CA2036874A1 - Bandgap radiation detector and method of fabrication - Google Patents

Abstract

A radiation detector device for radiation of the infrared region and regions of shorter wavelengths comprises a dipole antenna mounted on a substrate and connected through blocking contacts to a bandgap detector element. The dipole antenna has a length which is approximately one half the wavelength of the incident infrared radiation.
The bandgap detector element has linear dimensions which are each substantially smaller than the wavelength of the detected radiation. A
group of detector devices are combined to form an array which can produce a pixel signal for an image. Unlike conventional infrared radiation detectors, the disclosed detector device is capable of producing a usable output signal without the need for cooling below ambient temperature. A further infrared radiation detector device has an array of detectors each of which comprises a pattern of parallel detector elements. Each detector produces a pixel signal for an image.
The elements of the detector are photoconductive or photovoltaic bandgap materials, and the elements are spaced apart at a dimension which is equal to or less than the wavelength of the radiation to be received.
Additional layered structures above and/or below the detector elements provide impedance matching between free space radiation and the radiation impedance of the detector elements, to increase the capture of radiation.

Description

HANDLE AS CONFIDENTIAI_ BANDGAP RADIATION DETECTOR AND
METHOD OF FABRICATION
FIELD OF THE INVENTION
The present invention pertains in general to radiation detection devices and in particular to such devices for detecting radiation in the infrared region and shorter wavelength regions.
BACKGROUND OF THE INVENTION
Radiation detectors such as infrared detectors have long been used in thermal imagers for viewing at night, or for viewing through clouds, smoke and dust. A conventional infrared imager has an array of large-area detectors wherein each detector corresponds to a single picture element (pixel) for an image. Each detector is a planer structure which has both the length and width dimensions larger than the wavelength of the incident radiation such that the detector has an adequate collection area for the incident radiation. A conventional detector of this type is shown in "Semiconductors and Semimetals," Vol. 18, Mercury Cadmium Telluride, Academic Press, 1981, pp. 162-163.
A principal limitation in the use of conventional infrared imagers has been the requirement that the imaging device be enclosed within a very cold chamber. The cooling for such devices has most often been provided by the evaporation of liquid gases, such as nitrogen. However, the storage, piping and handling of coolants such as liquid nitrogen has proven to be difficult, time consuming and expensive.

Although conventional radiation detectors can successfully produce useful images, they have serious limitations in their operation. For a given input power of incident radiation, the resulting signal strength of these devices is relatively low.
Therefore, there is a great need for an improved infrared detector of radiation in the infrared region and shorter wavlength regions, which can produce a greater amplitude signal with less requirement for complex cooling apparatus.

SUMMARY OF THE INDENTION
A selected embodiment of the present invention is a device for detecting incident radiation. The device includes a support structure which has a dipole antenna mounted on it. The length of the antenna is equal to approximately one half the wavelength of the incident radiation.
The device further includes a quantum detector element which is mounted on the support structure and is connected to the dipole antenna. The detector element is fabricated to have each of the linear dimensions thereof substantially less than the wavelength of the incident radiation.
The incident radiation is captured by the dipole antenna and transferred to the detector element for producing an output signal to indicate detection of the incident radiation at the device.
Further embodiments of the present invention include an array of such devices and an imager comprising a plurality of such arrays.

BRIEF DESCRIPTION OF THE DRAWINGS
For a more complete understanding of the present invention reference is now made to the following detailed description taken in conjunction with the drawings in which:
Figure 1 is a plan view of an array of devices in accordance with the present invention for detecting incident radiation, Figure 2 is a plan view of an alternate embodiment of an array of infrared radiation detection devices in accordance with the present invention;
Figure 3 is a section view taken along lines 3-3 of the infrared radiation detection device shown in Figure 2;
Figure 4 is a perspective view of an infrared radiation imager in accordance with the present invention utilizing either of the infrared radiation detector arrays illustrated in Figures 1 and 2;
Figure 5 is a plan view illustrating a conventional infrared radiation detector;
Figure 6 is a plan view of an infrared radiation detector in accordance with the present invention;
Figure 7 is a section view taken along Lines 7-7 of the radiation detector shown in Figure 6;
2p Figure 8 is a section view of an alternative embodiment of the present invention;
Figure 9 is a section view of a further embodiment of the present invention employing orthogonal detector sets;

Figure 10 is a plan, partially cut away view, of the two detector set embodiment shown in Figure 9;
Figure 11 is a section view of a still further embodiment of the present invention;
Figure 12 is a plan view of a radiation imager utilizing the infrared radiation detector of the present invention;
Figure lx is a perspective view of a pixel element of an infrared detector fabricated in accordance with one embodiment the present invention;
Figure 2x is a section view taken along lines 2x-2x of the detector shown in Figure lx;
Figures 3Ax-3Kx illustrate steps in the process of making the detector illustrated in Figure lx;
Figure 4x is a chart illustrating the radiation absorption of mercury cadmium telluride at room temperature for two different concentrations of mercury;
Figure 5x is an illustration of the overall absorption of the infrared radiation detector illustrated in Figure lx;
Figure 6x is a perspective view of a nonpolarized infrared radiation detector in accordance with the present invention;
Figure 7x is a plan view for an alternate design for an infrared detector in accordance with the present invention;
Figure 8x is a section view taken along the lines 8x-8x of the detector shown in Figure 7x;

Figure 9x is a perspective view of the infrared detector shown in Figures 7x and 8x;
Figure lOx is a perspective view of an alternative embodiment of the present invention;
Figures llAx-llLx illustrate steps in the process of making the detector illustrated in Figures 7x-9x;
Figure 12x is an electrical schematic of a detector in accordance with the present invention;
Figure 13x is an illustration of an infrared imaging system having a detector array utilizing the detector elements in accordance with the present invention;
Figure 14x is a perspective view of an infrared detector having strips of photosensitive material disposed between parallel conductors;
and Figures l5Ax-l5Hx illustrate steps in the process of making the detector illustrated in Figure 14x.

DETAILED DESCRIPTION OF THE INDENTION
Referring now to Figure 1, there is illustrated an infrared (IR) radiation detector array 10 in accordance with the present invention.
The array 10 includes a support structure 12 which provides physical support and necessary electrical properties such as nonconductivity. A
plurality of identical detector devices 14, 16, 18, 20, 22, 24, 26, 28 and 30 make up a 3 x 3 matrix for the array 10. The detection device 16 will be described in detail, however, it is representative of the remaining devices within the array 10. The detecting device 16 includes a dipole antenna 36 which includes antenna elements 36a and 36b. The antenna 36 is fabricated of a conductive material such as, for example, aluminum. The wavelength of incident radiation is indicated by the symbol lambda (~). The length of the dipole antenna 36 is approximately one-half the wavelength of the incident radiation. Each of the detecting devices 14-30 has a capture area which is illustrated for detector device 22 by the dotted oval 38 surrounding device 22. It can be seen that the detecting devices 14-30 are positioned such that a substantial part of the incident radiation is captured by the dipole antennas.
Between the antenna elements 36a and 36b there is provided a bandgap detector element 40 which is mounted on the support structure 12 and is electrically connected to the antenna elements 36a and 36b.
Blocking contacts, described below, provide electrical connections between the antenna elements 36a, 36b and the detector element 40.
The support structure 12 has a thickness which is approximately one fourth the wavelength of the incident radiation. Selected materials for the structure 12 are zinc selenide or zinc sulfite.
The array 10 can serve as a pixel element in an infrared imager, as further described below in reference to Figure 4. In such an application the detector devices such as 16, for all of the detector devices in the array 10, are joined in common to produce a pixel signal.

