CA2024134C - Hydrogen plasma passivation of gaas - Google Patents
Hydrogen plasma passivation of gaasInfo
- Publication number
- CA2024134C CA2024134C CA 2024134 CA2024134A CA2024134C CA 2024134 C CA2024134 C CA 2024134C CA 2024134 CA2024134 CA 2024134 CA 2024134 A CA2024134 A CA 2024134A CA 2024134 C CA2024134 C CA 2024134C
- Authority
- CA
- Canada
- Prior art keywords
- plasma
- passivation
- gaas
- torr
- pressure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 229910052739 hydrogen Inorganic materials 0.000 title claims abstract description 13
- 239000001257 hydrogen Substances 0.000 title claims abstract description 13
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 title claims abstract description 12
- 238000002161 passivation Methods 0.000 title abstract description 24
- 239000004065 semiconductor Substances 0.000 claims abstract description 9
- 229910001218 Gallium arsenide Inorganic materials 0.000 claims description 18
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 10
- 239000000758 substrate Substances 0.000 claims description 6
- 239000000470 constituent Substances 0.000 claims description 2
- 239000007789 gas Substances 0.000 claims description 2
- 210000002381 plasma Anatomy 0.000 abstract description 29
- 230000001419 dependent effect Effects 0.000 abstract 1
- 238000011282 treatment Methods 0.000 description 15
- 230000004888 barrier function Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 235000012431 wafers Nutrition 0.000 description 7
- 238000005452 bending Methods 0.000 description 6
- 230000007423 decrease Effects 0.000 description 6
- 230000009467 reduction Effects 0.000 description 6
- 230000007774 longterm Effects 0.000 description 5
- 238000009832 plasma treatment Methods 0.000 description 5
- 238000005215 recombination Methods 0.000 description 5
- 230000006798 recombination Effects 0.000 description 5
- 150000002500 ions Chemical class 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000005284 excitation Effects 0.000 description 3
- 230000001965 increasing effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000002708 enhancing effect Effects 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 238000010849 ion bombardment Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- YZCKVEUIGOORGS-UHFFFAOYSA-N Hydrogen atom Chemical compound [H] YZCKVEUIGOORGS-UHFFFAOYSA-N 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012993 chemical processing Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000007598 dipping method Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 239000002019 doping agent Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 230000005669 field effect Effects 0.000 description 1
- 239000000383 hazardous chemical Substances 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 150000002431 hydrogen Chemical class 0.000 description 1
- 125000004435 hydrogen atom Chemical group [H]* 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000001782 photodegradation Methods 0.000 description 1
- 238000006862 quantum yield reaction Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006557 surface reaction Methods 0.000 description 1
- 238000004381 surface treatment Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
- H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
- H01L21/04—Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer
- H01L21/18—Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer the devices having semiconductor bodies comprising elements of Group IV of the Periodic Table or AIIIBV compounds with or without impurities, e.g. doping materials
- H01L21/30—Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26
- H01L21/3003—Hydrogenation or deuterisation, e.g. using atomic hydrogen from a plasma
- H01L21/3006—Hydrogenation or deuterisation, e.g. using atomic hydrogen from a plasma of AIIIBV compounds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Plasma & Fusion (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- General Physics & Mathematics (AREA)
- Manufacturing & Machinery (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Power Engineering (AREA)
- Formation Of Insulating Films (AREA)
- Semiconductor Lasers (AREA)
Abstract
Abstract of the Disclosure Hydrogen Plasma surface passivation of III-V Semiconductors is critically dependent on exposure time and pressure because of competition between plasma passivation and damage. Proper control of pressure according to the invention yields reproducible and stable passivation. Improved passivation is obtained using high pressure hydrogen plasmas, i.e. above 1 Torr.
Description
~2~134 HYDROGEN PLASMA PASSIVATION OF GaAs Field of the Invention The invention relates to surface passivation of III-V semiconductors by plasma processing.
