EP1454357A2

EP1454357A2 - Method for controlling blooming of a photodiode and corresponding integrated circuit

Info

Publication number: EP1454357A2
Application number: EP02799810A
Authority: EP
Inventors: François Roy
Original assignee: STMicroelectronics SA
Current assignee: STMicroelectronics SA
Priority date: 2001-12-12
Filing date: 2002-12-12
Publication date: 2004-09-08
Also published as: JP2005512341A; WO2003050874A3; WO2003050874A2; FR2833408A1; FR2833408B1; US7586172B2; US20050116270A1

Abstract

The invention concerns a photodiode comprising an upper junction PN (D1) formed on a top layer and an intermediate layer supported by a portion of a semiconductor substrate. A lower junction is formed between the intermediate layer and the substrate portion. The voltage activating direct conduction of the upper junction (D1) is lower than the voltage activating direct conduction of the lower junction (D2). The method consists in allowing storage of loads in the photodiode until activation of direct conduction of the upper junction so as to promote (F1) the recombination of carriers derived from the intermediate layer with carriers of the top layer.

Description

Method for controlling the over-illumination of a photodiode and corresponding integrated circuit.

The invention relates to microelectronics, in particular integrated circuits comprising photodiodes, and more particularly the control of over-illumination (“blooming” in English) of such a photodiode.

Image sensors based on semiconductor components take advantage of the principle of converting photons into electron / hole pairs in silicon. More precisely, the charges created in the photosensitive zones are stored in the photodiode and are then read using an electronic system. This electronic system, which controls the photodiode, comprises, in particular when the photodiode is a fully depleted photodiode (“fully depleted” in English), a transfer transistor authorizing the transfer of the charges stored in the photodiode.

If we do not take precautions, under conditions of over-illumination, the electrons which can no longer be stored in the photodiode can then diffuse in the semiconductor substrate to disturb all the photodiodes of the image sensor, this which visually leads to the generation of an increasingly important white halo on the image.

To remedy this problem, one solution consists in defining the saturation level of the photodiode by the voltage applied to the gate of the transfer transistor. More precisely, this voltage is chosen so that the charges will be evacuated towards the supply (via a conduction controlled by a reset transistor) before they diffuse in the substrate and come to degrade the signal received by neighboring photodiodes. However, this voltage must be adjusted by a specific power supply and must take into account the dispersions linked to the technology. In addition, these dispersions lead to a reduction in the storage dynamics on the photodiode. The invention aims to remedy this drawback and proposes a radically different solution for controlling the over-illumination of a photodiode.

The invention therefore provides an integrated circuit comprising at least one cell comprising a photodiode, this photodiode comprising a lower PN junction formed between a part of a semiconductor substrate and an intermediate layer situated on said substrate part, and an upper PN junction formed between the intermediate layer and an upper layer more heavily doped than the intermediate layer.

Furthermore, the direct conduction voltage of the upper junction is lower than the direct conduction voltage of the lower junction.

Thus, the photodiode will store charges until the surface diode (upper junction) goes into conduction and the carriers (electrons for example) “not storable” in the photodiode will then be recombined with the carriers (holes for example) of the surface layer (upper layer) instead of diffusing into the substrate. Thus, according to the invention, the maximum level of storage is no longer defined by the voltage of the channel of an MOS transistor (the transfer transistor), but by the voltage of direct conduction of a junction.

The storage dynamic on the photodiode is therefore increased. And we have solved the problems related to the dispersions of technology.

Furthermore, the over-lighting control means here comprise the upper junction and the upper layer. They therefore form an integral part of the photodiode, which is also a general characteristic of the integrated circuit according to the invention.

The doping of the upper layer must be greater than the doping of the intermediate layer, so as to contain more carriers (holes, for example) than the carriers (electrons) of the layer intermediate, and this in order to allow recombination during over-illumination.

Furthermore, even if the direct conduction voltage of the upper junction is lower than the direct conduction voltage of the lower junction of the photodiode, there is generally still a current in the substrate also resulting from the conduction direct from the lower junction.

Those skilled in the art will be able to adjust, depending on the applications and the maximum level of storage desired for the photodiode, on the one hand the doping of the upper layer compared to the doping of the intermediate layer and, on the other hand, the ratio between the direct conduction current for the upper junction compared to the direct conduction current for the lower junction.

This being the case, a doping of the upper layer is preferably chosen at least five times greater than the doping of the intermediate layer, and for example at least ten times greater. Furthermore, the respective voltages of the direct conductions of the two junctions will preferably be chosen so as to obtain a direct conduction current for the upper junction at least ten times greater than the direct conduction current for the lower junction.

