The present invention relates to a semiconductor imager and more particularly, to an interline transfer Schottky-barrier charge-coupled device (CCD) imager having improved fill-factor.
Background of the Invention
One type of solid state imager includes a substrate of single crystalline semiconductor material having along one surface thereof a plurality of detector areas for detecting radiation and converting the radiation to charge carriers, and transfer means along the detector areas for receiving the charge carriers and transferring them to an output terminal. One such type of detector commonly used is a Schottky-barrier detector formed of a layer of a conductive material forming a Schottky-barrier junction with the substrate. One type of transfer means used is a CCD. Such an imager is shown and described in the articles of W. F. Kosonocky et al., entitled "Design and Performance of 64.times.128 Element PtSi Schottky-barrier Infrared Charge-Coupled Device (IRCCD) Focal Plane Array", published in SPIE, Volume 344, Infrared Sensor Technology (1982), pages 66-77 and W. F. Kosonocky et al., entitled "160.times.244 Element PtSi Schottky-Barrier IR-CCD Image Sensor", published in IEEE Transactions On Electron Devices, Vol. ED-32, No. 8, August 1985, pp. 1564-1573. A problem with this type of imager is that the transfer means, such as the CCD, takes up room on the surface of the substrate so as to reduce the fill-factor of the device.
As described in U.S. Pat. No. 3,864,722 to J. E. Carnes, entitled "Radiation Sensing Arrays", issued Feb. 4, 1975, a CCD imager has been developed which has the Schottky-injection detector along one surface of a substrate of semiconductor material and a CCD transfer means along the opposite surface of the substrate. Radiation detected by the Schottky-barrier detector surface is converted to charge carriers and some of the carriers are injected into the substrate. The substrate is depleted so that the charge carriers move across the substrate to the channels of the CCD. Voltages are then clocked to the gates of the CCD to move the charge carriers along the channels to the output of the device. Although this device has an improved fill-factor, since the detector is across the entire surface of the substrate, this deiice has other problems resulting from the CCD being the collector of the charge carriers as well as the transfer means. As one set of charge is being transferred along the channels of the CCD, other charges are being collected. Thus, if a charge is not transferred along the channels fast enough, it can be interferred with by the new charges being collected thus causing smear.
Summary of the Invention
An imager includes a substrate of semiconductor material of one conductivity type having a pair of opposed major surfaces. Along one of the major surfaces of the substrate is means for converting detected radiation into charge carriers. A plurality of charge collecting regions for the charge carriers of the same conductivity type as the substrate are in the substrate at the other major surface. The substrate includes means extending along the other major surface for receiving the charge carrier from the collecting regions and for transferring the charge carriers region to an output. The substrate also includes means for isolating the collecting regions from each other and for preventing the charge carriers flowing from the connecting means to the collecting region from reaching the transfer means.
Brief Description of the Drawing
FIG. 1 is a schematic view of the imager of the present invention.
FIG. 2 is a top plan view of a portion of the imager of the present invention.
FIG. 3 is a sectional view taken along line 3--3 of FIG. 2.
FIG. 4 is a sectional view taken along line 4--4 of FIG. 2.
FIG. 5 is a profile of the potentials developed in one pixel of the CCD of the imager of the present invention.
FIG. 6 is a diagram of the potentials for operation of the imager of the present invention.
FIG. 7 is a sectional view of the edge portion of the imager of the present invention.
FIG. 8 is a sectional view of a portion of a modification of the imager of the present invention.
FIG. 9 is a sectional view of a portion of another modification of the imager of the present invention.
FIG. 10 is a sectional view of a portion of still another modification of the imager of the present invention.
FIG. 11 is a sectional view of a portion of a further modification of the imager of the present invention.
FIG. 12 is a block diagram of another modification of the imager of the present invention using an alternative transfer means.