5 Back~round of the Invention The well-known deficiencies in the surface properties of GaAs have stimulated many attempts to passivate the surface, reduce the surface state density, and unpin the surface Fermi level. There are many reasons why it is desirable toreduce the surface state density of III-V compound semiconductors and GaAs in 10 particular. A reduction in surface state density would lead to improved field effect transistor (FETs) by lowering surface-related leakage currents and by increasing the Schottky barrier height. This would allow the fabrication of digital logic circuits with increased noise margins and relaxed requirements on threshold voltage uniformities of the component Fl~Ts. In addition, unpinning the surface Fermi level 15 would allow advances in metal-insulator-semiconductor (MIS) capacitor and MISFET technology. For photonic devices, it is important for enhancing power output to reduce the surface recombination velocity and increase carrier lifetime.
Passivation may be carried out prior to processing as a cleaning step, or after processing to remove process-induced damage.
A variety of wet chemical and photochemical treatments have been used to passivate GaAs but controversy remains as to their effecdveness in unpinning the Fermi level. Nonetheless, it is clear ~that photoluminesence (PL) yield is enhanced by using these surface treatments. In fact, passivation is operationally defined in terms of this enhancement in quantum yield. Although large increases in PL yield25 have been observed, the enhancements tend to be short-lived. We are aware of only one case where the enhancement persisted for more than a few hours (See H. H. Lee, R. J. Racicot, and S. H. Lee, Appl. Phys. Lett. 54,724 (1989). Occasionally, real-time PL monitoring has been used to monitor the effectiveness and kinetics of these wet chemical treatments.
To improve process contro! and management of environmental hazards, it is desirable to replace wet chemical processing with dry processing. In particular, plasma methods are effective for not only etching but also deposition, cleaning, and passivadon. Recently, H2 plasmas operated at low pressure have been used to clean GaAs of residual As and AS2O3 . Similar plasnna treatments have also been reported 35 for InP surface cleaning. A variety of reactor configurations have been employed:
electron cyclrotron resonance (ecr) microwave, multi-polar, and rf diode. However, in these reports choice of optimal exposure times, which have ranged from S s to 30 .
^` 2024134 min, and operating pressures, which have ranged from 104 to 102 Torr, is not discussed. Similarly, the consequences o~ reactor geometry or excitation mode choice have not been addressed. In one case the properties oE Schottky diodes made on hydrogen plasma treated n-GaAs surfaces was studied as a function of the treatment temperature. The conclusion was that plasma passivation is ineffective Eor temperatures below 200C. (See A. Paccagnella). A similar conclusion has been made Eor hydrogen plasma treatment oE InP surfaces.
However, as we describe here, such conclusions depend on other plasma parameters. It is the recognition of the role of those parameters that leads to the inventive process.
Real-time monitoring of PL from GaAs during hydrogen plasma passivation clearly provides the means to investigate the efEects of processing conditions such as exposure time, pressure, and temperature. In addition, the kinetics of plasma-surface reactions can be monitored by measuring real-time changes in PL yield. In general, PL yield is sensitive to changes in surface recombination velocity (S) and band bending. Clearly, PL yield increases as S
decreases. In addition, if the bands are bent and a space-charge field exists near the surface, separation of electrons and holes can lead to a reduction in radiative recombination efficiency. Thus, a reduction in band bending can increase PL yield.
However, if S is reduced substantially, PL yield can increase despite an increase in band-bending. In n-GaAs, for example, passivation treatments tend to remove excess As from the surface and thereby reduce the As antisite deEect density that pins the Fermi level near mid-gap. The Fermi level shiEts to lower energy because the density of C~a antisite deEects remains constant or even increases. Thus, the band bending increases. Ordinarily this would cause a decrease in PL yield as carriers are separated by the space-charge Eield. However, reduction oE the near-mid-gap As antisite deEect density decreases the non-radiative recombination rate causing an increase in PL yield.
Statement of the In~ention In accordance with one aspect of the invention there is provided a method for passivating the surface oE a III-V semiconductor comprising the stepsti of exposing the surface oE a gas plasma having a major constituent of hydrogen for ; ~
.
.
2~2~134 a period sufficient to passivate the surface, the invention characterized in that said plasma has a pressure greater than 1.5 Torr, and the temperature of the surface is less than 100 degrees centigrade for the duration of said exposure to hydrogen plasma.
