, and the pressure is set to 0. A predetermined pressure value within the range of 01 to 1.0 Torr is maintained. In this state, the laser light source 52e irradiates the surface of the InN target 55 placed on the target holder 52d in the vacuum chamber 52 with a pulse laser. As a result, a laser ablation phenomenon occurs on the surface of the InN target 55, and InN ions or cluster-like neutral atoms are desorbed from the target 55. The detached clusters collide with the N ₂ gas and decelerate, collide with other clusters and the like, and the particle diameter gradually increases. As a result, ultrafine particles of several nm to several tens of nm are generated.

  A GaN target 56 is placed in the vacuum chamber 53, and ultrafine particles of GaN of several nm to several tens of nm are generated in the same manner as described above. Both InN and GaN may be separated into In and N, and Ga and N in the process of laser ablation. However, both of InN and GaN maintain the stoichiometric composition because laser ablation occurs in a nitrogen gas atmosphere. Be drunk.

  On the other hand, the semiconductor substrate 50 is placed in the vacuum chamber 51. Due to the differential pressure between the vacuum chamber 51 and the vacuum chambers 52 and 53, the InN ultrafine particles or the GaN ultrafine particles enter the vacuum chamber 51 through the communication passages 52f and 53f. Sprayed and deposited on the semiconductor substrate 50.

  For this reason, after forming the transparent electrode 21 of FIG. 5 on the surface of the semiconductor substrate 50 in the vacuum chamber 51, InN ultrafine particles and GaN ultrafine particles having the same particle size are simultaneously sprayed onto the semiconductor substrate 50 from the vacuum chambers 52 and 53. Thus, the photoelectric conversion layer 41 in FIG. 5 in which the InN ultrafine particles and the GaN ultrafine particles are uniformly dispersed is generated, and then only the GaN ultrafine particles are sprayed onto the semiconductor substrate 50, whereby the hole transport layer 42 is formed. Generated. Thereafter, by forming the transparent electrode film 24 on the surface of the semiconductor substrate 50 in the vacuum chamber 51, the photoelectric conversion film 23 sandwiched between the two transparent electrode films 21 and 24 is manufactured.

  The photoelectric conversion films 23, 18, 13, and 31 shown in FIGS. 1 and 2 are manufactured in the same process as described above. In the manufacture of each photoelectric conversion film, the particle size control of InN ultrafine particles and GaN ultrafine particles is performed. There is a need to do. This is performed by a mass separator provided in the communication passages 52f and 53f.

  The principle of this mass separator is the same as that of the mass spectrometer. First, the generated ultrafine particles are ionized and charged, and then accelerated by applying an acceleration electric field. By passing the accelerated particles through the electric field between the deflection electrodes, the accelerated particles take an arc orbit having a radius corresponding to the mass, that is, the particle diameter. As a result, only ultrafine particles having a particle size in a desired range can enter the vacuum chamber 51 through the communication passages 52f and 53f, and particles having other particle sizes hit the wall surface and adhere to the wall surface. That is, by adjusting the acceleration electric field and the deflection electric field, the particle diameter of the ultrafine particles entering the vacuum chamber 51 can be controlled, and the particle diameter of the quantum dots that generate the photoelectric conversion film can be controlled.

  In order to dope the GaN ultrafine particles with p-type impurities, the GaN target 56 itself may contain p-type impurities. Similarly, in order to dope n-type impurities, the GaN target 56 containing n-type impurities may be used. Is used.

  In order to generate the photoelectric conversion film shown in FIGS. 6 and 7, both a GaN target containing a p-type impurity and a GaN target containing an n-type impurity are prepared in the vacuum chamber 53 and used as necessary. What is necessary is just to switch a target.

  In addition, if the film thickness of the photoelectric conversion film 23 formed as described above is, for example, a photoelectric conversion film that photoelectrically converts blue light, the blue light is sufficiently absorbed into the next photoelectric conversion film. It is better to prevent light from entering. This is because if blue light is incident on the photoelectric conversion film for green light in the next layer and photoexcitation occurs, color separation becomes worse.

  As a method of moving signal charges from the photoelectric conversion films of FIGS. 1 and 2 to the corresponding high-concentration impurity regions 2 and the like, a method of extracting signals from a light receiving element of a normal CCD image sensor or CMOS image sensor is used. A similar method may be adopted. For example, a method of injecting a certain amount of bias charge into the high-concentration impurity region 2 or the like (storage diode) (refresh mode), accumulating the constant charge by light incidence (photoelectric conversion mode), and then reading the signal charge It is. The photoelectric conversion film itself can be used as a storage diode, or a storage diode can be additionally provided.