However, it is also possible to have each detector device, such as 16, individually generate a pixel signal.
Referring now to Figure 2, there is illustrated a second configuration for the present invention. This configuration is an array 50 which has a support structure 52. The structure 52 is similar to structure 12 described above. The array 50 is a 2 x 4 matrix comprising detector devices 54, 56, 58, 60, 62, 64, 66 and 68. Each of the detector devices, such as 56, includes a detection element and a dipole antenna. The detector devices 54-68 include respective detector elements 72-86. Each detector element is positioned between and connected to two planar metal antenna elements which serve as a dipole antenna for each detector element. The antenna elements are included in metal areas 94, 96 and 98. For example, device 56 includes detector element 74 which is connected via blocking contacts, described below, to a dipole antenna which comprises metal members 102 and 104 which are respectively parts of metal areas 94 and 96.
The configuration shown in Figure 2 permits an output signal to be produced between metal areas 94 and 98 due to the electrical responses produced by the detector elements 72-86 upon receipt of incident 2p radiation.
The horizontal cell spacing of the structure 50 is shown by line 106, the vertical cell spacing by line 108, the horizontal cell aperture by line 110 and the vertical cell aperture by line 112. For detecting infrared radiation having a wavelength of 10 microns, the horizontal and vertical cell spacings are approximately 5 microns and the cell apertures are approximately 4.5 microns. Each of the detector elements 72-86 has dimensions which are approximately .75 by .25 microns. Thus, the dimensions of the detector elements are substantially less than the wavelength of the intercepted radiation. For 5 micron wavelength IR
radiation, each of the above linear dimensions is reduced by approximately one half, but the detector element is still approximately the same size.

The array 50 likewise can serve as a pixel signal source within an imager having a large number of such arrays, as shown in Figure 4.
Array 50 is further illustrated in a section view in figure 3. A
reflective layer 116, such as metal, for example aluminum or a multilayer dielectric, is located between the support structure 52 and an insulating substrate 118. Layer 116 serves as a ground plane. A
preferred composition of the substrate 118 is a silicon substrate, which is a part of a silicon charge coupled device (CCD) chip that processes the signal produced by array 50.
The detector element 74 is electrically connected to the dipole antenna metal members 102 and 104 by means of blocking contacts.
Elements 72-86 are preferably an alloy of mercury cadmium telluride in which the fractional part of cadmium is represented by x and the fractional part of mercury is represented by 1-x. The preferred mercury cadmium telluride alloy for the detector elements 72-86 has x=.15. The array 50, referring to Figure 3, includes blocking contacts 120, 122, 124 and 126. Contact 120 is a layer which is on and in contact with the outer surface of detector 74. The metal member 102 is fabricated directly over the contact 120. Member 102 is electrically connected to contact 120 but is not directly connected to the detector 74. Contact 122 is similarly positioned between detector 74 and metal member 104.
Contact 124 is positioned between detector 82 and a part of the metal area 96. And likewise, blocking contact 126 is positioned between detector 82 and a part of metal area 98. the blocking contacts 120, 122, 124 and 126 comprise a mercury cadmium telluride alloy having X = .19.
The purpose of the blocking contacts 120, 122, 124, and 126 is to prevent the diffusion of change carriers from the detector elements, such as 74, into the aluminum metal members, such as 102 and 104. It is desirable to have a recombination velocity of less than 500 cm/sec, but a direct aluminum contact to the detector results in a recombination velocity which approaches infinity. By including the blocking contacts, such as 120, 122, 124 and 126, between the detector elements and the metal dipole members, the recombination velocity of the carriers is reduced to the range of 300-500 cm/sec. The blocking contacts and associated elements function much like high frequency coupling capacitors to provide do isolation of the carriers but present only a small impedance between the detector elements and the dipole antenna elements. The theory and detailed operation of a blocking contact of this type is described in "HgCdTe Heterojunction Contact Photoconductor," Applied Physics Letters 45(1), 1 July 1984, pp. 83-85 by D. L. Smith, D. K. Arch, P. A. Wood and M. Walter Scott.
The detector elements, such as 40 and 14 discussed above are direct bandgap detectors. A bandgap detector of this type produces an electrical signal as a result of interaction between the incident photons and the electrons (holes) in the material. A detector element of this type is not subject to thermal (phonon) exchange noise as are bolometer detectors and the like. Bandgap detectors are therefore not severely limited in performance at high (ambient) temperatures as are conventional bolometer and similar detectors.
A preferred material for the detector elements 40 and ~2-86 is a crystal of mercury cadmium telluride. A further material can be indium antimonide. Another suitable detector element can be a semiconductor 2p superlattice as described in Scientific American, "Solid State Superlattices," November 1983 by Gottfried H. Dohler. A still further detector element can be an organic material as described in Laser Focus, "Organic Crystals and Polymers - A new Class of Nonlinear Optical Materials," February 1982, pp. 59-64 by Anthony F. Garito and Kenneth H.
Singer. By using the organic material described 'in the reference to Garito and Singer, it is possible to have coherent detection of the incident radiation.
An imager 130 in accordance with the present invention is illustrated in Figure 4. the imager comprises a plurality of individual arrays, such as 132. The array 132 may, for example, be the same as array 10 or 50 described above. An infrared image is transmitted through a lens 134 onto the planar imager 130. Each of the arrays produces a pixel signal which, for array 132, is transmitted through a line 136. There is a corresponding line for each array within the imager 130. The collection and processing of all pixel signals forms a reproduction of the original image transmitted through lens 134. A line 138 can provide a common ground which extends throughout the imager 130.
One method in accordance with the present invention for fabricating the bandgap detector elements is described as follows. A layer of cadmium telluride is first epitaxially grown on a substrate to a thickness of 1-2 microns. This grown cadmium telluride layer is next exposed to mercury to form a surface layer crystal of mercury cadmium telluride. Last, the layer of mercury cadmium telluride is etched to form a selected array of detector elements such as shown in Figures 1 and 2. A further method for fabricating a mercury cadmium telluride crystal for the detectors and blocking contacts in the present application is described in "Molecular Beam Epitaxial Growth of High Quality HgTe and Hgl-xCdxTe onto GaAs (001) Substrates," Applied Physics Letters 45(12), December 15, 1984 by J. P. Fauve, S. Sivananthan, M. Boukerche and J. Reno.
A principal advantage for the structure of the present invention is the capability of producing an output signal for a detected infrared signal without the usual requirement of cooling the detector elements.
A conventional infrared detector element is a planar element which has an area with dimensions which are substantially greater than the wavelength of the incident radiation. These large area elements are needed to capture the incoming infrared radiation.
To summarize, one embodiment of the present invention includes a detector, array and imager designed principally for detecting infrared radiation by the use of dipole antennas which capture incident radiation and transfer it to a bandgap detector element. The detector element has dimensions which are substantially smaller than the wavelength of the incident radiation. A principal advantage of the present invention is the generation of an infrared imaging signal without necessarily cooling the detector elements.

A further embodiment of the present invention is an infrared radiation detector which produces an output signal when exposed to infrared radiation. A conventional, prior art, infrared radiation detector 200 is shown in Figure 5. Infrared radiation having a wavelength of 8-12 microns is of principal interest for detection due to its propagation characteristics through the atmosphere. The conventional detector 200 has large area detector elements, such as element 212, for capturing incident infrared radiation. The detector element 212 has typical length and width dimensions of 50 microns. The 50 micron dimensions are substantially greater than the 8-12 micron wavelength of the intercepted radiation. This large area detector configuration serves to capture the incident radiation over areas which approximately correspond to the size of a pixel (picture element) in an image. Each of the detector elements, such as 212, produces a pixel signal and these signals are used in combination to produce an image.
A further infrared radiation detector 214 in accordance with the present invention is illustrated in Figure 6. The detector 214 has a periodic, parallel pattern of photoconductive or photovoltaic, bandgap detector elements 216, 218, 220, 222, 224, 226, 228 and 230. These elements must be made of a radiation absorbing material. A preferred material for these detector elements is mercury cadmium telluride which is described as Hg(1_x) Cd(x) Te, where a selected value of x is .2.
The detector elements 216-230 are joined together at opposite ends by respective common lines 236 and 238, which are typically made of metal, such as aluminum. In a selected embodiment the detector elements 216 230 are etched from a single layer of mercury cadmium telluride.
The detector elements 216-230 and common lines 236, 238 are fabricated on a substructure 240 which serves multiple functions.
Substructure 240 provides a support for the elements 216-230 and lines 236, 238 and, quite significantly, it provides impedance matching between free space radiation and the radiation impedance of the pattern of detector elements 216-230. The substructure 240 contains layers with indices of refraction (n) different from that of air or free space. The substructure 240 increases the radiation absorption of the detector 214.

