Bipolar power supply for single-photon detectors

Abstract

The present invention relates to a single-photon detector arrangement for bipolar power supply for single-photon detectors using voltage converters, in particular for use in a monolithically integrated quantum random number generator (iQRNG). The present invention relates to a single-photon detector arrangement and an iQRNG based thereon, comprising at least one substrate (100) composed of a semiconductor material; a first p-n junction (50) formed in the substrate (100) and configured as a first single-photon detector (130), an operating voltage of the first single-photon detector (130) being above a supply voltage U0 provided with respect to an electrical zero potential; a first voltage converter (210) configured to invert the supply voltage U0, to amplify it with a first gain factor F1 greater than or equal to 1 and to provide it as a first converter voltage -U1 with respect to the electrical zero potential, the first p-n junction (50) being biased with a first differential voltage (∆V1.SPAD) with reference to the first converter voltage -U1.

Description

[0001] Bipolar power supply for single-photon detectors

[0002] The present invention relates to a single-photon detector arrangement for the bipolar voltage supply of single-photon detectors using voltage converters, in particular for use with a monolithically integrated quantum random number generator (iQRNG).

[0003] Technological background

[0004] In many areas of science and technology, random events and the determination of probabilities play a particularly important role. For example, Monte Carlo simulations and secure encryption methods rely heavily on the generation of random numbers. A general distinction is made between so-called pseudo-random numbers and true random numbers. While the former are generated using deterministic formulas by pseudo-random number generators (PRNGs) and are therefore not absolutely random, non-deterministic random number generators (TRNGs) for generating true random numbers are generally based on genuinely unpredictable processes such as thermal or atmospheric noise, and not on artificially generated patterns from deterministic algorithms.However, even the results of such non-deterministic random number generators based on external parameters can still tend slightly towards higher or even numbers due to weak correlations, depending on the underlying random element, thus enabling at least partial predictability of the random numbers generated in this way.

[0005] Quantum random number generators (QRNGs), a specific subset of TRNGs, are based on fundamental quantum processes for generating random numbers and are therefore, at least theoretically, not coupled to other external factors and effects that influence statistics. Quantum random number generators thus represent the best available source of true random numbers at present. Current digital QRNGs can deliver entropy rates (i.e., a sequence of bit values ​​with maximum randomness or entropy) of up to several hundred Mbps. The generated random numbers are required in both classical encryption methods and a multitude of quantum computing and quantum cryptography methods to ensure secure key distribution (SKD / QKD).Therefore, non-manipulable and fast QRNGs are absolutely necessary for generating secure keys in cryptography.

[0006] Many QRNGs are implemented as photonic QRNGs due to their particularly simple realization using random properties of photons. A simple concept for generating random numbers involves the behavior of a photon that is either reflected or transmitted at a semitransparent beam splitter independently of other photons. Another approach utilizes the random arrival times of photons at a single-photon detector. This distribution effect, based on an intrinsic photon statistics of the photons from a corresponding photon source that is not inherently deterministic, can also be used to provide truly random numbers.The random numbers can be provided, in particular, by varying the time differences between two successive detection events on the single-photon detector, with the arrival times generally exhibiting a distribution characteristic of the type of photon source (i.e., thermal, coherent, or non-classical). The combination of a photon source suitable for random number generation with a correspondingly sensitive single-photon detector is also referred to as the entropy source of the corresponding QRNG, where the random number generation is determined by the temporal sequence of the individual detection events in the entropy source.A corresponding QRNG requires at least one entropy source and may include at least one further component for evaluating the individual detection events and / or for providing a digital random number sequence suitable for further processing, based on the respective arrival times of the photons. Therefore, the terms entropy source and QRNG are used largely synonymously within the scope of this application.

[0007] Typically, in a single photon detector (SPD), a detector pulse is first generated by a single incident photon. This pulse is then converted by a time-to-digital converter (TDC) into a timestamped digital representation of the detection event, which can be further processed accordingly. The photon source in QRNGs is usually a laser diode (LD) attenuated to the single-photon level or a simple light-emitting diode (LED). The emitted photons from these LEDs can then be detected with temporal resolution by one or more highly sensitive single-photon avalanche diodes (SPADs).Such photon sources, which simultaneously provide only one or a few photons, are also referred to in this application as single photon sources (SPS). However, these do not necessarily have to be true single-photon emitters, for example, based on a single isolated two-level system.

[0008] SPADs are a type of photodetector similar to photodiodes (PDs) and avalanche photodiodes (APDs), but with significantly higher sensitivity. SPADs can be digitally read and analyzed, even within a single integrated circuit. When such an integrated detector circuit is excited by individual photons, only one electron-hole pair is generated per photon in the sensor's active region (absorption region). The excited electrons are drawn towards the cathode by electric fields, and the excited holes towards the anode. In a SPAD, the charge carriers drift through a so-called avalanche region, within which a charge avalanche is generated by enhanced impact ionization. These are therefore highly sensitive photon-receiver elements that, upon activation, generate a large amount of charge (approximately 105 - 10 6 can provide electrons) with high temporal resolution.

[0009] A SPAD is typically operated in Geiger mode above its breakdown voltage, detecting a single photon via the generated charge avalanche and subsequently recording it as an individual event. To reduce the dead time occurring during recording, active or passive suppression or quenching of further charge carrier amplification can be performed immediately after the avalanche formation begins. In addition to the SPAD, the integrated circuit can also include a so-called single-photon counter (SPC). In this case, instead of directly outputting a single detector pulse, a direct statistical evaluation of the temporal distribution of the individual detected single-photon events is generally performed.

