CA2235922A1

CA2235922A1 - Polarization sensitive photodetectors and detector arrays

Info

Publication number: CA2235922A1
Application number: CA002235922A
Authority: CA
Inventors: M. Selim Unlu; Bora Onat
Original assignee: Individual
Current assignee: Boston University
Priority date: 1995-10-26
Filing date: 1996-10-25
Publication date: 1997-05-01
Also published as: WO1997015812A1; JPH11514085A; AU7477396A

Abstract

A device for detecting polarization of light comprising a first photodetector tuned to absorb TE polarization, a second photodetector tuned to absorb TM
polarization, and a circuit for comparing an output from the first and second photodetector for generating a polarization output. The first photodetector comprises a first reflector, an absorption layer on top of the first reflector, and a second reflector on top of the absorption layer. The second photodetector comprises a third reflector on top of an absorption layer, and a fourth reflector disposed under absorption layer.

Description

POLARIZATION SENSITIVE
P~IOTODETECTORS AND DETECTOR ARRAYS
FIELD OF THE INVENTION
This invention relates to photodetection. More specifically, 5 this invention relates to detection of polarization of radiation.
BACKGROUND OF THE INVENTION
Present photodetectors detect the presence of light. Presence of light is usefu], but additional information may be desired. It may be desirable to detect the polarization of the light. Polarization sensing has 10 various applications ranging from magneto-optical data storage to im~gin,~
In im~gin~, the polarized light sensitivity is expected to underlie a visual quality similar to color vision that may permit the detection of objects that are blen-lç~l in the background. In magneto-optical (M-O) drives, the content of the stored data is coded as a change in the polarization of light. The 15 conventional M-O reading head configuration employs polatizing beam splitters and separate dedicated detectors for the two polarization components. The use of buLk discrete optical components of this t,ype requires individual ~1i nment in three spatial and three angular coordinates with narrow tolerances resulting in increasing cost and limite~l range of 20 applications.
Other devices have been ~lle~ ed to make the p~k~in~ of the optical system simpler and tolerant to alignment variations. For example, a c~c~(le~l hologra;phic sensor element has been proposed but is expected to be highly sensitive to ~lignment and wavelength variations as well. Each of 25 these devices also continues to suffer from the disadvantage of heavy and bulky optical elernents which limit access time of the data reading head.
Other disadvantages with t,he prior systems also exist.
SUMMARY OF THE INVENTION
~ Accordingly, a need has arisen for a device which provides for 30 intensity and polarization sensitivity and detection into a single device without requiring the use of beam splitters.

W O 97/lS812 PCT~US96/17188 Accordingly, it is an object of the ~leselll invention to meet the above-described need.
It is an object of the present invention to provide a device for detecting the absolute value or the variation in polarization for linearly 5 polarized light. It is also an object to provide a device in which bo~
intensity and polarization of light is ~imnlt~neously detected.
It is another object of the present invention to provide an integrated polarization selectivity and detection device in a single semiconductor device structure.
It is also an object of the present invention to provide a photodetector structure which is sensitive to the polarization of the incident radiation, the sensitivity being controlled by recessing the top surface of the detector allowing for the monolithic fabrication of detector arrays with varying sensitivities to TE and TM polarizations.
It is further an object of the present invention to provide a detector array comprising at least two photodetectors which can be used to measure the polarization of the incident light.
It is further an object of the present invention to provide two vertically integrated photodetectors which are sensitive to the polarization 20 of light.
It is another object of the present invention to provide a device for detecting absolute polarization and changes in polarization.
Accordingly, an embodiment of the present invention comprises a device for detecting polarization of light comprising a first 25 photodetector tuned to absorb TE pol~ri7~tion, a second photodetector tuned to absorb TM polarization, and a circuit for comparing an output from the f;rst and second photodetector for generating a polarization output.
According to another embodiment of the present invention, a device for detecting polarization of light comprises a substrate, a first 30 photodetector tuned to absorb TE polarization, a second photodetector -wo 97/15812 PCT~US96/17188 disposed on ~e substrate and tuned to absorb TM polarization, and a circuit for comparing an output from the first and second photodetector for generating a pola~ization output.
According to yet another embodiment of the present invention, S a method for ~letechn~ pol~ri7~tion of light comprises the steps of absorbing TE polarization using a first photodetector, absorbing TM polarization using a second photodetector, and comparing an output from the first and second photodetector for generating a polarization output.
According to another embodiment of the present invention, a 10 monolithic VLSI device for detecting polarization of light comprises a substrate, a first photodetector tuned to absorb TE polarization and formed on the ~,ul~ , a second photo~let~Pctor disposed on the substrate and tuned to absorb TM polarization and formed on the substrate, and a circuit for co.~ an outplut from the first and second photodetector for generating 15 a pol~ ion output and formed on the substrate.
According to yet another embodiment of the present invention, a photodetector system for detecting the positioning of a light source and the polarization of the light generated by the light source comprises a plurality of pairs of photodetectors for detecting light from the light source, each pair 20 comprising a first photodetector tuned to absorb TE polarization, and a second photodetector disposed on the substrate and tuned to absorb TM
polarization and formed on the substrate. The photodetector system also compri~çs a circuit for cc""l.~. ;"~ an output from at least one of the first and second photodetectors for generating a polarization output and a circuit for 25 cc,. ~ an output from each of the first photodetectors from the plurality of pairs with a predet~"~ ed output to d~te~ e whether the light source is in a pred~ ed desired position.
According to a fi~ther embodiment of the present invention, a m~gnPto-optical drive device compri~es a light source for generating a light 30 output directed onto an optical storage medium, and a mech~ni.cm for CA 0223~922 1998-04-27 W O 97tl5812 PCTAUS96/17188 receiving light reflected from the optical storage medium and detecting polarization of the light. In one embodirnent, the mech~ni.cm comprises a first photodetector tuned to absorb TE polarization, and a second photodetector tuned to absorb TM polarization. The optical drive system 5 also comprises a circuit for Co~ g an output from the first and second photodetector for generating an output indicating a value stored on the optical storage medium.
According to yet another embodiment of the present invention, a CCD array device for detecting polarization of an image comprises a 10 plurality of detectors arranged in rows and columns, each detector comprising a first photodetector tuned to absorb TE polarization, and a second photodetector tuned to absorb TM polarization. The CCD array device also cornprises a processor connected to the plurality of detectors for generating an array of polarization values for an image received by the 1 5 detectors.
Other objects and advantages of the present invention may be appreciated upon review of the specification and drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 depicts a schematic of a polarization sensitive detector 20 array according to an embodiment of the present invention.
Fig. 2 is a graph depicting reflectivity versus incidence angle for both TE and TM polarized light according to a typical dielectric-air incidence boundary device.
Fig. 3 is a graph depicting detector response versus surface 25 recess for both TE and TM polarized light according to one embodiment of the present invention.
Fig. 4 depicts a circuit for polarization sensitive detection according to an embodiment of the present invention.
Fig. 5 depicts a schematic of a polarization sensitive detector 30 according to an embodiment of the present invention.
-W O 97/15812 PCT~US96/17188 .