In our discovery we used PL to monitor the efEectiveness of an optimize processing conditions for hydrogen plasma passivation of native-oxide contaminated GaAs. We find that previously reported pressures and exposure times for passivation are far from optimal, and that low-temperature passivation is feasible. The effects of varying pressure and exposure time are critical.
Specifically, we discovered that at pressures above 1 Torr, permanent plasma-induced surface damage does not occur (regardless of exposure time). We also discovered that effective passivation can be obtained at processing temperaturesbelow 100C.
-2a-,~
.. ..
2~2~
Detai!ed Descri-ption FIG. 1 shows PL intensity (in arbitrary units) vs. time for a passivation treatment at 0.2 Torr.
FIG. 2 shows PL intensity vs. time for an indentical passivation 5 treatment except that the pressure is increased to 4 Torr.
FIG. 3 shows PL intensity vs. time for a passivation treatment optimized for treatment time (i.e., where monitored PL peaks).
FIG. 4 is a plot of PL intensity vs. time in days, showing the long-term stability of surfaces treated in accordance with the invention.
Samples were either semi-insulating or n-type (10l7cm~3) t}a~s (100) substrates. Initially, samples were inserted into a parallel-plate or hexode plasma reactor without any preprocess treatment. However, the magnitude of the PL
enhancement and its long-term stability are widely variable from wafer to wafer,although for a given wafer the reproducibility was excellent; within a wafer lot, 15 reproducibility was better than between lots. We found that when wafers are first dipped in concentrated (30%) HF or N~,OH for 30 s, excellent PL yield reproducibility and long-term stability are observed. Therefore, results presented here correspond to samples first treated by dipping for 30 s iQ HF.
For PL monitoring, a pulsed N2 pumped dye laser (Molectron) beam is 20 injected into a fiber optic cable and passed through a polarizer to select p-polarized light. The output from the fiber is focussed onto the sample to spot size of about 0.15cm2 (1 cm long x 0.15 cm wide). The angle of incidence is -83 from normal (close to Brewster's angle) so that most of the light is efficiently coupled into the substrate. The laser pulse energy is typically 100~J of 500 nm light. PL is collected, 25 collimated, and then focussed onto the entrance slit of a scanning monochromator.
Pulsed laser excitadon is preferrable to continuous-wave excitation since the PL signal is superlinear with laser pulse energy. The non-linearity isbelieved to result from high-level injection and flattening of the bands. Thus, short-pulse, high intensity excitation with gated electronic detection eliminates the plasma 30 background glow. In addition, a low duty-cycle, pulsed laser deposits less energy into the sample and thereby minimizes heating and photodegradation. Samples are plasma passivated in a parallel-plate, capacitively-coupled reactor.
When GaAs is exposed to an H2 plasma, H atoms created in the plasma react with native oxides and reduce the surface As concentration and thereby 35 increase the PL yield. The plasma also produces energetic ion bombardment of the surface and can create damage that increases non-radiative recombination and decreases PL yield. Both of these effects are evident in FIGS. 1 and 2. In FIG. 1 a . . . . .
.
': - - ' ' . , ' ' .
~ 20241 34 GaAs substrate is dipped in HF to thin the native oxide and then exposed to a hydrogen discharge with plasma parameters: 13 MHz, 20 W, 20C, 2 sccm H2, 0.2 Torr.
When the plasma is gated on, the PL yield first decreases the surface charges negatively, then increases rapidly as As is removed from the surface, and finally S plummets to a steady-state level below the initial value as the plasma irreversibly damages the surface. When the plasma is gated of~, a slight increase in PL yield occurs and is attributed to discharging of the surface.
