Patent Application
20180158849
This application is based upon and claims the benefit of priority from United Kingdom Patent Application No. 1620675.7, filed Dec. 5, 2016, the entire contents of which are incorporated herein by reference
The present invention relates to photodiode devices and methods of their manufacture. Specifically, but not exclusively, the present invention relates to methods of manufacturing avalanche photodiodes.
Optical sensors, such as photodiodes, are widely used in integrated complementary metal oxide semiconductor (CMOS) devices for conversion of light into an electrical current with an analogue or a digital signal output. Photodiode applications may include light meters, smoke detectors and optical communications.
Typically, a standard photodiode is formed by doping a silicon wafer to alter its electrical properties and to produce regions of different electrical conductivity, referred to as p-type and n-type regions. N-type (negative) regions contain mostly free electrons while p-type (positive) regions contain mostly “holes” (i.e. the absence of an electron). Where these regions meet a p-n junction is formed. An incoming photon of sufficient energy generates an electron-hole pair which is separated by the internal electrical field in the p-n junction, causing electrical current to flow. Such photodiodes may consist of standard CMOS implants and may have a wavelength dependent responsivity signal.
Stacked photodiodes can be used in order to obtain spectral information regarding incoming light. Stacked photodiodes are provided with multiple interleaved p-n junctions. In a typical top-illuminated CMOS stacked photodiode, light of shorter wavelength (such as visible light) is absorbed in the upper p-n junction, due to the larger absorption coefficient of silicon for light of short wavelength. Light of longer wavelength (such as infrared light) is absorbed in the larger volume of the lower p-n junction of the stacked photodiode. Each junction or independent photodiode of the stacked photodiode therefore responds to different wavelengths of light.
Stacked photodiodes have many applications, such as the ambient light sensor described in US2014263972A1 and the multispectral imager disclosed in US20150129747A1. US2014263972A1 also discloses a “dark” stacked photodiode structure, fully covered by metal layers to block all incident light, which may be used in conjunction with the stacked photodiode structure to improve accuracy at low ambient lights levels. US20150129747A1 further discloses the use of a colour filter array to provide wavelength discrimination between the visible and infrared detecting photodiodes.
Another type of photodiode is the avalanche photodiode device (APD). An APD allows for internal amplification of the electron-hole pair production by application of a high electrical field inside the p-n junction. This high field is generated by the application of a sufficiently large voltage between the cathode and anode of the avalanche photodiode, i.e. by reverse-biasing the photodiode. The electrons produced by an incoming photon are accelerated such that they produce further electron hole pairs through impact ionisation. In this way, a single photon can typically produce hundreds or thousands of electrons, producing a gain or multiplication effect of around ×100 or ×1000 in comparison with a normal photodiode. An APD operating in a linear mode can therefore provide a much higher level of sensitivity than a stacked photodiode. The doping level of the p-n junction in an APD is specifically designed to produce this response.
If the reverse-bias voltage applied to the avalanche photodiode is high enough to produce the above “avalanche” effect, but still below a threshold level, then the avalanche produced by the photon of light will eventually die down again due to friction losses. However, if the reverse-bias voltage increases beyond a critical or breakdown level, the electrical field in the device is high enough so that one photon can fire the “avalanche”. When operated with a reverse-bias voltage which is high enough above the breakdown voltage, an APD can operate in Geiger-mode as a single-photon avalanche photodiode (SPAD) with gain levels of up to ×106. In other words, an SPAD is sensitive enough to detect single photons of incident light. Again, the p-n junction is designed to be able to operate at these gain levels and a quenching circuit is typically required to reduce the voltage to the breakdown level. Quenching may be passive or active. Passive quenching includes the use of a series resistor to the APD, while active quenching includes the use of a transistor to control the voltage of the APD.
