WO2000063973A1

WO2000063973A1 - Msm optical detector with an active region heterojunction forming a two-dimensional electron gas

Info

Publication number: WO2000063973A1
Application number: PCT/US2000/010113
Authority: WO
Inventors: Michael O'hagan; James A. Mcadoo
Original assignee: National Aeronautics and Space Administration NASA; Silkroad Inc
Current assignee: National Aeronautics and Space Administration NASA; Silkroad Inc
Priority date: 1999-04-19
Filing date: 2000-04-14
Publication date: 2000-10-26
Anticipated expiration: 2001-10-19
Also published as: AU4461000A

Abstract

A metal-semiconductor-metal photonics detector (12) with a light-detecting aperture (26) is provided. The detector (12) includes a pair of electrodes (16, 18) disposed in a common plane on an outer surface of the detector (12) on either side of the aperture (26) for producing an electric field across the aperture (26) from a first to a second electrode of the pair of electrodes (16, 18). The detector (12) further includes a relatively thin active region (56) of a semiconductor material having a thickness of between 0.05 and 0.2 microns and adapted to form a two-dimensional electron gas in the electric field between the electrodes (16, 18) in the presence of light (24).

Description

MSM OPTICAL DETECTOR WITH AN ACTIVE REGION HETEROJUNCTION FORMING A TWO-DIMENSIONAL ELECTRON GAS

5 Field of the Invention

The field of the invention relates to photodetectors and more particularly to an improved metal-semiconductor-metal photodetectors .

10 Background of the Invention

Metal-semiconductor-metal (MSM) photodetectors are generally known. Such devices have found wide use in the field of communications.

MSMs are generally multilayer devices of metal

15 and a semiconductor. Generally, two electrodes are disposed on the semiconductor and a voltage potential is applied across the electrodes. When light strikes the semiconductor, the photons sometimes impart sufficient energy to dissociate electrons from atoms

20 of the semiconductor, thereby creating charge carriers made up of electrons and holes . The charge carriers flow to the electrodes under the influence of the voltage potential thereby producing current flow.

25 The electrodes are typically arranged in alternating rows of positive and negative electrodes . A space is provided between the electrodes for entry of light into the semiconductor substrate.

The electrodes are typically deposited on the

30 substrate using an appropriate metalization process. The junction of the metal with the semiconductor forms a Schottky contact. The Schottky contact forms a barrier layer resulting in a rectifying _. current- voltage relationship. MSM diodes are relatively easy to manufacture and naturally have a low capacitance per unit area. While MSM diodes are relatively fast, they are still unable to detect modulation energy in the 100 GHz range superimposed on a 1550 nanometer laser. Because of the importance of laser communication systems, a need exists for a photodiode which is capable of operating under these conditions .

Summary A metal-semiconductor-metal photonics detector with an light-detecting aperture is provided. The diode includes a pair of electrodes disposed in a common plane on an outer surface of the detector on either side of the aperture of the detector and an electric field disposed across the aperture from a first to a second electrode of the pair of electrodes . The diode further includes a relatively thin active region of a semiconductor material having a thickness of between 0.05 and 0.2 microns and adapted to form a two-dimensional electron gas in the electric field between the electrodes in the presence of light.

Brief Description of the Drawings FIG. 1 is a schematic of a photonic detector system in accordance with an illustrated embodiment of the invention;

FIG. 2 depicts a cross-section of the diode of the system of FIG. 1; FIG. 3 depicts charge-pair flow within the diode of FIG. 1;

FIG. 4 depicts an energy level diagram of the diode of FIG. 1;

FIG. 5 depicts to top view of the diode of FIG. 1;

FIG. 6 depicts a substrate of the diode of FIG. 1 under an alternate embodiment; and

FIG. 7 is a frequency response plot of the system of FIG. 1.

Detailed Description of a Preferred Embodiment FIG. 1 depicts a cut-away side view of a novel photonics detector assembly 10, generally, in accordance with an illustrated embodiment of the invention. Included within the assembly 10 is the detector itself (e.g., a photodiode) 12, a .power supply 22 and a modulation detector 20.