[0010] Statistical analysis performed in parallel with random number generation can, for example, be used to further secure the generation process against potential attacks. Particularly in non-integrated photonic QRNGs built from individual components, the necessary transmission paths within the system offer numerous attack vectors. Therefore, to increase security, such systems are designed to be as compact and isolated from their external environment as possible. Besides avoiding potential attack scenarios, another advantage of such compact QRNGs is that any external natural influences that might impair random number generation can also be minimized.Accordingly, compact, photon noise-based QRNGs are currently primarily provided as hybrid component systems (hybrid integrated QRNGs or hiQRNGs), in which a PLC and a SPAD are interconnected as independent elements, for example, via planar waveguide structures or free-jet coupling. However, various approaches for fully monolithically integrated QRNGs in a common substrate (monolithically integrated QRNGs or iQRNGs) are also known.

[0011] In both of the aforementioned integration approaches, a suitable PLC, a highly sensitive PLC, and associated control and evaluation electronics are typically combined in a single assembly. However, these components generally have very different supply voltage requirements. For example, in some embodiments of an iQRNG, the PLC may be a light-emitting avalanche Zener diode (also known as a Zener-avLED) operating at a point below or near its breakdown voltage. Zener-avLEDs, for example, those using silicon BCD technology (e.g., 130 nm technology), typically have breakdown voltages of less than 10 V, so the PLC requires an operating voltage of approximately 5 V to 10 V (e.g., approximately 7 V).However, in a SPAD also implemented using silicon BCD technology, the required operating voltage can be at least twice as high (typically between 20 V and 30 V). For example, a SPAD with a breakdown voltage at room temperature of approximately 17 V and an applied overvoltage of approximately 5 V during operation requires an operating voltage of approximately 22 V, which cannot be significantly reduced through simple design or layout measures. In contrast, the operating voltage requirements of the associated control and evaluation electronics are typically around 2 V to 3 V in the silicon BCD technology mentioned above. Thus, the individual voltage values ​​generally differ considerably, in the example given by a factor of approximately 8 to 10 relative to a typically provided low supply voltage of approximately 2 V to 3 V.While it is theoretically possible to supply the corresponding voltage values ​​separately from dedicated power supplies, for the sake of the simplest and most compact implementation of iQRNGs, it is preferred to supply only a single external voltage and derive all other required voltage values ​​from this base voltage by conversion. For practical reasons, a supply voltage with a predefined voltage level is preferred. Typically, these are discretely defined voltage levels of 1.8 V, 2.5 V, or 3.3 V.

[0012] In the prior art, charge pumps are primarily used for this purpose in integrated circuits. However, other concepts for DC-DC conversion are also known, such as boost converters. The latter will not be explicitly mentioned further, as their circuit components are generally difficult or impossible to integrate. Integrated circuits for voltage conversion will be referred to as voltage converters in the following. Specifically, voltage converters will be defined as charge pumps; however, the methods described are equally applicable to other voltage converter concepts or hybrid voltage converter concepts.

[0013] Charge pumps for increasing an input voltage are state of the art and are known, for example, from US 4,214,174 A. With a simple configuration, the corresponding input voltage can be almost doubled. However, depending on the circuit design, conduction losses occur equal to the diode voltage of the diodes used or the threshold voltage of the transistors used. To achieve higher voltages, several charge pump stages can be cascaded. Since the area requirement and power dissipation also increase with each additional stage, the achievable voltage cannot be chosen arbitrarily high. Charge pumps that invert an input voltage are also known in the art and can likewise be implemented with simple means. A corresponding converter circuit can be described, for example, by Bloch et al. (M. Bloch, C. Lauterbach, and W. Weber).“High efficiency charge pump circuit for negative high voltage generation at 2 V supply voltage.” Proceedings of the 24th European Solid-State Circuits Conference. IEEE, 1998.).

[0014] The high supply voltages required for operating single-photon detectors, especially SPADs for use as SPDs in integrated iQRNGs, cannot be efficiently and practically achieved with a conventional power supply by cascading multiple charge pumps due to their large footprint and increasing power losses. The approximately 8 to 10-fold difference in the required operating voltages presents a significant technical challenge.

[0015] The technical challenge to be solved therefore consists of providing high operating voltages from a small supply voltage in a circuit design without incurring high power losses or requiring a large footprint. In particular, to reduce the absolute terminal potentials, the operating voltages should be provided directly within an iQRNG without separate power supplies.

[0016] Description of the invention

[0017] The problem according to the invention is solved by the subject matter of the independent claims. Preferred embodiments are the subject matter of the respective dependent claims.

[0018] One aspect of the present invention relates to a single-photon detector arrangement comprising a substrate made of a semiconductor material; a first pn junction formed in the substrate and configured as a first single-photon detector, wherein an operating voltage of the first single-photon detector is provided above a supply voltage U relative to an electrical zero potential (e.g., ground, general substrate or reference potential). o lies; a first voltage converter, set up to supply the voltage U o to invert (inverting voltage converter), to amplify with a first gain factor Fi greater than or equal to 1 and to provide as the first converter voltage -Ui relative to the electrical zero potential, wherein the first pn junction is biased with a first differential voltage with respect to the first converter voltage -Ui.