Examples of avalanche photodiodes are described in U.S. Pat. No. 3,886,579 and WO2012032353A2. U.S. Pat. No. 7,547,872 discloses an integrated circuit comprising an array of single-photon avalanche diodes, and U.S. Pat. No. 8,779,543B2 describes a device having an avalanche photo diode and includes a method for sensing photons. “Epitaxial silicon single photon detector with reduced carrier diffusion effect and picosecond resolution”, M. Ghioni/Cova et al. IEDM 1987, describes a single-photon avalanche diode (SPAD).
The inventors have appreciated that additional filters or components may be required when using conventional stacked photodiodes to measure an exact ambient light level, for example. This adds costs and complexity to the photodiode device and may cause such a device to require a larger footprint.
The inventors have further appreciated that an avalanche photodiode, and in particular a single-photon avalanche photodiode, can provide a much higher degree of sensitivity than a stacked photodiode, and that it would be beneficial to be able to provide the functionality of both a stacked and an avalanche photodiode in the same device.
Aspects of the invention are set out in the independent claims.
Some embodiments of the invention will now be described by way of example only and with reference to the accompanying drawings, in which:
Described herein with reference to
The device 12 is further provided with electrical contacts, comprising a cathode (DNWELL) 24, an anode (PWELL) 22 and substrate contact (p− Epi) 30. The electrical contacts may be formed by diffusing p-type and n-type impurities into the diode structure. In this example, the electrical contacts 22, 24 are separated by a shallow trench isolation (STI) virtual guard ring 26.
The stacked photodiode device 12 illustrated in
In this context, the term “multiplication implant” refers to an implant which creates a single multiplication or acceleration region at an appropriate location within the photodiode device. In other embodiments, one or more implants may be used to create a single multiplication region within the photodiode device. The one or more implants may be implanted in a single step.
The implant 28 is configured to be located within the DNWELL 18, and adjacent to the junction with the PWELL 20 (the p-n junction). In order to increase effectiveness, the minimum implant 28 size is generally defined by the active area of the photodiode. Using a smaller implant size than this minimum implant size may still result in a suitable device but the effectiveness would be reduced as it would be expected that some photons will not be multiplied. The maximum implant 28 size is limited by the need to avoid edge breakdown.
The multiplication implant 28 creates an avalanche or multiplication region within the stacked photodiode 12 structure, and hence an avalanche photodiode device is formed. To achieve this, the dose of the implant 28 should be high enough to produce, upon application of reverse bias, an electrical field of, for example, greater than 3.5×105 volts per cm (V/cm) in the avalanche region, i.e. corresponding to the region of the p-n junction between the PWELL 20 and DNWELL 18.
It will be appreciated that the device 10 functions as an avalanche photodiode 32 only when a suitable reverse bias is applied. A suitable reverse bias may be in the range 14 to 20 V, for example. This has the advantage that the device 10 can function as a normal stacked photodiode 12 until the application of sufficient reverse bias. If the reverse bias sufficiently exceeds the breakdown voltage then the device 10 can operate as the second type of avalanche photodiode described above, i.e. a single photon avalanche diode (SPAD), capable of detecting single photons. A suitable circuit and quenching is used for SPAD operation. As discussed above, quenching may be passive or active. Effectively, increasing the reverse bias voltage increases the sensitivity of the device 10. Upon sufficient decrease of the reverse bias voltage (for example, a decrease down to around 2 V reverse bias) the device 10 will resume operation as a simple stacked photodiode device 12.
The inventive method of manufacture described herein therefore provides a single device 10 which is operable as a stacked photodiode, an avalanche photodiode or an SPAD. The stacked photodiode 12 and the avalanche photodiode 32 share the same guard ring 26 periphery and contact structure. Both stacked photodiode 12 and avalanche photodiode 32 devices share a bottom p-n junction (the junction between the p− epi layer 16 and the DWELL 18) and identical parasitic capacitances and leakage current. In other words, apart from the multiplication implant 28 the layout of both devices 12, 32 is the same. The addition of the multiplication implant 28 has no effect upon the footprint of the original stacked photodiode 12.