Referring also to Figure 7, the substructure 240 comprises separate layers 242 and 244. Layer 242 is preferably indium antimonide and layer 244 is preferably cadmium telluride. Layer 242 has an index of refraction n=4 and layer 244 has an index of refraction n=2.7, where n=1 for free space.
The pattern of detector elements 216-230 as shown in Figure 6 comprises a pixel having overall dimensions of 50 microns by 50 microns.
This structure is designed for receiving 8-12 micron infrared radiation.
Each of the detector elements 216-230 has a width of approximately .5 micron and a length of approximately 50 microns. The preferable period of centerline to centerline spacing between the elements 216-230 is 3 microns. A preferred thickness for each of the elements 216-230 is .5 micron. A preferred thickness for each of the layers 242 and 244 is in the range of .1-10 microns.
It has been determined that there is a limiting criteria for effective operation of the present invention as shown in Figures 6 and 7.
For normal incident radiation, this criteria has two aspects, first, the wavelength (~) of the incident radiation must be greater than or equal to the product of the periodic spacing (p) between the detector elements and the index of refraction (n2) for the lower layer, that is, layer 244 as shown in Figure 7. This is expressed at ~ = n2p. Second, the upper layer 242 must have a greater index of refraction (n1) than the index of refraction (n2) for the lower layer 244. This is expressed as n1 > n2.
When these two aspects have been met, the absorption of incident radiation for the present invention can approach 100%. When this criteria is not met, a detector such as 214 shown in Figure 7, will be limited to a maximum absorption of less than 50%.
While the detailed theoretical operation of the present invention is not yet fully understood, it appears that the incident radiation, which is not directly absorbed by the detector elements 216-230, is essentially trapped in the layer 242 due to the different indices of refraction between layer 242 and layer 244 on one side and layer 242 and free space on the opposite side.
The incident radiation is likely diffracted by the detector elements 216-230 to alter its propagation direction away from the normal incidence path. The trapped radiation is absorbed when it ultimately strikes the detector elements 216-230 after possibly many reflections.
It appears that radiation which does escape from layer 242 back into free space through the plane of the detector elements 216-230 is cancelled by the incoming incident radiation, thereby contributing to the overall absorption of incident radiation.
All of the detector elements 216-230 are connected in parallel between the lines 236 and 238. Line 236 is connected through a conductor 246 to the terminal of a DC source or battery 248. Line 238 is connected through a conductor 250 to a terminal 252. A resistor 254 is connected between terminal 252 and a terminal 256. The remaining terminal of battery 248 is connected to terminal 256. The battery 248 applies a bias across the detector elements 216-230 and the resistor 254 serves as a series load. When infrared radiation is captured by the detector 214, electrons in the detector elements 216-230 are boosted to higher energy bands which alters the current flow produced by the battery 248. This translates to changes in the current through resistor 254 which alters the voltage between the terminals 252, 256. Thus, the detector 214 produces a pixel signal at terminals 252, 256. An array of detectors, such as detector 214, produces a complete image by generating a signal for each pixel.
The detector of the present invention has greater responsivity because of the higher captured radiation power density in the sensitive material as compared to a conventional infrared detector. For example, the detector 212 in Figure 5 and the detector 214 in Figure 6 have the same overall planar dimensions. Detector 212 has an active area of 2,500 square microns with a typical thickness of 10 microns, but detector 214 has an active area of only about 425 square microns with a typical thickness of 0.5 microns. With equal incident radiation intensity, the detector 214 will have approximately 120 times greater power density in the sensitive elements, which provides a substantial increase in performance. As the widths of the elements of the detector 214 become smaller, the increase in power density becomes greater. The detector of the present invention therefore offers a substantial performance advantage over conventional large area detectors.
The substructure 240 serves to provide impedance matching between free space radiation impedance and the radiation impedance of the detector 214 elements 216-230. A basic measure of performance for a radiation detector is the percent of absorption for incident radiation.
Without the substructure 240, the elements 216-230 have a radiation absorption of less than 5% but with the addition of the substructure 240, which meets the above criteria, the absorption is increased to over 80%, as indicated by computer simulations.
The various detectors illustrated for this embodiment of the present invention employ parallel detector strips, but in general, periodic elements of any shape can be used provided that the spacing between the elements is less than or equal to the wavelength of the incident radiation.
2p A still further embodiment of the present invention is shown in Figure 8. A detector 260 is similar to the detector 214 shown in Figures 6 and 7 but with the additional of a superstructure to provide additional impedance matching between the detector elements and free space impedance. Detector 260 has a set of parallel detector elements 262, 264, 266, 268, 270, 272, 274 and 276 which are the same as the detector elements 216-230 in detector 214. Detector 260 has a substructure 280 comprising layers 282 and 284 which correspond to the layers 242 and 244 in detector 214. The detector 260 also includes a superstructure 286 comprising layers 288 and 290. Layer 288 is similar to layer 242 in detector 214 and layer 290 is similar to layer 244 in detector 214. the superstructure 286 functions like the substructure 240 for improving the impedance match between elements 262-276 and free space radiation impedance.

A further embodirr~nt of the present invention is a detector 296, which is shown in Figures 9 and 10. The detector 296 has detector element sets 298 and 300 each of which is the same and electrically connected in the same manner as the detector elements 216-230 in detector 214 shown in Figure 6. However, the detector elements in set 298 are orthogonal to the detector elements in set 30U. The detected signals from the two sets can be electrically combined. The detector 296 has a substructure 302, a superstructure 304 and a midstructure 306.
The substructure 302 comprises layers 308 and 310 and the superstructure 304 comprises layers 312 and 314. Substructure 302 corresponds to substructure 280 and superstructure 304 corresponds to superstructure 286. The midstructure 306 has layers 316 and 318 each of which is preferably a material such as cadmium telluride having a thickness of approximately .1 to 10 microns. Detector set 298 is in layer 316 and detector set 300 is in layer 318. The two detector element sets 298, 300 are oriented orthogonally to capture orthogonal polarizations of the incoming radiation. The structure shown in Figure 6 captures only one polarization. A sectioned planar view of detector 296 is shown in Figure 10 with the superstructure 304 removed and the detector elements in set 300 shown by dotted lines.
A still further embodiment of the present invention is a detector 320 shown in Figure 11. This detector has a substructure 322, which is, for example, a dielectric layer or plate 330 preferably made of cadmium telluride having a thickness of 0.1 to 10 microns. On the surface of the substructure 322 there are positioned a plurality of parallel detector elements such as 324, 326 and 328. These elements are arranged and connected in the manner shown in Figure 6 for elements 216-230. The elements 324-328 are made of the same material as elements 216-230. On the lower surface of plate 330 there is provided a layer 332 comprising a metal, such as aluminum. Layer 332 has a preferable thickness of .5 microns. The substructure 322 comprises the dielectric plate 330 and t he metal layer 332.