[0019] Preferably, the single-photon detector is a single-photon avalanche diode (SPAD). These are detectors that, due to their particularly high sensitivity, high gain, and low (dark) noise, are in principle capable of detecting and identifying single photons. In contrast to the prior art, the first single-photon detector is thus connected bipolarly with respect to a supply voltage U provided relative to an electrical zero potential (e.g., a substrate-related zero voltage). o . A required high operating voltage is thus achieved through both positive and negative voltage potentials with a lower absolute voltage compared to the operating voltage.

[0020] Since the first pn junction is biased with a first differential voltage relative to the first converter voltage -Ui, the high gain factors required with conventional voltage supplies can be reduced to provide a specific required operating voltage. With a low supply voltage, a maximum reduction of approximately half the otherwise required voltage gain can be achieved. That is, in the example mentioned at the beginning, where a gain of a factor of 8 to 10 is required, a bipolar configuration of the pn junction can reduce the required gain to approximately a factor of 4 to 5, whereby a corresponding gain of both the provided supply voltage and the first converter voltage -Ui relative to the electrical zero potential is required.Even when implemented with two independent voltage converters, the area consumption and power loss are still lower than when using a single voltage converter with an amplification factor of 8 to 10.

[0021] The substrate may preferably be a silicon substrate, and the single-photon detector arrangement may preferably be built using a modern and particularly widely applicable bipolar CMOS-DMOS technology (BCD technology).

[0022] The magnitude of the first gain factor Fi depends not only on the magnitude of the supplied voltage Uo, but also, and especially, on the respective voltage requirements of the single-photon detector arrangement and other circuit components to be powered. Preferably, the first gain factor Fi lies in a range between approximately 2 and 3 and / or between approximately 4 and 5. For an exemplary assumed supply voltage of approximately 2 V to 3 V, this results in first converter voltages -U1 in the range of approximately -4 V to -9 V or -8 V to -15 V. Such relatively low gain factors can be efficiently and effectively implemented with cascaded voltage converters. A single voltage converter can also be configured to simultaneously provide several different converter voltages.The present invention enables, in particular, a reduction in area requirements by allowing the selection of suitable components with a lower voltage / insulation class. This also makes it possible to completely or partially replace the coupling capacitors in charge pumps with the MOM (Metal-Oxide-Metal, MOM) class, which is otherwise common in the prior art, with the MOS (Metal-Oxide-Semiconductor, MOS) class. By reducing the number of pump stages in charge pumps as voltage converters, power losses can be reduced compared to direct amplification from a supplied voltage. The concept of a bipolar circuit is generally applicable to a variety of single-photon detector designs (e.g., conventional nSPAD or pSPAD) when the substrate-related operating voltage is too high or the supplied voltage is too low.This is especially true when purely unipolar concepts may not be sufficient to provide the operating voltage (above the breakdown voltage) of the single-photon detectors used.

[0023] Preferably, the first differential voltage results from the magnitude of the difference between the supply voltage U o and the first converter voltage -Ui. With a supply voltage U provided relative to electrical zero potential o For example, 2 V and an operating voltage of 22 V required to operate the first single-photon detector (relative to the electrical zero potential, e.g., a substrate-related zero voltage) would be compared to the provided supply voltage U. oA bias voltage of the first pn junction with a first converter voltage -Ui of -20 V is also possible. In this simple case, the first gain factor Fi can already be reduced from 11 times to 10 times the provided supply voltage U. o This can be reduced. A bipolar power supply can therefore significantly reduce the required amplification, and consequently the space requirements and power loss of the voltage converters used.

[0024] Preferably, the single-photon detector arrangement further comprises a second voltage converter, configured to supply the provided supply voltage U oto amplify with a second gain factor F2 greater than 1 and provide it as the second converter voltage U2 relative to the electrical zero potential, wherein the first differential voltage results from the magnitude of the difference between the second converter voltage U2 and the first converter voltage -Ui. In contrast to the previously described embodiment, the first pn junction is thus not biased relative to the supply voltage U. o but compared to a supply voltage Uo amplified by the second amplification factor F2 as the second converter voltage U2.

[0025] Preferably, the second gain factor F2 lies in a range between approximately 2 and 3 and / or between approximately 4 and 5. With the described circuit, a higher second gain factor F2 allows the first gain factor Fi to be reduced accordingly, while maintaining the same requirements for the operating voltage of the first single-photon detector. A maximum reduction in the required gain can be achieved with a bipolar circuit when the first gain factor Fi and the second gain factor F2 are approximately equal. The second voltage converter can be designed independently of the first voltage converter. However, it can also be a single, integrated voltage converter arrangement that can provide both the first converter voltage Ui and the second converter voltage U2. The first and second voltage converters can thus also be combined into a single voltage converter arrangement.A common voltage converter can be implemented.

[0026] Preferably, the differential voltage for biasing the first pn junction is symmetrical or asymmetrical with respect to the electrical zero potential. A symmetrical configuration has the advantage that, with a high required operating voltage for the first single-photon detector and a relatively low supply voltage, the required gains can be kept as low as possible. Even with an implementation using only the first voltage converter, the required overall gain can be approximately halved if the supply voltage, while too low, is still relatively high.For example, with a supply voltage of 11 V relative to electrical zero potential and a required operating voltage of 22 V for the first single-photon detector, a sufficiently high first converter voltage -Ui of -11 V relative to electrical zero potential can be provided by a single inverting first voltage converter with a first gain factor Fi of 1. The total gain factor of 2 required in this case can thus be achieved simply by inverting the supplied voltage.

