Electromagnetic radiation spectrum detection system

Abstract

1. A system (100) for detecting the spectrum of electromagnetic radiation, comprising: a sensor device (20,1) having two photodiodes in a "back-to-back" configuration and configured to convert incident electromagnetic radiation (EMR) into an electric current as a function of relative spectral responsivity that depends on an electrical control voltage; and an electronic control and processing module (50) configured to selectively provide a plurality of control voltage (VB) values ​​to the sensor device (1) by varying the spectral responsivity to obtain a corresponding plurality of detection currents (Iph) related to the incident electromagnetic radiation; and to process the values ​​of the detection currents (Iph) based on the values ​​of the control voltage (VB) and the spectral responsivity to obtain information about the spectrum of the incident electromagnetic radiation (SP-ANS; IPST-IM).

Description

[Technical Field]

[0001] This invention relates to a system for detecting the spectrum of electromagnetic radiation. In particular, this invention relates to a spectral analysis system and a hyperspectral imaging system. [Background technology]

[0002] As is well known, spectral analysis (spectroscopy and spectroscopy) concerns the measurement of radiation intensity as a function of wavelength.

[0003] Hyperspectral imaging is a measurement technique that makes it possible to acquire spectral information for each point in an image.

[0004] Known methods for measuring spectrally resolved images include placing one or more bandpass filters or dispersive elements between the object and the camera, or using a multi-channel detector to acquire images of each spectral band. The number of bands measured and the spectral width of each band depend on the characteristics of the filters or dispersive elements and the detector used. This technique acquires spectral information in a discrete set of optical bands and is also called multispectral imaging technique.

[0005] International Publication No. 2014063117A1, "Single-sensor hyperspectral imaging device," describes a multi-pixel sensor in which an array of nine filters placed in front of the pixels was developed to reconstruct multispectral images of different bands of the electromagnetic spectrum.

[0006] Another known technique based on Fourier transform spectroscopy uses an interferometer between the object and the detector. The paper J. Craven-Jones et al. “Infrared hyperspectral imaging polarimeter using birefringent prisms,” Applied Optics, Vol. 50, No. 8, March 10, 2011 describes a near-infrared and mid-infrared hyperspectral imaging polarimeter, which includes a pair of sapphire Wollaston prisms and several higher-order retarders to form a Fourier transform imaging spectrophotometer.

[0007] The Wollaston prism is used as a birefringence-based interferometer with reduced vibration sensitivity compared to path-division interferometers such as the Michelson interferometer. Polarization data is acquired using a channel spectropolarometer to modulate the spectrum with Stokes parameter information. The acquired interferogram is filtered and reconstructed by a Fourier transform to extract the spatially and spectrally resolved Stokes vectors of the image.

[0008] The paper ARHarvey et al. “Birefringent Fourier-transform imaging spectrometer”, OPTICS EXPRESS 5368 No.22, Vol.12, 1 November 2004 describes a birefringent Fourier transform imaging spectrometer based on an input polarizer, two Wollaston prisms arranged in a cascade (one fixed, the other movable), a second polarizer, and a lens for image formation.

[0009] U.S. Patent No. 7800067B1, "Electronically tunable and reconfigurable hyperspectral photon detector," describes an optical sensor that can collect spectral information by utilizing the depletion zone of a diode in which the material composition changes linearly along the depletion zone.

[0010] The paper "Voltage-tunable dual-band Ge / Si photodetector operating in VIS and NIR spectral range" by E. Talamas Simola et al., Vol. 27, No. 6; 03 / 18 / 2019, OPTICS EXPRESS 8529 describes a device with an epitaxial germanium-on-silicon structure having two photodiodes connected back-to-back and operating as a photodetector for a wide range of wavelengths.

Prior Art Documents

Patent Documents

[0011]

Patent Document 1

Patent Document 2

Non-Patent Documents

[0012]

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

[0013] The applicant noted that prior art spectral detection systems (e.g., hyperspectral imaging systems) appear structurally complex in that they require the use of optical components (e.g., filters, prisms, dispersive elements) or mechanical components (e.g., MEMS micro-electromechanical systems) to reconstruct the incident spectrum.

[0014] The present invention addresses the problem of providing a spectral detection system that is structurally less complex than systems of known technology (for example, one that operates particularly as a spectral analysis or hyperspectral imaging system).

[0015] The present invention relates to the spectral detection system described in claim 1 and preferred embodiments thereof as defined by claims 2 to 15.