Detailed Description of Preferred Embodiment
Referring initially to FIGS. 2-4, one form of the imager of the present invention is generally designated as 10. Imager 10 includes a substrate 12 of a semiconductor material, such as single crystalline silicon, having a pair of opposed surfaces 14 and 16. The substrate 12 is preferably about 25 microns in thickness and of very lightly doped P type (P--) silicon, typically about 1000 ohm-cm. On the surface 16 of the substrate 12 is a thin layer 18 of a conductive material which forms a Schottky-barrier junction with the silicon of the substrate 12, such as platinum silicide. The conductive layer 18 may be coated with a layer 20 of an anti-reflective material, such as silicon oxide.
A N-type conductivity well 22 extends into the substrate 12, typically about 6 microns, from the surface 14. The well 22 has a plurality of openings 24 therethrough to the substrate surface 14. As shown in FIG. 2, the openings 24 are arranged in spaced, parallel rows and columns over the substrate surface 14 with the spacing between the columns being greater than the spacing between the rows. A charge collecting electrode 26 extends into the substrate 12 from the surface 14 in each of the openings 24 in the well 22. The charge collecting electrodes 26 are regions of the same conductivity type as the substrate 12 but of higher conductivity, i.e. P+ type conductivity. This provides a plurality of rows and columns of the charge collecting electrodes 26.
Between each adjacent pair of columns of the charge collecting electrodes 26 and within the well 22 is a buried channel 28 of a CCD. Each buried channel 28 is a region of P type conductivity extending into the substrate 12 from the surface 14 and extending longitudinally along the substrate surface 14 between and parallel to the columns of the charge collecting electrodes 26. Each of the buried channels 28 is spaced from its adjacent columns of the charge collecting electrodes 26. However, at each of the collecting electrodes 26 in one of the adjacent columns the buried channel 28 has a connecting region 30 extending to and contacting the respective collecting electrode 26. Thus, each buried channel 28 is electrically connected to all of the charge collecting electrodes 26 in one adjacent column. The connecting regions 30 can be of the same conductivity type as the buried channels 28 but more lightly doped, i.e. P- type, or can be of the opposite conductivity type, i.e. N type. Between each of the buried channels 28 and its other adjacent column of collecting electrodes 26 is an overflow drain 32. Each overflow drain 32 is a region of the same conductivity type as that of the buried channel 28 but of higher conductivity, i.e. P+ type conductivity, extending into the substrate 12 from the surface 14 to a depth greater than the depth of the buried channel 28. Between each of the overflow drains 32 and its adjacent column of collecting electrodes 26 is a barrier region 34 of the same conductivity type as the channel 28 and of a conductivity the same as or less than that of the buried channels 28.
Covering the substrate surface 14 is a layer 36 of an insulating material, such as silicon oxide or a layer of silicon nitride over a layer of silicon oxide. On the insulating layer 36 are a plurality of spaced, parallel field-shield gates 38 which extend over and parallel to the columns of the charge collecting electrodes 26. The field-shield gates 38 are of a conductive material, such as doped polycrystalline silicon. Each field-shield gate 38 is of a width to extend from the edge of one buried channel region 28, across the width of its respective charge collecting electrode 26, the adjacent barrier region 34, overflow drain 32, and a portion of the buried channel 28 adjacent the overflow drain 32. However, as shown in FIG. 2, the field-shield gate 38 has recesses in its edge at the connecting regions 30 so that it does not extend over the connecting regions 30. Each of the charge collecting electrode gates 38 is covered with a layer 40 of an insulating material, such as silicon oxide.
Two sets of CCD transfer gates 42a and 42b extend transversely across the buried channels 28 and the charge collecting electrode gates 38. The transfer gates of each se are in spaced, parallel relation with the transfer gates of one set being between adjacent gates of the other set. Each of the gates 42b extends across a connecting region 30 in each CCD. As shown in FIGS. 1 and 4, the CCD transfer gates 42b each overlap portions of the adjacent gates 42a and are insulated therefrom by a layer 44 of an insulating material, such as silicon oxide, which is on the gates 42a. The CCD transfer gates 42a and 42b are of a conductive material, such as doped polycrystalline silicon.