  For reading the signal charge moved to the high concentration impurity region 2 or the like, a normal CCD image sensor or CMOS image sensor read method can be applied as it is.

  Conventionally, a solid-state imaging device such as a CCD has a light receiving element having a photoelectric conversion function, a function of storing a converted signal, a function of reading out a stored signal, a function of selecting a pixel position, and the like. The signal charge or signal current photoelectrically converted by the light receiving unit is stored in the light receiving unit itself or an attached capacitor. The stored charges are read out together with the selection of the pixel position by a so-called charge coupled device (CCD) or a MOS type image pickup device (so-called CMOS sensor) using an XY address system.

  The CCD image sensor includes a charge transfer unit that transfers a charge signal of a pixel to an analog shift register by a transfer switch, and sequentially reads the signal to the output terminal by the operation of the register. Examples include a line address type, a frame transfer type, an interline transfer type, and a frame interline transfer type method. The CCD has a two-phase structure, a three-phase structure, a four-phase structure, and a buried channel structure. The structure of the vertical transfer path in the photoelectric conversion film stacked solid-state imaging device of the present invention is one of these. Any structure can be adopted.

  In addition, as an address selection method, there is a method in which each pixel is sequentially selected by a multiplexer switch and a digital shift register and read as a signal voltage (or charge) to a common output line. An image sensor for XY address scanning that is two-dimensionally arrayed is known as a CMOS sensor. This is because the switch provided in the pixel connected to the intersection of XY is connected to the vertical shift register, and is read from the pixel provided in the same row when the switch is turned on by the voltage from the vertical scanning shift register. The signal is read out to the output line in the column direction. This signal is sequentially read from the output through a switch driven by a horizontal scanning shift register.

  For reading out the output signal, a floating diffusion detector or a floating gate detector can be used. Further, S / N can be improved by providing a signal amplifying circuit in the pixel portion, a correlated double sampling method, or the like.

  For signal processing, gamma correction by an ADC circuit, digitization by an AD converter, luminance signal processing, and color signal signal processing can be performed. Examples of the color signal processing include white balance processing, color separation processing, and color matrix processing. When used for NTSC signals, RGB signals can be converted to YIQ signals. These are the same as conventional CCD image sensors and CMOS image sensors.

  In the above-described embodiment, the microlens, the infrared cut filter, and the ultraviolet cut filter have not been described. However, the infrared cut filter can be provided in the lowermost layer or the uppermost layer in the configuration of FIGS. It is also possible to increase the light collection rate using a microlens. In addition, an ultraviolet cut filter may be provided in the uppermost layer, or may be placed at an appropriate position between the lens and the photoelectric conversion film.

  Furthermore, the photoelectric conversion film laminated solid-state imaging device of the present embodiment can have various advantages by making the photoelectric conversion film have a three-layer structure or a four-layer structure. For example, there is no moiré in the captured image, R, G, and B can be detected together with one pixel, so an optical low-pass filter is not required and high resolution is obtained. An imaging lens that has simple processing and does not generate pseudo signals, improves the reproducibility of hair, etc., facilitates pixel mixing and facilitates partial reading, and has an aperture ratio of 100% without using a microlens. There is no shading because there is no restriction on the exit pupil distance with respect to the lens, which makes it suitable for interchangeable lens cameras and contributes to the thinning of the lens, and solves the problems of conventional CCD type and CMOS type image sensors. be able to.

  The photoelectric conversion film laminated solid-state imaging device according to the present invention can be used in place of a conventional CCD type or CMOS type image sensor, and has the advantage that the sensitivity can be increased because one pixel can be made larger than the conventional one. It is useful when mounted on a digital camera or the like.

It is a cross-sectional schematic diagram for 1 pixel of the photoelectric conversion film laminated | stacked solid-state image sensor of the 3 layer structure which concerns on one Embodiment of this invention. It is a cross-sectional schematic diagram for 1 pixel of the photoelectric conversion film laminated | stacked solid-state image sensor of the 4 layer structure which concerns on one Embodiment of this invention. It is a graph which shows human visibility. FIG. 3 is a circuit diagram of a signal readout circuit configured by a MOS circuit. It is a cross-sectional schematic diagram of the photoelectric conversion film which concerns on one Embodiment of this invention. It is a cross-sectional schematic diagram of the photoelectric conversion film which concerns on another embodiment of this invention. It is a cross-sectional schematic diagram of the photoelectric conversion film which concerns on another embodiment of this invention. It is explanatory drawing of the manufacturing apparatus of the photoelectric converting film in embodiment of this invention.

