CN103165628B - Multifunctional exposure imaging method based on composite dielectric grating metal-oxide-semiconductor field-effect transistor (MOSFET) light-sensitive detector - Google Patents

Abstract

Provided is a multifunctional exposure imaging method based on a composite dielectric grating metal-oxide-semiconductor field-effect transistor (MOSFET) light-sensitive detector. The exposure imaging method includes the steps that negative bias pulse Vb is added to a substrate, a forward voltage pulse Vp is added to one end of a source electrode or a drain electrode, the other end of the source electrode or the other end of the drain electrode is floated, or the source electrode and the drain electrode are respectively provided with one forward voltage pulse Vp, the source electrode and the drain electrode are additionally provided with a bias pulse larger than bias voltage of the substrate, meanwhile zero bias voltage or a forward bias pulse Vg is added to a control grid, and the substrate and a source and drain region generate a depletion layer. The number range of the Vg is 0-20V, the number range of the Vb is -20-0V, and the number range of the Vp is 0-10V. By means of adjustment of voltage, the detector can collect photoelectrons so that exposure imaging of the device is carried out. The multifunctional exposure imaging method has the advantages of having low-voltage operation, being free of dark current interference and accurate in imaging, detecting in low light, and being fast in imaging speed and the like.

Description

Multifunctional Exposure Imaging Method Based on Composite Dielectric Gate MOSFET Photosensitive Detector

technical field

The invention relates to a signal acquisition method of an imaging detection device, in particular to the working mechanism of an imaging detection device based on a composite dielectric gate MOSFET for infrared, visible light to ultraviolet bands, and is a multifunctional exposure imaging method for a composite dielectric gate MOSFET photosensitive detector.

Background technique

Imaging detectors are widely used in various fields such as military and civilian use. The main imaging detectors currently developed are CCD and CMOS-APS. CCD appeared earlier and the technology is relatively mature. Its basic structure is a series of MOS capacitors connected in series The generation and change of the semiconductor surface potential well is controlled by the voltage pulse sequence on the capacitor, and then the storage and transfer readout of the photogenerated charge signal is realized. It is precisely because of this signal transfer characteristic that the charge transfer speed is very limited, so the imaging speed is not high. In addition, since capacitors are connected in series, a problem with one capacitor will affect the transmission of the entire line of signals, so the process requirements are extremely high, and the yield and cost are not ideal. Each pixel of CMOS-APS is composed of diodes and transistors. Each pixel is independent of each other. There is no need to serially move charges during the entire signal transmission process. A problem with a certain pixel does not affect the performance of other pixels, so it overcomes CCD. Due to the shortcomings in this aspect, the process requirements are not so harsh. COMS uses single-point signal transmission and uses simple X-Y addressing technology to allow data to be read from the entire arrangement, part or even unit, thereby increasing the addressing speed. Achieve faster signal transmission. However, each pixel of CMOS-APS is composed of multiple transistors and a photosensitive diode (including amplifier and A/D conversion circuit), so that the photosensitive area of each pixel only occupies a small surface area of the pixel itself, and the sensitivity and resolution are relatively small. .

Through comparison, it is found that the two traditional imaging detection technologies have their own advantages and disadvantages. CMOS-APS has achieved rapid development along with the continuous progress of CMOS technology in recent years, showing us its great prospects. It can be seen that a CMOS-based technology and can be proposed Imaging detectors that overcome the shortcomings of traditional CMOS-APS as much as possible are of great significance. Therefore, the present applicant proposed a composite dielectric gate photosensitive detector based on CMOS technology in patent WO2010/094233.

Contents of the invention

The object of the present invention is to propose a multi-functional exposure imaging method based on a composite dielectric gate MOSFET photosensitive detector, and adjust the working mode of the device through the source-drain bias voltage. More photoelectrons can be collected to obtain signal amplification. When the electric field in the depletion region reaches an avalanche electric field, even weak light can stimulate the generation of enough photoelectrons.

