GB2107930A

GB2107930A - Photoconductive strip detectors

Info

Publication number: GB2107930A
Application number: GB08229682A
Authority: GB
Inventors: Charles Thomas Elliott; Anthony Michael White
Original assignee: UK Secretary of State for Defence
Current assignee: UK Secretary of State for Defence
Priority date: 1981-10-21
Filing date: 1982-10-18
Publication date: 1983-05-05
Also published as: GB2107930B

Abstract

In a photoconductive strip detector comprising a strip (1) of photoconductive material having bias contacts (3 and 5) and between these a read-out region (9), to improve detector resolution the strip (1) is adapted so that when biassed, the bias field is of modified distribution along the length of the strip (1) and affords a significant reduction in deviation of the photocarrier drift velocity along the strip (1). This may be achieved by grading the width or thickness of the strip (1), by grading the bulk donor concentration of the strip (1), by grading the surface treatment applied to the strip material, or by using ancillary electrodes either side of an exposed length of the strip (1). <IMAGE>

Description

SPECIFICATION Photoconductive strip detectors Technical field This invention concerns photoconductive strip detectors; detectors of photoconductive material, the material being formed in a strip and having two bias contacts and a read-out region defined between these contacts. These detectors are used in scanning thermal imaging systems.

Background art A scanning thermal imaging system incorporating a photoconductive strip detector is described in UK Patent No 1,488,258 (US Patent No 3,995,159). The strip detector, as described there, comprises a parallel sided strip filament of photoconductive material, n-type Cadmium Mercury Telluride (CMT), having bias contacts and between these a read-out region. in one case the read-out region is provided by a diode contact to the strip, located near to one of the bias contacts. In the system, this diode is connected to a diode bias circuit. Flow of minority carriers through this circuit produces a response signal that can then be used to reconstruct the thermal image. In another case the read-out region is defined by a pair of spaced conductor contacts, one of which may also serve as a bias contact. A response signal is developed across these read-out contacts.In use, the strip detector is located in the scanned image plane of the system and a thermal image is scanned along its length at a constant velocity. Bias is applied to the detector so that the photocarriers generated in the strip, in response to the incident thermal radiation, are driven towards the read-out region and the negative bias contact. Ideally, as is in practice the case in materials having the property of ordinary carrier lifetime, the bias field is uniform along the length of the rectilinear detector, and the carrier mobilities are constant. The photocarriers thus drift towards the read-out region at a constant velocity, an ambipolar drift velocity.The bias magnitude is carefully chosen so that the scan velocity and drift velocity are accurately matched, and there is thus throughout the scan cycle unique correspondence between the photocarrier density profile and the image intensity profile, constructive integration taking place. However, the spatial resolution that can be attained in the reconstructed image depends on several factors including carrier thermal diffusion, strip width, and upon the diode bias circuit integration time constant or the spacing of the read-out conductors. These are usually chosen to give good resolution. Techniques have also been developed to limit the diffusive spread of photocarriers, improving detector resolution-eg the strip may be slotted to form a meander path (see GB 2,019,649, US 4,258,254).

Further advances have since been made in the fabrication of cadmium mercury telluride materials. There is now the potential for developing detectors with much improved detectivity (D*) and responsivity (R). In fact detectors with cut-off wavelength of circa 11 ym having a detectivty D* (5000K) of circa 2.0x1011 cm Hz112 m#1, (in a background photon flux of 2.4x 10'6cm2s~' and at a scan speed of 1.1 x 104 cm s-:) and extraordinarily high carrier lifetime (Tm)-3.6 tjs can now be produced However in detectors, 0.7 mm long of this material, the resolution, which has been determined experimentally from line spread function measurement, appears to be anomalously poor.This resolution cannot be predicted accurately in terms of thermal diffusion alone, it is considerably worse and in general it is unacceptably poor.

Disclosure of the invention The invention is intended to provide a remedy; to improve the resolution of photoconductive strip detectors of relatively short active length, of material having the property of long carrier life-time.

It has now been found, following extensive investigation, that the resolution degradation can be attributed to the effects of progressive accumulation of carriers due to background illumination. In rectilinear strip detectors of long lifetime material the bias field and carrier mobilities are therefore nonuniform over a considerable proportion of the strip length, with the result that the drift volocity is far from constant (at least over a region of length-2vz,) and accurate velocity matching is no longer possible. [If the strip is of relatively long length (L 2VTm) variation of velocity over the initial portion of the strip is of no consequence, as such carriers recombine before detection].