In operation, the power supply 22 imposes an electric field between a first and second electrode 16, 18. In the presence of light (e.g., a modulated laser signal) , photons of light 24 enter the photodiode 12 through an aperture 26 of the photodiode 12 between the two electrodes 16, 18. The photons 24 penetrate a substrate 14 of the photodiode 12 and cause the disassociation of charge pairs

(e.g., electrons and holes) in an active layer. Due to the novel features of the active region, the incident radiation 24 forms a two-dimensional electron gas in the active layer beneath (and parallel to a plane of) the electrodes 16, 18, resulting in an extremely fast response of current to radiation 24. Changes in current through the photodiode 12 may be detected within a detector 20 and provided as a signal output. 100 FIG. 2 depicts a cross-section of a substrate 14 that has been found to work successfully for a 1550 n laser signal. The substrate 14 may be fabricated using a semiconductor insulator (S.I.) (e.g., an InP substrate) . A distributed-Bragg reflector (DBR) 105 layer 52 made up of alternating layers of semiconductor material with dissimilar indexes of refraction may be disposed on top of the substrate 14 using well-known processes.

The alternating layers may be that of InP and 110 InAlAsP. The alternating layers may have a thickness of 8 microns and may form 40 periods of alternating material of approximately 0.23 μm each. The DBR layer 52 functions to double the effective length of the active region 56. 115 An InAlAs isolation layer 54 may be disposed over the DBR layer 52 by an appropriate deposition process. The isolation layer 54 may have depth of 0.4 microns. The isolation layer 54 functions to isolate and prevent charge carriers generated below 120 the active layer 56 from contributing to current flow out of the diode 12. The isolation layer 54 improves frequency response by eliminating tail current from slower electrons reaching the electrodes 16, 18 much later than photon arrival . 125 The active layer 56 may be formed of InGaAs disposed over the isolation layer 54 by another appropriate deposition process . Under one illustrated embodiment, an active layer 55 of 0.075 microns has been found to perform well. 130 Alternatively, it is believed that the depth of the active layer 56 may be varied anywhere within the range of from 0.05 to 0.2 microns.

A graded layer 58 of In(GaAl)As of a thickness of 200 Angstroms may be deposited over the active 135 layer 56. The graded layer 58 may be provided to avoid carrier pile-up from an abrupt herojunction with the upper layers .

An InAlAs layer 6O of a depth of 150 Anstroms is provided over the graded layer 58. The InAlAs layer 140 60 has been found to be very effective in suppressing a dark current. Dark current, as generally understood, is a residual current which persists even in the absence of light.

Over the InAlAs layer 60, a 50 Angstrom thick 145 layer of InP is provided as a cap layer. The cap layer 62 functions to improve the reliability of the diode 14 by reducing oxidation of the InAlAs layer 60.

The electrodes 16, 18 may be disposed above the 150 cap layer 62. The electrodes 16, 18 may be formed of a graded mixture of Ti/Au by an appropriate metal deposition process.

A passivation layer of silicon nitride (not shown) may be disposed over the electrodes 16, 18 and 155 aperture to an appropriate thickness (e.g., 3000

Angstroms) . A photolithographic/etch process may be used to expose the electrodes 16, 18 for connection to the external circuit.

FIG. 5 shows an interdigitated structure that 160 may be used to further increase speed and current output. An overall diameter of the interdigitated electrodes may be kept to a relative small diameter (e.g. , 6-8 microns) . As shown, the interdigitated structure of the

165 electrodes 16, 18 is in the form of a series of concentric annular rings . A central subelectrode of the first electrode 16 is in the form of a solid circle. A central subelectrode of the second electrode 18 forms an annular ring around the central

170 subelectrode of the first electrode 16. A second subelectrode of the first electrode 16 forms an annular ring around the central electrode of the second electrode 18, and so on.

Under the illustrated embodiment of FIG. 5, the

175 aperture 26 of the diode 12 is a composite structure. The area of the composite structure (and an estimate for a driving current potential) may be found by integrating the area between opposing electrodes 16, 18.

180 Connection of the diode 12 with the circuit of

FIG. 1 may be made using a number of structures. For example, a 50 Ohm parallel conductor transmission line structure may be used. Similarly a 75 Ohm transmission line may be used.