[0016] The present invention will be described in detail below with reference to the accompanying drawings, not as an limitation but as an example. [Brief explanation of the drawing]

[0017] [Figure 1] This diagram schematically shows a spectrum detection system comprising an electromagnetic radiation sensor device and an electronic control and processing module. [Figure 2] This is a cross-sectional view of a first realization of a semiconductor sensor included in the above-mentioned device, which includes two photodiodes. [Figure 3] This figure shows an example of the trend of current generated by a sensor as a function of wavelength and control voltage. [Figure 4]This is a cross-sectional view of a second realization of the sensor described above. [Figure 5] This is a cross-sectional view of the third realization of the sensor described above. [Figure 6] This is a cross-sectional view of the fourth realization of the sensor described above. [Figure 7] This shows a set of sensor response curves as a function of the voltage difference applied to the sensor when a specific incident radiation is applied. [Modes for carrying out the invention]

[0018] In this specification, similar or identical elements or components are indicated in the drawings by the same identification symbol.

[0019] Figure 1 schematically shows a spectrum detection system 100 comprising an electromagnetic radiation sensor device 20 and an electronic control and processing module 50. More specifically, the electronic control and processing module 50 is configured to extract information about the spectrum of radiation illuminating the sensor 1 and provide it on the output OU. The electronic control and processing module 50 includes a readout module 2 (RD), a regulating module 3 (CND), and an analysis and control device 4 (CNT-AN).

[0020] According to the first embodiment described below, the spectral detection system 100 may be configured to perform spectral analysis of electromagnetic radiation by providing SP-ANS, information relating to the measurement of the intensity of incident radiation as a function of wavelength, on the output OU.

[0021] The electromagnetic radiation sensor device 20 includes two photodiode (i.e., dual photodiode) type sensor devices or photodetector devices 1 (hereinafter, for brevity, "sensors") in a back-to-back configuration. As will be described later, the sensors 1 are configured to convert incident electromagnetic radiation into electric current depending on a relative spectral response that depends on an electrically controlled voltage.

[0022] For the purpose of spectral analysis of electromagnetic radiation, the sensor device 20 can include a single sensor 1.

[0023] As schematically shown in FIG. 1, the sensor 1 includes a first photodiode PD1 and a second photodiode PD2 (hereinafter also simply referred to as the first and second "diodes") arranged in a "back-to-back" configuration, i.e., the cathodes (or anodes) are electrically connected to each other. These first and second diodes PD1 and PD2 are adjusted by applying an appropriate voltage (V B ) to the remaining anodes (or cathodes) connected to the control and processing circuit 50.

[0024] In particular, the first photodiode PD1 is obtained from a semiconductor material having an energy gap E g1 , and the second photodiode PD2 is obtained from another semiconductor material having an energy gap E g2 <E g1 .

[0025] The first diode PD1 is configured to collect radiation having a wavelength λ between a first minimum wavelength λ min (PD1) and a first maximum wavelength λ max (PD1) = hc / E g1 (hc is a universal constant) and convert it into an electrical signal (i.e., a photocurrent Iph).

[0026] The second photodiode PD2 is configured to collect radiation having a wavelength in the range from a second minimum wavelength λ min (PD2) to a second maximum wavelength λ max (PD2) = hc / E g2 , λ min (PD2) < λ[[ID=۳۹]] max (PD1) and convert it into an electrical signal (the corresponding photocurrent Iph).

[0027] For example, in the case of silicon, λ min (PD1) and λ maxThe index values ​​for (PD1) are 400 nm and 1100 nm, respectively, and in the case of germanium, λmin(PD2) and λ max The index values ​​for (PD2) are 400nm and 1800nm, respectively.

[0028] Figure 2 shows a first exemplary embodiment of sensor 1, described with respect to silicon and germanium as examples of materials. According to the example described, the first photodiode PD1 is a pin (p-type, intrinsic, n-type) diode made of silicon, and the second photodiode PD2 is a pin (p-type, intrinsic, n-type) diode made of germanium.

[0029] More specifically, the sensor 1 includes a substrate 5 made of a first semiconductor material (silicon, for example) which sequentially comprises a first p-type doped layer 6 and a first intrinsic layer 7(i) superimposed on the first layer 6. A second n+-type (i.e., highly doped) doped layer 8 is superimposed on the first intrinsic layer 7.

[0030] According to the embodiment shown in Figure 2, the first doped layer 6 (anode), the first intrinsic layer 7, and the second doped layer 8 (cathode) form a pin-type first PD1 diode.