[0027] In other cases, however, an asymmetrical differential voltage configuration can also be advantageous. This is particularly true if the voltage converter is intended to supply other components with voltage requirements that differ from the operating voltage of the first single-photon detector. An asymmetrical configuration may also be preferred if additional circuit characteristics (e.g., capacitances and resistances within the circuit) need to be taken into account.

[0028] Preferably, the single-photon detector arrangement further comprises a second pn junction formed in the substrate and configured as a second single-photon detector, wherein a first region located directly between the first pn junction and the second pn junction is electrically conductive. The second single-photon detector can, in particular, be configured as a single-photon avalanche diode (SPAD). Since the two pn junctions are electrically connected to each other via the first region, the pn junctions in the first region have a common electrical potential. In order for both to function as independent single-photon detectors, they must be coupled to the first region with the same polarity.

[0029] Preferably, the operating voltage of the second single-photon detector is higher than the supply voltage U. oThe second single-photon detector is biased with a second differential voltage with respect to a common electrical potential in the first region. This second differential voltage can, in particular, result from the difference between two supplied voltages, i.e., the supply voltage Uo, the first converter voltage Ui, the second converter voltage U2, or other supplied voltages. This means that the first and second single-photon detectors are each biased with a differential voltage that results with respect to the common electrical potential in the first region.

[0030] Preferably, the second pn junction is formed below the first pn junction. This means that the pn junctions are arranged vertically above one another in the substrate. Preferably, the first and second pn junctions can be deep-lying pn junctions generated in a BCD process. These can be created, for example, by partially forming p-doped regions in a deep-lying n-type buried layer (NBL) of a substrate (or by reversing the charge types).

[0031] Another aspect of the present invention relates to a monolithically integrated quantum random number generator (iQRNG) comprising a single-photon detector arrangement according to the invention, wherein a third pn junction configured as a photon source is formed above the first pn junction, and wherein an intermediate region located directly between the first pn junction and the third pn junction is electrically conductive. Since the two pn junctions are electrically connected to each other via the intermediate region, the pn junctions in the intermediate region have a common electrical potential. In order for both elements to function as independently operated elements, they must be coupled to the intermediate region with the same polarity.

[0032] Preferably, the third pn junction is biased with a third differential voltage with respect to a common electrical potential in the intermediate region. This third differential voltage can, in particular, be the magnitude of a difference between two applied voltages, i.e., the supply voltage Uo, the first converter voltage Ui, the second converter voltage U2, or other applied voltages. This means that the photon source and the first single-photon detector are each biased with a differential voltage that results with respect to the common electrical potential in the intermediate region.

[0033] Preferably, the photon source is a single photon source (SPS) configured to provide only one or a few photons simultaneously. Such photon sources providing only one or a few photons simultaneously are also referred to as single-photon emitters (SPS) within the scope of this application. However, these need not be true single-photon emitters, for example, based on a single isolated two-level system; rather, conventional light sources (e.g., silicon LEDs) can also be configured as (quasi-)SPS by sufficiently attenuating the emission or the supplied current.

[0034] Preferably, the photon source is a simple diode junction, an avalanche diode, or a single-photon avalanche diode (SPAD), each configured as an emitter or quasi-emitter, respectively. The photon source can be made of a direct semiconductor material (e.g., a 13 / V semiconductor) or, more preferably, of an indirect semiconductor material (e.g., silicon). The construction of radiating pn junctions is known to those skilled in the art. By supplying them with only a small current and optionally further attenuating the intensity of the generated radiation significantly using optical filters and absorbers, these pn junctions can be used for QRNG applications. With appropriate attenuation, operation as a PLC is even possible.The training in or from an indirect semiconductor material offers the particular advantage that photon generation is greatly hampered in such materials for energetic reasons, and thus the effective generation rate for photons is low even with well controllable current strengths during control, and operation as a PLC is possible without further additional attenuation measures.

[0035] It is also known that under certain circumstances, even non-radiating pn junctions (e.g., pn junctions in indirect semiconductors) can generate photons and thus emit radiation. However, the electro-optical conversion rate is generally so low that these rather randomly generated photons can usually be neglected in the design of typical circuits. In combination with a highly sensitive detector, however, such extremely weakly radiating pn junctions can also be used as photon sources in an iQRNG. What has been said about non-radiating pn junctions applies accordingly to avalanche diodes and SPADs, which are also typically assumed to be non-radiating. By appropriately selecting the operating parameters, a small amount of photon emission can also be observed and used for QRNG applications.In all the examples mentioned, the most important factor is the provision of a stable photon stream with extremely low intensity.

[0036] Preferably, the photon source can be a light-emitting avalanche Zener diode (Zener-avLED) operated at an operating point below or near the breakdown voltage. Preferably, the Zener-avLED has a breakdown voltage of < 10 V, more preferably a breakdown voltage of < 8 V, and even more preferably a breakdown voltage of < 7 V. This type of single-photon source allows a high single-photon rate at a relatively low operating voltage, even below and in the region of the Zener breakdown voltage. Therefore, Zener-avLEDs are particularly suitable for use as a single-photon source in an iQRNG according to the invention.

[0037] Further aspects of the present invention are disclosed in the dependent claims or in the following description of the drawings.