The dielectric plate 330 has a preferred thickness depending on the wavelength of the incident radiation. The preferred thickness is an odd multiple of a quarter wavelength of the received radiation . For a 12 micron radiation wavelength, a thickness of up to 10 microns is acceptable. Computer simulations indicate that the detector 320, with the illustrated dimensions, will have a radiation absorption of almost 100%.
The detector 320 operates in the same manner as the detector 214 described above. The metal layer 332 provides the lower reflective surface just as the interface between the layers 242 and 244 provides a reflective surface.
An infrared imager 330 in accordance with the present invention is shown in Figure 12. The imager 330 has an array of detectors, such as detector 332. Each of the detectors, such as 332, produces a pixel signal and the collection of pixel signals taken together form an image.
Each of the detectors in the imager 339 has a separate output line for that detector's pixel signal. The imager 330 can utilize for detectors 332 any one of the detectors described herein including detectors 214, 260, 296 or 320 described above.
Further embodiments of the present invention are illustrated in Figures lx through l5Ax-l5Hx.
An infrared detector 20x fabricated in accordance with the present invention is illustrated in a perspective view in Figure lx and as a section view in Figure 2x. These views are not necessarily to scale. A
detailed description for the steps of fabricating the detector 20x in accordance with the present invention is presented in Figures 3Ax-3Kx.
Detector 20x includes a substrate 22x, which is preferably made of sapphire but optionally may be cadmium telluride or silicon. The preferred thickness of substrate 22x is approximately 2 millimeters. On the upper surface of the substrate 22x there is provided a reflecting plane 24x, which is preferably a layer of aluminum having a thickness of approximately 500 - 1,000 angstroms. An epoxy bonding layer 23x binds the reflecting plane 24 to the substrate 22x.
On the surface of the plane 24x, there is provided a rectangular array of insulating blocks represented by the blocks 26Ax-26Ex. These blocks are preferably made of cadmium telluride and have approximate lateral dimensions of 4 microns by 1 micron and are approximately .3 microns thick. The center to center spacing of these blocks is approximately 8 microns. All of the dimensions set forth for the detector 20x are based upon a design having optimal response over a wavelength band of 8-12 microns for the incident infrared radiation.
The dimensions would be proportionately scaled for a different wavelength.
Immediately above each of the insulating blocks 26A-26E is a set of segments which are photosensitive to infrared radiation in the 8-12 micron range. These are photosensitive segments 28Ax-28Ex which have essentially the same lateral dimension as segments 26Ax-26Ex and a thickness of approximately .5 micron. These segments comprise mercury cadmium telluride (MCT) having an X ratio of approximately .15, corresponding to an operating temperature of 300°K. Mercury cadmium telluride is specified by fractional parts in which the fractional part of the cadmium is represented by the alloy ratio X and the fractional part of mercury is represented by 1-X.
The junction between the blocks 26Ax-26Ex and the corresponding segments 28Ax-28Ex is a blocking junction which prevents the transfer of all carriers, both majority and minority. This junction can be made by having a short transition between the photosensitive and non-photosensitive segments.
Positioned immediately above the photosensitive segments 28Ax-28Ex are bridging, non-photosensitive segments 30Ax-30Fx. the bridging segments 30Ax-30Fx comprise mercury cadmium telluride 'in which the X
alloy ratio is greater than or equal to .2. With this X ratio, the segments 30Ax-30Fx are not photosensitive to infrared radiation in the wavelength band of 8-12 microns at an operating temperature of approximately 300 K. Each of the segments 30Ax-30Fx bridges across a pair of the segments 2$Ax-28Ex. For example, segment 30Bx bridges across segments 28Ax and 28Bx. For each of the segments 30Ax-30Fx a preferred length is approximately 6 microns, a preferred width is 1 micron and a preferred thickness is .25 microns. The gap between the segments 30Ax-30Fx is approximately 2 microns.
Each of the segments 30Ax-30Fx is in contact with two of the segments 28Ax-28Ex. The junction between these segments is a selective blocking junction, a heterojunction. This junction blocks minority carriers but allows majority carriers to pass through. In the preferred embodiment, the minority carriers are P-type holes and the majority carriers are electrons. Therefore, the electrons pass freely through the junction while the P-type holes are blocked. One method for achieving this type of junction is to grade the transition of the alloy ratio between the two segments over a distance of approximately one thousand angstroms. Other techniques are well known in the art.
The combination of the blocks 26Ax-26Ex, segments 28Ax-28Ex and segments 30Ax-30Fx comprises a structure 32x which is repeated with identical structures 34x, 36x, 38x and 40x. Each of these structures is an elongate, segmented, electrically conductive member. The structures 32x-40x are parallel and are spaced apart by a distance which is less than the wavelength of the incident infrared radiation of interest. A
selected centerline to centerline spacing is 8 microns.
The combination of the segments 30Ax-30Fx and 28Ax-28Ex comprises an elongate, segmented, electrically conductive member which serves to capture incident infrared radiation, transfer the energy of the radiation to the photosensitive segments 28Ax-28Ex wherein a detection signal is generated and electrically conveyed through the member. The plurality of detection signals generated by the multiple photosensitive segments 28Ax-28Ex are summed along the segmented, electrically Conductive member.

The photosensitive segments 28Ax-28Ex, and the corresponding segments in the other structures, are preferably spaced apart from each other, in this embodiment, by less than the wavelength of the incident radiation. The reflecting plane 24x is spaced apart from the plane of the photosensitive segments 28Ax-28Ex by less than the wavelength of the incident radiation and preferably at one quarter wavelength from the photosensitive segments.
The segment 30Ax and the corresponding segments in the structures 34x, 36x, 38x and 40x are connected to a conducting member 42x which comprises the same material as the segment 30Ax and is preferably an extension thereof. A similar conducting member 44 is connected to segment 30Fx and the corresponding segments of the structures 34x, 36x, 38x and 40x. A conductive connection pad 46x, preferably an indium layer, is formed on the surface of the member 42x for providing electrical contact to the member 42x. A similar pad 48x is provided on the member 44x. The pads 46x and 48x are connected to a voltage bias, as described further below, and serve to collect detection signals which are generated within the structures 32x-40x.
When the material for the members 42x and 44x is n-type, indium is the preferable material for the pads 46x and 48x. But if the material for the numbers 42x and 44x is p-type, gold is the preferred material for the pads 46x and 48x.
Conductors 50x, 52x, 54x, 56x, 58x and 60x extend transverse to the structures 32x-40x and are positioned immediately above respective segments 30Ax, 30Bx, 30Cx, 30Dx, 30Ex and 30Fx, and the corresponding segments within the structures 34x-40x. Each of the conductors 50x-60x is electrically isolated from any other circuit element in the detector 20x. These conductors are preferably aluminum having a width of 2 microns and a thickness of .1 microns. The center to center spacing is approximately 8 microns. These conductors extend across the entire array made up of a plurality of the detectors 20x. These conductors function to couple a greater amount of the energy of the incident infrared radiation into the photosensitive segments, such as 28Ax-28Ex.

The detector 20x in Figure 1 is shown with substantial open spaces between the blocks and segments of the various structures. However, the open spaces shown in Figure 1 below the plane of the conductors 50x-60x is filled with a nonconductive material, such as zinc sulfide. This filler has been omitted from Figure lx to make possible a better view of the structure of the detector 20x. The filler material is shown in Figures 3Dx-3Kx.
Further referring to Figure lx, in operation, incident infrared radiation indicated by the arrows is received by the detector 20x. The infrared radiation is incident to the top surface of detector 20x as shown in Figure lx. The incident infrared radiation is substantially captured by the structural combination of the reflecting plane 24x, the non-photosensitive segments and the photosensitive segments, together with the conductors 50x-60x. The infrared energy is transferred to the photosensitive elements 28Ax-28Ex, and corresponding elements, with the structure as a whole providing a substantial impedance match to that of the incident field. The purpose of the non-photosensitive elements 30Ax-30Fx is to enhance the impedance matching and to provide a continuous DC current path to extract the photogenerated signal current.
The photo current detection signal produced by the photosensitive elements is extracted by the DC biased electrode pads 46 and 48.
The conductors 50x-60x can extend across the top of the detector 20x and are preferably spaced 8 microns apart. The conductors 50x-60x can extend across an array of the detectors 20x and serve to increase the collection of incident infrared radiation. Without the conductors 50x-60x the detector 20 collects approximately 50% of the incident infrared radiation across the wavelength band of 8-12 microns. But, with the inclusion of the conductors 50x-60x, the collection of incident radiation is increased to approximately 70% across the band of interest.
'fhe conductors 50x-60x reduce the polarization sensitivity of the detector 20x. These percentages have been determined through computer simulations for the described structure.

