CN106229324A - Imageing sensor and preparation method thereof - Google Patents

Abstract

The present invention discloses an image sensor and a manufacturing method thereof, comprising: providing a silicon substrate, the silicon substrate including a photodiode region and an isolation region; selectively part of the silicon substrate, in the photodiode region A trench is formed; a semiconductor layer is filled in the trench, the forbidden band width of the semiconductor layer is smaller than that of the silicon substrate, and the semiconductor layer is used to form a photodiode. Wherein, a semiconductor layer is formed in the photodiode region, the semiconductor layer has a band gap smaller than that of silicon, can better absorb near-infrared light, and can effectively improve the quantum conversion efficiency of the CMOS image sensor. Further, the silicon germanium epitaxial process is well compatible with the current silicon single crystal manufacturing process, so it is relatively easy to apply to the current CMOS image sensor process.

Description

Image sensor and manufacturing method thereof

technical field

The invention relates to the technical field of image sensors, in particular to an image sensor and a preparation method thereof.

Background technique

With the rapid development of the mobile Internet, people's demand for smart terminals is increasing, and image sensors, known as the "eyes" of smart terminals, have also ushered in unprecedented development space. The traditional CCD (Charge-coupled Device, charge-coupled device) image sensor is limited in high-performance digital cameras due to its high power consumption; Moreover, it is easy to be compatible with existing semiconductor processes, and the production cost is low, which makes CMOS image sensors occupy half of the image sensor market.

The main problem encountered by CMOS image sensors is the low quantum efficiency (QE, quantum efficiency) of near-infrared light. Quantum efficiency refers to the probability of a photon being transformed into a photogenerated electron in a PD. In order to improve the quantum efficiency of infrared light, the overall thickness of the silicon substrate is often increased in the prior art, however, the CMOS image sensor obtained by this method has poor performance.

Contents of the invention

The object of the present invention is to provide an image sensor and a preparation method thereof, which can improve the quantum efficiency of near-infrared light and improve the performance of the CMOS image sensor at the same time.

In order to solve the above technical problems, the present invention provides a method for preparing an image sensor, comprising:

providing a silicon substrate comprising a photodiode region and an isolation region;

selectively portioning the silicon substrate to form a trench in the photodiode region; and

A semiconductor layer is filled in the groove, the forbidden band width of the semiconductor layer is smaller than the forbidden band width of the silicon substrate, and the semiconductor layer is used to form a photodiode.

Further, in the manufacturing method of the image sensor, the semiconductor layer is filled in the trench by using an epitaxy process.

Further, in the manufacturing method of the image sensor, germanium-doped single crystal silicon is grown in the trench by epitaxial vapor deposition.

Further, in the manufacturing method of the image sensor, the semiconductor layer is a silicon germanium layer, and the mass percentage of germanium in the silicon germanium layer is 40%-60%.

Further, in the preparation method of the image sensor, before filling the semiconductor layer in the trench, the preparation method of the image sensor further includes:

forming a sacrificial layer on the surface of the trench;

The sacrificial layer is removed.

Further, in the preparation method of the image sensor, the material of the sacrificial layer is oxide, and the thickness of the sacrificial layer is

Further, in the manufacturing method of the image sensor, the manufacturing method of the image sensor further includes: performing ion doping on the semiconductor layer to form a photodiode.

Further, in the manufacturing method of the image sensor, the depth of the groove is 1 μm˜5 μm.

According to another aspect of the present invention, there is also provided an image sensor prepared by any one of the image sensor preparation methods above, including a silicon substrate, the silicon substrate includes a photodiode region and an isolation barrier for isolating the photodiode region region, there is a groove in the region of the electric diode, the groove is filled with a semiconductor layer, the forbidden band width of the semiconductor layer is smaller than the forbidden band width of the silicon substrate, and the semiconductor layer is used to form a photodiode , the material of the isolation region is silicon.

Further, in the image sensor, the semiconductor layer includes a first type doped layer and a second type doped layer on the first type doped layer, the first type doped layer and the second type doped layer The second type doped layer forms a photodiode.