[0038] Brief description of the characters

[0039] The invention is explained in more detail below with reference to an exemplary embodiment and accompanying drawings. The figures show: Figure 1 an exemplary embodiment of an iQRNG according to the invention with three pn junctions arranged one above the other and requiring two different operating voltages;

[0040] Figure 2 shows an equivalent circuit diagram for the arrangement of pn junctions shown in FIG. 1;

[0041] Figure 3 shows a unipolar power supply according to the state of the art for an iQRNG;

[0042] Figure 4 shows an exemplary first embodiment of a bipolar power supply according to the invention for an iQRNG;

[0043] Figure 5 shows an exemplary second embodiment of a bipolar power supply according to the invention for an iQRNG;

[0044] Figure 6 shows an exemplary third embodiment of a bipolar power supply according to the invention for an iQRNG;

[0045] Figure 1 shows an exemplary embodiment of a bipolar power supply according to the invention for a single-photon detector arrangement; and

[0046] Figure 8 shows a preload of the pn junctions according to the invention with different differential voltages.

[0047] Detailed description of the invention

[0048] Detailed embodiments are now described, illustrated by way of example in the accompanying drawings. The effects and features of these embodiments are described with reference to the accompanying drawings. In the drawings, identical reference numerals denote identical elements, and redundant descriptions are avoided. The present disclosure can be implemented in various forms and is not to be understood as being limited only to the embodiments shown here. Rather, these embodiments are examples to ensure that this disclosure is thorough and complete and fully conveys the aspects and features of the present disclosure to the person skilled in the art.

[0049] Methods, elements, and techniques that are not necessary for a person skilled in the art to fully understand the aspects and features of this disclosure are therefore not described where applicable. In the drawings, the relative sizes of elements, layers, and areas may be exaggerated for clarity.

[0050] As used herein, the term “and / or” includes all combinations of one or more of the elements listed. Furthermore, the use of “may” in the description of embodiments of the present disclosure refers to “one or more embodiments of the present disclosure.” In the following description of embodiments, singular terms may also include the plural unless the context clearly indicates otherwise.

[0051] Although the terms "first" and "second" are used to describe different elements, these elements should not be restricted by these terms. These terms are used only to distinguish one element from another. For example, a first element may be called a second element, and likewise a second element may be called a first element, without this deviating from the scope of the present disclosure. Expressions such as "at least one of," when placed before a list of elements, modify the entire list and not just the individual elements of the list.

[0052] Terms such as "essentially", "approximately", and similar expressions are used as approximations, not as degrees, and are intended to account for the inherent variations in measured or calculated values ​​that are recognized by those skilled in the art. When the term "essentially" is used in conjunction with a characteristic that can be expressed by a numerical value, the term "essentially" refers to a range of at least + / - 5% of the value centered on that value.

[0053] Figure 1 shows an exemplary embodiment of an iQRNG 200 according to the invention with three pn junctions 50, 52, 60 arranged one above the other and requiring two different operating voltages. A single-photon detector arrangement according to the invention comprises a substrate 100 made of a semiconductor material; a first pn junction 50 formed in the substrate 100 and configured as a first single-photon detector 130, wherein an operating voltage of the first single-photon detector 130 (e.g., 1. SPAD) is provided above a supply voltage U relative to an electrical zero potential. o lies; a first voltage converter 210, set up to supply the voltage U oto invert, amplify with a first gain factor Fi greater than or equal to 1, and provide as the first converter voltage -Ui relative to electrical zero potential, wherein the first pn junction 50 is biased with a first differential voltage AV-I.SPAD with respect to the first converter voltage -Ui. The first differential voltage AV-I.SPAD can, for example, be the magnitude of the difference between the supply voltage U o and the first converter voltage -Ui.

[0054] A second voltage converter 220 is widely shown, configured to amplify the provided supply voltage Uo with a second gain factor F2 greater than 1 and to provide it as a second converter voltage U2 relative to the electrical zero potential, whereby the first differential voltage AV-I.SPAD can result from the amount of the difference between the second converter voltage U2 and the first converter voltage -Ui.

[0055] In the substrate 100, a second pn junction 52 is formed and configured as a second single-photon detector 140 (e.g., 2nd SPAD), wherein a first region 22 located directly between the first pn junction 50 and the second pn junction 52 is electrically conductive. The second single-photon detector 140 is biased with a second differential voltage AVZSPAD with respect to a common electrical potential in the first region 22. The second differential voltage AV 2 S PAD can be determined in particular from the magnitude of a difference between two applied voltages, i.e., the supply voltage Uo, the first converter voltage -Ui, the second converter voltage U2, or other applied voltages. This means that the first and second single-photon detectors 130, 140 are each supplied with a differential voltage AV-I.SPAD, AV 2 SPAD is pre-tensioned, which results with respect to the common electrical potential in the first region 22.

[0056] The second pn junction 52 is located below the first pn junction 50. Above the first pn junction 50, a third pn junction 60, configured as a photon source 120, is located, with an electrically conductive intermediate region situated directly between the first pn junction 50 and the third pn junction 60. The third pn junction 60 is connected to a third differential voltage AV. 3V The LED is biased with respect to a common electrical potential in the intermediate range. The third differential voltage AV 3VThe LED can be generated in particular by the magnitude of a difference between two supplied voltages, i.e., the supply voltage Uo, the first converter voltage -Ui, the second converter voltage U2, or other supplied voltages. This means that the photon source 120 and the first single-photon detector 130 are each supplied with a differential voltage AV. aV i_ED, AVi SPAD are biased, which results with respect to the common electrical potential in the intermediate region. The photon source 120 can in particular be a light-emitting avalanche Zener diode, Zener-avLED, operated at an operating point below or near the breakdown voltage.