Similar results are obtained over a range oE substrate temperatures (10 -50C), ~low rates (0.4 - 40 sccm), and rf powers (4 - 20 W). Only pressure has asubstantial effect on the magnitude and time-dependence of the PL enhancement (compare FIGS. 1 and 2). At higher pressures, sheath voltages are smaller, sheaths are more collisional, ion energies are smaller, and damage does not readily occur. As a result, PL enhancement is much larger at higher pressure and the decrease in PL that occurs on over-exposure is attributed solely to reversible charging of the surEace by the plasma. Exposure time at pressures above about 1 Torr is not critical as can be seen by comparing final PL levels in FIGS. 2 and 3. The exposure time in the treatment yielding the data of FIG. 3 is optimized using PL monitoring. In the treatments yielding the data of FIGS. 1 and 2 the surfaces were "overexposed". The overexposure permanently damages the surface when processed at 0.2 Torr. At 4 Torr no permanent damage occurs even though the surface may be overexposed. The pressureat which this critical transition occurs is between 0.5 and 1.5 Torr and the pressure recommended is above 1.0 Torr. Thus according to the invention surface passivation is optimized and is unexpectedly made independent of exposure time through the use of pressure levels above those used in the prior art.
For plasma treatments under the conditions of FIGS. 2 and 3 - high pressure, short times, low temperature only the surface should be passivated (i.e. ~ 1000 A). Bulk passivation is not likely to play an importnnt role in enhancing the PL yield. For a peak to peak voltage of 380V, typical oE our operating conditions, the time-averaged sheath voltage that ions can respond to in being accelerated to the surface is only #120V. At 4 Torr, ions experience a signiEicant number oE chargeexchange collisions as they traverse the ~s0.5mm thick sheath and their average energy is reduced further to ~50V. With such low ion energies, the penetration of ionichydrogen into the bulk beyond a few hundred A can be neglected. Similarly, at the low temperature employed here (< 50C), diffusion of atomic hydrogen beyond 100 ~ can be neglected. These conditions are quite different Erom those typically employed for bulk dopant or defect passivation - long exposures, high temperature, ~ s _ 4 _ 1,., . .
.. .
... .
\
~24~3~
energetic ion bombardment.
To further evaluate the consequences of low-temperature (less than 100C) H2 plasma passivation, we treated n-type GaAs with an ohmic contact on the back side of the wafer. On the front side, Schottky diodes were fabricated both prior 5 to and after plasma treatment. No change was observed in the chracteristics of the diodes forrned prior to plasma treatment. However, for diodes fabricated on passivated areas of the same wafer (i.e. after plasma treatment), the barrier heights showed small but significant increases (Table 1). Consistent with a reduction insurface state density, we found barrier heights to depend slightly on the metal used to 10 make the contact. T~e barrier heights are also close to those reported in the literature for clean GaAs prepared under UHV conditions. Note that such clean surfaces are still thought to be "pinned".
Schottky Barrier Heights for n-GaAs 15 Metal n ~ (eV) ~ (eV) (cleanGaAs)b Au 1.06 0.89 0.88 Ti 1.01 0.84 0.83 Pt 1.02 0.85 a For untreated n-GaAs: n=1.02 and ~ = 0.75 eV, where n = ideality factor, ~ = barrier height.
b E. H. Rhoderick and R. H. Williams, Metal-Semiconductor ContactsLClaredon 25 Press (Oxford 1988), p. 70.
The increases in Schottky barrier height shown in Table I are significant and useful for reducing leakage current in MES devices. The increases are also consistent with an increase in band-bending as the Fermi level shifts to a lowerpinning position closer to the Ga antisite defect. Evidently, the increase in band 30 bending is more than compensated for by the reduction in the near mid-gap As antisite defect state density.
With few exceptions, most plasma passivation treatments reported in the literature do not exhibit good long-term stability. In most cases, PL intensity degrades within hours after treatment. Seven samples were treated in accordance 35 with the invention and mentioned for long term stability. The results appear in FIG.
5 Back~round of the Invention The well-known deficiencies in the surface properties of GaAs have stimulated many attempts to passivate the surface, reduce the surface state density, and unpin the surface Fermi level. There are many reasons why it is desirable toreduce the surface state density of III-V compound semiconductors and GaAs in 10 particular. A reduction in surface state density would lead to improved field effect transistor (FETs) by lowering surface-related leakage currents and by increasing the Schottky barrier height. This would allow the fabrication of digital logic circuits with increased noise margins and relaxed requirements on threshold voltage uniformities of the component Fl~Ts. In addition, unpinning the surface Fermi level 15 would allow advances in metal-insulator-semiconductor (MIS) capacitor and MISFET technology. For photonic devices, it is important for enhancing power output to reduce the surface recombination velocity and increase carrier lifetime.