Compared with the prior art, the image sensor provided by the invention and its preparation method have the following advantages:

In the image sensor and its preparation method, a semiconductor layer is formed in the photodiode region, the semiconductor layer has a band gap smaller than that of silicon, can better absorb near-infrared light, and can effectively improve the CMOS Image sensor quantum conversion efficiency. Further, the silicon germanium epitaxial process is well compatible with the current silicon single crystal manufacturing process, so it is relatively easy to apply to the current CMOS image sensor process.

Description of drawings

FIG. 1 is a flowchart of a method for preparing an image sensor in an embodiment of the present invention;

2 to 9 are schematic diagrams of device structures in a method for manufacturing an image sensor according to an embodiment of the present invention.

detailed description

In existing image sensors, the quantum efficiency of near-infrared light is improved by increasing the overall thickness of the silicon substrate. However, the performance of the CMOS image sensor obtained by this method is not good. The inventors have studied the existing image sensors and found that the quantum efficiency of near-infrared light is increased from the current 2um-3um to 5um-10um by increasing the overall thickness of the silicon substrate in the existing image sensors. Simply increasing the thickness of single-crystal silicon can improve the quantum conversion efficiency, but the accompanying process challenges are also intensified, such as deeper ion implantation, which requires thicker photoresist, and thicker photoresist. The resist will reduce the resolution of the minimum size, and finally affect the performance of the CMOS image sensor. In addition, thick single crystal silicon will bring about process problems in photolithography alignment, and additional processes need to be added to achieve the alignment process.

The inventors have further studied and found that due to the inherent characteristics of the energy band structure, silicon single crystal materials have problems such as low absorption coefficient and long absorption length for near-infrared light. Especially as the feature size of semiconductor devices enters the sub-micron and deep sub-micron range, the working voltage becomes smaller and smaller, the P-N junction of the transistor becomes shallower, the depletion region is closer to the surface, and the thickness is getting thinner and thinner. The incident light signal is effectively absorbed, and the photogenerated carriers generated deep in the substrate will quickly recombine without being pulled by the electric field, and will not contribute to the photocurrent, resulting in a very low quantum conversion efficiency of the fabricated CMOS image sensor.

The inventors conducted in-depth research and found that the bandgap width of the silicon germanium material is smaller than that of silicon. If the silicon germanium material is used in CMOS image sensor products, it can better absorb near-infrared light and effectively improve the quantum conversion of CMOS image sensors. efficiency.

According to the above research, the present invention provides a method for preparing an image sensor, providing a method for preparing an image sensor, as shown in Figure 1, comprising the following steps:

Step S11, providing a silicon substrate, the silicon substrate including a photodiode region and an isolation region;

Step S12, selectively parting the silicon substrate to form a trench in the photodiode region; and

Step S13 , filling the trench with a semiconductor layer, the semiconductor layer having a band gap smaller than that of the silicon substrate, and the semiconductor layer being used to form a photodiode.

A semiconductor layer is formed in the photodiode region, the semiconductor layer has a band gap smaller than that of silicon, can better absorb near-infrared light, and can effectively improve the quantum conversion efficiency of the CMOS image sensor. At the same time, since the silicon germanium epitaxial process is well compatible with the current silicon single crystal manufacturing process, it is relatively easy to apply to the current CMOS image sensor process.

The image sensor of the present invention and its preparation method will be described in more detail below in conjunction with schematic diagrams, wherein preferred embodiments of the present invention are represented, and it should be understood that those skilled in the art can modify the present invention described herein while still realizing the present invention beneficial effect. Therefore, the following description should be understood as the broad knowledge of those skilled in the art, but not as a limitation of the present invention.

In the interest of clarity, not all features of an actual implementation are described. In the following description, well-known functions and constructions are not described in detail since they would obscure the invention with unnecessary detail. It should be appreciated that in the development of any actual embodiment, numerous implementation details must be worked out to achieve the developer's specific goals, such as changing from one embodiment to another in accordance with system-related or business-related constraints. Additionally, it should be recognized that such a development effort might be complex and time consuming, but would nevertheless be merely a routine undertaking for those skilled in the art.

In the following paragraphs the invention is described more specifically by way of example with reference to the accompanying drawings. Advantages and features of the present invention will be apparent from the following description and claims. It should be noted that all the drawings are in a very simplified form and use imprecise scales, and are only used to facilitate and clearly assist the purpose of illustrating the embodiments of the present invention.