[0031] The first diode PD1 is made of silicon and collects visible and near-infrared radiation, converting it into an electrical signal (i.e., current Iph), and then, as shown, at a minimum wavelength of 400 nm λ min (PD1) and the maximum wavelength λ at 1100 nm max Radiation with (PD1) is collected and converted into an electrical signal. The VIS (visible) band, as is well known, is included in the wavelength range of 400 nm to 700 nm. The near-infrared (NIR) band extends from 700 nm to 1100 nm.

[0032] For the first diode PD1 (silicon), it is advantageous for the thickness of the first intrinsic layer 7 to be between 50 μm and 300 μm. Note that thinner thicknesses do not actually allow silicon to absorb all incident radiation, while thicker thicknesses impair the efficiency of photogenerated charge collection.

[0033] In addition, to ensure proper operation of the sensor 1, it is preferable that the first intrinsic silicon layer 7 has n-type background doping. In this way, the photogenerated charges in silicon can be effectively transported along the structure at a low recombination rate, and thus high efficiency can be ensured.

[0034] In a specific example, the following size values ​​can be used for the first PD1 diode: - The first doped layer 6 (Si-p) has a thickness of approximately 200 nm, and 10 19 cm -3 It is doped with boron atoms at a concentration equal to; - The first intrinsic silicon layer 7 has a thickness of approximately 300 μm; - The second doped layer 8 (Si, n+) is 10 19 cm -3 It has a thickness of approximately 150 nm and a dopant concentration of [specified value].

[0035] In addition, sensor 1 includes a layer 9 made of a second semiconductor material (for example, made of germanium) superimposed on a second doping layer 8 of the first diode PD1, into which a portion of the second diode PD2 is integrated. More specifically, the germanium layer 9 includes a second intrinsic layer 10 placed on the second doping layer 8 and a third n-type doping layer 11 superimposed on the second intrinsic layer 10. The second diode PD2 includes the second doping layer 8 (i.e., a cathode common to the first diode PD1), the second intrinsic layer 10, and the third doping layer 11 acting as an anode.

[0036] Note that, according to the alternative structure, the n-type layer of the second PD2 diode is not shared with the first PD1 diode, as shown in Figure 2 (i.e., the second doped layer 8), but can be realized by the associated germanium layer.

[0037] The second diode PD2, fabricated at least partially from germanium, has a minimum wavelength λ as shown. min (PD2) 400 nm and maximum wavelength λ max (PD2) is configured to collect radiation having an wavelength of 1800 nm and convert it into an electrical signal (i.e., current Iph).

[0038] As is well known, the near-infrared band (NIR) extends from 700 nm to 1100 nm, while the short-wave infrared band (SWIR) is included in the wavelength range of 1100 nm to 3000 nm.

[0039] In other words, if sensor 1 is manufactured using silicon for photodiode PD1 and germanium for photodiode PD2, as shown in a specific implementation in Figure 2, it can operate in the visible near-infrared range up to a portion of the short-wave infrared range.

[0040] It is advantageous for the germanium layer 9 of the second PD2 diode to have a thickness of more than 1 μm. In fact, it should be noted that thinner thicknesses do not allow the germanium to absorb all of the incident radiation, resulting in an overall decrease in responsiveness.

[0041] For example, the germanium layer of the second PD2 diode can be made to have the following size values: -The thickness of the second intrinsic layer 10(i-Ge) is approximately 3 μm; -The thickness of the third dope layer 11 (p-Ge) is 150 nm; the dopant concentration is 10 18 cm -3 That is the case.

[0042] It should be noted that, in order to increase the variability of the response curve in response to changes in the voltage applied to sensor 1, it is also possible to manufacture a second germanium PD2 diode having a pn diode in which the n-type germanium layer is less doped than the p-type germanium layer. Furthermore, in this case, the total thickness of the p-type layer is 200 nm or less. In practice, a thicker thickness reduces the overall efficiency of sensor 1.

[0043] Sensor 1 also has a first ohmic contact BC fabricated on the free surface (i.e., not in contact with the first intrinsic layer 7) of the first doped layer 6. A second ohmic contact TC is formed on the surface of the third doped layer 11 that is not in contact with the second intrinsic layer 10. The first ohmic contact BC and the second ohmic contact TC may be fabricated as, for example, a metal contact, particularly gold.

[0044] The first ohmic contact BC is fabricated to define the exposed area 12 of the surface of the first doped layer 6 that may be affected by EMR electromagnetic radiation.

[0045] Sensor 1 can be manufactured using semiconductor device manufacturing techniques known to those skilled in the art, including but not limited to injection techniques, spin-on dopant and deposition techniques (including epitaxy, sputtering, and evaporation), and transfer as wafer bonding.