Fig. 6 is a graph depicting reflectivity versus incidence angle for both TE and TM polarized light according to an embodiment of the present invention.
Fig. 7A depicts a sçllem~ic of a pol~ri7~ion sensitive detector S according to an embodiment of the present invention.
Fig. 7B is a graph depicting reflectivity versus incidence angle for both TE and TM polarized light according to an embodiment of the present invention Fig. 8 is a graph illustrating detector response versus surface 10 recess for an embodiment of the present invention.
Fig. 9A depicts a srhem~tic of a po1~ti7~1ion sensitive detector according to an embodiment of the present invention.
Fig. 9B is a graph depicting reflectivity versus incidence angle for both TE and TM polarized light according to an embodiment of the 15 present invention.
Fig. 10 is a graph illu~l~a~ g detector response versus surface recess for an embodiment of the present invention.
Fig. llA depicts a schematic of a polarization sensitive detector according to an embodiment of the present invention.
Fig. 1 lB is a graph depicting re~ectivity versus incidence angle for both TE and TM pol~n7~d light according to an embodiment of the present invention.
Fig. 12 is a graph illustrating detector response versus surface recess for an embodiment of the present invention.

Fig. 13 is a graph illustrating qll~ntl-m efficiency for vanous top and bottom reflectivities according to embo(1imentc of the present mventlon.
Fig. 14 is a graph illustrating detector ~ L~ at ~lirre ell~
incidence angles om a device according to an embo&ent of the present 30 invention.

W O 97/15812 PCTrUS96/17188 Fig. 15 depicts an embodiment of a monolithic VLSI circuit according to the present invention.
Fig. 16A depicts a magneto-optical data storage drive using conventional photodetectors.
Fig. 16B depicts an embodiment of a magneto-optical data storage drive according to an embodiment of the present invention.
Fig. 17 depicts a quadrant detector according to an embodiment of the present invention.
Fig. 18 depicts a quadrant detector according to an 10 embodiment of the present invention.
Fig. 19 depicts a CCD array according to an embodiment of the present invention.
Fig. 20 depicts an example of a object within a scene which may be detected using a CCD array according to an embodiment of the 15 present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The principle of operation is based on the resonant cavity enhanced (RCE) photodetectors. The RCE s~ucture consists of a thin absorption region placed in an asyrnmetric Fabry-Perot cavity. The cavity 20 is formed by top and bottom reflectors which may be fabricated by various me~ods (for example, ~ltern~ting layers of quarter-waveleng~ dielectrics, i.e., Distributed Bragg Reflectors). For P~mple, the top reflector may be formed of a semiconductor/air interface and the bottom reflector may comprise a DBR. When the optical length of the cavity satisfies ~e 25 resonance condition, the cavity enhances the optical fields and the detector response is drastically increased.
The 4~ .. efficiency for a RCE detector can be expressed as _ _ W O 97/15812 PCT~US96/17188 (I + R e ad) 2 ~ X (1 - R~ e - 2~RlR2e-adcos(2~L + ~ + ~2) + RlR2e~d~

where a and d are the absorption coefficient and thickness for the thin absorber"B is the optical propagation constant, 1, is the length of the cavity, and R~ and R2, ~2 are the amplitude and phase of the top and bottom reflectors, respectively. If the light is incident to the surface at an angle 5 with the normal, L, is replaced by L/ cos ~9 . When the optical length of the cavity satisfies the resonance condition, the cavity enhances the optical fieldsand the cletect~r response is drastically increased. The peak ~ at the resonant wavelçn~h~ can be derived by imposing the resonant condition in Eqn. 1.