Passivation may be carried out prior to processing as a cleaning step, or after processing to remove process-induced damage.
A variety of wet chemical and photochemical treatments have been used to passivate GaAs but controversy remains as to their effecdveness in unpinning the Fermi level. Nonetheless, it is clear ~that photoluminesence (PL) yield is enhanced by using these surface treatments. In fact, passivation is operationally defined in terms of this enhancement in quantum yield. Although large increases in PL yield25 have been observed, the enhancements tend to be short-lived. We are aware of only one case where the enhancement persisted for more than a few hours (See H. H. Lee, R. J. Racicot, and S. H. Lee, Appl. Phys. Lett. 54,724 (1989). Occasionally, real-time PL monitoring has been used to monitor the effectiveness and kinetics of these wet chemical treatments.
To improve process contro! and management of environmental hazards, it is desirable to replace wet chemical processing with dry processing. In particular, plasma methods are effective for not only etching but also deposition, cleaning, and passivadon. Recently, H2 plasmas operated at low pressure have been used to clean GaAs of residual As and AS2O3 . Similar plasnna treatments have also been reported 35 for InP surface cleaning. A variety of reactor configurations have been employed:
electron cyclrotron resonance (ecr) microwave, multi-polar, and rf diode. However, in these reports choice of optimal exposure times, which have ranged from S s to 30 .
^` 2024134 min, and operating pressures, which have ranged from 104 to 102 Torr, is not discussed. Similarly, the consequences o~ reactor geometry or excitation mode choice have not been addressed. In one case the properties oE Schottky diodes made on hydrogen plasma treated n-GaAs surfaces was studied as a function of the treatment temperature. The conclusion was that plasma passivation is ineffective Eor temperatures below 200C. (See A. Paccagnella). A similar conclusion has been made Eor hydrogen plasma treatment oE InP surfaces.
However, as we describe here, such conclusions depend on other plasma parameters. It is the recognition of the role of those parameters that leads to the inventive process.
Real-time monitoring of PL from GaAs during hydrogen plasma passivation clearly provides the means to investigate the efEects of processing conditions such as exposure time, pressure, and temperature. In addition, the kinetics of plasma-surface reactions can be monitored by measuring real-time changes in PL yield. In general, PL yield is sensitive to changes in surface recombination velocity (S) and band bending. Clearly, PL yield increases as S
decreases. In addition, if the bands are bent and a space-charge field exists near the surface, separation of electrons and holes can lead to a reduction in radiative recombination efficiency. Thus, a reduction in band bending can increase PL yield.
However, if S is reduced substantially, PL yield can increase despite an increase in band-bending. In n-GaAs, for example, passivation treatments tend to remove excess As from the surface and thereby reduce the As antisite deEect density that pins the Fermi level near mid-gap. The Fermi level shiEts to lower energy because the density of C~a antisite deEects remains constant or even increases. Thus, the band bending increases. Ordinarily this would cause a decrease in PL yield as carriers are separated by the space-charge Eield. However, reduction oE the near-mid-gap As antisite deEect density decreases the non-radiative recombination rate causing an increase in PL yield.
Statement of the In~ention In accordance with one aspect of the invention there is provided a method for passivating the surface oE a III-V semiconductor comprising the stepsti of exposing the surface oE a gas plasma having a major constituent of hydrogen for ; ~
.
.
2~2~134 a period sufficient to passivate the surface, the invention characterized in that said plasma has a pressure greater than 1.5 Torr, and the temperature of the surface is less than 100 degrees centigrade for the duration of said exposure to hydrogen plasma.
In our discovery we used PL to monitor the efEectiveness of an optimize processing conditions for hydrogen plasma passivation of native-oxide contaminated GaAs. We find that previously reported pressures and exposure times for passivation are far from optimal, and that low-temperature passivation is feasible. The effects of varying pressure and exposure time are critical.
Specifically, we discovered that at pressures above 1 Torr, permanent plasma-induced surface damage does not occur (regardless of exposure time). We also discovered that effective passivation can be obtained at processing temperaturesbelow 100C.