[0046] Regarding the operation of sensor 1, the bias voltage V B It should be noted that this has traditionally been understood as the voltage difference between the second ohmic contact TC and the first ohmic contact BC. According to this definition, the positive bias voltage V B In this case, the second diode PD2 is directly biased, and the first diode PD1 is reverse-biased, and a negative voltage V B In the case of [this], the opposite is true.

[0047] Sensor 1 has an energy gap E g1When illuminated from the side of the semiconductor having (i.e., in the exposed region 12 of diode PD1, for example), the first photodiode PD1 absorbs some of the higher energy light radiation, and the second photodiode PD2 absorbs the gap semiconductor E g1 It is illuminated only by light radiation that cannot be absorbed by the light source.

[0048] Therefore, gap semiconductor E g1 (That is, the photodiode realized using the first diode PD1) has a low wavelength (λ <hc / E g1 The first photodiode PD2 can generate a photocurrent only in the presence of light radiation (where hc is a universal constant), but conversely, the second photodiode PD2 can generate a photocurrent (where λ is a universal constant). max (PD1) < λ < λ max It responds only to photons having (PD2). The total photocurrent generated by sensor 1 is given by the difference between the photocurrents generated by two different photodiodes.

[0049] More specifically, when the first diode PD1 is reverse-biased, it can generate a photocurrent Iph when irradiated with radiation having a wavelength λ between λmin(PD1) and λmax(PD1), while the second diode PD2 is directly biased and therefore optically inert, but allows for the circulation of the current Iph.

[0050] Conversely, when the second diode PD2 is reverse-biased, it can generate a photocurrent Iph (when irradiated with a radiation wavelength λ between λmax(PD1) and λmax(PD2)), while the first diode PD1 is directly biased and therefore optically inert, but allows for the circulation of current Iph.

[0051] When the bias voltage VB is applied, the spectral response of sensor 1 can be electrically selected via the analysis and control device 4. Figure 3 shows an example of the bias voltage V BThis shows the trend of the current Iph as a function of wavelength for the two extreme values.

[0052] The applied bias voltage V B Depending on the situation, the collection efficiency of the two junctions can be changed, especially when the gap E g1 As the collection efficiency at the junction increases, the gap E g2 A decrease in collection efficiency was observed at the junction, and vice versa. This method allows the response spectrum of sensor 1 to change continuously.

[0053] Figure 7 shows a set of response curves for sensor 1 as a function of the voltage difference applied to the sensor when some incident radiation is applied. The curves in Figure 7 are experimentally obtained for a structure similar to the curves in Figure 2.

[0054] While the example in Figure 1 mentions the use of silicon and germanium, it should be noted that sensor 1 can be manufactured from other semiconductor materials. For example, other possible materials that can satisfy the above-mentioned relationship between the photoresponse bands of the first diode PD1 and / or the second diode PD2 include semiconductor materials selected from one of the following types: a) III-V semiconductors (e.g., GaAs, InAs, InP) and their alloys; b) Group II-VI semiconductors (e.g., ZnSe, ZnTe, CdSe, CdTe, HgTe, PbS, PbSe) and their alloys; c) Group IV semiconductors (e.g., Si, Ge, GeSn) and their alloys.

[0055] Regarding the control and processing module 50, note that the readout module 2 is configured to convert the current signal Iph into a voltage signal Vph, and includes, for example, a transimpedance amplifier (not shown). The current signal Iph is converted into a bias voltage Vph. B It is advantageous that these values ​​are obtained sequentially as the bias voltage V changes. BA single readout module 2 (having a single transimpedance amplifier) ​​can be used to convert all Iph current signals obtained as a result of the change.

[0056] The adjustment module 3 is configured to process the voltage signal Vph in a way that is suitable for subsequent processing. For example, the adjustment module 3 performs operations such as amplification, filtering, level adaptation, dark current cancellation, and analog-to-digital conversion of the voltage signal Vph to a digital value Vphi. Note that the bias voltage V B A single adjustment module 3 is suitable for processing the voltage signal Vph obtained by changing the value.

[0057] The analysis and control device 4 uses various bias voltages V B The system is configured to receive a digital value Vphi representing a voltage signal Vph corresponding to the photocurrent signal generated by sensor 1, and to perform spectral analysis of the radiation entering sensor 1. The analysis and control device 4 may include programmable logic (e.g., ASIC / FPGA) configured to perform such spectral analysis.

[0058] In addition, the analysis and control device 4 applies a bias voltage V to the sensor 1 through appropriate adjustment (which can be performed by the adjustment module 3). B It is configured to generate a digital voltage value Vbi that is converted to an analog value and changes its spectral response.

