Variable reticulation time delay and integrate sensor

Details Variable reticulation time delay and integrate sensor Variable reticulation time delay and integrate sensor

1999-07-26

2003-10-14

Wilson, Allan R. (Department: 2815)

Active solid-state devices (e.g., transistors, solid-state diode

Field effect device

Charge transfer device

C257S225000, C257S231000

Reexamination Certificate

active

06633058

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to charge coupled device image sensors and specifically to architectural implementations that enhance the short wavelength radiation response in TDI (time delay and integration) CCD sensors.
2. Description of Related Art
Charge-coupled devices (CCD) are metal insulator semiconductor (MIS) devices which belong to a general class of semiconductor devices that store and transfer information in the form of electrical charge. Charges in the CCD can either be electrically introduced, thermally generated, or photo-generated. The photosensitive property of the CCD is used in imaging applications.
There are three general classes of CCD imaging arrays: area arrays, linear arrays, and TDI (time delay and integration) arrays. An area CCD is comprised of a two-dimensional array of photoelements. Like a conventional analog camera, an area CCD captures snapshots of two-dimensional images. A linear CCD, on the other hand, is comprised of a one-dimensional array of photoelements. Linear arrays are usually used in applications where there is relative motion between the sensor and the source of the image. The linear array is oriented in the direction perpendicular to the direction of motion, and successively scans lines out of an area image. The lines can later be reconstructed to constitute the two-dimensional image. Like a linear array, a TDI array also scans successive lines of a two-dimensional image. Unlike a linear array and like an area array, however, a TDI array is comprised of a two-dimensional array of photoelements. A TDI array does not take snapshots of two-dimensional images. Instead, the vertical dimension of the TDI array is operated to transfer photogenerated charges within the array so as to follow the motion of the image source. Line outputs are then read out after the photocharges have tracked the image for some time.
The TDI concept is illustrated in
FIGS. 9A
,
9
B and
9
C. In
FIG. 9A
, a moving image source is projected onto a fixed TDI array. At time t
1
, section s
1
of the two-dimensional image is imaged onto line l
1
of the TDI array, while the next section, section s
2
, of the same two-dimensional image is imaged onto the next line, line l
2
, of the TDI array (see FIG.
9
B). The pixel pitch of the TDI array (the center-to-center distance between l
1
and l
2
) is denoted by p. If the image is moving relative to the TDI array at a velocity v, then at time t
2
(t
2
=t
1
+p/v) section s
1
will be imaged onto line l
2
of the TDI array, while section s
2
will be imaged onto line l
3
of the TDI array (see FIG.
9
C). The TDI array is clocked in a fashion such that the photogenerated charges that correspond to each section of the image follow the image along the array. Line data are read out after charges have integrated for a specified number of stages.
There are three main types of photoelements out of which such arrays are made: p-n junction photodiodes, photogates, and pinned photodiodes. A p-n junction photodiode is merely a reversed-biased p-n junction diode. In CCDs, the potential of the p side of the diode (e.g., a lightly doped p type substrate) is fixed, usually by the bias applied to the silicon substrate. The bias on the n side of the diode (e.g., an n type implant in the substrate) is either floating or is set prior to photocharge integration by an adjacent gate through which a predetermined voltage has been applied. The reversed biased diode forms a depletion layer between the p side and the n side. Photogenerated electrons are swept by an electric field in the depletion layer of the diode to the n region (due to its positive bias), where they are either temporarily stored or transferred to an adjacent gate. Photogenerated holes are usually swept to the substrate.
A photogate is essentially a sandwich comprised of an electrode layer, an insulator (e.g., silicon oxide and/or silicon nitride), and a semiconductor (e.g., lightly doped silicon). When an appropriate bias is applied to a poly-crystalline silicon gate electrode (called a polysilicon gate electrode) with respect to the semiconductor, a potential well forms in the semiconductor. For example, a positive potential with respect to the semiconductor may be applied to a polysilicon gate electrode insulatively spaced over a lightly doped n type semiconductor. The positive potential repels negative charges in the conduction band of the semiconductor and induces a positive potential well beneath the gate electrode. Photogenerated charges are collected in this potential well.
In many CCDs, photogates include a doped semiconductor layer between the insulator and the semiconductor substrate. For example, an n type implant layer may be formed on a surface of a lightly doped p type semiconductor substrate. A positive potential with respect to the p type semiconductor substrate may be applied to a polysilicon gate electrode insulatively spaced over the n type implant. The positive potential induces a positive potential well beneath the gate electrode that is buried under the gate electrode so that the potential well is separated from the insulator interface by a potential barrier. Because the potential well is buried in the semiconductor, such photogates are commonly referred to as buried-channel photogates.
The maximum potential in the collecting well is a function of the gate bias. As the gate bias increases, the maximum potential in the collecting well normally also increases by approximately the same amount. In buried-channel photogates however, when the gate bias is sufficiently negative, electrons in the n implant are repelled and a layer of holes will form at the insulator-semiconductor interface. When this occurs, biasing the gate even more negatively will have virtually no effect on the collecting well potential since the layer of holes shields the electric field from the substrate. At this point, the photogate is said to be “pinned”. Pinned photodiodes are similar to pinned photogates, with the exception that the layer of holes is implanted rather than induced by a gate potential.
In area and TDI CCD sensors formed with photogates, a pixel will ordinarily have two, three or four photogates architected so that the gate electrodes are coupled to respective two, three or four phase clocking signals.
The spectral response of photoelements depends on whether the photoelement is a photodiode or a photogate. In silicon based image sensors, spectral response in the silicon typically peaks at the red, yellow or green spectrum, and levels off both at the longer infrared wavelengths and at the shorter blue and ultra-violet wavelengths.
FIG. 10
is a typical plot of the quantum efficiency (the proportion of photogenerated charge collected by the CCD) as a function of wavelength. Photons are absorbed in the semiconductor at a distance determined by the absorption depth in the material for given wavelength. At shorter wavelengths, most absorption occurs either close to the semiconductor surface or in overlying layers such as the polysilicon electrodes. Because most of these electrons do not reach the CCD photoelement collecting well, quantum efficiency is poorer at short wavelengths than at long wavelengths. At 400 nm, for example, the average absorption depth of photons in silicon is only 200 nm. This is thinner than the typical polysilicon layers, and as a result, the short wavelength response of photogates will suffer relative to the response achieved with photodiodes. The quantum efficiency is lower in the infrared because the energy (E=h&ngr;) of the photons tapers off at longer wavelengths. In the near infrared spectrum, some loss in quantum efficiency also occurs due to loss of electrons that are generated beyond the depletion region of the collection well. These electrons either diffuse to the substrate or recombine with holes. Silicon is mostly transparent to incoming photons with wavelengths greater than 1,100 nm.
In photogate CCDs, a large fraction of the surface of the photoelement is cover

Affiliated with

Also associated with

Rating

Variable reticulation time delay and integrate 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 Variable reticulation time delay and integrate sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Variable reticulation time delay and integrate sensor will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3115513