~max = 1 + R2e~ad ~ X (1 - R~ - e~a~

2e ) ~ (2) lf the length of the cavity is altered (for example, by surface recessing) then the sensitivity is reduced below that of a conventional detector structure without the cavity. The enhancement/reduction in the detector response is a strong function of the top reflectivity. The origin of the drastic enhancement in rl is the greatly increased amplitude of the 15 electric field inside a high Q resonant cavity which causes more energy to be absorbed in the active region. An equivalent intel~relaLion is that an individual photon is multiply-reflected at the n~illol~ and therefore makes many passes through the absorption region for varying mirror reflectivities as a function of c~d. For small ~in absorbers (ocd about equal to 0.1), i.e., a 20 low loss cavity, the enhancement factor exceeds 10. Fig. 13 illustrates the W O 97/15812 PCT~US96/17188 qll~n~lm efficiency, ~, at resonance as a function of ad for various top (Rl) and bottom (R2) mirror reflectivities. In Fig. 13, solid lines represent R2 =
0.9 and Rl = 0.9, 0.7, 0.5, and 0.3 as indicated. Dashed lines represent R2 = 0.99 and Rl = 0.7. Dotted dashed lines represent the conventional detector S case in which R2 = O and Rl = 0.3. For a typical native semiconductor surface (Rl ~ 0.3), the R2 = ~ and 0.9 curves illustrate the contrast between the conventional and the RCE cases. RCE detection improves 1l by a factor of about 6.5 for a 0.1 micron thick absorption layer (a = 104 cm~l) Qll~n1llm efficiency may further be enhanced by higher reflectivity n~ ol ,. The R2 10 = 0.99, Rl = 0.77 curve reaches a ma~ in excess of 98%.
For off-normal incidence of light, the reflectivity of a semiconductor/air interface may be significantly different for TE and TM
polarizations. At Brewster's angle, for example, TM reflectivity vanishes and TE reflectivity is approximately 0.75 for the GaAs-air interface.
15 Therefore, sensitivity, i.e., q~nt~lrn efficiency, is a strong function of the cavity length for TE polarization while sensitivity for TM is invariant.
A pair of monolithically integrated RCE photodetectors with cavity lengths tuned for resonance and anti-resonance for TE polarization provide a large contrast. A comparison of the current from these two 20 detectors under equal illllmin~tion yields the absolute polarization of the incident light. In an embodiment in which the detectors are horizontally arranged, equal illl~min~tion may be used to ensure proper functioning. In addition, however, vertically arranged photodetectors may be used in which u~ llmin~tion is less essenti~l. The described invention is applicable 25 to most m~t~ri~l systems and detector st~uctures and various waveleng~ (~) reglons.
Fig. 1 comprises a photodetector structure 10 according to an embodiment of the present invention. Photodetector structure 10 comprises a substrate 12, a bottom reflector 14, a bottom layer 16, an absoIption layer 30 18, and a top layer 20. A bottom electrical contact 22 and a first top CA 0223~922 1998-04-27 W O 97/15812 PCT~US96/17188 electrical contact 24 and a second top electrical contact 26 may also be provided on bottom layer 16 and top layer 20 respectively.
In one embo-limPnt, a cavity 28 may be formed in bottom layer 16, absorption layer 18, and top layer 20. Two detectors may thus be 5 formed: first detector 30 and second detector 32, which are sep~led by cavity 28. First detector 30 and second detector 32 may have di~elelll cavity 1en~hc First detector 30 may have a cavity length of Ll while secorld detector 32 may have a cavity length of L2. In a pler~lled embodiment, a cavity length ma~r be provided such that the m~xi..,~ . cavity length at any point equals about 2.5 micr~ mPtPr~ for example. Top layer 20 and air form a top reflector 34 on both first detector 30 and second detector 32. Bottom reflector 14 provides reflectance from the bottom.
In one embodiment, top layer 20 may comprise GaAs.
Absorption layer 18 may comprise InGaAs, for example, such as In0 lGa0 9As and have a depth, d, of about 0.1 microns. Other depths may also be used. Bottom layer 16 may comprise AlGaAs, for example. These layers may also comprise GaAs, AlGaAs, InGaAs, InP, InGaAs, InAlAs, InGaAs, GaAs, AlAs, Ge, Si, SiO2, SiGe, GaP, AlP, GaN, AlGaN, InGaN, or InAlGaN or other semiconductor or dielectric materials.
By selecting recess values for each of the detectors in the array, for example, by selecting two recess values corresponding to the maxi~ and miniml-m of the TE q~l~ntllm efficiency, a shalp contrast in TE-response may be achieved. For example, at normal incidence, the GaAs/air interface provides a reflectivity of 0.32. This reflectivity is a strong function of incidence angle and polarization as shown in Fig. 2. As ~ Fig. 2 illustrates, the reflectivity of TE and TM diverge up to the Brewster angle (which in this case is about 74 degrees) as the incidence angle increases away from zero. The present detector structure lltili7es the polarization dependent variation of the reflectivity of this interface under oblique incidence. Fig. 2 shows the incidence angle dependence of the W O 97/15812 PCT~US96/17188 surface reflectivity for GaAs/air interface for light having a wavelength of ~ = 900 nm for TE and TM polarizaaons. At a wavelength of about 900 nm, the refractive index of GaAs is n = 3.55. At 74c incidence angle (Brewster angle), TM-reflectivity vanishes, but TE-reflectivity is quite large, e.g., 5 about 0.75. Therefore, sensitivity is a strong function of the cavity length for TE polarization. Fig. 3 illustrates that sensitivity is a strong function of thecavity leng~ for TE pol~ tion (TE is repr~sente~l by a solid line in Fig. 3).
Therefore, according to the present invention, since cavity length may be controlled by recessing the top surface, the sensitivity of a device to TE
10 polarization may likewise be controlled. As for TM sensitivity, Fig. 3 illu~l~al~s that TM sensitivity is invariant to cavity length (TM is representedby the dashed line in Fig. 3 - TM remains con~t~nt). In Fig. 3, the wavelength of light is 900 n~n, the incidence angle is 74 ~, the m;.x;...~
length, Lm~ = 2.5 microns, the absorption coefficient and dep~, ocd = 0.2, 15 the amplitude of the bottom reflector, R2, is about 1.0, and the refractive index, n = 3.55 for the GaAs.
According to tne present invention, f~rst and second detectors 30 and 32 may be constructed to have very different sensitivities for TE
polarization while their responses to TM polarization are equal. For 20 example, in Fig. 1, first detector 30 may be ~ cte~l to achieve the maximum sensitivity of the resonant cavity for TE polarization for ~e selected incidence angle. For example, for light having a waveleng~ of about 900 nm, the detector may have a recess of about 0.07 microns, such that the length of first detector 30, L" = 2.43 microns. In this embodiment, ~he 25 surface of second detector 32 may be recessed such that the incident TE
polarized light is rejected and TM polarized light is permitted to pass through for detection. For example, for light having a wavelength of about 900 nm, second detector 32 may be recessed by about 0.175 microns, such that the length of second detector 32, L2, = 2.325 microns. The detector , W O 97115812 PCT~US96/17188 responses of first detectnr 30 and second detector 32, D~ d D2 respectively, may be expressed as Dl = sl rE ~ TE + SI TM ~ TM (3) D2 = S2TE ~ TE + S~TM ~ TM (4) that is:
D~ 5l TE s~ rM [TE] [S] [TE] (5) S where SiTE and SiTM re~ ~senl the sensitivity of first and second detectors 30and 32 (represented by subscripts i = 1 and i =2, respectively) to ~liL~r~l-l perpendicular polarizations. In the matrix of Eqn. 5, Dl and Dz represent d~tt?ctor CUll~ ou.tput and TE and TM represent the incident power of the corresponding pol~ri7~tions. If the two detectors have the same sensitivity 10 for diL[er~lll pol~ri7~1ions as in conventional ~lesi~n~, then Eqns. 3 and 4 are identical. For the described device structure, there is a big contrast in the response to dirre~ t polalizations. Therefore, TE and TM powers can be evaluated from the detector signals as:

[TM] = [S]-l ~ D (6) For the device parameters of Fig. 3, for example if Sl TE = 0.98, 15 SZTE = 0.03, and Sl,TM = S2,TM = 0.33, ~e following equation is obtained:

W O 97/15812 PCT~US96/17188 IrTEl _ 1.053 -1.053 Dl LTM~ -0.096 3.126 D2 In one embo-1iment the RCE detector structures may be formed in GaAs/(In, Al)GaAs m~ten~l systerns. Bottom reflector 14 may be formed by a 15 period GaAs/AlAs DBR mirror. Bottom reflector 14 serves to reflect light which passes through top layer 20, absorption layer 18 and 5 bottom layer 16 back up through bottom layer 16, absorption layer 18 and top layer 20. Likewise, top reflector 34 may reflect light back through top layer 20, absorption layer 18, and bottom layer 16, thereby increasing ~e amount of the light which is absorbed in absorption layer 18. For the GaAs/AlAS DBR milTor, the R2 = about 1Ø The thin InGaAs absorption 10 region placed in the GaAs cavity extends the photosensitivity beyond the GaAs absorphon edge where optical losses in the other layers are negligible.
At around the waveleng~ of about 900 nm, only the InGaAs region absorbs light, thus the device provides a low-loss resonant cavity. It is further important to note that although dirrelellces in TE and TM reflectivity are 15 greatest at the Brewster angle, a large diLr~ ce also exists at many other incidence angles as illustrated in Fig. 3.
To determine the absolute polarization with this embodiment, one only needs to have a device for solving two equations and two unknowns. Because each detector has a predetennined sensitivity to one of 20 the pol~n7~hon components, each detector's output is directly related to ~e amount of the other type of polarization received. Therefore, because there are two pol~ri7~1ion.c and two unknowns, a device which solves this type of problem may be used.
For example, a circuit as shown in Fig. 4 may be used to 25 evaluate ~ese equations. In Fig. 4, the variable resistor values are set , W O 97/15812 PCT~US96/17188 according to the matrix coefficients of [S]-l as in Eqn. 5. The circuit of Fig.
4 comprises a f;rst detector input 36 and a second detector input 38 which are connected to tlhe output of first detector 30 and second detector input 32, respectively. A first subcircuit 60 comprises a first variable resistor 40, a 5 second variable resistor 42, an amplification circuit 48, and an output 54.
Ampli~cation cir~uit 48 may c~ mpri~e a third resistor 50 and an operational amplifier 52. A second subcircuit 62 may comprise a third variable resistor 44, a four~ variable resistor 46, amplification circuit 48, and an output 56.
Output 54 provides an output representing the value of the TE component 10 and output 56 provides an output representing the value of the TM
component. As discussed above, the value of first through fourth resistors 40-46 may be dt;te....i~ed by the matrix coefficients in Eqn. 5. Therefore, for example, first resistor 40 may value a value of about 980 ohrns, second lt;si~lor 42 = about 30 ohms, third resistor 44 = about 330 ohms and four~
15 resistor 46--about 330 ohms.
To d~l~....;l-e the polarization, the ratio of the power in the two polarization complonents can be then evaluated as:
TE~ 1.053 ~ Dl - 1.053 ~ D2 (8) ~ TM) 3.126 ~ D2 - 0.096 ~D, The described detector array of two detectors measures the relative value of the TE and TM polarization components accurately. The 20 described invention is applicable to most material systems and detector structures and various wavelength (~) regions. For example, a ~ GaAs/AlGaAs/InGaAs and SitSiO2/Si3N4 photodetector structure may be used The present invention may also be used for dt;le...~ g 25 variations in polarization. By using the ratio of photocu~ (TEtTM) derived and storing this result, any change in polarization may be detected _ W O 97/15812 PCT~US96/17188 by mol~iLo~ g the photocurrent ratio for changes. Thereby, for example, in a magneto-optical data device, the present detector system may be used to detect the presence of either a one or a zero, a one representing one level of pol~n7~tion and a zero representing another, the change in polarization 5 indicating a change from one to zero or zero to one.
According to anot~er embodiment of the present invention, the two detectors may be stacked. This invention provides a detector which operates effectively with less reliance on the illumin~tion area (spot size) of the light. As with the embodiment of Fig. 1, the basic principle is based on 10 a Distributed Bragg Reflector (DBR) rmirror placed between the two detector structures. This DBR mirror has si~uficantly diLrele.lt reflectivities for TE
and TM polarizations for large incidence angles.
Fig. 5 depicts a photodetector structure 70 according to an embodiment of the present inven~ion. Photodetector structure 70 comprises 15 a substrate 72, a second absorption region 74, a mirror structure 76, a separating layer 78, a first absorption region 80, and a top layer 82.
Photodetector structure 70 also comprises a first electrical contact device 84, a second electrical contact device 86, and a third electrical contact device 73.Top layer 82, first electrical contact device 84, first absorption region 80, and third electrical contact device 73 form a first detector 88. Second absorption layer 74, second electrical contact device 86, and third electrical contact device 73 form a second detector 90. First detector 88iS situ~te~l in a resonant cavity bounded by mirror structure 76, which may be, for example, a DBR device, and the interface between top layer 82 and air. In a ~l~;felled embodiment, top layer 82 may comprise a semiconductor or dielectric m~t~,Ti~l The present inven~on utilizes the polarization dependent variation of the reflectivity (Rl) of the int~ ce between top layer 82 and air under oblique incidence. In a preferred embodiment, top layer 82 may comprise GaAs. Other substances may also be used including for example, 30 GaAs, AlGaAs, InGaAs, InP, InGaAs, InAlAs, InGaAs, GaAs, AlAs, Ge, Si, SiO2, SiGe, GaP, AlP, GaN, AlGaN, InGaN, or InAlGaN or other semiconductor or dielectric materials.
Fig. 6 shows the incidence angle dependence of the surface reflectivity for GaAs/air interface ~=900 nm for TE and TM polarizations.
S Note that, at 74~ incidence angle, TM-reflectivity vanishes (Br~wslel angle).
For the same angle, TE-reflectivity is quite large. Similarly, a contrast in the bottom reflectivity can also be achieved. For example, in a ~lert;lled embo-liment as depicted in Fig. 7A, mirror structure 76 may comprise a Z0 pair Al0 5GaO 5As/AlAs DBR mirror. Mirror structure 76 may thus comprise 10 alternating layers of a first layer 92 and a second layer 94, in this embodiment, first layer 92 comprising Al05GaO5As and second layer 94 comprising AlAs. Also, according to an embodiment of the present invention, a sep;~ g layer 78, a first absorption region 80, and a top layer 82 may instead compri~e a single first absorption region 80 with absorption 15 layer 80 and air forming the first reflector 96. Fig. 7B shows the angle dependence of TE (solid) and TM (dashed) reflectivities for a 20 pair Al05GaO5As/AlAs DBR mirror which provides the reflectively for second detector 90 and may be repres.-nt.o~l by R2. In this embo-1iment the refractive index of the AlAs is about 2.9, the refractive index of the GaAlAs is about 20 3.2 and the refractive index of the GaAs absorptive mat~ri~l is about 3.55.
The wavelength of light used in Fig. 7B is about 900 with an initial incidence angle, ~0 = 30. As Fig. 7B illustrates, at larger incidence angles, a greater divergence of mirror reflectively occurs.
In this emborliment because both the interface between top 25 layer 82 and air, which forms a top reflector 96, and mirror structure 76 have ~ a large contrast in reflectively at around Brew~Lel angle, the annount of light captured in the top detector is a strong function of its polarization. Second detector 90 is disposed below mirror structure 76 and thereby its responsivity is proportional to the overall tran~mi.csion of light through the top layers 30 including the DBR nni~ror. Therefore, the polarization dependent reflectivity W O 97115812 PCT~US96/17188 of mirror structure 76 results in a contrast in TE v. TM detection in second detector 90. Thereby, the s~ucture is designed to capture as much TE light as possible in first detector 88, which is disposed on top, and transmit most of the TM light to second detector 90, which is disposed on bottom.
In other words, because top reflector 96 is polarization depend~nt the resulting cavity for first detector 88 provides resonance enhancement for TE thus ca~lwi~g the TE polarized light in first detector 88. For TM, both reflectivities of the top and bottom reflectors 96 and 76, respectively, are low and therefore, light is tr~n~mi~ted to the bottom de~ ol. For a thin absorbing layer in the RCE detector, a large contrast in TE/TM response of first and second detectors 88 and 90 is achieved and ~e linear polarization may be colllpul~d from their relative responses using a circuit as depicted in Fig. 4, for example.
The contrast in the TE/TM reflectivity of the bottom DBR
l 5 mirror is relatively small for a AlAs/AlGaAs structure because the incident beam is strongly refracted due to the large refractive index (rl = 2.9 for AlAs). The beam inside the semiconductor is therefore always at a small angle with the normal resulhng in a small dirr~le1~ce in reflectivities at ~e AlAs/GaAs interfaces. The DBR reflec~ivity contrast may be improved significantly if smaller refractive index materials are used to construct the multi-layer structure. For the GaAs material system, the use of native oxide ol~ (with ~ as small as 1.7) may be used as an alternative.
The detector responses Dl and D2, of first and second detectors 88 and 90, respectively, may be expressed by Eqns. 3-5 given above.
To calculate the elements of the [S] matrix, we use the reflectivities for top (Rl) and bottom (R2) ~ lols for given material combin~ti. n~. For the GaAs/AlGaAs structure, R, is given by Fig. 6 and R2 is shown in Fig. 7B for a specific DBR design as indicated. Fig. 8 shows the ed g~ nt~lm efficiencies for TE and TM in first and second detectors , W O 97/15812 PCT~US96/17188 88 and 90. At resonance (peak of TEI) a significant contrast for the two detectors can be obtained. For this case Sl TE Sl TM 0.19 0.04 S2 TE S2 TM 0.24 . (9) According to another embotlim~nt, an improved structNre may be provided by using a diLre~ t material system with smaller refractive 5 inrlices For ~ mple, Si m~tlori~l systems using dielectric DBR n~ o.~ offer more drastic enhancement in TE/TM contrast for even fewer periods of the DBR mirror. As an added advantage, an Si/SiO2/ Si3N4 m~teri~l system offers monolithic integration of the polanzation detectors with vertical cavity polarization detection cil-iuilly implementing smart pixels and arrays for 10 pol~ri7~tion sensi;ng and im~gin~ These structNres may be formed on Si VLSI circuits by depositing dielectric films for DBR ~ lOl~ and using SOI
(silicon on insulator) for the top absorption layer in the RCE detector. The VLSI circuit may be accessed by mesa processing of the detector structure and detectors may be integrated with electronic devices for proces.ein~ the 15 two detector outputs to colllpule the inci-lent polarization.
For example, Fig. 15 depicts an embodiment of a monolithic VLSI circuit according to the present invention. Fig. 15 comprises two detectors 88 and 90 formed on a substrate 72. First and second electrical contact devices 84 and 86 are connected to the VLSI circuit using layer 95.
20 Layer 91 may comprise a dielectric film used to provide electrical isolation of layer 95 and photodetectors 88 and 90. The detectors may also be connected to other electrical on-chip components 93. Other electrical on-chip components 93 may comprise a microprocessor, for example. In this embo-liment, the microprocessor may be used to calibrate the operation of 25 the detectors in connection with external devices.