According to the invention there is provided a photoconductive strip detector comprising:~ a strip of photoconductive material, the material having as a property thereof a long carrier lifetime; the strip having at least two bias contacts and a read-out region located between these contacts; the region of strip material between a first one of the bias contacts and the read-out region being of a length that is relatively short; wherein the detector is adapted to develop, on application of bias, a bias field of magnitude dependent on distance from the first bias contact and such as to produce a significantly reduced deviation in value of the corresponding carrier drift velocity in a major portion of the region of the strip, thus giving an improvement in detector resolution.

Conveniently the detector may be so adapted, in that at least one of the lateral dimensions of the strip, the width or the thickness thereof, is of different value at points along the length of the strip.

Preferably this lateral dimension is varied gradually and continuously, the width, the thickness, or both, being tapered along substantially the entire length of the strip between the first bias contact and the read-out. Ideally, this taper may be profiled so as to modify the bias field in manner producing negligible variation of the carrier drift velocity.

Alternatively the detector may be adapted by tailoring the bulk donor concentration along the strip material. For example, dopant species may be introduced in graded amount along the length of the strip by a controlled ion implant technique. This has the advantage of conserving detector area, but is technically difficult to realise in practice.

Alternatively the strip may be given a graded surface treatment.

Alternatively, insulated tapered metal conductors or insulated resistive contacts could be located over the surface of the strip material, either side of an exposed length of the material. The surface charge profile can then be changed on charging the conductors or contacts, changing the effective bias field as desired.

Brief introduction of the drawings In the drawings accompanying this specification:~ Figure 1 is a plan view of a strip detector of ordinary rectilinear geometry; Figure 2 is a graph depicting by plots and curves the variation of excess carrier density and of drift velocity with distance measured from detector end, calculated for the detector shown in the preceding figure; Figure 3 is a plan view outline of a model strip detector, the detector having a tapered width; Figure 4 is a graph depicting the variation of carrier density and drift velocity calculated for the model detector shown in preceding figure 3; and Figure 5 is a plan view of a strip detector, a detector similar to the detector of figure 3, but one having a preferred tapered profile.

Description of examples Embodiments of the invention will now be described, by way of example only, with reference to the drawings.

There is shown in Figure 1 a strip detector of ordinary rectilinear geometry. It is formed of a strip 1 of cadmium mercury telluride material, of thickness (t)-8 #m. The material is n-type and is characterised by an extrinsic electron (majority carrier) density nO at 800K of typical value 4x 1014 cm-3. The hole (minority carrier) mobility yh is of value 500 cm2 V- s- for the material used. The strip 1 is provided with bias contacts 3 and 5, one at each end of the strip and between these contacts 3 and 5 there is a conductor contact 7 spaced a short distance away from the second bias contact 5.

Together, the conductor contact 7, the second bias contact 5 and the photoconductive material inbetween form a read-out region 9 of the detector. When installed in a thermal imaging system, bias is applied to the bias contacts, positive to the first bias contact 3, and negative to the second bias contact 5, as shown by the + symbols in Figure 1.

The plots for Figure 2 have been derived numerically using, as a basis, the continuity equation and the identities appearing below, with the boundary condition that the excess carrier density due to background flux (nb) is zero at the positive bias contact (x=O). The one-dimensional continuity equation ##/t -R(nb)- (nb)~E(nb)#nb/#x=0 Equation 1 neglecting diffusion effects and assuming that terms in the field gradient and mobility gradient are small.

In this equation:~ is is the photo-electric conversion efficiency; is is the background flux of radiation incident on the detector; is isthe thickness of the detector; R(nb): is the Auger recombination rate; ,u(nb): is the ambipolar mobility of the carriers; E(nb): is the bias electric field; is isthe electron density (nanO+nb); is isthe excess carrier density due to background flux; and is isthe distance measured from one end, the positive bias end of the detector; and in the equations that follow:- To is the excess carrier lifetime measured in zero background; Tm is the lifetime measured at the negative bias end of a very long filament, in situ; Eo is the bias field measured at the positive bias contact (x=0); and, is is an effective bulk carrier density which takes account of the surface shunt conductance, and is usually2n0.

The Auger recombination rate R(nb), the ambipolar mobility X{nb) and the bias field E(nb) all vary with the excess carrier density nb:~ nb R(nb)~ (1 +2nb/n0) Equation 2; #o hn0 (nb)~ Equation 3; n0+nb and, nee E(nb)~E0 Equation 4.

nb+neff The detector is used with a cold shield of aperture F/2.5 and for this the background radiative flux # is taken as 2.7 x1016cm-2s-1 for a background temperature T=2950K and radiation wavelength #c=1 1.7 #m. Typical values have also been taken for the quantum efficiency #=0.7, and the bias field E0=30 Vcm-'.

In Figure 2 the excess carrier density nb(x) is shown in broken outline, and the drift velocity v(x) in continuous outline, for three different values of the carrier lifetime #0=2,4 and 6 ys.