The manufacturing method of the image sensor of the present invention will be described in detail below with reference to FIG. 2 to FIG. 9 . FIG. 2 to FIG. 9 are schematic diagrams of device structures in the manufacturing method of the image sensor according to an embodiment of the present invention.

First, step S11 is performed, as shown in Figure 2 and Figure 3, wherein Figure 2 is a top view, and Figure 3 is a cross-sectional view of Figure 2 along the line AA'. A silicon substrate 100 is provided, and the silicon substrate 100 may be a doped silicon substrate 100, for example, the silicon substrate 100 has P-type dopant ions.

The silicon substrate 100 includes a photodiode region 101 and an isolation region 102 , wherein the isolation region 102 is used to isolate adjacent photodiode regions 101 , and the photodiode region 101 is used to form a photodiode. Other regions may also be formed in the silicon substrate 100. For example, the isolation region 102 may further include a gate region 103, and the gate region 103 is used to form a gate on the silicon substrate 100. It can be understood by those skilled in the art and will not be repeated here.

In FIG. 2 , a group of four photodiode regions 101 are schematically shown, which are respectively used to form R sub-pixels, G sub-pixels, B sub-pixels and NIR sub-pixels, which can be understood by those skilled in the art. I won't go into details here.

Generally, in the stacked image sensor manufacturing process, two wafers are required, one is a logic operation circuit wafer, and the other is a pixel circuit wafer, and then the two wafers are bonded together. The silicon substrate 100 in this embodiment is used to prepare a pixel circuit wafer.

Then, step S12 is performed to selectively partially form the silicon substrate 100 to form trenches in the photodiode region 101 . Specifically, as shown in FIG. 4, the photodiode region 101 is first defined using a patterned photoresist 110, that is, the patterned photoresist 110 exposes the photodiode region 101 and covers the isolation region 102 ; then, as shown in FIG. 5 , etching the silicon substrate 100 to form a trench 110 . Preferably, the trench 110 is formed by a dry etching process, so that the trench 110 with a better shape can be formed. Preferably, the depth H1 of the groove 110 is 1 μm˜5 μm, such as 2 μm, 3 μm, 4 μm, which is beneficial to improve the quantum efficiency of the image sensor.

In order to improve the morphology of the trench 110 so that the subsequently formed semiconductor layer has a better crystal form, preferably, before filling the trench 110 with a semiconductor layer, the method for preparing the image sensor further includes:

As shown in FIG. 6, a sacrificial layer 200 is formed on the surface of the trench 110. Preferably, the material of the sacrificial layer 200 is oxide, which has a good surface shaping effect. The sacrificial layer 200 The thickness is For example In order to achieve a better shaping effect. Preferably, the sacrificial layer 200 is grown by a furnace tube process, and the sacrificial layer 200 is also formed on the upper surface of the silicon substrate 100;

As shown in FIG. 7 , the sacrificial layer 200 is removed. Generally, a wet etching process is used to completely remove the sacrificial layer 200 .

Next, step S13 is performed. As shown in FIG. 8 , the trench 110 is filled with a semiconductor layer 300 whose forbidden band width is smaller than that of the silicon substrate 100 , that is, the semiconductor layer The band gap of the material 300 is smaller than that of single crystal silicon, and the semiconductor layer 300 can better absorb near-infrared light, which can effectively improve the quantum conversion efficiency of the CMOS image sensor.

Preferably, the semiconductor layer 300 is a silicon germanium layer, that is, the material of the semiconductor layer 300 is silicon germanium, and the forbidden band width of silicon germanium is smaller than that of single crystal silicon. In the silicon germanium layer, the mass percentage of germanium is 40%-60%, for example 50%. Preferably, the semiconductor layer 300 is filled in the trench 110 by an epitaxial process. In this embodiment, germanium-doped single crystal silicon is grown in the trench 110 by an epitaxial vapor deposition method to form the The semiconductor layer 300 described above. Since the silicon germanium epitaxial process is well compatible with the current silicon single crystal manufacturing process, it is relatively easy to apply to the current CMOS image sensor process.