[0057] The three pn junctions 50, 52, 60 can thus be configured, in particular, to form a light-emitting avalanche Zener diode (Zener-avLED) operating at an operating point below or near the breakdown voltage as a photon source 120, a first single-photon detector 130 as the first single-photon avalanche diode (1st SPAD), and a second single-photon detector 140 as the second single-photon avalanche diode (2nd SPAD). In particular, the three pn junctions 50, 52, 60 can be arranged vertically one above the other in a common BCD substrate 100 made of a semiconductor material (e.g., silicon) using BCD technology. Preferably, the photon source 120 is configured as a single-photon source (SPS) so that only one or a few photons 128 are provided simultaneously.

[0058] Preferably the BCD substrate 100 comprises a support substrate; and an epitaxial layer grown on the support substrate, wherein a low-lying pn junction 50 lying in the epitaxial layer was generated between the support substrate and the epitaxial layer by diffusion of dopants introduced into a surface of the support substrate below the epitaxial layer. Preferably, the first single-photon detector 130 forms an avalanche region in a region around the deep-lying pn junction 50 and comprises an absorption region (PW / HPW) with a high-voltage p-type well HPW and a p-type well PW for converting photons into electron-hole pairs, wherein the absorption region (PW / HPW) is directly adjacent to the regions forming the deep-lying pn junction 50, a deep-lying n-layer NBL, and a deep-lying p-layer PBL."n-type buried layer"). The fully developed high-voltage p-tub HPW enables optimal connection of the deep-lying pn junction 50 from the anode side.

[0059] Preferably, the deep-lying pn junction 50 is formed at least partially between a deep-lying n-layer NBL as cathode 132 and a deep-lying p-layer PBL immediately adjoining the deep-lying n-layer NBL, the absorption region (PW / HPW) is immediately adjoining the deep-lying p-layer PBL and is essentially formed as a p-region (optionally comprising an intrinsic region), and a p- + -Area P +The formed anode 134 is directly adjacent to the absorption region (PW / HPW). However, the forward directions and thus the arrangement of the anode and cathode terminals of the individual pn junctions can also be arranged differently. In particular, the individual pn junctions can be inverted with respect to their respective forward direction, so that the deep-lying pn junction 50 is at least partially formed between a deep-lying p-layer PBL as the anode and a deep-lying n-layer NBL immediately adjacent to the deep-lying p-layer PBL as the cathode.

[0060] In the illustrated embodiment, the respective anodes 124 and 134 of the photon source 120 and the first single-photon detector 130 are combined. These can then be electrically contacted, for example, via a common second metallization MET2 on the surface O of the substrate 100. A common and continuous second metallization MET2 can also be used to create a shading effect for shielding the entire space below. The corresponding cathodes 122 and 132 of the photon source 120 and the first single-photon detector 130 are each implemented separately as an example and can be electrically contacted via a corresponding first metallization MET1. In the illustrated embodiment, the cathode 132 of the first single-photon detector 130 is combined with the cathode 142 of the second single-photon detector 140 and can also be electrically contacted via the corresponding first metallization MET1.The associated anode 144 of the second single-photon detector 140 is also individually configured and can be electrically contacted via the associated first metallization MET1. The iQRNG 200 according to the invention can preferably be configured as a circular structure (corresponding to a spatial rotation of the illustrated plane about an imaginary central axis in the vertical direction). However, other configurations of the structure shown are also possible.

[0061] Figure 2 shows an equivalent circuit diagram for the arrangement of pn junctions shown in FIG. 1. This equivalent circuit diagram represents a general arrangement and is not limited to the embodiment of an iQRNG shown in FIG. 1. The forward directions, and thus the arrangement of the anode and cathode terminals of the individual pn junctions, can also be different. In particular, the individual pn junctions can be configured with inverted forward directions. The vertical structure is supplied with power at the top via a first terminal C. aVi_ED (cathode connection, see cathode 122 in FIG. 1) and on the underside via a second connection SX (anode connection, see anode 144 in FIG. 1), whereby the second connection SX can be routed over the substrate. The area between the photon source avLED and the first single-photon avalanche diode 1. SPAD is contacted via a common anode connection (see common anode 124, 134 in FIG. 1). Furthermore, the area between the first single-photon avalanche diode 1. SPAD and the second single-photon avalanche diode 2. SPAD is contacted via a common SPAD cathode connection CSPAD (see common cathode 132, 142 in FIG. 1).These connections allow the individual pn junctions to be supplied with different voltages, whereby when the second single-photon avalanche diode is omitted, the SPAD cathode connection CSPAD coincides with the second connection SX (which in this embodiment must then be implemented as the cathode connection).

[0062] Figure 3 shows a unipolar voltage supply according to the prior art for an iQRNG. In particular, the iQRNG can be a structure corresponding to the equivalent circuit shown in Figure 2. In this and the following figures, it is assumed that voltage converters are charge pumps. However, this is not intended to be a fundamental restriction to charge pumps; in particular, all other corresponding voltage converters are also intended to be included in the examples shown, instead of just the charge pumps shown. In the voltage supply shown, a supply voltage U is provided relative to an electrical zero potential. oThe signal is amplified unipolarly via two voltage converters (charge pumps 2 and 4) with different gain factors (F2 and F4) greater than or equal to 1, to the required converter voltages (U2 and U4). The provided converter voltages (U2 and U4) can then be used to bias the individual pn junctions of the iQRNG relative to the electrical zero potential. Due to the typically high values ​​required for the gain factors (F2 and F4), such a unipolar power supply exhibits the technical disadvantages already described in detail above.

