Method of fabricating an image sensor

Details Method of fabricating an image sensor Method of fabricating an image sensor

2002-05-28

2003-09-09

Le, Dong (Department: 2818)

Semiconductor device manufacturing: process

Making device or circuit responsive to nonelectrical signal

C438S070000, C438S058000, C438S059000, C438S065000, C257S233000, C257S294000

Reexamination Certificate

active

06617189

ABSTRACT:

BACKGROUND OF INVENTION
1. Field of the Invention
The present invention relates to a method of fabricating an image sensor, and more particularly, to a method of fabricating a complementary metal-oxide semiconductor (CMOS) image sensor.
2. Description of the Prior Art
Charge-coupled devices (CCDs) have been the mainstay of conventional imaging circuits for converting light into electrical signals. The applications of CCDs include monitors, transcription machines, and cameras. Although CCDs have many advantages, CCDs also suffer from high costs and the limitations imposed by their size. To overcome the weaknesses of CCDs and to reduce costs and dimensions, CMOS photodiode devices have been developed. Since a CMOS photodiode device can be produced using conventional techniques, both the cost and the size of the sensor can be reduced.
Whether an image sensor device is composed of CCD or CMOS photodiode, incident light striking the image sensor must be separated into combinations of light of different wavelengths in order to properly sense color images. The intensities of the different wavelengths of the light is received by sensor devices and is transformed into electrical signals which are used to determine the original color of the incident light. To accomplish this feat, a color filter array (CFA) must be formed on each photosensor device.
FIG.
1
through
FIG. 3
are schematic diagrams in a conventional method of fabricating a CMOS image sensor
38
on a semiconductor substrate
10
. The CMOS image sensor
38
comprises a P-well
12
and a sensor array region I positioned on the P-well
12
. The sensor array region I comprises a plurality of photodiodes (not shown) positioned on the P-well
12
and a plurality of shallow trench isolations (STI)
14
formed in the P-well
12
surrounding the photodiode. Each photodiode comprises a CMOS transistor (not shown) electrically connected to a photosensor area
16
. The STI
14
acts as a dielectric insulating material to prevent short-circuiting due to contact between the photosensor area
16
and other components.
First, a planarizing layer
18
is coated on the semiconductor substrate
10
and covers each photosensor area
16
. Then, a plurality of metal barriers
20
is formed on the planarizing layer
18
in the sensor array region I of the semiconductor substrate
10
. The metal barriers
20
are located above each STI
14
and are used to prevent scattering of incident light
39
. A patterned metal layer is formed on the semiconductor substrate
10
outside the sensor array region I and is used as a bonding pad metal layer
22
. Afterwards, a planarizing layer
24
is coated on the semiconductor substrate
10
. Then a patterned photoresist layer (not shown) is formed on the planarizing layer
24
outside the sensor array region I to define a pattern of a bonding pad opening
26
. Thereafter, an etching process is performed to form the bonding pad opening
26
down to the bonding pad metal layer
22
.
A red color filter layer
28
is formed on the sensor array region I of the semiconductor substrate
10
. The color filter layer is composed of a positive type photoresist containing a red dye in an amount of 10 to 50 wt % (dry weight). Then, a photo-etching process (PEP) is performed to form a red color filter array (CFA)
28
in the red color filter layer corresponding to the respective photodiode. To increase the effectiveness and reliability of the red CFA
28
, an ultra-violet (UV) light-irradiation process and a heating process can be performed after the formation of the red CFA
28
. The UV light used has a wavelength of about 320 nm or less, and a quantity of about 20 J/cm
2
or less. The heating process is best performed in an inert atmosphere, for example in nitrogen (N
2
), to suppress the oxidation of the photoresist material. The starting temperature of the heating process is between 60 and 140° C. Then, an average increasing temperature rate used in the heating process is 1.5° C./sec. The end temperature of the heating process is between 160 and 220° C. A green CFA
30
and a blue CFA
32
are formed by repeating the above-mentioned processes with dyes of different colors. Thus, an R/G/B CFA comprises a red CFA
28
, a green CFA
30
and a blue CFA
32
.
A spacer layer
34
is formed on the R/G/B CFA, and a polymer layer (not shown) composed of acrylate material is formed on the spacer layer
34
. Further, an exposure, development, and annealing process is performed to form a plurality of U-lenses
36
in the polymer layer corresponding to the respective R/G/B CFA, and fabrication of the CMOS image sensor
38
is completed.
When incident light
39
entering the CMOS image sensor
38
is focused by the U-lens
36
, it passes through the R/G/B CFA, which only transmits light of a specific wavelength. The incident light
39
is transferred to the corresponding photosensor area
16
, which transforms the incident light
39
into electrical signals so as to obtain the original color of the incident light
39
.
In a conventional CMOS image sensor
38
, the U-lens
36
, the spacer layer
34
and the R/G/B CFA are all made of photoresist materials with low-temperature flash points of around 300° C. or less. Therefore, a passivation layer cannot be formed on the U-lens
36
to protect the U-lens
36
and the R/G/B CFA from loose particles or other contamination sources. Other drawbacks of the conventional CMOS image sensor
38
are listed below:(1) Because the U-lens
36
and the R/G/B CFA are made of photoresist materials, the bonding pad has to be formed prior to the formation of the R/G/B CFA and the U-lenses
36
. However, the R/G/B CFA development process will attack the bonding pad metal layer
22
and creates a risk of pitting. (2) Since the bonding pad is formed prior to the R/G/B CFA steps, a large trench on a scribe line will induce some color wave images on the CFA. (3) Dropped particles cannot be removed using a jet clean process because there is no passivation layer on the U-lens
36
. This means that contamination during the manufacturing process requires that the whole U-lens
36
and the R/G/B CFA be removed and recreated. (4) The conventional technique utilizes the high curvature U-lens
36
to adjust the focal plane of the incident light
39
passing through and focused by the U-lens
36
. As the process size in semiconductor manufacturing decreases, formation of the high curvature U-lens
36
becomes increasingly difficult. (5) A space exists between each U-lens
36
, so that scattered light easily enters the neighboring photosensor area
16
, resulting in cross talk effects. This increases the noise levels of the CMOS transistor image sensor
38
and reduces sensitivity.
SUMMARY OF INVENTION
It is therefore an objective of the claimed invention to provide a method of fabricating an image sensor having a passivation layer.
It is another objective of the claimed invention to provide a method of fabricating an image sensor in which the focal plane of incident light can be adjusted.
The claimed invention involves providing a semiconductor substrate comprising a sensor array region. First, a planarizing layer is formed on the semiconductor substrate. An R/G/B color filter array (CFA) is formed on portions of the planarizing layer corresponding to the sensor array region, and a spacer layer is formed on the R/G/B CFA. A plurality of U-lenses is formed on the spacer layer corresponding to the R/G/B CFA, with a space between each U-lens. Finally, a buffer layer is applied to fill the space between the U-lens, and a low-temperature passivation layer is deposited on the buffer layer and the U-lens.
The present invention utilizes the buffer layer and the low-temperature passivation layer sequentially formed on the U-lens to prevent damage to the U-lens. Because the buffer layer has a predetermined index of refraction (IR), an optical path of incident light can be changed by adjusting the IR of the buffer layer. Simultaneously, cross talk effects, caused when incident light is refracted in a way that causes it to stri

Affiliated with

Also associated with

Rating

Method of fabricating an image sensor does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method of fabricating an image sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of fabricating an image sensor will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3040682