After the semiconductor layer 300 is formed, subsequent processes can be performed, for example, ion doping is performed on the semiconductor layer 300 to form a photodiode. Generally, two ion implantations are performed in the semiconductor layer 300 respectively, and N Type ions and P-type ions to form a photodiode. For example, as shown in Figure 9, the first type of ions (for example, N-type ions) are first carried out to form the first type doped layer 301, and then the second type of ions (for example, P-type ions) are carried out to form the first type doped layer 301. The second-type doped layer 302, the first-type doped layer 301 and the second-type doped layer 302 form a photodiode.

After the above steps, an image sensor 1 as shown in FIG. 9 is formed. The image sensor 1 includes a silicon substrate 100, and the silicon substrate 100 includes a photodiode region 101 and an isolation for isolating the electric diode region 101. A region 102, the electric diode region 101 has a trench 110, the trench 110 is filled with a semiconductor layer 300, the forbidden band width of the semiconductor layer 300 is smaller than the forbidden band width of the silicon substrate 100, the The material of the isolation region 102 is silicon.

Preferably, the trench 110 has a depth of 1 μm˜5 μm, and the semiconductor layer 300 is an epitaxial silicon germanium layer. In this embodiment, the semiconductor layer 300 includes a first type doped layer 301 and a second type doped layer 302 located on the first type doped layer 301, the first type doped layer 301 and The second type doped layer 302 forms a photodiode 300 .

In the image sensor 1, a semiconductor layer 300 is formed in the photodiode region 101. The semiconductor layer 300 has a forbidden band width smaller than that of silicon, can better absorb near-infrared light, and can effectively improve the CMOS image. Sensor 1 quantum conversion efficiency. Further, since the silicon germanium epitaxial process is well compatible with the current silicon single crystal manufacturing process, it is relatively easy to apply to the current CMOS image sensor process.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims (10)

1. the preparation method of an imageing sensor, it is characterised in that including:

A silicon substrate, described silicon substrate is provided to include photodiode area and area of isolation；

Remove the described silicon substrate of part, in described photodiode area, form groove；And

Filling semiconductor layer in described groove, the energy gap of described semiconductor layer is less than the energy gap of described silicon substrate, Described semiconductor layer is used for forming photodiode.

2. the preparation method of imageing sensor as claimed in claim 1, it is characterised in that use epitaxy technique at described groove The described semiconductor layer of interior filling.

3. the preparation method of imageing sensor as claimed in claim 2, it is characterised in that use extension vapour deposition process in institute The monocrystal silicon of doped germanium is grown in stating groove.

4. the preparation method of imageing sensor as claimed in claim 1, it is characterised in that described semiconductor layer is germanium-silicon layer, And the mass percent of germanium is 40%～60% in described germanium-silicon layer.

5. the preparation method of the imageing sensor as described in any one of claim 1-4, it is characterised in that fill out in described groove Before filling semiconductor layer, the preparation method of described imageing sensor also includes:

A sacrifice layer is formed on the surface of described groove；

Remove described sacrifice layer.

6. the preparation method of imageing sensor as claimed in claim 5, it is characterised in that the material of described sacrifice layer is oxidation Thing, the thickness of described sacrifice layer is

7. the preparation method of imageing sensor as claimed in claim 1, it is characterised in that the preparation side of described imageing sensor Method also includes: described semiconductor layer carries out ion doping and forms photodiode.

8. the preparation method of imageing sensor as claimed in claim 1, it is characterised in that the degree of depth of described groove is 1 μm～5 μm。

9. one kind utilizes imageing sensor prepared by claim 1-8 any one imageing sensor preparation method, it is characterised in that Including silicon substrate, described silicon substrate includes photodiode area and for isolating the isolation area of described photodiode area Territory, has groove in described photodiode area, be filled with semiconductor layer, the energy gap of described semiconductor layer in described groove Less than the energy gap of described silicon substrate, described semiconductor layer is used for forming photodiode.

10. imageing sensor as claimed in claim 9, it is characterised in that described semiconductor layer includes first kind doped layer And it is positioned at the Second Type doped layer on described first kind doped layer, described first kind doped layer and Second Type doping Layer forms photodiode.