Explanation of symbols

1 P well layer 2, 4, 6, 36 High concentration impurity region 3, 5, 7, 37 MOS circuit 8 Gate insulating film 9, 10 Insulating film 11, 14, 16, 19, 21, 24, 32, 33 Transparent electrode Films 12, 17, 22, 35 Electrodes 13, 18, 23, 31 Photoelectric conversion film 25 Transparent insulating film 26 Light shielding film 41 Photoelectric conversion layer 42 Hole transport layer 43 InN quantum dots (ultrafine particles)
44 GaN quantum dots (ultrafine particles)
44pp GaN quantum dots doped with p-type impurities 44n GaN quantum dots doped with n-type impurities 50 Semiconductor substrates 51, 52, 53 Vacuum chamber 52e, 53e Pulse laser light source 55 InN target 56 GaN target

Claims (12)

  A photoelectric conversion film is stacked on a semiconductor substrate on which a signal readout circuit is formed, and a signal corresponding to the amount of incident light photoelectrically converted by the photoelectric conversion film is read out to the outside by the signal readout circuit. In the device, the photoelectric conversion film has a photoelectric conversion layer in which first quantum dots contributing to photoelectric conversion and second quantum dots having a larger band gap than the first quantum dots are substantially uniformly dispersed. A photoelectric conversion film laminated solid-state imaging device.   2. The photoelectric conversion film stacked solid-state imaging device according to claim 1, wherein the photoelectric conversion film is configured by stacking a hole transport layer including the second quantum dots on the photoelectric conversion layer.   The photoelectric conversion film stacked solid-state imaging device according to claim 2, wherein the second quantum dots of the hole transport layer are doped with a p-type impurity.   4. The photoelectric conversion film stacked solid-state imaging device according to claim 2, wherein the second quantum dot of the photoelectric conversion layer is doped with an n-type impurity. 5.   The p-type impurity is doped in the second quantum dot of the hole transport layer, and the p-type impurity is doped in the second quantum dot on the hole transport layer side of the photoelectric conversion layer. The photoelectric conversion film laminated solid-state imaging device according to claim 2.   6. The photoelectric conversion film stacked solid-state imaging device according to claim 5, wherein an n-type impurity is doped in the second quantum dot on the opposite side of the photoelectric conversion layer to the hole transport layer.   The photoelectric conversion film stacked solid-state imaging device according to claim 1, wherein the first quantum dot is InN, and the second quantum dot is GaN.   The photoelectric conversion film stack according to claim 1, wherein the photoelectric conversion film sandwiched between two transparent electrode films is stacked in three layers via a transparent insulating film. Type solid-state imaging device.   Among the photoelectric conversion films of the three layers, the maximum value of light absorption of the photoelectric conversion film of the first layer is a wavelength of 400 to 500 nm, and the maximum value of light absorption of the photoelectric conversion film of the second layer is a wavelength. The average particle diameter of the first quantum dots provided in each photoelectric conversion film is determined so that the maximum value of light absorption of the photoelectric conversion film of the third layer is 500 to 560 nm and the wavelength is 560 to 640 nm. The photoelectric conversion film laminated solid-state imaging device according to claim 8, wherein   The photoelectric conversion film stack according to any one of claims 1 to 7, wherein the photoelectric conversion film sandwiched between two transparent electrode films is stacked in four layers through a transparent insulating film. Type solid-state imaging device.   Among the four layers of photoelectric conversion films, the maximum value of light absorption of the photoelectric conversion film of the first layer is a wavelength of 420 to 480 nm, and the maximum value of light absorption of the photoelectric conversion film of the second layer is a wavelength. The maximum value of light absorption of the photoelectric conversion film of the third layer is 480 to 520 nm, and the maximum value of light absorption of the photoelectric conversion film of the fourth layer is the wavelength of 560 to 620 nm. The photoelectric conversion film stacked solid-state imaging device according to claim 10, wherein an average particle diameter of the first quantum dots provided in each photoelectric conversion film is determined.   The red signal amount is obtained by subtracting the signal amount detected by the photoelectric conversion film of the second layer from the signal amount detected by the photoelectric conversion film of the fourth layer, The photoelectric conversion film laminated solid-state imaging device according to claim 11.

Images