According to the technical solution of the present invention, the composite dielectric gate MOSFET photosensitive detector single-tube structure (as shown in Figure 1) includes: a semiconductor substrate (P-type) 1; an underlying insulating medium 5, an optoelectronic storage layer 4, and Top layer insulating dielectric 3, control gate 2; N-type source 6 and drain 7 are formed by ion implantation in semiconductor substrate 1 (close to both sides of stacked dielectric); the charge storage layer 4 is polysilicon, Si ₃ N ₄ or other electronic conductors or semiconductors; the control grid 2 is polysilicon, metal or transparent conductive electrode, and at least one of the control grid surface or base layer is a transparent or semi-transparent window for the detection wavelength of the detector. The two layers of insulating medium effectively isolate the charge storage area, so that the storage function is realized in the charge-limited charge storage layer 4, which is generally a broadband semiconductor, so as to ensure that electrons can enter the charge storage layer 4 from the P-type semiconductor substrate 1 across the potential barrier. The bottom dielectric material can be silicon oxide, SiON or other high dielectric constant dielectric; the top dielectric material can be silicon oxide/silicon nitride/silicon oxide, silicon oxide/alumina/silicon oxide, silicon oxide, aluminum oxide or other high The dielectric constant of a dielectric material. The dimensions of the bottom insulating medium 5 , the optoelectronic storage layer 4 , the top insulating medium 3 and the control gate 2 are 22nm*22nm~0.18μm*0.18μm.

Multifunctional exposure imaging method based on composite dielectric gate MOSFET photosensitive detector: exposure imaging process steps are: add a negative bias voltage pulse Vb to the substrate, add a forward voltage pulse Vp to one end of the source electrode 6 or the drain electrode 7 and another One end is floating or the source 6 and the drain 7 are simultaneously applied with a positive voltage pulse Vp, the source and the drain are applied with a bias pulse greater than the substrate bias voltage, and the control gate is applied with zero bias or 0 ~20V forward bias pulse Vg. The substrate and source and drain regions will generate depletion layers. The photoelectrons generated by photon excitation are driven by the control gate and the source-drain electric field to accelerate toward the interface between the P-type semiconductor substrate 1 and the bottom layer dielectric 5, and when the energy of the electrons reaches a certain level, they will cross the bottom layer dielectric 5 and the P-type semiconductor layer. The potential barrier of the substrate 1 is injected into the charge storage layer 4; some photoelectrons are accelerated to move toward the high field region of the source-drain PN junction under the drive of the source-drain junction electric field, and the values of Vp and Vb are adjusted so that the voltage difference between Vp and Vb is 0~11V, can make this PN junction reach deep depletion, can generate more photoelectrons when illuminated, these photoelectrons will be injected into the charge storage layer 4, the change of the charge amount of the charge storage layer 4 leads to the change of the threshold voltage of the photosensitive detector, Since a bias pulse greater than the substrate bias voltage is added to the source and drain of the photosensitive detector device, when no photons provide energy, electrons are discharged from the source and drain, and almost no electrons are injected into the charge storage layer 4, so there is almost no dark current. At the same time, because this is a hot electron process, the collection of photoelectrons is almost not limited by the electric field in the underlying medium, and the gate voltage can be imaged at a small time, so it can be operated at low voltage. When there is light, the photons provide energy to the electrons, so that the electrons can cross the potential barrier between the underlying dielectric 5 and the P-type semiconductor substrate 1 and inject into the charge storage layer 4 . This change can be obtained through the reading process, and then the number of photoelectrons in the charge storage layer 4 can be known.

Adjust the values of Vp and Vb so that the voltage difference between Vp and Vb is greater than 11V, and the PN junction of the source and drain of the photosensitive detector reaches the avalanche voltage, and more electron-hole pairs will be excited near the junction during the electron acceleration process. At the same time, these electron-hole pairs are endowed with sufficient energy to cross the potential barrier between the underlying dielectric 5 and the P-type semiconductor substrate 1, so that more charges will be injected into the charge storage layer 4, and the location where the avalanche occurs can be adjusted by adjusting the voltage (pn The junction area or the depletion layer directly below the bottom layer), when collected in an avalanche state, the photoelectric signal is amplified and the gain is very large, even a very weak light intensity can also stimulate enough photoelectrons to generate. Because the avalanche process is very fast, only a short voltage pulse is needed to realize programming.