The detector 20x is further shown in a section view in Figure 2x.
This section view is taken along lines 2x-2x in Figure lx.
The detector 20x shown in Figures lx and 2x can comprise a single pixel within an image. A two-dimensional array of detectors 20x, as shown in Figure 13x, can be used to produce an infrared image.
A sequence of steps in accordance with the present invention for making the detector 20x is shown in Figures 3Ax-3Kx. As shown in Figure 3Ax, there is provided a substrate 70x, which is preferably cadmium zinc telluride having a crystal orientation of 2 degrees off <100>. The substrate 70x has a thickness of approximately 2 millimeters. On the surface of the substrate 70x there is grown a layer 72x of mercury cadmium telluride having an alloy ratio X=.2 and a thickness of approximately 2.0 micron. On the surface of the layer 72x there is grown a layer 74x of mercury cadmium telluride having an alloy ratio x=.15 with a thickness of approximately .5 micron. On the surface of the layer 74x there is provided a layer 76x of cadmium telluride. The layer 76x contains no mercury and therefore has an alloy ratio X = 1Ø
Layer 76x has a maximum preferred thickness of 1.0 micron. Each of these layers 72x, 74x and 76x is preferably formed through a process of epitaxial growth using Metal-Organic Chemical Uapor Deposition (MOCUD) or Molecular Beam Epitaxy (MBE).
Referring to Figure 3Bx, there is shown a step of precision thinning of CdTe layer 76x by either wet etching using dilute bromine methanol or by free methyl radical dry plasma etching. The preferred approach is dry plasma etching. The final thickness is determined by near infrared (0.8-2.5 micron) interference spectroscopy. The dry plasma etching can be carried out by use of a secondary afterglow reactor. In such a reactor there is a microwave discharge into a fluorine source gas. This produces a flowing afterglow. Methane is injected into the flowing afterglow to produce methyl radicals, which etch the cadmium telluride. Equipment for performing this process is produced by PlasmaQuest, Inc. of Richardson, Texas.

Referring to Figure 3Cx, a photoresist is applied to the surface of layer 76x and by use of photolithography techniques, the layers 76x and 74x are etched to produce insulating blocks 76Ax and 768x and photosensitive segments 74Ax and 748x. A selected photoresist is AZ5214 and a selected etchant is a free methyl radical, as noted above. The insulating blocks 76Ax and 76Bx correspond to the insulating blocks 26Ax-26Ex shown in Figure lx. Likewise, the photosensitive segments 74Ax-74Bx correspond to the photosensitive segments 28Ax-28Ex shown in Figure 1.
Referring now to Figure 3Dx, there is applied a zinc sulfide filler 7$x which fills the open areas between the etched stacks comprising the blocks 76Ax-768x and segments 74Ax-748x. The filler 78x extends up to the surface of blocks 76Ax and 768x. The filler 78x is preferably applied by a process of electron beam evaporation.
Referring now to Figure 3Ex, the reflective plane 24x is applied by electron beam evaporation of aluminum on the top surface of filler 78x and the surface of the blocks 76Ax and 768x. The aluminum layer comprising plane 24x preferably has a thickness of approximately 500-1,000 angstroms. This plane is reflective to infrared radiation.
Referring to Figure 3Fx, there is applied to the surface of the plane 24x an adhesive layer 80x which is preferably an epoxy, such as Epoxy Technology 301-2. The layer 80x has a thickness of approximately .5 micron. The epoxy layer 80x corresponds to the layer 23x shown in Figure lx. A superstrate, which is the substrate 22x shown in Figure 1x, is applied to the epoxy adhesive layer 80x so that the superstrate, substrate 22x, is bonded to the structure comprising the reflective plane 24x, filler 78x, blocks 76Ax, 76Bx, segments 74Ax, 748x, layer 72x and substrate 70x.
Referring now to Figure 36x, there is shown a further step in the manufacturing operation for producing the detector 20x., In the step shown in Figure 3Gx, the substrate 70x has been removed, preferably by an etching process. The orientation of the device has been changed by a 180° rotation. This is done in Figure 36x for the purpose of making the description of the process more understandable and to position the resulting device in the same orientation as detector 20x shown in Figure lx.
Substrate 70x can be removed by any one of a number of techniques including lapping or conventional etching. A preferred technique is etching by use of a technique described in an article entitled "Selective Etching of CdTe and ZnCdTe Substrate from HgCdTe Epilayers"
by G. M. Metze, d. L. Spears and N.P. Walsh of Lincoln Laboratory, MIT, published in the Proceedings of the 1985 Meeting of the IRIS Specialty Group on Infrared Detectors, held on 6-8 August, 1985 in Vol. 2, pp.
123-132 and dated August 7, 1985.
In Figure 3Hz, the layer 72x is etched by photolithographic processes to form bridging segments 72Ax, 72Bx and 72Cx. A selected photoresist is AZ5214 and a selected etchant is free methyl radical, as noted above. These segments correspond to the non-photosensitive segments 30Ax-30Fx shown in Figure lx.
Referring to Figure 3Ix, there is shown the step of depositing aluminum conductors 82x, 84x, and 86x which correspond to the conductors 50x-60x. This is done by use of conventional aluminum electron beam evaporation photolithography. Indium conductors for pads 46x and 48x are formed in a subsequent step.
Referring now to Figure 3Jx, a passivating layer 88x is applied over the surface of the conductors 82x-86x, the exposed surfaces of the segments 72Ax, 72Bx and 72Cx as well as the exposed surfaces of the segments 74Ax and 74Bx. Thus, the entirety of the exposed surface of the device is covered with the passivating layer 88x. The layer 88x is preferably zinc sulfide having a thickness of approximately 0.1 micron.
Referring now to Figure 3Kx, there is shown symbolically the step of attaching leads to the appropriate surface conductors of the device followed by packaging of the device. This step represents conventional lead attachment and packaging for semiconductor devices.
The infrared absorption characteristics for the photosensitive and non-photosensitive segments illustrated in Figures lx--3x are shown in Figure 4x. The term "photosensitive" is relative to the infrared band of interest. The chart in Figure 4x shows the infrared radiation absorption characteristics for mercury cadmium telluride (MCT). Curve 87x illustrates the absorption characteristics for MCT having alloy ratio of approximately X = .2. Note that with this ratio the MCT is absorptive for infrared radiation over essentially the range of 4-8 microns. Curve 89x illustrates the absorption of MCT having X = .15.
Note that the MCT with this ratio has high absorption over the range of 8-12 microns. Thus, MCT having X = .2 is substantially not photosensitive over the 8-12 micron range. These absorption curves are for MCT at room temperature. Curve 87x corresponds to the non-photosensitive bridging segments 30Ax-30Fx. The curve 89x represents the photosensitive characteristic of segments 2$Ax-28Ex.
Referring to Figure 5x, there is shown an infrared absorption curve 91x which represents the overall infrared absorption of the device 20x shown in Figures lx and 2x. As shown, this device has a very high percentage of absorption in the 8-12 micron range. The absorption reaches near 100% at one wavelength in the band of interest. It is this range of wavelengths that are of principle interest for thermal imaging.
This absorption curve has been determined by computer modeling for the detector 20x.
A further detector embodiment is a radiation detector 90x shown in Figure 6x. Detector 90x is a nonpolarized embodiment of the detector 20x shown in Figure lx. The detector 90x has similar base structures 22x, 23x, and 24x. However, the elongate conductors 50x-60x are replaced by additional elongate structures to make the detector 90x polarization insensitive. The detector 90x collects both horizontally and vertically polarized infrared radiation. The detector 90x has a plurality of elongate structures 92x, 93x, 94x, 95x and 96x. Structure 92x will be described in detail as representative of the other strucutres. The structure 92x includes rectangular insulating blocks 98Ax, 98Bx, 98Cx, 980x and 98Ex. These correspond to the blocks 26Ax-26Ex shown in Figure lx.
On the surface of the blocks 98Ax-98Ex there are provided similarly shaped photosensitive segments 100Ax, 100Bx, 100Cx, 1000x and 100Ex.
These correspond to the photosensitive segments 28Ax-28Ex shown in Figure lx. The detector 90x further includes x-shaped nonphotosensitive bridging segments 101x, 102x, 103x and 104x. These comprise the same material as the segments 30Ax-30Ex shown in Figure lx. The segment lOlx bridges across the segments 100Ax and 100Bx. The segments 102x, 103x and 104x likewise bridge across corresponding photoconductive segments.
Transverse to the structures 92x-96x, there are provided structures 105x, 106x, 107x and 108x. Structure 108x will be described in detail as representative of the remaining parallel structures 105x-107x.
Structure 108x includes insulating blocks 110Ax, 110Bx, 110Cx, 1100x, 110Ex and 110Fx. These correspond in size and material to the blocks 26Ax-26Ex shown in Figure lx.
On the surface of the blocks 110Ax-110Fx there are corresponding photosensitive segments 112Ax, 112Bx, 112Cx, 1120x, 112Ex and 112Fx.
These correspond in size and material to the photosensitive segments 28Ax-28Ex shown in Figure lx.
The nonphotosensitive bridging segment lOlx likewise bridges across photosensitive segments 112Ax and 112Bx. Corresponding bridging segments 113x, 114x, 115x and 116x in structure 108x bridge across corresponding elements 112Bx-112Fx.
The fabrication of the detector 90x is virtually the same as that described for the detector 20x but with appropriately altered masks to produce the additional elements and altered shapes.