[0059] The control and processing module 50 is, for example, a ROIC (readout integrated circuit) that can be directly integrated into a silicon substrate used to fabricate the first photodiode PD1. Alternatively, the sensor 1 and ROIC 50 may be manufactured on two different substrates and then connected via bump bonding or wafer bonding technology. Alternatively, the sensor 1 and ROIC 50 may be manufactured on different substrates and then connected via a PCB electronic substrate.

[0060] As described above, the sensitivity spectrum of sensor 1 can be continuously changed. In particular, it is possible to obtain a number of spectral responses intermediate between the extreme spectral responses shown in Figure 3 (obtained by applying the maximum or minimum bias voltage). For example, bias voltage V B The value varies in steps of 7mV to 16mV, preferably 8mV to 13mV. Note that using a 10mV step can yield a spectral resolution of less than 50nm. This characteristic applies to the bias voltage V. B It is dynamically adjustable by applying a certain parameter, offering the possibility of obtaining a large number of spectra extended into the visible and near-infrared ranges, and enabling the realization of a compact spectral analysis system.

[0061] Note that the spectral analysis is based on knowledge of the spectral responsiveness of sensor 1, and this spectral responsiveness is determined in the characterization step.

[0062] The spectral analysis performed by the analysis and control device 4 is based on the following considerations.

[0063] We consider a finite number n bias voltage and a finite number m analysis wavelength for sensor 1. The following matrix equation (1) is valid:

number

number

number

number

[0064] Photocurrent vector

number

[0065] Matrix obtained from the characterization of sensor 1

number

[0066] According to one mode of analysis, the matrix

number

number

number

number

number

[0067] Both elements of Equation 1

number

number

number

number

number

[0068]

number

number

[0069] As a result, the wavelengths within the sensitivity spectrum of sensor 1 are centered

number

[0070] If you do not have a device that can generate a set of linearly independent response curves, the Gaussian function

number

number

[0071] Using this method, even when it is not possible to calculate the coefficients in a closed form, the coefficients can be calculated instead.

number

number

number

[0072] The sensor 1 based on the germanium-silicon structure described above allows the response spectrum to change continuously, resulting in multiple sparsely correlated response curves.

[0073] Therefore, as can also be expressed by equation (3), the spectral density vector

number

number

number

[0074] Required for the precise execution of linear combination of photocurrents

number

[0075] In particular, during the preliminary calibration stage of sensor 1, the response matrix

number

number

number

number

[0076] The mathematical methods used to extrapolate spectral information by equation (3) include the use of addition and multiplication operations.

[0077] To summarize the above, the sensor device 20 is subjected to a bias voltage V applied to it. B The system undergoes a calibration procedure designed to measure the spectral response by continuously varying the parameter. The results of this calibration procedure are stored in the analysis and control device 4 as a spectral response matrix for use in subsequent stages.

number

number

[0078] In the operation of the sensor device 20, the sensor device 20 is illuminated with radiation that requires the power spectrum to be reconstructed, and a bias voltage V B This changes continuously, just as it did in the calibration step. The acquired photocurrent signal (i.e., the photocurrent vector)

number

[0079] The photocurrent signal is a vector of weights obtained through calibration.

number

number

[0080] According to a second embodiment of system 100, the system may be configured to operate as a hyperspectral imaging system, i.e., a hyperspectral imaging system IPST-IM. In this case, the sensor device 20 includes a plurality of sensor devices 1 organized according to a matrix, similar to the sensors described above. For example, the sensor device 20 may include millions of sensors 1, each corresponding to a pixel in the acquired hyperspectral image.

[0081] In this case, the electronic control and processing module 50 is configured to receive the photocurrent Iph provided by each sensor 1 (corresponding to a pixel) and process them in the same manner as described above to acquire a hyperspectral image, providing the image spectrum of the scene for each pixel.

[0082] Furthermore, other configurations of the sensor 1 are also possible, some of which are described below.

[0083] According to an alternative embodiment of sensor 1 (not shown), in order to simplify the construction process, the first silicon diode PD1 has an n-type region of 10 15 cm -3 ~10 16 cm -3 It is formed by a PNN diode with a dopant density. Higher dopants reduce the overall efficiency of the sensor, while good low dopants provide acceptable performance but are more technically complex to implement.

[0084] According to another alternative embodiment (similarly not shown), in order to enhance the responsive modulation capability with applied voltage, the n-layer of germanium (e.g., the second doped layer 8 in Figure 1) is intended to be replaced with a layer formed of a SixGe1-x germanium-silicon alloy having a reduced germanium ratio and n-type doping.