_ _ _ ~ _ _ _ As another example of a photodetector using Si/SiO2/ Si3N4, a photodetector 100 may be provided with a mirror structure 76 comprising an 8 pair DBR mirror con.eietin~ of SiO2 and Si3N4, as depicted in Fig. 9A.
First layer 92 comprises SiO2 and second layer 94 comprises Si3N4 In this 5 embo-lim~ont Fig. 9B shows the TE and TM reflectivities for an 8 pair DBR
mirror comprising layers of SiO2 and Si ~J .4 In this embodiment, the refractive index of SiO2 is about 1.457 ~e refractive index of Si3N4 is about 2.05 and the refractive index of the absorptive layer 80 which is Si, is about 3.1. Absorptive layer 80 may cl mprice Si and may have a thickness of about 10 1.5 microns. The light me~sllred in Fig. 9B had a waveleng~ of about 900 nm and an initial incidence angle, ~0 - 30. At around the Brewster angle of Si/air interface (about 72~), a very large contrast for R2 may be achieved.
Therefore, the ql~ntl~m efficiency for TE and TM for first and second detectors 88 and 90 is very dirre~ L, as illustrated in Fig. 10. The TE and 15 TM values for both first and second detectors 88 and 90 are indicated by the subscript 1 and 2, respectively. [S] matrix becomes:

Sl,TE Sl,TM 0.88 0.18 S2.TE S2.TM 0 90 0.04 (10) According to another embodiment, an irnproved structure may be provided by using a dirre~ t m~t~ri~l system with smaller refractive indices. For example, a photodetector 100 may be provided with a mirror 20 structure 76 comprising a 5 pair DBR mirror consisting of SiO2 and Si3N4, as depicted in Fig. 1 lA. First layer 92 comprises SiO2 and second layer 94 comprises Si3N4 In this embodiment, Fig. llB shows the TE and TM
reflectivities for a S pair DBR mirror comprising layers of SiO2 and Si3N4.
In this embo~1im~nt the refractive index of SiO2 is about 1.45, the refrac~ive 25 index of Si3N4 is about 2.05 and the refractive index of absorptive layer 80 (Si) is about 3.1. The light measured in Fig. 1 1B had a waveleng~ of about W O 97/15812 PCTrUS96/17188 900 nm and an initial incidence angle, ~ = 30. Therefore, the qll~nhlm efficiency for TE and TM for first and second detectors 88 and 90 is very l, as illustrated in Fig. 12. The TE and TM values for both first and second detectors 88 and 90 are indicated by the subscript 1 and 2, 5 respectively.
In the embo-limPnt.c of Figs. 5, 7A, 9A, and 1 lA, the ~ nment of the incident beam is not critical. Because first and second detectors 88 and 90 are vertically integrated, the spatial distribution of incident radiationdoes not affect the accuracy of polarization measurement. Therefore, the 10 light beam may be smaller than the detector area allowing for ca~lu.il~g all of the light in the detector. The incident light may be focused onto the top surface of the photodetector using a large incidence angle. Therefore, the beam shape of the light may be spread out in one climen~ion. For highest efficiency, the surface area of the device matches the beam shape.
15 Consequently, because of the incidence angle and the creation of a beam shape being larger in one dimension, the detector surface may likewise be larger in one ~limPn.~ion to m~ximi7:e detector efficiency. Also, because the devices work under spatially varying light intensities, im~ing array detectors that can register the polarization distribution in addition to the 20 intensity of light are re~li7e~1 according to the present invention without requiring any beam splitters and/or polarization filters.
For each of the embodiments of the present application, detector response cullellls, Dl and D2, may be expressed by Eqns. 3 and 4.
Fig. 14 depicts a graph illustrating the detector cLL~ Ls at two different 25 incidence angles as a function of polarization. In Fig. 14, the solid lines represent D~ and D2 at Brewster's angle and the dashed lines represent D, and D2 at an angle of about 60 degrees. The cavity length of the RCE device is o~ s~l separately for dirrel~ incidence angle cases. The increase in the total variation oiF detector response ratios observed at smaller incidence W O 97/lS812 PCT~US96/17188 angles indicates improving polarization resolution. This improvement is accomplished by reduction in the linearity of the response.
According to another embodiment of the present invention, a quadrant detector may be provided. According to this embodiment, S polarization and positioning sensing may be detected ~imllltaneously. Fig.
17 depicts a quadrant detector according to an embodiment of the present invention. Fig. 17 comprises a substrate 70 with four pairs of detectors vertically aligned on the substrate 70. Monolithic horizontally arranged detectors may also be used. According to this embodiment, quadrant device 10 200 comprises first detector 88 and second detector 90, which may be vertically ~ligne-l Quadrant device 200 also comprises a third detector 130 and fourth detector 140 which may be vertically aligned. Quadrant device 200 also comprises a flfth detector 150 and a sixth detector (not shown) which may be vertically aligned and a seventh detector 160 and an eighth 15 detector (not shown) which may be vertically aligned. Also, each of ~e pairs may be separate within the quadrant according to the embodiment of Fig. 1, for example.
A light source 210 may be provided for directing a light beam 220 at quadrant detector 200. In this arrangement, first, third, fifth and 20 seventh detectors 88, 130, 150, and 160, respectively, may be used to detect whether light source 210 is properly positioned in a predete~nined desired position. For example, in a magneto-optical drive, the laser source may be positioned such that light is to be evenly distributed across the four top detectors. The position of even distribution may be the prede~ nlled 25 desired position, for example. If the output from each of the detectors is not equal, then quadrant detector 200 detects that light source 210 is out of position and may cooperate with other ~;il~; ~ill y and/or devices to move lightsource 210 back into the predete~ ed desired position.
At the same time, however, each of the vertically aligned pairs 30 may be ~imlll~neously used for polarization detection. In ~is arrangement, CA 0223~922 1998-04-27 W O 97/15812 PCT~US96/17188 , any one of the four pairs may be used for the output, or, some combination of the four oul~uls may be used to avoid errors. For example, a norm~ t on sclheme may be employed which uses all four detector pair outputs to generate one pol~ tion output to ensure accuracy in operation 5 of the system.
Fig. 18 depicts another embodiment of a quadrant detector 200 according to the present invention. Each quadrant comprises a vertically stacked pair of detectors. Each pair of detectors comprises a first electrical contact device 300 and a second electrical contact device 302. Further, each 10 detector pair shares a third electrical contact device 304. According to this embotlimPnt, each cletector may be rect~n~ r. A mesa shaped embodiment as in Fig. 17 may also be used.
The devices described according to several embodiments of the present invention may be used in a number of applications. According to 15 one embodiment of the present invention, an erasable optical data storage using magneto-optical media may be provided. Erasable optical data storage using magneto-optical media rely on the state of polarization of the read beam to detect the data. Current technology uses a polarizing beam splitter (cube or plate) to separate the TM and TE component of polarization and 20 direct them to separate diode detectors.
For example, Fig. 16A depicts a magneto-optical data storage drive using conventional photodetectors. A magneto-optical drive 800 operates to read information stored on an optical disk drive 408. Magneto-optical drive 800 comprises a light source 400, a first lens 402, a beam 25 splitter mech~ni~m 404, a first focusing lens 406, a second focusing lens 410, a pol~ri~ing beam splitter mech~ni~m 420, two photodetectors 424 and 422, and a comparator circuit 412. Photodetectors 424 and 422 may compri~e conventional photodetectors. Magneto-optical drive 500 provides an output 414 representing data read from optical disk device 408. In this 30 a~lus, polarizat3ion sensing is performed by ~e polarizing beam splitter W O 97/15812 PCT~US96/17188 mech~ni~m 420 and photodetectors 422 and 424. The light coming from beam splitter 404iS focused via second focusing lens 410. The polarization beam splitter mech~ni.cm 420 separates the light such ~at one polarization is reflected and absorbed by photodetector 422 and the other polarization is 5 transmitted and absorbed by photodetector 424. The com~ator 412 compares the outputs from photodetectors 422 and 424 and thus provides an output indicating a change in polarization of the light which signals a change from a zero to a one or vise versa.
The apparatuses, for example, the device of Fig. 16A, which employ a 10 separate pol~ri7ing beam splitter, suffer from weight and m~n~lf~cturing problems. The weight makes the optical head slower and reduces data transfer times. M~m-f~cturing with ~ree components adds complexity and creates demanding tolerances. Any progress towards an integral polarization detector ~at provides a polarization di~elel-ce readout while reducing 15 weight and parts count would be welcomed by ~e data storage industry.
Cost is also a problem with the present erasable optical data storage devices because polarizing cube beam splitters are expensive.
A photodetector array according to the present invention may be used to detect data by sensing the polarization of the read beam. As such, 20 there is no need for a beam splitter and ~us a significant reduction in cost may be achieved. Additionally, the optical and electrical properties of silicon allow for fabrication of polari ation detectors in visible to near -IR
wavelength range. Therefore, for magneto-optic data storage applications, the capability of fabricating polarization sensors in the visible spectrum is 25 particularly important since dle storage capacity scales inversely with the waveleng~.
For example, Fig. 16B depicts an embodiment of a magneto-optical data storage drive according to the present invention. A magneto-optical drive 350 operates to read information stored on an optical disk 30 device 408. Magneto-optical drive 350 comprises a light source 400, a first W O 97/15812 PCTrUS96/17188 lens 402, a beam splitter mPch~ni~m 404, a first focusing lens 406, a second focusing lens 410, a detector 411, and a co~ alalor circuit 412. Magneto-optical drive 350 provides an output 414 representing data read from optical disk device 408. Light source 400 may preferably comprise a laser such as 5 a GaAs laser, for example. Other laser sources or light sources may also be used. First lens 402 serves to direct the output of light source 400 through beam splitter mech~ni~m 404. Beam splitter mech~ni.cm 404 directs light from light source 400 toward first focusing lens 406 and at the same time directs light coming through first focusing lens 406 toward second focusing 10 lens 410. Beam splitter assemblies for serving this purpose are known. Any beam splitter which performs this function may be used. Light p~ccing through beam split~er mech~nicm 404 is directed by first focusing lens 406 toward optical disk device 408.
The light directed at optical disk device 408 is reflected in a 15 polarization depending on the surface of the optical disk device 408. I
other words, optical disk device 408 reflects the stored data in terms of a change in polarization as optical disk device 408 rotates, and indicates a change from a one to a zero or vise versa. Therefore, by detecting the polarization of light reflected from optical disk device 408, the magneto-20 optical drive 350 may read the data stored on optical disk device 408.Operation of magneto-optical drives is known in the art.
The present invention provides the advantage of having a polarization detecl:or 411 according to the presemt invention without requiring a secondl polarization beam splitter mech~ni.cm, for example 25 pol~ri7inp beam splittermec~l~nicm 420 of Fig. 16A. Thereby, the weight, size, and cost of the system may be significantly reduced. Polarization detector 411 may comprise an embodiment of polarization detector according to the present invention. For example, polarization detector may comprise apol~ri7~t;on detector 10, 70, 100, or 200. Other embo~limentc of pol~ri7~tion detectors according to the present invention may also be used.