Thus, as schematically shown in FIG. 1, the imager 10 has along the surface 14 of the substrate 12 a plurality of spaced, parallel columns of the charge collecting electrodes 26. The charge collecting electrodes 26 in each column are electrically connected to a CCD 46 which extends longitudinally between the columns of the charge collecting electrodes. Each of the CCDs 46 is made up of a buried channel 28 and the transfer gates 42a and 42b. An output CCD register 48 extends across the ends of the columns of the charge collecting electrodes 26 and is electrically connected to the CCDs 46. The channel of the output CCD register 48 is also formed in the well 22. The output CCD register 48 is adapted to receive charges from the CCDs 46 and transfer the charges to an output of the imager 10.
In the operation of the imager 10, radiation is directed at the surface 14 of the substrate 12 as indicated by the arrows 50 in FIGS. 3 and 4. The radiation is absorbed in the conductive layer 16, and is converted to hot charge carriers which are injected across the Schottky-barrier junction into the substrate 12. The wavelength of the radiation absorbed is determined by the barrier height of the Schottky-barrier junction which, in turn, is determined by the particular material of the substrate 12 and the conductive layer 16. The Schottky-barrier formed between a platinum silicide conductive layer 16 and a silicon substrate 12 will absorb infrared radiation having a wavelength up to about 5 microns and also ultra-violet radiation. The type of charge carriers generated depends on the conductivity type of the substrate 12. For a substrate 12 of P type conductivity "holes" are generated in the substrate 12 as indicated by the plus marks 52 in FIGS. 3 and 4.
A potential difference is generated between the conductive layer 16 and the charge collecting electrodes 26 which depletes the substrate 12 completely between the conductive layer 16 and the charge collecting electrodes 26. This causes the charge carriers 52 to follow the electric field lines as indicated by the dash lines 54 from the conductive layer 16 to the charge collecting electrodes 26. The charge collecting electrodes 26 are electrically isolated from each other by the wells 22 of opposite conductivity type which also act as potential barriers for the charge carriers so that the charge carriers will flow to the charge collecting electrodes 26. Thus, the charge collecting electrodes 26 serve to collect the charge carriers which are generated by the incoming radiation at the substrate surface 16.
As shown in the diagrams of FIGS. 5 and 6, when the transfer gates 42b are pulsed to a potential .phi..sub.BT, a potential .psi..sub.T2 is applied to the transfer region 30 and a potential .psi..sub.S2 is applied to the portion of the buried channel 28 under the gate 42b. This lowers the barrier provided by the transfer region 30 to the level of the potential well in the charge collecting electrode 26 and also lowers the potential well in the buried channel 26. Thus, the charge signal (Q.sub.D), which collected in the charge collecting electrode 26, will flow into the buried channel 26. The gates 42a and 42b are then clocked to transfer the charge signal (Q.sub.D) along the buried channel 26 to the output CCD register 48 where the charges are clocked to the output of the imager 10. During the clocking of the transfer gates 42a and 42b the clock voltage .theta..sub.B results in a buried channel potential changes of A.psi..sub.B =.psi..sub.S1 -.psi..sub.S10 while maintaining a barrier potential, .psi..sub.T1, between the buried channel 28 and the potential well of the charge collecting electrode 26.
Instead of using a potential applied to the transfer gates 42b of the CCD column registers to cause the charge to be transferred fromthe charge collecting electrodes 26 to the buried channel 28 of the CCD, a separate transfer gate can be formed over and insulated from the edge portion of the field-shield 38 which extends between the charge collecting electrodes 26 and the buried channel 28. The transfer gate would extend parallel to the field-shield 38, a column of the charge collecting electrode 26 and the buried channel 28 so as to extend over a column of the connecting regions 30. The connecting region 30 would then be formed of the same conductivity type and conductivity as the buried channel 28. The transfer gates 42a and 42b of the CCD column register would extend over and be insulated from the separate transfer gates.