[0085] In addition to the use of several possible semiconductor materials, sensor 1 can be manufactured to present several possible structures. Possible alternative structures for sensor 1 are described below.

[0086] Figure 4 shows a second embodiment of the sensor 1 described as an example, similarly relating to silicon and germanium.

[0087] In the example shown in Figure 4, within the substrate 5, the first diode PD1 is formed by a first layer 21 (silicon, p-type), on which a second layer 22 (silicon, n-type) is placed, and there is no intrinsic silicon layer. The second diode PD2 of the sensor 1 in Figure 4 consists of the same first layer 22 (common cathode) and a third layer 23 (intrinsic within germanium) (anode of the second PD2 diode) which behaves as a p-type semiconductor. The structure of the sensor 1 in Figure 4 is particularly compact.

[0088] Figure 5 shows a third, non-exemplary, compact realization of silicon and germanium. The first diode PD1 is formed by a first silicon layer 24 (p-type) and a second silicon layer 25 (n-type), with no intrinsic silicon layer present. The second diode PD2 in Figure 5 consists of a first germanium layer 26 (n-type) and an intrinsic layer 27 made of germanium acting as a p-type semiconductor.

[0089] Figure 6 shows a fourth realization of Sensor 1, which is illustrated with germanium and silicon as examples, but to which the same considerations described above apply to possible alternative materials that may be used.

[0090] According to this embodiment, the sensor 1 comprises a relative substrate 5 within a first semiconductor material (silicon in the illustrated example) having, for example, p-type doping, which defines a first surface 30 and an opposing second surface 32. The first surface 30 constitutes an exposed region 12 through which EMR radiation penetrates. Note that the substrate 5 of the sensor 1 in Figure 6 may also be n-type.

[0091] The first doped region 31 (n-Si) is formed within the substrate 5 using the same material as the substrate, but with a doping opposite to that of the substrate 5, i.e., n-type doping. The first doped region 31 extends from the second surface 32 of the substrate 5 toward the interior of the substrate itself and does not reach the first surface 30.

[0092] It should be noted that the first doped region 31 described above is intended to act as a common electrode (cathode, for example) for the first diode PD1 and the second diode PD2.

[0093] The second doped region 33 is formed within the substrate 5 and, for example, is of the same type as the substrate 30 but has higher doping, i.e., p+ type doping. The second doped region 33 extends inward from the second surface 32 into the substrate 5, does not reach the first surface 30, and is shallower in depth than the first doped region 31. For example, the second doped region 33 extends into the substrate 5 so as to laterally surround the first doped region 31 in a ring shape.

[0094] It should be noted that the second doped region 33 is intended to act as an additional electrode (as an anode in this example) for the first diode PD1.

[0095] Furthermore, the sensor 1 according to the fourth embodiment includes a layer 34 made of a second semiconductor material (for example, made of germanium) which is disposed on the second surface 32 of the substrate 5 and is positioned to be in contact with the first doped region 31. The germanium layer 34 is, for example, intrinsic germanium. This includes defining a third surface 35 that faces the germanium layer 34 and the second surface 32.

[0096] As a result, the intrinsic germanium layer 34 faces a portion of the first doped region 31 (particularly its central portion) and does not completely cover the first doped region 31.

[0097] In this example, the germanium layer 34 is intended to function as the intrinsic layer of the second diode PD2.

[0098] On the third surface 35 of the intrinsic germanium layer 34, a doped layer 36 (inside the germanium) having high p-type doping (i.e., p+ doping) is, for example, positioned. The doped layer 36 is intended to act as the anode of the second PD2 diode.

[0099] According to the example in Figure 6, the sensor 1 has a first metal layer 37 positioned above the second doped region 33 (i.e., the anode of the first PD1 diode) to form a first ohmic contact BC.

[0100] Furthermore, the sensor 1 includes a second metal layer 38 that is electrically connected to the doped layer 36 (forming a second ohmic contact TC).

[0101] The connection between the second ohmic contact TC and the doped layer 36 (i.e., the anode of the second diode PD2) is made by the first highly doped well 39 (i.e., a highly conductive well) and the second highly doped well 40.

[0102] The first highly doped well 39 extends inward from the second surface 32 into the first doped region 31 (n-type) and, for example, has p+ doping. The second ohmic contact TC is positioned on the second surface 32 in contact with a portion of the highly doped well 39.

[0103] The second highly doped well 40 extends within the germanium layer 34, from the doped layer 36 (in contact with it) to the first highly doped well 39 on the second surface 32. In this example, the second highly doped well 40 is located within p+ doped germanium.