As described above, a colllp~lol circuit 414 may be provided for generating a polarization output 414 indicating either the polarization 5 ratio or a change in polarization. According the present invention may provide a magneto-optical drive without requiring two beam splitters.
Therefore, a smaller and rnore compact drive device may be provided.
Additionally, according to another embodiment of the present invention polarization detector 411 may be connected to a computer. In one 10 embo-1im~nt, the coml~ul~l may comprise a microprocessor and memory and may be provided on the same chip as the pol~ri7~tion detector. Additionally, a calibration control mech~ni.cm 416 may be provided on polarization detector 411. In this embo-1im~nt, the operation of the magneto-optical drive may be calibrated. For example, an optical disk having a known sequence 15 of values may be used. Magneto-optical drive device 350 may then be operated to read the known optical disk sequence. If the detected value differs from the known value, calibration control mech~ni.~m 416 may be operated to adjust the angle of incidence of light onto polarization detector 411. Also, the entire magneto-optical drive device 350 may be adjusted to 20 reposition the device in relation to optical disk device 408. This process may be repeated until the incidence angle of the polarization detector 411 and positioning of magneto-optical drive device 350 are calibrated.
Calibration may be necessary over time due to physical disruption of the device or due to any other change in the arrangement of components in the 25 magneto-optical drive device 350.
According to another embodiment of ~e present invention, polarization multipleDg may be provided for increasing the capacity of optical tr~ncmi~sion by a factor of two. By using the low cost polarization sensitive detectors according to the present invention, polarization 30 multiplexing may be achieved which may be used to increase the capacity W O 97/15812 PCT~US96/17188 of undersea fiber cables, for example. The proposed devices can be employed as fast~ reliable ~letectors in any polarization multiplexing system.
According to another embodiment of the present invention, the present photodetecl:or structures may be used in polarization sensing arrays.
5 Tm~gin~ array detectors which are sensitive to polarization may be used in remote sensing, especially for naval applications and medical im~in~ for example.
Fig. 19 depicts a polarization CCD array according to an embodirnent of the present invention. CCD array 550 comprises a plurality 10 of pol~ri7~tion detectors 500 arranged into a plurality of columns 502 and rows 504. Each colurnn and row may comprise a plurality of polarization detectors 500, such as four, for example. A larger number of polarization detectors 500 may be provided. Each polarization detector 500 may compri~e a p-contact 506, an n-contact 508 and a subs~ate detector contact 15 510. Each contact may be connected to a bus via a MOS transistor 512, for example. Substrate detector contact 510 is connected to a substrate contact bus 514, p-contact 506 may be connected to a p-bus 516, and n-contact 508 may be connected to an n-bus 518. Each of the buses may be connected to cil.;uiLIy for providing outputs from each of ~e polarization detectors such 20 that a pol~ri7~*on array is generated. A con,~ulel or other processing device may be connected to the buses and c~-;uiLly for generating information inl1ir?~tive of the pol~ri7~tion of images which are detected by the CCD array 550. In such an arrangement, pol~ri7~tion im~ing may be provided into a single chip device because beam splitters are not necessary for polarization 25 sensing.
This type of detector may be particularly useful for detecting objects having a dirre~ polarization than the background in which the object is viewed. For example, Fig. 20 depicts a sketch of an image of a ship on the horizon. The image comprises the ocean, the ship, and the sky. In 30 ordinary circl1m.ct~nces in daylight, the ocean and the sky are both blue.