185 Under the illustrated embodiment, the diode 12 is constructed with an improved structure (FIG. 3) which predominantly restricts the generation and transportation of the charge pairs to a relatively shallow region, essentially confining the holes and 190 electrons to a thin two-dimensional layer (i.e., within active layer 56) . The voltage potential (and shape and position of the electrodes 16, 18 forming a single parallel plane over the active region 56) creates a tightly confined gas of electrons and 195 holes. Placing the electrodes 16, 18 in close proximity on a common surface plane provides an aperture of a relatively small diameter (e.g., 50 nm) . Placing a relatively low voltage across the electrodes 16, 18 200 (e.g., ±2 volts) provides a relatively large field gradient (i.e., 8 X 10^ volts/cm) along the length of the active region. The large field gradient causes rapid movement of the charge pairs to their respective electrodes 16, 18.

205 The Bragg reflecting structure 52 is provided to enhance the probability of additional pair generation within a short time span. Typically, this allows only a single reflection of an unconverted IR photon 24. A single bounce of an IR photon 24 back into the

210 active region still maintains the fast response time of the diode 12, yet still significantly improves detector sensitivity.

Confining the holes and electrons to a thin channel of high mobility is similar to the technology

215 used in the non-analogous art of the manufacture of high electron mobility transport (HEMT) devices. While the HEMT technology used in construction of the diode 12 is somewhat similar to other prior art HEMT devices, the mechanism of operation is different.

220 For example, HEMT is typically used to carry, charge across an active region, not parallel to it. As a consequence, it is believed that the concept of a two-dimensional electron gas has not been recognized or used in prior art HEMT devices . 225 Further, the creation of the two-dimensional gas is dependent upon a number of additional factors not recognized in the HEMT technology as it currently exists. For example, the creation and control of the electron gas has been found to be dependent upon the 230 thickness of the active layer and what is underneath it . The InAlAs of the isolation layer 54 was chosen for use with the InGaAs of the active layer 56 to provide a bandgap discontinuity which prevents charge carriers generated in the isolation layer 54 from 235 contributing to the diode current.

Another factor important to the generation of the electron gas is the boundary conditions of the electrostatic potential. It has been found that the primary location for creation of the electron gas is 240 at the bottom of the active region 56 adjacent the isolation region 54. An expeditious selection of overlying layers 58, 60, 62 is made to increase speed by avoiding sharp heterojunctions between the electron gas and the metal electrodes 16, 18. 245 Note that half of the E-field generated by the differing polarity of the metal electrodes 16, 18 (i.e., the Schottky diodes) is in the isolation layer, thus potentially lowering the effective capacity per unit area. Even though the E-field 250 generated by the metal electrodes may penetrate into the isolation layer, the electrons (and somewhat slower holes) are confined to move along an HEMT-like interface channel with high mobility due to available energy sub-bands. The energy diagram of FIG. 4 255 illustrates this concept.

In an alternate embodiment, the invention may be extended to a laser signal in the 810 nm region by the use of the substrate 14 shown in FIG. 6. The substrate 14 of FIG. 6 also relies upon the 260 principles discussed above and upon the use of the two-dimensional electron gas. As shown, the substrate 14 of FIG. 6 is based upon a S.I. substrate 70 of GaAs . A buffer layer 72 of GaAs may be disposed on the substrate 70 to a 265 depth of 500 Angstrom. An AlGaAs isolation layer 74 may be disposed on top of the buffer layer 72 to a depth of 0.5 microns. An active layer 76 of GaAs may be disposed on the isolation layer 74 to a depth of 0.12 microns. As above, the AlGaAs of the isolation 270 layer 74 and the GaAs of the buffer layer were chosen for use with the GaAs of the active layer 76 to provide a bandgap discontinuity which prevents charge carriers generated in the isolation layer 74 from contributing to the diode current. 275 A cap layer 78 of AlGaAs may be disposed over the active layer 76 to a depth of 100 Angstroms. Finally, an overlayer 80 of GaAs may be disposed over the cap layer 80 to a depth of 50 Angstroms. As above, the overlying layers 78, 80 are selected to 280 facilitate speed by avoiding sharp heterojunctions between the area of the electron gas and the metal electrodes 16, 18.