[0104] In summary, the first diode PD1 includes a first doped region 31, a portion of the substrate 5, and a second doped region 33. The second diode PD2 includes a first doped region 31, an intrinsic germanium layer 34, a doped layer 36, and first and second highly doped wells 39 and 40.

[0105] The first doped region 31, the second doped region 33, the first highly doped well 39, and the second highly doped well 40 can be formed by injection techniques and / or by spin-on dopant techniques and / or by deposition techniques including epitaxy, sputtering, and vapor deposition.

[0106] The intrinsic germanium layer 34 can be fabricated by chemical and / or physical deposition techniques such as epitaxy, sputtering, vapor deposition, or even transfer as wafer bonding. The intrinsic germanium layer 34 is entirely within the first doped region 31 of the silicon substrate 5, and therefore it is possible to deposit the intrinsic germanium layer 34 by techniques such as selective deposition (e.g., oxide window deposition), or to deposit the germanium layer 34 over the entire substrate 5 and then define the geometric shape of the layer 34 itself by selective removal techniques (photolithography).

[0107] The embodiment shown in Figure 6 has the advantage of being easy to manufacture and also presenting the sensor's metal contacts on the same side of the substrate 5 where they are integrated, thus resulting in a "planar" design.

[0108] The spectral detection system 100 can be used, for example, in the automotive sector (fog, night vision, increased field of view), the machine vision sector (Industry 4.0, improved machine vision), or the plastic recycling sector.

[0109] The system 100 described above is very advantageous in its various embodiments.

[0110] In particular, it allows for broad modulation of the spectral response, enabling the realization of effective algorithms for spectral reconstruction without relying on complex optical and mechanical structures as in known techniques, thus enabling the creation of numerous different spectral responses and achieving superior performance. In fact, structural simplification results in greater robustness and greater compactness.

[0111] Furthermore, the possibility of integrating electronic modules onto the same silicon substrate further reduces the structural complexity and associated costs of system 100. [Explanation of Symbols]

[0112] -Spectrum detection system 100 - Electromagnetic radiation sensor device 20 - Electronic control and processing module 50 -Sensor 1 -Output OU - Transparent Module 2 - Adjustment Module 3 - Analysis and control device 4 -First photodiode PD1 - Second photodiode PD2 -Photocurrent Iph - Circuit board 5 -First doping layer 6 -First true layer 7 - Second doping layer 8 - Epitaxial layer 9 -Second true layer 10 - Third doping layer 11 - First BC ohmic contact - Second Ohmic Contact TC -Exposed area 12 - Bias voltage VB -Voltage signal Vph - Digital voltage value Vbi - Layer 1, Layer 21 - Second layer 22 -Third layer 23 -First layer silicon 24 -Second layer silicon 25 - First germanium layer 26 -Intrinsic layer 27 - First side 30 -First doping area 31 -Second side 32 -Second doping area 33 - Layer 34 in the second semiconductor material - Third side 35 - Dope layer 36 -First metal layer 37 -Second metal layer 38 - First high-doping pocket 39 - Second high-doping pouch 40

Claims

1. It features two photodiodes in a "back-to-back" configuration, which regulates incident electromagnetic radiation (EMR) by applying an electrically regulated voltage (V). Ｂ A sensor device (20,1) configured to convert the spectral response of each of the following, which depends on the respective spectral response, into a current, An electronic control and processing module (50), - During the calibration stage, The aforementioned electrical adjustment voltage (V Ｂ By changing the optical response of the sensor device (1), the response matrix of the sensor device (1) is measured. A coefficient vector defined by multiple weight values, wherein the coefficient vector is configured such that a composite response matrix of the sensor device (1), which is bandpass type and in vector form, is obtained during operation by linearly combining the rows of the response matrix and the coefficient vector, and the coefficient vector is determined and stored. - During operation, The aforementioned electrical adjustment voltage (V Ｂ ) is continuously changed according to predetermined change steps so that the electrical adjustment voltage (V B) takes on multiple electrical control voltage values. The plurality of electrical control voltage values ​​are selectively provided to the sensor device (1), thereby changing the spectral response in a continuous mode according to a predetermined spectral resolution to obtain a plurality of corresponding detection currents (Iph) related to the incident electromagnetic radiation. The values ​​of the detected current (Iph) obtained from the plurality of electrical control voltage values, and the plurality of weight values ​​are processed to obtain information about the spectrum of the incident electromagnetic radiation (SP-ANS; IPST-IM), and the power spectral density of the incident electromagnetic radiation for a plurality of optical wavelengths is determined. An electronic control and processing module (50) configured as such and An electromagnetic radiation spectrum detection system (100) comprising the following:

2. The electronic control and processing module (50) changes the electrical adjustment voltage (V) that changes the responsiveness of the sensor device (1). Ｂ The system (100) according to claim 1, configured to select the wavelength of the incident electromagnetic radiation detectable by the sensor device (1) according to the value of ).