According to one embodiment of the invention, the upper layer is formed of silicon doped with indium, while the intermediate layer is formed of silicon doped with phosphorus and that said portion of substrate is formed of silicon doped with boron. However, other materials are possible. One can thus replace phosphorus by arsenic or antimony. And, in general, we can choose any material which reduces the difference between the valence band and the silicon conduction band.

The invention also relates to an image sensor comprising at least one integrated circuit as defined above.

The invention also proposes a method for controlling the over-lighting of a photodiode produced in a semiconductor substrate. According to a general characteristic of this method according to the invention, the photodiode comprises an upper junction PN formed between an upper layer and an intermediate layer supported by a portion of the substrate and, storage of the charges in the photodiode is authorized until direct conduction of said upper junction so as to promote the recombination of the carriers coming from the intermediate layer with the carriers of the upper layer.

Other advantages and characteristics of the invention will appear on examination of the detailed description of embodiments and implementations, in no way limiting, and of the appended drawings, in which:

FIG. 1 schematically illustrates an image sensor according to the invention, formed of several cells equipped with photodiodes, according to the invention,

FIG. 2 schematically illustrates a semiconductor structure of a photodiode, according to the invention;

FIG. 3a schematically illustrates an example of doping of the different layers forming a photodiode, according to the invention;

FIG. 3b schematically illustrates an example of self-polarization of a photodiode, according to the invention, completely depleted;

FIG. 4 is another schematic representation of a photodiode, according to the invention, illustrating an embodiment of the method according to the invention; and FIG. 5 represents two voltage-current curves concerning the two junctions of a photodiode according to the invention.

In FIG. 1, the reference CIM generally designates an image sensor formed from a matrix of PX _i5 cells (or pixels) each comprising a photodiode PD as well as means for controlling this photodiode.

In the example which is described here, the photodiode is a totally depleted photodiode and the control means are means with four transistors TG, RST, SF and RS. That said, the invention is not limited to a completely depleted diode and applies to all photodiodes, especially those whose associated control means only have three transistors.

The structure and operation of the control means with four transistors of a totally depleted photodiode are conventional and well known to those skilled in the art. We recall here, with reference to Figure 1, the main features.

These control means include a transfer transistor

TG, connected on the one hand to the photodiode PD and, on the other hand, to the gate of a follower transistor SF. The gate of the follower transistor SF is also connected to the supply voltage Vdd by means of a reset transistor, referenced RST.

Finally, an RS transistor for selecting the cell

PX _; connects the follower transistor SF to a capacitive output register.

As regards the operation of the cell, the first phase is a phase of integration or storage of the charges in the photodiode during which the transfer transistor TG is blocked. During most of the integration phase, the reset transistor is on. Then, just before the transfer of the charges from the photodiode to the output, the transistor RST is blocked, then the selection transistor RS is made conductive, which makes it possible to determine the level of charge in the photodiode before the transfer.

During the transfer, the transfer transistor TG is made conductive, then, while the reset transistor is still blocked, the transistor TG is again blocked, marking the end of the transfer. The selection transistor RS is then again made conductive so as to be able to determine the charge level after transfer.

After this second measurement, the reset transistor RST is again made conductive, while a new integration phase has already started.

The amount of light received by the photodiode is then determined by the difference between the two charge levels, measured before and after transfer respectively. Whereas in the prior art, the means for controlling the over-illumination comprise a specific supply for adjusting the low level of the voltage applied to the gate of the transfer transistor TG, the control means are, according to the invention, part integral of the PD photodiode, as detailed below.

The CIM image sensor, and more particularly each cell or pixel PX} is produced in CMOS technology.

FIG. 2 illustrates in more detail the semiconductor structure of the photodiode PD of a cell PXj. In this FIG. 2, the reference SB denotes a silicon semiconductor substrate, doped here P. This substrate SB can for example be a P doped well and located within a P ⁺ doped semiconductor plate. The substrate SB is connected to ground.

The transfer transistor TG is a MOS transistor whose source 2, N ^" doped, forms for the photodiode PD an intermediate layer which extends above part 1 of the substrate SB.

Above this intermediate layer 2 is produced, for example by implantation, an upper layer 3, P ⁺ doped, which extends to an isolation region IS, for example a region of the LOCOS type, according to a name well known to those skilled in the art. Furthermore, part 1 of the substrate SB comes into contact with the upper layer 3, P ⁺ doped.

The photodiode PD is therefore here formed of these three layers, which define two PN junctions (diodes), namely an upper junction formed by layers 2 and 3, and a lower junction formed by layer 2 and the underlying part of substrate 1.

An example of the doping profile of this photodiode PD is illustrated in FIG. 3a. More precisely, the upper layer 3, P ⁺ doped, is more strongly doped than the intermediate layer 2, itself more strongly doped than the underlying part of the substrate 1.