The field-shield gates 38 serve several functions. A voltage applied to the field-shield gates 38 increases the capacitance, and thus the charge handling capacity, of the charge collecting electrodes 26. They also provide the barriers between the blooming drain 32 and the charge collecting electrodes 26 on one side and the buried channel 28 on the other side and between the charge collecting electrodes 26 and the buried channel 28 to which the charge collecting electrodes are connected by the connecting regions 30.
As shown in FIG. 7, the imager 10 can be formed with the charge collecting electrodes 26 being close to an edge 54 of the substrate 12. A P+ drain region 56 is provided between the substrate edge 54 and the adjacent charge collecting electrodes 26 to collect any edge leakage current. A N type passivation region 58 is provided along the edge 54 between the surfaces 14 and 16 to minimize any edge leakage current. This structure allows the charge collecting electrodes 26 to be close to the substrate edge 54 so that two of the imagers 10 can be placed in edge abutting relation to provide a large imager with a minimum of loss space between the columns of the charge collecting electrodes 26 on the two imagers. By making the portions 58a and 58b of the passivation region 58 which are adjacent the major surfaces 14 and 16 of the substrate 12 highly conductive, i.e. N+ type, the passivation region 58 can also serve as a contact to the conductive layer 18 on the substrate surface 16. This allows all of the terminals for the imager 10 to be at one side of the imager and also allows the imager 10 to be mounted on a support for providing greater strength to the imager.
A modification of the CCD imager of the present invention, generally designated as 100, is shown in FIG. 8. The CCD imager 100 is similar in structure to the CCD imager 10 shown in FIGS. 3 and 4 except that the charge collecting electrodes 126 have extensions 126a of P type conductivity of a conductivity greater than that of the substrate 112 but less than the charge collecting electrodes 126. The extensions 126a extend into the substrate 112 a distance at least equal to and preferably greater than the depth of the well 122, typically about 7 to 8 microns. This extension 126a provides a more gradual doped transistion from the heavily doped charge collecting electrode 126 to the very lightly doped substrate 12 to reduce leakage currently preventing the depletion of the substrate 12 from reaching the collecting electrode 126. places the charge collecting region closer to the Schottky-barrier detection junction at the surface 116 and also increases the area of the charge collecting region so as to improve the charge collecting efficiency.
Another modification of the imager of the present invention is shown in FIG. 9 and is generally designated as 200. Imager 200 is similar in construction to the imager 100 shown in FIG. 8 except that the extension 226a of the charge collecting electrodes 226 extends completely across the substrate 212 and under the wells 222. In making the imager 200, the charge collecting electrode extensions 226a can be formed by embedding P type conductivity modifiers, such as boron, into the substrate surface 214 across the entire substrate 212 either by diffusion or ion implantation. The well 222, charge collecting electrodes 226 and other features of the device would then be formed in the P type region. For example, the well 122 can be formed by implanting phosphor ions into the substrate 112 at 100 KeV and a dosage of 3.times.10.sup.12 cm.sup.-2 and the extension 226a can be formed by implanting boron into the substrate 112 at 50 KeV and a dosage of 10.sup.12 cm.sup.-2. After the imager 200 is completed, the phosphor and boron ions will have been subjected to a heat treatment equivalent to 4 hours at 1200.degree. C. and 8 hours at 1200.degree. C. respectively. This will provide an extension 226a which extends about 7 to 8 microns below the substrate surface 114 and a well 122 which extends about 4 microns below the substrate surface 114. Alternatively, the charge collecting electrode extension 226a can be formed by epitaxially growing a layer of P type semiconductor material on the substrate 212 and then forming the various elements of the device in the epitaxial layer.