[0063] Figure 4 shows an exemplary first embodiment of a bipolar power supply according to the invention for an iQRNG. In contrast to the unipolar power supply according to FIG. 3, two additional voltage converters (charge pumps 1 and 3) provide two further converter voltages (-Ui, -U3), wherein the additional voltage converters (charge pumps 1 and 3) invert the supply voltage Uo (they are thus inverting charge pumps) and amplify it with gain factors (F1 and F3) greater than or equal to 1. The provided converter voltages (-Ui, U2, -U3 and U4) can then be individually combined with each other to bias the individual pn junctions of the iQRNG by generating suitable differential voltages.In the illustrated embodiment, a first differential voltage between the first and fourth converter voltages (-Ui and U4), a second differential voltage between the first and second converter voltages (-Ui and U2), and a third differential voltage between the second and third converter voltages (U2 and -U3) are provided to bias the individual pn junctions of the iQRNG. This embodiment allows for both a symmetrical and an asymmetrical voltage supply to the two SPAD pn junctions of the iQRNG with respect to the electrical zero potential.

[0064] Figure 5 shows an exemplary second embodiment of a bipolar power supply according to the invention for an iQRNG. The second embodiment corresponds essentially to that shown in Figure 4, except that this example also includes a total of four voltage converters (charge pumps 1 to 4). However, the provided converter voltages (-Ui, U2, -U3, and U4) do not have to be derived exclusively from the supply voltage Uo; rather, the converter voltage of a first voltage converter can also be used as the input voltage for a downstream second voltage converter instead of the supply voltage Uo (cascaded arrangement). Such an amplifier concept also enables the provision of relatively high voltages with a modular design of the individual voltage converters.In the illustrated embodiment, the second converter voltage (U2) of the second voltage converter (charge pump 2) is inverted by the first voltage converter (charge pump 1) and amplified to the first converter voltage (-Ui) by a first gain factor (Fi) greater than or equal to 1. This embodiment also allows for both a symmetrical and an asymmetrical voltage supply to the two SPAD pn junctions of the iQRNG with respect to the electrical zero potential.

[0065] Figure 6 shows an exemplary third embodiment of a bipolar power supply according to the invention for an iQRNG. The third embodiment represents a simplified version of the first embodiment according to FIG. 4. Only two voltage converters (charge pump 1 and 2) are required. This embodiment allows only an asymmetrical power supply to the two SPAD pn junctions of the iQRNG with respect to the electrical zero potential.

[0066] Figure 7 shows an exemplary embodiment of a bipolar voltage supply according to the invention for a single-photon detector arrangement according to the invention. The first voltage converter (charge pump 1) is configured to provide a supply voltage U relative to an electrical zero potential. o to invert, amplify with a first gain factor Fi greater than or equal to 1, and provide as the first converter voltage -Ui relative to electrical zero potential, wherein the pn junction configured as a SPAD is biased with a differential voltage with respect to the first converter voltage -Ui. The first differential voltage is the magnitude of the difference between the supply voltage Uo and the first converter voltage -Ui.

[0067] Figure 8 shows a bias voltage of the pn junctions according to the invention with different differential voltages AV. aVi_ED, AVi SPAD, AV2SPAD. The equivalent circuit diagram described above, based on the exemplary embodiment shown in FIG. 1, is shown again on the left side in FIG. 2. However, the equivalent circuit diagram shown is not limited to this specific embodiment of an iQRNG and is intended to encompass a variety of possible circuit arrangements in which high operating voltages for operating an integrated single-photon detector are to be generated from a small supply voltage without high power losses or a large area requirement. In this specific example, two different operating voltage levels are to be provided from a small supply voltage.The presented concept for the bipolar voltage supply of single-photon detectors is also applicable to more than two different operating voltage levels for operating one or a multitude of single-photon detectors, with or without additional voltage levels for other circuit elements and components. Accordingly, the voltage values ​​given for the photon source avLED, the first single-photon detector 1. SPAD, and the optional second single-photon detector 2. SPAD are to be understood as examples only.

[0068] Based on the values ​​mentioned for the specific example above, two alternative power supply concepts are explained in more detail below.

[0069] In the symmetrical embodiment, the first SPAD (with the highest voltage level) is supplied with voltages of equal magnitude but opposite polarity to zero potential via the anode terminal (-11 V) and the SPAD cathode terminal (CSPAD) (+11 V). The required operating voltage for the photon source avLED (with a lower voltage level) can then be provided by simply applying a small negative voltage to the first terminal C. aVi_ED (approx. -4 V). Alternatively, the photon source avLED can also be supplied via current regulation. The optional second SPAD from the aforementioned embodiment can also be supplied with the required operating voltage (in this example, the same as the first SPAD) by providing a voltage at the second terminal SX (-11 V) that is symmetrically inverted from the voltage applied to the SPAD cathode terminal CSPAD (+11 V). The voltage applied to the second terminal SX allows the second SPAD to be switched independently of the first SPAD.

[0070] In the asymmetrical embodiment, the voltage supply to SPAD 1 (with the highest voltage level) is provided via voltages with different values ​​and polarities through the anode terminal (-7 V) and the SPAD cathode terminal (CSPAD) (+15 V). In this case, the required operating voltage of the photon source avLED (with a lower voltage level) can be achieved simply by connecting the first terminal C. aVi_ED can be supplied with zero electrical potential (0 V / GND). Alternatively, the power supply to the photon source avLED can also be provided via current regulation. The optional second SPAD from the aforementioned embodiment can then be supplied with the required operating voltage by applying the voltage present at the anode terminal to the second terminal SX (-7 V). In this embodiment as well, the second SPAD can be switched independently of the first SPAD via the voltage applied to the second terminal SX. The negative bias of the substrate typically does not pose a problem for the area below the SPADs, as this area is generally isolated from the environment and other electronic components.