The beneficial effects of the present invention are: the composite dielectric gate MOSFET photosensitive detector works under the source-drain bias condition, and the device can be operated in different modes by adjusting the operating voltage, so that the lower device exhibits different performances and expands its application Range. By increasing the substrate voltage Vb or Vp, the depletion region increases, and more photoelectrons can be collected. At the same time, the electric field on the surface of the P-type semiconductor substrate 1 increases, and the generated photoelectrons move toward the P-type semiconductor substrate 1 and the underlying medium. 5. During the process of moving at the interface and in the direction of source and drain, the energy will excite more electron holes and obtain signal amplification. When the electric field in the depletion region reaches the avalanche electric field, even weak light can excite enough Generation of photoelectrons.

Apply a negative bias pulse Vb to the P-type semiconductor substrate 1, apply a positive voltage pulse Vp to one end of the source 6 or drain 7 (the other end is floating), or apply a forward voltage to the source 6 and drain 7 at the same time Pulse Vp, while the control gate 2 is to be applied with zero bias or forward bias pulse Vg, the P-type semiconductor substrate and the source and drain regions will produce a depletion layer. The photoelectrons generated by the photon excitation are driven by the control gate and the source-drain voltage to accelerate toward the interface between the P-type semiconductor substrate 1 and the bottom dielectric 5, and when the electron energy reaches a certain level, they will cross the bottom dielectric 5 and the P-type semiconductor substrate. The potential barrier of the bottom 1 is injected into the charge storage layer; and some photoelectrons flow to the source and drain regions driven by the electric field of the source and drain junctions.

Increase the substrate voltage Vb or Vp, the depletion region increases, and more photoelectrons can be collected. At the same time, the surface electric field of the P-type semiconductor substrate 1 increases, and the generated photoelectrons move toward the interface between the P-type semiconductor substrate 1 and the underlying medium 5 In the process of moving in the direction of source and drain, the energy will excite more electron holes, and give these electrons enough energy to cross the potential barrier between the bottom dielectric 5 and the P-type semiconductor substrate 1, and obtain signal amplification. When exhausted When the electric field in the region reaches an avalanche electric field, even a very weak light can stimulate the generation of enough photoelectrons.

The working mode according to the present invention has the following characteristics: low-voltage operation: under the working conditions, the photoelectron collection process can adjust the control gate 2 voltage and adopt the thermal electron injection mode, without tunneling, and can reduce the operating voltage. For example, the grid voltage Vg can be 5v About 10 volts, which is far lower than the high voltage required by tunneling. No dark current: under the working conditions, the source and drain are in forward bias, when there is no photon to provide energy, and when the detector works at a lower voltage of the PN junction, the electrons are almost all discharged from the source and drain, so there is almost no Dark current, when photons are injected, the photons provide enough energy for electrons to cross the potential barrier between the underlying dielectric 5 and the P-type semiconductor substrate 1 and inject them into the charge storage layer 4 . Accurate imaging: The described working condition adopts the injection method of hot electrons, and the electrons collected by the photoelectron storage layer are hardly affected by the gradual decrease of the electric field in the bottom medium 5, and have a linear relationship with the light intensity, so they can be read by reading The electrons collected by the charge storage layer 4 can obtain accurate light intensity information, thereby obtaining accurate images. High-speed detection: Under the above working conditions, the source-drain PN junction can be in a near-breakdown state during the exposure process by adjusting the voltage, and photons trigger avalanche storage of photoelectrons to realize light detection. The avalanche speed is very fast, and the exposure can be completed in a short time process, which helps to achieve high-speed detection. Weak light detection: the working conditions can make the device work under the avalanche condition, the optical signal triggers the avalanche to realize the multiplication effect, the gain is very large, and it also has a good response to weak light, and can realize weak light detection.