", CA 02036874 1991-02-22 ,..
The detector 90x further includes conducting members 114x and 116x.
Member 114x comprises the same material as the bridging structures, such as 101x-104x and is an extension of these members at the edge of the detector 90x. On the surface of the member 114x there is provided a connecting pad 115x, which is preferably an indium contact. the conducting member 116x corresponds to the member 114x and has a similar conducting pad 117x thereon. The members 114x and 116x are supported by respective insulating members 118x and 119x. A QC bias for the detector 90x is applied between the pads 115x and 117x and the detected signal is likewise taken therefrom.
The detector 90x operates in the same manner as described above for the detector 20x but has enhanced performance due to the collection of transverse polarized incident radiation. This eliminates the need for the conductors 50x-60x shown in Figure lx.
A still further radiation detector 200x fabricated in accordance with the present invention is illustrated in Figures 7x, 8x, and 9x.
Detector 200x is likewise designed to capture infrared radiation in the wavelength range of 8 to 12 microns. Elements of the detector 200x are fabricated on a dielectric substrate 212x which is, for example, cadmium telluride just as for the substrate 22x shown in Figure lx. A plurality of electrically conductive, segmented, elongate members 214x, 216x, 218x, 220x, 222x, and 224x are fabricated on the surface of the substrate 212x. A representative size for each of these members is a width of 1.0 micron, a thickness of 0.5 micron, and a length of 50 microns.
At the opposite ends of the members 214x-220x, there are provided electrically conductive end members 226x and 228x which connect the ends of the members 214x-224x in parallel. The members 226x and 228x are preferably the same material as the nonphotosensitive but conductive segments such as 30Ax-30Fx shown in Figure lx. Each of the members 226x and 228x preferably has a width of approximately 2-5 microns and a thickness of approximately 0.5 micron. Member 226x has a conducting pad 227x, preferably an indium layer on the surface thereof. Member 228x similarly has a conducting pad 229x thereon.
Each of the members 214x-224x has a plurality of segments along its length. A preferred material for these elongate members is mercury cadmium telluride. The photosensitive characteristics of this material are determined by the ratios of the mercury and cadmium elements. Each of the members 214x-224x comprises mercury cadmium telluride, but alternating segments have different alloy ratios which change the photosensitive nature of the segments with respect to 'the wavelength of the incident radiation. For this embodiment, each photosensitive segment has a length of approximately 3 microns and each of the nonphotosensitive segments has a length of approximately 5 microns.
Member 224x is described in detail as representative of all the members 214x-224x. Member 224x comprises segments 224Ax-224Kx connected in series. Segment 224Ax is electrically connected to the conductive member 226x. Segment 224Kx is likewise connected to the electrically conductive member 228x.
Each of the members 214x-224x is made of mercury cadmium telluride, but the alloy ratio of the segments is different. For room temperature operation, segments 224Ax, 224Cx, 224Ex, 224Gx, and 224Ix have X greater than or preferably equal to .2, which is high enough to make the material transparent to infrared radiation over the wavelength band of interest, that is, 8-12 microns. For the segments 2248x, 224Dx, 224Fx, 224Hx and 224Jx, the value of the alloy ratio of X is approximately .15 to make the material absorbing, that is, photosensitive, over the 8-12 micron wavelength band. As a result, the segments 2248x, 224Dx, 224Fx, 224Hx and 224Jx are photosensitive, while the remaining segments are not photosensitive, over the wavelength band of interest. It can therefore be seen that the segments 224Ax, 224Cx, 224Ex, 224Gx, 224Ix and 224Kx correspond to the nonphotoconductive segments 30Ax-30Fx. shown in Figure lx. Likewise, the composition and function of the segments 2248x, 224Dx, 224Fx, 224Hx and 224Jx corresponds to the photosensitive segments 28Ax-28Ex shown in Figure lx.

The detector 200x is illustrated in a section view in Figure 8x. A
reflecting plane of layer 236x, preferably a layer of aluminum having a thickness of approximately 500-1,000 angstroms, is offset from the members 214x-224x by a distance of less than 0.5 microns, which is less than the wavelength of the radiation of interest. A layer 235x of zinc sulfide is located between the layer 236x and the substrate 212x. The preferred offset spacing is one quarter of the optical wavelength for the radiation at the center of the band of interest. The radiation wavelength within the detector material is substantially shorter than in free space.
An additional substrate section 238x may be provided below the reflecting layer 236x to enhance structural integrity . Substrate section 238x is bonded to the reflecting layer 235x by an epoxy layer 237x. The substrate 238x material can be the same as that of substrate 212x.
The reflecting plane, layer 236x, may optionally be a dielectric discontinuity between the substrates 212x and 238x, which discontinuity serves to reflect infrared radiation. Such a discontinuity can be provided by the adjacent substrate layers having different dielectric indices. In such a configuration, the aluminum layer 236x would not be needed.
In reference to Figures 7x-9x, a selected thickness for the members 214x-224x is 0.5 microns. A selected thickness for the lower substrate 238x is 2 millimeters.
Operation of the detector 200x is now described in reference to Figures 7x-9x. Infrared radiation is directed through a lens (shown in Figure 13x) to the surface of the detector 200x. The objective of the present invention is to capture a very high percentage of the incident radiation and transfer the energy of the radiation to the photosensitive detector elements. These include the photosensitive segments such as segments 2248x, 224Dx, etc. which produce a detection signal that is proportional to the amplitude of the incident radiation. Infrared radiation, for the preferred example, is captured by the combination of the structure comprising the reflective layer 236x and the structure of the elongate members 214x-224x which comprise both non-photosensitive and photosensitive segments.
The photosensitive segments 28Ax-28Ex (Figure lx) and segments 2248x, 224Dx, 224Fx, 224Hx and 224Jx (Figure 6x) have the following physical properties:
1. Non-zero conductivity, i.e., they conduct DC current.

2. Infrared radiation conductivity is finite, not zero.

3. Dielectric with preferred index n = 3.6-3.8.

4. Alloy ratio X is preferably .15 at room temperature.

5. Non-zero infrared radiation absorption.

The non-photosensitive segments 30Ax-30Fx (Figure lx) and the segments 224Ax, 224Cx, 224Ex, 224Gx, 224Ix and 224Kx (Figure 6x) have the following physical properties:

1. Non-zero conductivity, i.e., they conduct DC current.

2. Infrared radiation conductivity is zero.

3. Alloy ratio x is preferably greater than .2 at room temperature.
4. A dielectric having a preferred constant of n = 3.6.
5. No infrared absorption.
A still further embodiment fabricated in accordance with the present invention is illustrated in Figure 10x. A detector 300x has a structure as shown in Figure 9x with the addition of conductive lines 302x, 304x, 306x and 308x. The remaining structural elements are the same as those shown in Figure 9x and are notated by the same reference numerals. The conductive lines 302x, 304x, 306x and 308x are preferably aluminum and extend transversely across the elongate members 214x-224x.
The aluminum conductive lines 302x-308x are each independent and are not electrically connected to each other or to any other element of the detector 300x. The lines 202x-208x serve the function of enhancing the CA 02036874 1991-02-22 ,, collection of radiation by the detector 300x, just as the conductors 50x-60x shown in Figure 1.
For detectors 200x and 300x, a DC bias signal is applied between the electrically conductive members 226x and 228x. The photosensitive detector segments produce charge carriers and therefore change impedance upon receipt of the infrared radiation energy. These impedance changes modify the applied bias signal. Amplitude changes in the bias signal comprise the detected signal.
For detector 300x, the capture structure also includes the collection of conductive lines which includes lines 302x-308x. This structural combination can capture a very high percentage of the overall incident radiation in a given band. A graph of the capture of such radiation for the detector 300x is also shown in Figure 5x, as projected by theoretical modeling. The intercept percentage approaches 100% for the design wavelength.
The process for making the detectors 200x and 300x in accordance with the present invention is shown in Figures llAx-llLx. Referring to Figure llAx, there is shown a substrate 250x which is preferably cadmium telluride having a thickness of 2 millimeters. A mercury cadmium telluride layer 213x is grown by the process of MOC1~D or MBE on the surface of the substrate 250x. A layer 212x of cadmium telluride is grown on the surface of the layer 213x. The layer 213x has a preferable thickness of 2 microns and the layer 212 has a preferable thickness of 0.5 micron.
Referring now to Figure llBx, the cadmium telluride layer 212x is thinned by one of several processes. The material comprising layer 212x may be thinned by wet etching or dry etching. The wet etching can be done with dilute bromine methanol. The dry etching can be carried out by use of free methyl radicals, as described above in reference to Figure 3Bx. The wet etching is typically faster in removing material but the dry etching can be controlled for a more precise etching of the layer 212x. The ultimate desired thickness of layer 212x is approximately .3 microns. This can be measured by use of near infrared {0.8-2.5 microns) interference spectroscopy. The layer 2.2 is precisely thinned to adjust the distance between the photosensitive elements and the reflecting plane.
Referring to Figure llGx, an insulating layer 235x of zinc sulfide is deposited on the surface of the thinned cadmium telluride layer 212x.
A layer 236x of aluminum, serving as a reflective mirror, is deposited on the surface of the zinc sulfide layer 235x. The preferred thickness of the zinc sulfide layer 235x is 0.1 microns and the preferred thickness of the aluminum layer 236x is 500-1,000 angstroms. The zinc sulfide layer serves as an additional insulator to prevent the leakage of any currents from the photosensitive segments and conductors into the substrate. If the cadmium telluride layer 212x is of a sufficiently pure quality, it is a very good insulator and the supplemental layer 235x of zinc sulfide is therefore not required.
Referring to Figure llDx, a superstructure, which comprises the substrate 23$x, is bonded by use of an epoxy layer 237x to the surface of the aluminum layer 236x.
Referring to Figure llEx, the substrate 250x is removed from the 2p overall structure by one of several possible processes. This is the same as described above for the removal of substrate 70x in reference to Figures 3Fx and 3Gx. As noted therein, the layer 70x can be removed by mechanical lapping or etching by use of the described processes. In Figure llEx, the structure is rotated 180° to enhance the description of the subsequent steps and correspond to the orientation of the illustrated detectors 200x and 300x.
In Figure llEx, the layer 213x is thinned to a desired thickness of approximately 0.5 microns. The material can be removed by use of any one of several processes including mechanical lapping and etching, either wet or dry. A selected wet etchant is dilute bromine methanol.
Dry etching can be carried out as described above for layer 76x in ""," CA 02036874 1991-02-22 Figure 3Bx. The thickness of the layer 213x can be measured by the use of infrared interference spectroscopy.
Referring to Figure llFx, the layer 213x is etched in a photolithographic process utilizing AZ5214 as a selected photoresist and free methyl radicals, as described above, as an etchant. This process produces a plurality of photosensitive segments 213Ax, 213Bx and 213Cx.
These correspond to the photosensitive segments 2248x, 224Dx, etc. shown in Figure 9x. A perspective view of the structure produced in the step shown in Figure llFx is illustrated in Figure llGx.
Referring to Figure llHx, a layer of mercury cadmium telluride 240x having an alloy ratio x = .2 is grown by MOOD or MBE on the surface of the structure. The layer 240x covers the surface of the layer 212x as well as the detector segments 213Ax, 213Bx and 213Cx.
Referring to a top view of detector 200x or 300x in Figure llIx, the layer 240x is patterned and etched by photolithographic techniques to remove the material of layer 240x which is positioned between the previously formed rows of photosensitive segments, such as a row comprising segments 213Ax, 213Bx, and 213Cx. The photosensitive segments are outlined between the dashed lines.
Referring to Figure llJx, the layer 240x is further etched where it directly covers the previously formed photosensitive segments, such as 213Ax, 213Bx, and 213Cx. The remaining intermediate material comprises nonphotosensitive, conductive segments 240Ax and 240Bx. A top view of the structure shown in Figure llJx is illustrated in Figure llKx. There are now formed continuous strips which comprise alternate segments that are photosensitive with other segments that are conductive but non-photosensitive for the wavelength of infrared radiation of interest.
Referring to Figure lllx, there are shown the steps of adding a passivating layer 242x on a surface of the structure. This layer is preferably a material such as zinc sulfide. Finally, contacts are formed to the appropriate conductive portion of the detector as shown, for example, a contact 244x. Such contacts are preferably indium.
Finally, the overall device has lead attachment and is packaged in a conventional manner. If an individual detector is required, an IR
window is provided in the package. In a focal plane array, a plurality of devices are provided in an evacuated environment where they receive an infrared image.
A schematic circuit illustration of the operation of the detector 20x, and similarly detectors 90x, 200x and 300x, is shown in Figure 12.
In Figure 12x, the detector segments, the photosensitive segments in the infrared detectors, are represented as signal sources such as 28Ax-28Ex which are connected between the conductive pads 46x and 48x (Figure lx).
A bias signal is applied by a DC source 314x which is connected in series with a load resistor 321x between the conductive pads 46x and 48x.
When the detectors, which include the segments 28Ax and 28Ex, receive the energy of the captured infrared radiation, this energy is translated into an impedance variation that alters the amplitude of the DC bias signal and produces a detection signal between output terminals 320x and 322x. This is the output signal for a single pixel element in an array of such circuits.
A detector array 324x is illustrated in Figure 13x. The array 324x comprises a plurality of single pixel detectors, as represented by detectors 326x. The detectors 326x can be any of the detectors 20x, 90x, 200x or 300x as shown in Figures lx, 6x, 7x, 9x and 10x. All of the detectors within the array 324x can have a common bias line but each must have a separate output signal line, with lines 328x for the detectors 326x. Each of the pixel detectors within the array 324x have separate signal lines.
The array 324x is a part of an infrared imaging system 325x. The collection of all of the pixel detectors within the array 324x can produce an image as a result of the focusing of infrared radiation onto the surface of the array 324x by a lens 330x. The image is in the signal at the output signal lines, such as 328x. Further, all of the individual pixel detectors, such as 326x, may be fabricated on a single, common substrate, such as substrate 22x shown in Figure lx.
Referring to Figure 14x, there is illustrated an infrared detector 400x which has a plurality of photosensitive strips 402x, 404x, 406x, 408x, 410x and 412x positioned on a substrate layer 418x. The strips 402x-412x comprise mercury cadmium telluride (MCT) having an x ratio of approximately .15, corresponding to an operating temperature of 300°K.
The layer 418x is preferably cadmium telluride.
The strips 402x-412x have a thickness of approximately .5 micron, a width of 1 micron and a length of 50 microns. The layer 418x is preferably approximately .3 microns thick.
At opposite ends of the strips 402x-412x are conductive members 420x and 422x which are preferably mercury cadmium telluride having an x alloy ratio equal to or greater than .2. With this ratio the members 420x and 422x are electrically conductive, but not photosensitive, in the 8-12 micron band at 300 K. Indium contacts 424x and 426x are positioned respectively above the conductive members 420x and 422x and are in ohmic contact with members 420x and 422x.
The layer 418x is positioned on a layer 430x which comprises zinc sulphide having a thickness of approximately 0.1 micron.
An aluminum layer 432x is deposited between layer 430x and an epoxy bonding layer 434x. Layer 432x is an infrared reflecting plane and has a thickness of approximately 500-1,000 angstroms.
A substrate 436x, preferably sapphire, has a thickness of approximately 2 millimeters. The epoxy layer 434x bonds the aluminum layer 432 to the substrate 436x.
Referring to Figures l5Ax-l5Hx, there is shown a process for making the detector 400x shown in Figure 14x. This is very much like the fabrication process described in Figures 3Ax-3Kx. In Figure lSAx, a layer 442x of mercury cadmium telluride having x=.15 is grown on the surface of a dielectric plate 440x of cadmium zinc telluride. The layer 442x will be etched, as described below, to become the strips 402x-412x.
A layer 418x of cadmium telluride is grown on the surface of layer 442x.
In Figure l5Bx the layer 418x is thinned in the same manner as described above for layer 76x in Figure 3Bx.
Referring to Figure l5Cx there is grown the layer 430x on the surface of layer 418x. The aluminum layer 432x is formed on the surface of layer 430x as described above for plane 24x shown in Figure 3Ex.
In Figure l5Dx, an epoxy layer 434x is applied to the exposed surface of the aluminum layer 432x for bonding substrate 436x to the remainder of the structure.
Referring to Figure l5Ex, the plate 440x has been removed in the same manner as the substrate 70x shown in Figure 3Fx. The structure has been inverted in Figure l5Ex from that shown in Figure l5Dx.
The layer 442x is thinned as shown in Figure l5Fx, by a process of methyl radical dry etching to gain the desired thickness for the strips 402x-412x.
In Figure l5Gx, a resist 450x, as described above, is applied to the layer 442x and patterned for selectively etching layer 442x to produce the strips 402x-412x. The resist 450x is then removed.
In Figure l5Hx, a passivating layer 454x is applied on the exposed surface of the detector structure for protection. The detector is completed by conventional processes steps of indium contacting, lead attachment and packaging.
The detector 400x, shown in Figure 14x, in comparison to the detector 200x shown in Figure 9x, may have lesser detectivity than detector 200x for similar size and geometry; but can be more easily fabricated due to lesser complexity and fewer manufacturing steps.
Otherwise, the functionability is essentially the same.
Infrared detectors fabricated as described herein have substantially increased detectivity over prior designs. This increased detectivity can be traded off to reduce the need for cooling equipment, while maintaining standard sensitivity, or, by using cooling equipment, a detector made in accordance with the present invention can have substantially enhanced sensitivity.
The photosensitive segments described herein for the disclosed embodiments are fabricated of mercury-cadmium-telluride having a specified alloy ratio. This material is photoconductive, that is, a bandgap material which produces charge carriers in response to incident radiation. The photosensitive segments may also be made of a photovoltaic structure, such as a mercury-cadmium-telluride p-n junction, which produces a voltage in response to the incident radiation.
In summary, the present invention includes a method for making infrared detectors. The detectors have a plurality of electrically conductive elongate members comprising photosensitive segments separated by, but contacting, non-photosensitive conductive segments. In a further aspect, electrically isolated, parallel conductive lines are positioned immediately above the detector surface, and spaced apart by less than the radiation bandwidth, for enhancing the capture of infrared radiation.
Although several embodiments of the invention have been illustrated in the accompanying drawings and described in the foregoing detailed description, it will be understood that the invention is not limited to the embodiments disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the scope of the invention.