[0071] With reference to a preferred supply voltage between 2 V and 3 V, a bipolar voltage supply according to the invention significantly reduces the requirements for providing higher voltages, particularly for operating SPADs integrated into a semiconductor substrate. While the supply voltages differ by a factor of approximately 8 to 10 in conventional voltage supplies, these requirements can be roughly halved by a bipolar voltage supply according to the invention, so that the higher supply voltages provided from the base voltage differ by only a factor of approximately 4 to 5 in the case shown. This enables the use of simpler multi-stage charge pumps to generate higher supply voltages, although a corresponding inversion of the provided supply voltage must be performed.

[0072] Reference symbol list

[0073] 22 first area (e.g. NBL)

[0074] 32 second area (e.g. PBL)

[0075] 50 first pn junction (e.g. upper pn junction of a double structure)

[0076] 52 second pn junction (e.g. lower pn junction of a dual structure)

[0077] 60 third pn junction (e.g. light-emitting pn junction)

[0078] 100 substrate

[0079] 120 / avLED photon source (e.g. single photon source, PLC, especially Zener-avLED)

[0080] 122 / Ca LED cathode (n + , photon source)

[0081] 124 / Anode Anode (p + , photon source)

[0082] 128 photons

[0083] 130 / 1. SPAD first single-photon detector (e.g. single-photon avalanche diode, SPAD)

[0084] 132 / CSPAD Cathode (n + , first single-photon detector)

[0085] 134 / Anode Anode (p +, first single-photon detector)

[0086] 140 / 2. SPAD second single-photon detector (e.g. single-photon avalanche diode, SPAD)

[0087] 142 I CSPAD Cathode (n + , second single-photon detector)

[0088] 144 / SX Anode (p + , second single-photon detector)

[0089] 200 QRNG (integrated Quantum Random Number Generator)

[0090] 210 First voltage converter (e.g. charge pump)

[0091] 220 second voltage converter (e.g. charge pump)

[0092] NBL deep n-layer (engl. “n-type buried layer”)

[0093] PBL deep p-layer (English: "p-type buried layer")

[0094] H(V)PW high-voltage p-type well

[0095] H(V)NW high-voltage n-type well

[0096] PW p-well

[0097] PBODY p-type body

[0098] NEPI weakly n-doped or (approximately) intrinsic epitaxial region

[0099] MET1, MET2 Metallization

[0100] CONT Contact

[0101] VIA1 Via

[0102] STI isolation area (English: "shallow trench isolation") polysilicon layer

[0103] P + p + -Area N + n + -Area

[0104] Surface (substrate, e.g., BCD substrate)

Claims

Patent claims 1. Single-photon detector arrangement comprising: a substrate (100) made of a semiconductor material; a first pn junction (50) formed in the substrate (100) and configured as a first single-photon detector (130), wherein an operating voltage of the first single-photon detector (130) is provided above a supply voltage U relative to an electrical zero potential o lies; a first voltage converter (210), configured to invert the supply voltage Uo, amplify it with a first gain factor Fi greater than or equal to 1 and provide it as the first converter voltage -Ui relative to the electrical zero potential, wherein the first pn junction (50) is biased with a first differential voltage (AV-I.SPAD) relative to the first converter voltage -Ui.

2. Single-photon detector arrangement according to claim 1, wherein the first differential voltage (AV-I.SPAD) is derived from the magnitude of the difference between the supply voltage U o and the first converter voltage -Ui results.

3. Single-photon detector arrangement according to claim 1, further comprising a second voltage converter (220) configured to convert the provided supply voltage U o to amplify with a second gain factor F2 greater than 1 and to provide as a second converter voltage U2 relative to the electrical zero potential, wherein the first differential voltage (AV-I.SPAD) results from the amount of the difference between the second converter voltage U2 and the first converter voltage -Ui.

4. Single-photon detector arrangement according to one of the preceding claims, wherein the first differential voltage (AV-I.SPAD) is symmetrical or asymmetrical with respect to the electrical zero potential.

5. Single-photon detector arrangement according to one of the preceding claims, further comprising a second pn junction (52) formed in the substrate (100) and configured as a second single-photon detector (140), wherein a first region (22) located directly between the first pn junction (50) and the second pn junction (52) is electrically conductive.

6. Single-photon detector arrangement according to claim 5, wherein an operating voltage of the second single-photon detector (140) is above the supply voltage U o lies and the second single-photon detector (140) is biased with a second differential voltage (AVZSPAD) with respect to a common electrical potential in the first region (22).

7. Single-photon detector arrangement according to claim 5 or 6, wherein the second pn junction (52) is formed below the first pn junction (50).

8. Monolithically integrated quantum random number generator, iQRNG, comprising a single-photon detector arrangement according to one of the preceding claims, wherein a third pn junction (60) configured as a photon source (120) is formed above the first pn junction (50), wherein an intermediate region located directly between the first pn junction (50) and the third pn junction (60) is electrically conductive.

9. iQRNG according to claim 8, wherein the third pn junction (60) is connected to a third differential voltage (AV) aV i_ED) is biased with respect to a common electrical potential in the intermediate range.

10. iQRNG according to claim 8 or 9, wherein the photon source (120) is a light-emitting avalanche Zener diode, Zener-avLED, operated at an operating point below or near the breakdown voltage.

Images