Description of drawings

Figure 1 shows the basic structure of the composite dielectric gate MOSFET photosensitive detector.

Figure 2 shows the formation of the depletion layer when the detector works under exposure conditions.

Figures 3–6 are schematic diagrams of four types of photoelectron transfer and storage.

Figure 7 is a schematic diagram of threshold voltage changes during exposure.

Figure 8 shows the relationship between the threshold voltage change and the substrate voltage before and after exposure of the detector.

Figure 9 shows the relationship between the threshold voltage change and light intensity of the detector before and after exposure.

Specific implementation method

The working process and physical mechanism of the present invention will be described below with reference to the accompanying drawings as follows

Figure 1 shows the basic structure of a composite dielectric gate MOSFET photosensitive detector, a semiconductor substrate (P-type) 1, a bottom insulating medium 5, an optoelectronic storage layer 4, a top insulating medium 3, and a control gate 2 are arranged in sequence directly above the semiconductor substrate. N-type source 6 and drain 7 are formed in the semiconductor substrate 1 (close to both sides of the stack medium) by ion implantation doping.

The optoelectronic storage layer is polysilicon, Si ₃ N ₄ or other electronic conductors or semiconductors; the control grid 2 is polysilicon, metal or transparent conductive electrodes, and at least one of the control grid surface or base layer is transparent or semi-transparent to the detection wavelength of the detector. transparent window. The two layers of insulating medium effectively isolate the charge storage region 4, so that the storage function is realized in the charge-limited charge storage layer 4, which is generally a broadband semiconductor, so as to ensure that electrons can pass through the potential barrier from the P-type semiconductor substrate 1 and enter the charge storage layer 4. The bottom dielectric material can be silicon oxide, SiON or other high dielectric constant dielectric; the top dielectric material can be silicon oxide/silicon nitride/silicon oxide, silicon oxide/alumina/silicon oxide, silicon oxide, aluminum oxide or other high Dielectric constant dielectric material. The dimensions of the bottom insulating medium 5 , the optoelectronic storage layer 4 , the top insulating medium 3 and the control gate 2 are 22nm*22nm~0.18μm*0.18μm.

Photoelectric conversion process: the programming process shown in Figure 2, add a 0~10V positive voltage pulse Vp to one end of the source 6 or drain 7 and leave the other end floating or add a 0 to the source 6 and drain 7 at the same time For the same forward voltage pulse of ~10V, a negative bias pulse Vb of -20~0V is added to the P-type semiconductor substrate 1, and a forward bias pulse Vg of 0~20V is added to the control gate 2, because the source 6 The drain 7 is a PN junction with respect to the substrate, and a deep depletion layer region is formed under reverse bias, and a voltage of -20~0V applied to the substrate can form a depletion layer on the substrate surface, so that the entire substrate under the A continuous depletion region is formed as shown by the dotted line. When a photon reaches the depletion region, if the photon energy photon h v > semiconductor E _g (or E _g + Δ E _c ), the photon is absorbed by the semiconductor and excites an electron-hole pair.

Electron transfer and storage: There are mainly 4 ways for electron transfer and storage. If the energy h v of photons > semiconductor E _g + Δ E _c between the semiconductor and the underlying medium, the excited photoelectrons will directly enter the underlying medium 5, and then in the underlying medium 5 Migrate to the charge storage layer 4 under the action of an electric field, as shown in FIG. 3 . If the photon energy is not large enough, the photoelectrons generated by the photon excitation will accelerate towards the interface between the P-type semiconductor substrate 1 and the underlying medium 5 and the direction of source and drain under the drive of the electric field of the depletion layer. Δ E _c , the photoelectrons can cross the potential barrier and enter the bottom medium 5 , and then migrate to the charge storage layer 4 under the action of the electric field of the bottom medium 5 , as shown in FIG. 4 . When the electric field in the bottom medium is high, electrons can tunnel into the bottom medium 5 and migrate to the charge storage layer 4 under the action of the electric field in the bottom medium 5, as shown in FIG. 5 . When the electric field in the depletion region reaches the avalanche electric field, the generated photoelectrons and holes will ionize more electron-hole pairs during the movement process, and endow the electrons with enough energy to cross the bottom dielectric 5 and the P-type semiconductor The potential barrier of the substrate 1 enters the bottom medium 5, and migrates to the charge storage layer 4 under the action of the electric field in the bottom medium, resulting in a multiplication effect, as shown in Figure 6; since there will be thermal excitation even in the absence of light, etc. Dark current will be generated, so we need to program once in the absence of light and once in the presence of light, and make a difference between the threshold changes of the two programs again, and the resulting difference is called threshold change (as shown in Figure 7 ), the threshold change multiplied by the capacitance is the amount of charge that characterizes the light intensity.