Claims (25)

1. A detector for radiation in the infrared region and regions of shorter wavelengths, comprising:
an array of bandgap photodetector elements located in a plane and spaced apart by less than approximately the wavelength of said radiation, a plurality of periodic radiation collecting structures connected respectively to said photodetector elements, and a radiation reflecting plane offset from the plane of said photodetector elements by less than the wavelength of said radiation, wherein the combination of said radiation collecting structures and said reflecting plane captures and conveys said radiation to said detector elements for generating a detection signal therein.

2. A detector as recited in Claim 1 wherein each of the linear dimensions of said detector elements is less than the wavelength of said radiation.

3. A detector as recited in Claim 1 including conductive lines connected to said radiation collecting structures for transmitting said detection signal.

4. A detector as recited in Claim 1 wherein said photodetector elements comprise mercury cadmium telluride.

5. A detector as recited in Claim 1 wherein said photodetector elements comprise indium antimonide.

6. A detector as recited in Claim 1 wherein said photodetector elements comprise a semiconductor superlattice.

7. A detector as recited in Claim 1 wherein said radiation reflecting plane is a metal layer.

8. A detector as recited in Claim 1 wherein said radiation reflecting plane is a multilayer dielectric.

9. A detector as recited in Claim 1 wherein each of said radiation collecting structures comprise a dipole antenna.

10. A detector as recited in Claim 1 including blocking contacts for connecting said photodetector elements to said radiation collection structures.

11. A detector for radiation, comprising:
a periodic pattern of parallel, elongate, photoconductive or photovoltaic, bandgap detector elements spaced apart at a period which is less than approximately the wavelength of said radiation, said pattern of detector elements having a given radiation impedance, substructure means for supporting said detector elements and for providing impedance matching between the radiation impedance of said detector elements and free space radiation impedance, and means for electrically connecting said detector elements for producing a detection signal when said detector is exposed to said radiation.

12. A detector as recited in Claim 11 wherein each of the linear dimensions of said detector elements is less than the wavelength of said radiation.

13. A detector as recited in Claim 11 wherein said substructure means comprises a dielectric plate having said detector elements on one surface thereof and a metal layer on an opposite surface thereof.

14. A detector as recited in Claim 11 including a superstructure means adjacent said pattern of detector elements and opposite said substructure means for providing impedance matching between said detector elements and free space impedance.

15. A detector as recited in Claim 11 wherein said detector elements comprise mercury cadmium telluride.

16. A detector as recited in Claim 11 wherein said substructure means comprises a first and a second layer having different indices of refraction.

17. A detector as recited in Claim 16 wherein said first layer is indium antimonide and said second layer is cadmium telluride.

18. A detector as recited in Claim 11 including a second periodic pattern of parallel, elongate, photoconductive or photovoltaic, bandgap detector elements spaced apart at a period which is equal to or less than the wavelength of said radiation, said second pattern of detector elements being orthogonal to said first pattern of detector elements and located in a plane offset from and parallel to said first pattern of detector elements and means for electrically connecting said second pattern of detector elements for producing a detection signal when said detector is exposed to said radiation.

19. A method for fabricating a device for detecting radiation in the infrared region and regions of shorter wavelengths, comprising the steps of:
forming a plurality of groups of photosensitive segments in a planar array, said photosensitive segments being sensitive to said radiation and said photosensitive segments having a thickness less than the wavelength of said radiation, said photosensitive segments comprising a plurality of groups, each group having a plurality of said photosensitive segments positioned in an elongate pattern, the photosensitive segments in each group being offset from each other by a distance which is less than approximately said wavelength, and the lateral dimensions of each said photosensitive segments being less than said wavelength, forming a plurality of electrically conductive segments for interconnecting adjacent ones of said photosensitive segments in each of said groups, said electrically conductive segments not being photosensitive to said radiation, whereby each group of said photosensitive segments together with the corresponding conductive segments is electrically conductive along its length, forming a plane which is reflective to said radiation, said planar array of photosensitive segments and said reflective plane being offset from each other by less than said wavelength, and electrically connecting a plurality of said groups of photosensitive segments in parallel to provide a conduction path for detection signals produced by said photosensitive segments in response to said infrared radiation.

20. A method for fabricating a device for detecting infrared radiation as recited in Claim 19 wherein said electrically conductive segments are positioned in a plane parallel to but offset from the plane of said photosensitive segments.

21. A method for fabricating a device for detecting infrared radiation as recited in Claim 19 wherein said electrically conductive segments are positioned in a plane coplanar with the plane of said photosensitive segments.

22. A method for fabricating a device for detecting infrared radiation as recited in Claim 19 including the step of forming a heterojunction at each interface between said photosensitive segments and said electrically conductive segments.

23. A method for fabricating a device for detecting infrared radiation as recited in Claim 19 including the step of fabricating insulating material for separating said reflective plane from said photosensitive segments, said insulating material in contact with said photosensitive segments.

24. A method for fabricating a device for detecting infrared radiation as recited in Claim 23, further including the step of forming a blocking junction at each interface between said photosensitive segments and said insulating material.

25. A method for fabricating a device for detecting infrared radiation as recited in Claim 19 wherein the step of forming a reflecting plane comprises forming an aluminum layer.