Since the manufacturing process uses only a single mask implantation step and creates only a single multiplication region, the resulting device 10 can have a small footprint and a high fill factor. In one embodiment, the only additional step beyond the provision of the stacked photodiode device 12 is the single implantation step in order to form implant 28. The addition of only one further step to the standard process for manufacturing a stacked photodiode device keeps manufacturing costs down. The inventors envisage that an annealing step would normally be carried out after implantation of implant 28. However, since annealing is typically carried out as part of the stacked photodiode manufacturing process no separate, additional annealing step is required in relation to implant 28, i.e. the annealing of implant 28 can be carried out as part of the annealing that would normally be required during the manufacture of the stacked photodiode device. To this end, the annealing step (as part of the manufacture of the stacked photodiode device) is deferred until after the implantation step which forms implant 28.
A further benefit of the manufacturing process herein described is that the configuration of the multiplication implant 28 can be varied in order to produce avalanche photodiodes 32 having different breakdown characteristics, thus allowing different voltage application ranges, as required. Thus, the point at which the device 10 begins to function as an APD can be controlled.
Step 1: Providing a stacked photodiode device, the stacked photodiode device comprising: a substrate having a first conductivity type; a first well having a second conductivity type, within the substrate; and a second well having the first conductivity type, within the first well.
Step 2: Modifying the stacked photodiode device by implanting a multiplication implant within the first well, so as to manufacture an avalanche photodiode device.
The method further comprises modifying at least one of the stacked photodiode devices 12 by implanting a multiplication implant 28 within the first well 18 associated with the at least one stacked photodiode device 12 adjacent the respective second well 20, so as to manufacture at least one avalanche photodiode device 32. It will be appreciated that more than one stacked photodiode device 12 may be modified using this method, each being provided with its respective multiplication implant 28. In the array of
The array 36 of stacked 12 and avalanche 32 photodiode devices produced by the above-described method of manufacture share the same guard ring 26 periphery and bottom p-n junction. This may limit or reduce crosstalk between the avalanche photodiode devices 32. It will be appreciated that the implantation process as described above can be continued such that every other stacked photodiode 12 on the wafer is converted into an avalanche photodiode 32. Alternatively, stacked photodiode devices 12 formed upon the wafer can be converted to avalanche photodiode devices 32 as required so that, for example, every fifth stacked photodiode is provided with a multiplication implant and hence converted to be operable as an avalanche photodiode.
As illustrated in the plan view of
An array 36 so produced can advantageously act as a photosensor or the like to detect the wavelength of incoming light using two or more separately optimized devices. In this way, the ambient light signal level can be (more) accurately determined without the need for additional filters or components.
Step 1: Providing two or more stacked photodiodes, each of the stacked photodiodes located within a common first well, the first well located on or within a substrate, the substrate having a first conductivity type and the first well having a second conductivity type; each of the two or more stacked photodiodes having a respective second well having the first conductivity type, within the first well.
Step 2: Modifying at least one of the stacked photodiodes by implanting a multiplication implant within the first well adjacent the respective second well, so as to manufacture an avalanche photodiode.
As discussed above, in further embodiments a single multiplication region may be created within a stacked photodiode by implanting more than one multiplication implant.
The term “device” as described herein may refer to an individual photodiode or may refer to a number of photodiodes of the same or different types located on the same wafer or otherwise associated with one another, depending upon the context.
Although the invention has been described in terms of certain embodiments as set forth above, it should be understood that these embodiments are illustrative only and that the claims are not limited to those embodiments. Those skilled in the art will be able to make modifications and alternatives in view of the disclosure which are contemplated as falling within the scope of the appended claims. Each feature disclosed or illustrated in the present specification may be incorporated in the invention, whether alone or in any appropriate combination with any other feature disclosed or illustrated herein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1620675.7 | Dec 2016 | GB | national |