Increasing the voltage of the detector P-type semiconductor substrate 1 can increase the width of the depletion region, thereby increasing the photon collection area. When the voltage of the P-type semiconductor substrate 1 is adjusted to the avalanche voltage, the threshold value changes rapidly due to the multiplication effect of photoelectrons increase, as shown in Figure 8. Figure 8 nicely demonstrates the existence of the doubling process.

When the voltage of the P-type semiconductor substrate 1 does not reach the avalanche voltage, that is, when the multiplication does not occur, the electrons collected by the charge storage layer 4 are proportional to the number of incident photons, so under the same exposure time, the threshold value change is proportional to the light intensity, as shown in the figure 9, so this programming mechanism can get the light intensity data very well, so as to get an accurate image.

Claims (1)

1., based on the multi-functional exposure image method of compound medium grid MOSFET light-sensitive detector, it is characterized in that:

The step of exposure image is: add a back bias voltage pulse Vb at P type semiconductor substrate, and one end of source electrode or drain electrode adds a forward voltage pulse Vp and other end floating; Or source drain adds a potential pulse Vp simultaneously, P type semiconductor substrate ( 1)add a back bias voltage pulse Vb, control gate adds zero-bias or adds forward bias pulse Vg simultaneously, and substrate and source-drain area can produce depletion layer; Vg number range 0 ~ 20v, Vb number range-20 ~ 0v, Vp number range 0 ~ 10V;

Photon excitation produce photoelectron cross underlying dielectric ( 5)with P type semiconductor substrate ( 1)potential barrier iunjected charge accumulation layer ( 4); Some photoelectron accelerates mobile under the driving of source-and-drain junction electric field towards source and drain PN junction high field region, regulate Vp and Vb numerical value, the voltage difference of Vp and Vb is made to be 0 ~ 11V, this PN junction can be made to arrive deeply exhaust, can more photoelectron be produced when illumination, these photoelectrons meeting iunjected charge accumulation layer ( 4), charge storage layer ( 4)the change of the quantity of electric charge causes the change of light-sensitive detector threshold voltage;

When there being light, photon provides energy to electronics, thus enable electronics cross underlying dielectric ( 5)with P type semiconductor substrate ( 1)potential barrier thus iunjected charge accumulation layer ( 4); This change obtains by reading process, so can know charge storage layer ( 4)middle photoelectron number.

2. the multi-functional exposure image method based on compound medium grid MOSFET light-sensitive detector according to claim 1, it is characterized in that exposure image procedure regulation Vp and Vb numerical value, the voltage difference of Vp and Vb is made to be greater than 11V, the PN junction of light-sensitive detector source electrode and drain electrode arrives avalanche voltage, more electron hole pair can be inspired near PN junction in Accelerating electron process, give the potential barrier that the enough energy of these electron hole pairs cross underlying dielectric (5) and P type semiconductor substrate (1) simultaneously, thus have more electric charge meeting iunjected charge accumulation layer (4), by regulation voltage thus regulate snowslide occur position, depletion layer immediately below PN junction area or underlying dielectric (5), when collecting photoelectron under avalanche condition, photosignal is exaggerated, gain is very large, even if very weak light intensity can excite enough photoelectrons to produce equally, because avalanche process is very rapid, only need very short potential pulse can realize programming.