W O 97/15812 PCT~US96/17188 Moreover, the ship may be grey. In general, ordinary array detectors may not be able to detect the presence of the ship on this horizon. However, a pol~ri7~tion CCD array may be used because the ocean and the ship reflect di~el~nl pol~ri7~tions of light and the sky is randomly polarized. Therefore, 5 a detector which senses cll~n~s in polarization on a horizon may be used for detecting the presence of a ship or other object on that horizon. O~er uses for the polarization CCD array detector also exist. Moreover, by using vertically arranged pairs of photodetectors according to the present invention, for example, as depicted in Fig. 19, a compact and cost-effective 10 polarization sensing CCD array may be provided.
According to another embodiment of the present invention, research laboratory instrllment~1ion may be provided. A detector according to the present invention may be used to replace the bulky and expensive eqllipmPnt used in ~:~e~ Pn~l laboratories working on fiber-optics, lasers, 15 im~in~ optical storage, etc.
Other uses and embodiments are also wi~in the scope of the present invention. Other photodetector structures may also be used wi~out departing from the scope of the present inven~on. For example, photodetector devices for use in the present invention may be formed from ~0 GaAs, AlGaAs, InGaAs, InP, InSb, InGaAs, InAlAs, InGaAs, GaAs, AlAs, Ge, Si, SiC, SiGe, GaP, AlP, GaN, AlGaN, InGaN, or InAlGaN. Other photodetector device surfaces may also be used as would be recognized by one of ordinary skill in the art.

Claims

What is claimed is:

1. A device for detecting polarization of light comprising:
a first photodetector tuned to absorb TE polarization;
a second photodetector tuned to absorb TM polarization; and a circuit for comparing an output from the first and second photodetector for generating a polarization output.

2. The device of claim 1 wherein the first photodetector has a first length and the second photodetector has a second length and wherein the first length and second length are not equal.

3. The device of claim 1 wherein the first photodetector has a first length and the second photodetector has a second length;
wherein the first length is selected to achieve maximum sensitivity for TE polarization for a predetermined incidence angle; and wherein the second length is selected to achieve maximum rejection of TE polarization for the predetermined incidence angle.

4. The device of claim 1 wherein the first and second detectors comprise RCE detectors.

5. The device of claim 1 wherein the first and second detectors are vertically stacked.

6. The device of claim 1 wherein the first photodetector comprises:
a first reflective means;
an absorption layer on top of the first reflective means; and a second reflective means on top of the absorption layer.

7. The device of claim 6 wherein the first reflective means comprises a DBR mirror.

8. The device of claim 6 wherein the second photodetector comprises:
a third reflective means;
an absorption layer the reflective means being on top of the absorption layer; and a fourth reflective means, the absorption layer being disposed between the third reflective means and the fourth reflective means.

9. The device of claim 8 wherein the first reflective means comprises the third reflective means.

10. The device of claim 1 wherein the second photodetector comprises:
a third reflective means; and an absorption layer, the reflective means being on top of the absorption layer.

11. The device of claim 8 wherein the second photodetector also comprises a fourth reflective means, the absorption layer being disposed between the third reflective means and the fourth reflective means.

12. The device of claim 10 wherein the third reflective means comprises a DBR mirror.

13. A method for detecting polarization of light comprising the steps of:
absorbing TE polarization using a first photodetector;
absorbing TM polarization using a second photodetector; and comparing an output from the first and second photodetector for generating a polarization output.

14. The method of claim 13 wherein the first photodetector has a first length and the second photodetector has a second length and wherein the first length and second length are not equal.

15. The method of claim 13 wherein the first photodetector has a first length and the second photodetector has a second length;
wherein the first length is selected to achieve maximum sensitivity for TE polarization for a predetermined incidence angle; and wherein the second length is selected to achieve maximum rejection of TE polarization for the predetermined incidence angle.

16. The method of claim 13 wherein the first and second detectors comprise RCE detectors.

17. A device for detecting polarization of light comprising:

a substrate;
a first photodetector tuned to absorb TE polarization;
a second photodetector disposed on the substrate and tuned to absorb TM polarization; and a circuit for comparing an output from the first and second photodetector for generating a polarization output.

18. The device of claim 17 wherein the first photodetector is disposed on the second photodetector.

19. The device of claim 18 wherein the first photodetector comprises a top surface which has a size larger in one dimension to match a shape of a beam of the light being detected.

20. A monolithic VLSI device for detecting polarization of light comprising:
a substrate;
a first photodetector tuned to absorb TE polarization and formed on the substrate;
a second photodetector disposed on the substrate and tuned to absorb TM polarization and formed on the substrate; and a circuit for comparing an output from the first and second photodetector for generating a polarization output and formed on the substrate.

21. The monolithic device of claim 20 wherein the circuit comprises a microprocessor formed on the substrate.

22. A photodetector system for detecting the positioning of a light source and the polarization of the light generated by the light source comprising:
a plurality of pairs of photodetectors for detecting light from the light source, each pair comprising:
first photodetector tuned to absorb TE polarization; and a second photodetector disposed on the substrate and tuned to absorb TM polarization and formed on the substrate; and a circuit for comparing an output from at least one of the first and second photodetectors for generating a polarization output; and a circuit for comparing an output from each of the first photodetectors from the plurality of pairs with a predetermined output to determine whether the light source is in a predetermined desired position.

23. The photodetector system of claim 22 wherein the plurality of pairs of photodetectors comprise four pairs of photodetectors.

24. A magneto-optical drive device comprising:
a light source for generating a light output directed onto an optical storage medium;
means for receiving light reflected from the optical storage medium and detecting polarization of the light comprising:
a first photodetector tuned to absorb TE polarization; and a second photodetector tuned to absorb TM polarization; and a circuit for comparing an output from the first and second photodetector for generating an output indicating a value stored on the optical storage medium.

25. The magneto-optical drive device of claim 24 further comprising a calibration means for calibrating the operation of the magneto-optical drive device.

26. The magneto-optical drive device of claim 25 wherein the calibration means comprises a microprocessor.

27. A CCD array device for detecting polarization of an image comprising:
a plurality of detectors arranged in rows and columns, each detector comprising:
a first photodetector tuned to absorb TE polarization; and a second photodetector tuned to absorb TM polarization, and a processor means connected to the plurality of detectors for generating an array of polarization values for an image received by the detectors.

28. The CCD array device of claim 27 each of the first and second photodetectors comprises a surface having a shape which matches a shape of the image being detected.