The drift velocity V is given by the identity: r(n)=y(n) ~ E(n) Equation 5.

The in situ lifetime Too given by: #oo=(#o . n02)/(n0+nb)(n0+2nb) Equation 6 has the values 1.5, 2.5, and 3.4 S corresponding to the three values of carrier lifetime #o=2,4 and 6 FLS.

From Figure 2 it can be seen that there is considerable variation in drift velocity, and this is most pronounced for the device with the longest carrier lifetime (To=6,as). If a long strip (~1200 m) is chosen, it can be seen that the variation of drift velocity is small over the end region 750-1200 form.

But for shorter strips eg 2vTm long, the variation of drift velocity is a problem.

Significant improvement, however, can be achieved by changing the bias field profile. Thus in Figure 3 a detector of modified geometry is shown. This detector, also of n-type CMT material and of thickness 8 m, is in the shape of a truncated circle segment, having a width at its widest point of 62.5 m, at which end a positive bias contact 3 is formed. The strip 1 thus has tapered sides and these subtend an angle of approximately 1.50 at the circle centre, a distance R=2,400 #m from the positive bias contact.

For this the continuity equation becomes:~

The differential equation 7 has been solved numerically with the same parameters as given earlier, except that the bias field Eo has been reduced to match a scan speed of 1.1 x 104 cm Eo=22.6 V cm-'.

As can be seen from Figure 4 the spread of drift velocity (ie deviation) is very much less than found for the device of rectilinear geometry (see Figure 2, of different scale), though for the particularly simple form of taper chosen there is some undercorrection near the positive bias contact, and some overcorrection at points of the detector furthest from this contact. Nevertheless, if the detector length is limited to 400 m, the ambipolar velocity at no point along the length, for any of the three cases considered, departs by more than 5% from a value of 1.13 x 104 cm s-1 .A rough estimate of the pulse broadening due to velocity mismatch indicates that this is less than 8 m for the worst case (#o=2 s) and compares very favourably with the pulse broadening estimated for the device of rectilinear geometry~41 Mm. The resolution therefore is much improved.

Further impovement may also be achieved by modifying the taper profile. Thus as shown in Figure 5 the width of the detector is profiled to give a constant drift velocity along the entire length of the strip. This ideal profile has been calculated from the continuity equation with the constraint that the velocity is constant:~ v=y(x)E(x)=vc (constant). Equation 9 The variation of width W is given by the equation

with:~

where:~

This profile may be produced using a photomask made by computer aided design, shaping the photoconductive material photolithographically using a chemical etchant.

For other recombination processes there would be different relationships for nine, thus for example for a device in which radiative recombination predominates the relationship for ndn, would be: nb =α(1 -exp(-X/Vc#0). equation 14 nO For the general case including Auger and radiative recombination or indeed any other recombination process for which the recombination process varies with carrier concentration in the same way, there are corresponding formulae derivative from the basic equations given above.

Where the taper needed is large, minor corrections to the parameters f, and O may be made to account for incomplete interception of the total photon flux by the narrower parts of the detector.

These may be calculated in a straight forward manner from the last equation given.

Claims

1. A photoconductive strip detector comprising:~ a strip of photoconductive material, the material having as a property thereof a long carrier lifetime; the strip having at least two bias contacts and a read-out region located between these contacts; the region of strip material between a first one of the bias contacts and the read-out region being of a length that is relatively short; wherein the detector is adapted to develop, on application of bias, a bias field of magnitude dependent on distance from the first bias contact and such as to produce a significantly reduced deviation in value of the corresponding carrier drift velocity in a major portion of the region of the strip, thus giving an improvement in detector resolution.

2. A detector as claimed in claim 1 wherein at least one of the lateral dimensions of the strip, the width or the thickness thereof, is of different value at points along the length of the strip.

3. A detector as claimed in claim 2 wherein the width of the strip, the thickness of the strip, or both, are tapered along substantially the entire length of the strip between the first bias contact and the read-out.

4. A detector is claimed in claim 3 wherein the taper is profiled to give a substantially constant drift velocity throughout the strip.

5. A detector as claimed in any one of the preceding claims wherein the bulk donor concentration of the strip material is graded along the length of the strip.

6. A detector as claimed in any one of the preceding claims, wherein the strip has been given a graded surface treatment.

7. A detector as claimed in any one of the preceding claims wherein insulated tapered metal conductors or insulated resistive contacts are provided over the surface of the strip material, each side of an exposed length of the material.

8. A detector constructed, adapted, and arranged to operate substantially as described hereinbefore with reference to and as shown in either one of Figures 3 and 5 of the accompanying drawings.