As an indication, a doping of the upper layer 3 will be chosen between 10 ¹⁸ and 10 ¹⁹ at./cm ³ . The doping of the intermediate layer 2 could be chosen between 10 ¹⁷ and 10 ¹⁸ at./cm ³ , while the doping of the underlying part 1 of the substrate could be chosen less than 10 ¹⁶ at./ cm ³ .

With such doping of the intermediate layer 2, the photodiode is then of the totally depleted type, that is to say that it self-polarizes at a determined voltage when the concentration of electrons in layer 2 is zero at the end of transfer. In the example described, the photodiode self-polarizes at 1.5 volts (Figure 3b).

As illustrated in FIG. 4, and as indicated above, the photodiode PD consists of two junctions (diodes) D1 and D2. These two diodes Dl and D2 are chosen so that the voltage VI for direct conduction of the diode Dl (FIG. 5) is less than the voltage for direct conduction V2 of the diode D2. As a result, the photodiode PD will store charges during its illumination until the upper junction D1 goes into direct conduction.

The “non-storable” electrons in the N-doped intermediate layer 2 will then be preferentially transferred (arrow FI, figure

4) in the upper layer 3 to recombine with the holes in this upper layer 3, and will not diffuse into the substrate to disturb the neighboring photodiodes.

However, depending on the characteristics CD1, CD2 (figure

5) of the junctions, there may also exist, simultaneously with the current Il of direct conduction of the junction D1, a weak current 12 of direct conduction of the junction D2. However, in the example described here, a ratio 10 between the current II and 12 generally allows a good recombination of the electrons with the holes of the upper layer without the electrons which diffuse in the substrate (arrow F2, figure 4) being troublesome. for neighboring photodiodes. As an indication, if we take a direct conduction voltage V2 for the diode D2 equal to 0.6 volts, we could for example choose a direct conduction voltage VI equal to 0.4 volt for the junction Dl. Materially, to obtain this difference in direct conduction voltage, it will be possible for example to choose a P ⁺ implantation obtained with indium, while the P doping of the substrate part will be obtained with boron. Region 2 can be doped N by phosphorus.

Claims

1-Integrated circuit, comprising at least one cell comprising a photodiode and means for controlling over-illumination of the photodiode, characterized in that the means for controlling over-illumination (2,3) are an integral part of the photodiode (PD).

2-Device according to claim 1, characterized in that the photodiode comprises a lower PN junction (D2) formed between a part of a semiconductor substrate (1) and an intermediate layer (2) located on said substrate part, and an upper PN junction (Dl) formed between the intermediate layer (2) and an upper layer (3) more heavily doped than the intermediate layer, and by the fact that the direct conduction voltage of the upper junction is lower than the direct conduction voltage of the lower junction, the over-illumination control means comprising the upper junction (Dl) and the upper layer (3).

3-Integrated circuit comprising at least one cell comprising a photodiode, characterized in that the photodiode comprises a lower PN junction (D2) formed between a part of a semiconductor substrate (1) and an intermediate layer (2) located on said substrate part, and an upper PN junction (Dl) formed between the intermediate layer (2) and an upper layer (3) more heavily doped than the intermediate layer, and by the fact that the direct conduction voltage of the junction upper is lower than the direct conduction voltage of the lower junction.

4-Integrated circuit according to claim 2 or 3, characterized in that the doping of the upper layer (3) is at least 5 times greater than the doping of the intermediate layer (2).

5-Integrated circuit according to one of claims 2 to 4, characterized in that the respective direct conduction voltages of the two junctions are chosen so as to obtain a direct conduction current for the upper junction (D l) at least ten times greater than the direct conduction current of the lower junction (D2).

6-Integrated circuit according to one of claims 2 to 5, characterized in that the upper layer (3) is formed of silicon doped with indium, in that the intermediate layer

(2) is formed of silicon doped with phosphorus, and by the fact that said substrate portion (1) is formed of silicon doped with boron.

7-Integrated circuit according to one of the preceding claims, characterized by the fact that the photodiode is a completely depleted photodiode, and by the fact that the cell further comprises means for controlling the photodiode with four transistors (TG, RST, RS, SF).

8-Integrated circuit according to one of the preceding claims, characterized in that it comprises a matrix of cells (PXj).

9-Image sensor, characterized in that it comprises at least one integrated circuit according to claim 8.

10-Method for controlling the over-illumination of a photodiode produced in a semiconductor substrate, characterized in that the photodiode comprises an upper PN junction (D l) formed between an upper layer and an intermediate layer supported by a part of the substrate , and by the fact that the storage of charges is authorized in the photodiode until said upper junction is put into direct conduction so as to promote (FI) the recombination of the carriers from the intermediate layer with the carriers of the upper layer.