Another method of making the imager 200 is shown in FIG. 10. For this method a layer 227 of P type conductivity is formed in the substrate 212 across the major surface 214 either by ion implantation or diffusion. Barrier regions 223 of N type conductivity are formed in the substrate 214 across the layer 227 in the areas between where the charge collecting electrodes 226 are to be formed. An epitaxial layer of N type conductivity is deposited on the substrate surface 214 and regions of P type conductivity are formed through the epitaxial layer to the P type layer 227, such as by ion implantations or diffusion, to form the extension 226c. The P type regions are heavily doped at the surface of the epitaxial layer to form the charge collecting electrodes 226. The portion of the epitaxial layer around the charge collecting electrodes 226 form the well 222. The parts of the CCD are then formed in the well 222.
Although the operation of the imager of the present invention has been described with the radiation impinging directly on the side of the imager having the Schottky-barrier detector, the imager can be operated with the radiation impinging on the side of the imager having the CCD transfer device as indicated by the arrows 60 in FIG. 11. For such an operation the conductive layer 16 would be coated with a dielectric layer 62, such as of silicon oxide, about one-quarter wavelength in thickness, and a layer 64 of a reflective material, such as aluminum, over the dielectric layer 62. The dielectric layer 62 and reflective layer 64 form an optical cavity which would reflect any radiation passing through the conductive layer 18 back to the substrate 12 to be absorbed at the Schottky-barrier junction. This greatly increases the absorption of the radiation in the conductive layer 18. This construction also allows the imager 10 to be easily mounted on a cooling means, such as the cold plate of a Dewar assembly, which improves the operation of the device. However, this manner of operating the imager has a disadvantage in that the radiation must pass through the gates 38 and 42 and the substrate 12 before reaching the Schottky-barrier detector. This could result in some loss of the radiation.
Although the imagers of the present invention have been described as using a CCD to transfer the charges from the charge collecting electrodes to the output of the imager, other types of transfer means can be used instead of the CCD. For example, as shown in FIG. 12, MOS transistor transfer means can be used. The particular MOS transistor transfer means shown in FIG. 12 includes a pair of MOS transistor 66 and 68 connected in series to each charge collecting electrode 26. All of the MOS transistors 66 and 68 are formed in the well 22 so as to be isolated from the charge collecting electrodes 26. The source of the transistor 66 of each pair is connected to its respective charge collecting electrode 26 and the drain of the other transistor 68 of each pair is connected to the output 70. The gates of each of the transistors 66 for each row of charge collecting electrodes 26 are connected by a common line to a horizontal line selector 72 and the gates of the transistor 68 for each column of the charge collecting electrodes are connected by a common line to an element scanner 74. This provides an x-y scanner for the charge collecting electrodes 26 so that the charges from each of the charge collecting electrodes 26 can be transferred in the proper order to the output 70.
The imager of the present invention with interline transfer, buried channel CCD, has been described as a storing type of sensing device. However, this device can be used as a time delay integration line sensing device as is generally known in the art of visible imager sensors. Time delay integration line sensing can be achieved by synchronizing the scanning motion of the imager with the clocking of the CCD column registers and transferring the detected charges from the charge collecting electrodes to the buried channel of the CCD column registers over for each clock of the register.
Thus, there is provided by the present invention an imager which includes a detector along one major surface of a substrate for receiving radiation and converting the radiation to charge carriers within the substrate, and along the opposed major surface of the substrate collecting electrodes for collecting the charge carriers and transfer means for transferring the charge carriers from the collecting electrodes to an output. By having the detector along one entire surface of the substrate and the collector electrodes and transfer means along a different surface, the fill-factor of the image is improved for improved sensitivity and reduction of aliasing of the image being detected: By having the means for collecting the charge carriers separate from the means for transferring the charge carriers, there is no interference between the carriers being transferred and the new carriers being collected. This reduces the possibility of smear.