The diode 12 has been found to provide a structure of extremely low capacitance and 285 inductance. The low capacitance and inductance has been found to offer the potential of a six-fold speed increase over prior art devices. FIG. 7, for example, shows a plot of diode performance as a function of speed for a diode with a 0.1 urn aperture 290 gap and an active layer of 0.12 micron. As shown, the diode 12 offers a significant improvement over prior art devices .

A specific embodiment of a method and apparatus for detecting photons of light according to the 295 present invention has been described for the purpose of illustrating the manner in which the invention is made and used. It should be understood that the implementation of other variations and modifications of the invention and its various aspects will be

300 apparent to one skilled in the art, and that the invention is not limited by the specific embodiments described. Therefore, it is contemplated to cover the present invention any and all modifications, variations, or equivalents that fall within the true

305 spirit and scope of the basic underlying principles disclosed and claimed herein.

Claims

Claims 310 1. A metal-semiconductor-metal photonics detector with an light-detecting aperture, such detector comprising: a pair of electrodes disposed in a common plane on an outer surface of the detector on either side of 315 the aperture of the detector; an electric field disposed across the aperture from a first to a second electrode of the pair of electrodes; and a relatively thin active region of a 320 semiconductor material having a thickness of between 0.05 and 0.2 microns and adapted to form a two- dimensional electron gas in the electric field between the electrodes in the presence of light.

325 2. The detector as in claim 1 wherein the pair of electrodes further comprise a plurality of concentric circular structures.

3. The detector as in claim 1 wherein the plurality 330 of concentric circular structures wherein the first electrode further comprises a substantially solid circle and the second electrode an annular ring.

4. The detector as in claim 1 wherein the aperture 335 further comprises a diameter of 6-8 micrometer.

5. The detector as in claim 1 further comprising a 50 ohm transmission line coupled to the pair of electrodes .

340

6. The detector as in claim 1 further comprising a one-bounce Bragg reflector below the active region.

7. The detector as in claim 6 wherein the Bragg 345 reflector further comprises 40 periods of 0.23 microns per period.

8. The detector as in claim 6 further comprising an isolation layer of InAlAs disposed between the active

350 layer and Bragg reflector.

9. The detector as in claim 8 wherein the isolation layer further comprises a thickness of 0.4 micron.

355 10. The detector as in claim 1 wherein the relatively thin active layer further comprises InGaAs .

11. The detector as in claim 10 wherein the relatively thin active layer further comprises a

360 thickness of 0.075 microns.

12. The detector as in claim 1 further comprising a graded layer of In(GaAl)As disposed adjacent the active layer between the pair of electrodes and the

365 active layer.

13. The detector as in claim 12 wherein the graded layer further comprises a thickness of 200 Angstrom.

370 14. The detector as in claim 1 further comprising a dark current suppression layer of InAlAs disposed adjacent the In(GaAl)As layer between the In(GaAl)As layer and the pair of electrodes .

375 15. The detector as in claim 14 wherein the dark current suppression layer further comprises a thickness of 150 Angstrom.

16. The detector as in claim 1 further comprising a 380 cap layer of InP disposed adjacent the InAlAs layer between the InAlAs layer and the pair of electrodes .

17. The detector as in claim 16 wherein the cap layer further comprises a thickness of 50 Angstrom.

385

18. The detector as in claim 1 wherein the relatively thin active layer further comprises GaAs .

19. The detector as in claim 18 wherein the

390 relatively thin active layer further comprises a thickness of 0.12 microns.

20. The detector as in claim 1 further comprising a cap layer of AlGaAs disposed adjacent the active

395 layer between the pair of electrodes and the active layer .

21. The detector as in claim 20 wherein the cap layer further comprises a thickness of 100 Angstrom.

400

22. The detector as in claim 1 further comprising an overlayer of GaAs disposed adjacent the cap layer between the cap layer and the pair of electrodes.

405 23. The detector as in claim 22 wherein the overlayer further comprises a thickness of 50 Angstrom.