3. The system (100) according to claim 1, wherein the electronic control and processing module (50) is configured to determine the plurality of weight values ​​by a least-squares fitting procedure in order to obtain the bandpass type composite response set.

4. The control and processing module (50) A readout module (2) configured to receive the aforementioned multiple detection currents (Iph) and convert them into multiple detection voltages (Vph), An adjustment module (3) configured to receive the aforementioned multiple detection voltages (Vph) and convert them into corresponding digital detection values ​​(Vphi), An analytical device (4) configured to receive the digital detection value (Vphi) and extrapolate the information (SP-ANS; IPST-IM) related to the spectrum of the incident electromagnetic radiation, and The system (100) according to claim 1, including the above.

5. The system (100) according to at least one of claims 1 to 4, wherein the spectral detection system (100) is configured to perform spectral analysis (SP-ANS) of the incident electromagnetic radiation.

6. The system (100) according to claim 5, comprising a single sensor device (1).

7. The system comprises a number of additional sensor devices (1) configured according to the sensor device (1), and further comprising a number of additional sensor devices (1) arranged together with the sensor device according to a matrix of sensor devices that identify their pixels, The system (100) according to at least one of claims 1 to 5, wherein the system is configured to operate as a hyperspectral imaging system.

8. The system (100) according to at least one of claims 1 to 7, wherein the sensor device (1) comprises a first photodiode (PD1) in one semiconductor material and a second photodiode (PD2) in at least partially another semiconductor material, and the first photodiode (PD1) and the second photodiode (PD2) have their respective cathodes or anodes electrically connected to each other.

9. The system (100) according to claim 8, wherein the first photodiode (PD1) comprises a substrate (5) in a first semiconductor material, and a first layer (6) having a first doping type and a second layer (8) having a second doping type opposite to the first doping type are integrated within the substrate (5).

10. The system (100) according to at least claim 9, wherein the substrate (5) of the first photodiode (PD1) further comprises an intrinsic layer (7) within the first semiconductor material.

11. The second photodiode (PD2) includes the second layer (8) shared with the first photodiode (PD1), and the second photodiode (PD2) The system (100) according to claim 9 or 10, further comprising an integrated layer (9) in a second semiconductor material that overlaps with the substrate (5), wherein an intrinsic layer (10) and a first doped layer (11) having doping opposite to that of the second layer (8) contained in the substrate (5) are integrated within the integrated layer (9).

12. The first and / or second photodiodes (PD1, PD2) are of the following types: (a) Semiconductors of Group III-V and their alloys, (b) Semiconductors of Group II-VI and their alloys, (c) Semiconductors of Group IV and their alloys The system (100) according to claim 8, which is made of a semiconductor material selected from one of the following.

13. The sensor device (1) A substrate (5) in a first semiconductor material that defines a first surface (12, 30) exposed to electromagnetic radiation (EMR) and a second surface (32) facing the first surface, A first doped region (31) having a first doping type is included in the substrate (5) so as to extend to the second surface (32), A second doped region (33) is included in the substrate so as to extend to the second surface (32), is separated from the first doped region (31) by a portion of the substrate (5), and has a second doping type, A layer (34) of a second semiconductor material is disposed on the second surface (32) such that it is in contact with the first doped region (31) and defines a third surface (35) facing the second surface (32), A doped layer (36) in the second semiconductor material having the second doping type and overlapping with the third surface (35), A metal contact is positioned to contact the second doped region (33) of the second surface and the doped layer (36). Includes, The first doped region (31), a part of the substrate (5), and the second doped region (33) are part of the first photodiode (PD1) of the sensor device (1), The first doped region (31), the layer (34) within the second semiconductor material, and the doped layer (36) within the second semiconductor material are part of the second photodiode (PD2) of the sensor device (1). The system (100) according to claim 8.

14. The system (100) according to claim 8, wherein the control and processing module (50) is integrated with the substrate (5) on which the first photodiode (PD1) is integrated.

15. The system (100) according to claim 8, wherein the first and second semiconductor materials are selected so that the sensor device (1) operates in the visible near-infrared range up to a portion of the short-wave infrared range.

Images