-2a-,~
.. ..
2~2~
Detai!ed Descri-ption FIG. 1 shows PL intensity (in arbitrary units) vs. time for a passivation treatment at 0.2 Torr.
FIG. 2 shows PL intensity vs. time for an indentical passivation 5 treatment except that the pressure is increased to 4 Torr.
FIG. 3 shows PL intensity vs. time for a passivation treatment optimized for treatment time (i.e., where monitored PL peaks).
FIG. 4 is a plot of PL intensity vs. time in days, showing the long-term stability of surfaces treated in accordance with the invention.
Samples were either semi-insulating or n-type (10l7cm~3) t}a~s (100) substrates. Initially, samples were inserted into a parallel-plate or hexode plasma reactor without any preprocess treatment. However, the magnitude of the PL
enhancement and its long-term stability are widely variable from wafer to wafer,although for a given wafer the reproducibility was excellent; within a wafer lot, 15 reproducibility was better than between lots. We found that when wafers are first dipped in concentrated (30%) HF or N~,OH for 30 s, excellent PL yield reproducibility and long-term stability are observed. Therefore, results presented here correspond to samples first treated by dipping for 30 s iQ HF.
For PL monitoring, a pulsed N2 pumped dye laser (Molectron) beam is 20 injected into a fiber optic cable and passed through a polarizer to select p-polarized light. The output from the fiber is focussed onto the sample to spot size of about 0.15cm2 (1 cm long x 0.15 cm wide). The angle of incidence is -83 from normal (close to Brewster's angle) so that most of the light is efficiently coupled into the substrate. The laser pulse energy is typically 100~J of 500 nm light. PL is collected, 25 collimated, and then focussed onto the entrance slit of a scanning monochromator.
Pulsed laser excitadon is preferrable to continuous-wave excitation since the PL signal is superlinear with laser pulse energy. The non-linearity isbelieved to result from high-level injection and flattening of the bands. Thus, short-pulse, high intensity excitation with gated electronic detection eliminates the plasma 30 background glow. In addition, a low duty-cycle, pulsed laser deposits less energy into the sample and thereby minimizes heating and photodegradation. Samples are plasma passivated in a parallel-plate, capacitively-coupled reactor.
When GaAs is exposed to an H2 plasma, H atoms created in the plasma react with native oxides and reduce the surface As concentration and thereby 35 increase the PL yield. The plasma also produces energetic ion bombardment of the surface and can create damage that increases non-radiative recombination and decreases PL yield. Both of these effects are evident in FIGS. 1 and 2. In FIG. 1 a . . . . .
.
': - - ' ' . , ' ' .
~ 20241 34 GaAs substrate is dipped in HF to thin the native oxide and then exposed to a hydrogen discharge with plasma parameters: 13 MHz, 20 W, 20C, 2 sccm H2, 0.2 Torr.
When the plasma is gated on, the PL yield first decreases the surface charges negatively, then increases rapidly as As is removed from the surface, and finally S plummets to a steady-state level below the initial value as the plasma irreversibly damages the surface. When the plasma is gated of~, a slight increase in PL yield occurs and is attributed to discharging of the surface.
Similar results are obtained over a range oE substrate temperatures (10 -50C), ~low rates (0.4 - 40 sccm), and rf powers (4 - 20 W). Only pressure has asubstantial effect on the magnitude and time-dependence of the PL enhancement (compare FIGS. 1 and 2). At higher pressures, sheath voltages are smaller, sheaths are more collisional, ion energies are smaller, and damage does not readily occur. As a result, PL enhancement is much larger at higher pressure and the decrease in PL that occurs on over-exposure is attributed solely to reversible charging of the surEace by the plasma. Exposure time at pressures above about 1 Torr is not critical as can be seen by comparing final PL levels in FIGS. 2 and 3. The exposure time in the treatment yielding the data of FIG. 3 is optimized using PL monitoring. In the treatments yielding the data of FIGS. 1 and 2 the surfaces were "overexposed". The overexposure permanently damages the surface when processed at 0.2 Torr. At 4 Torr no permanent damage occurs even though the surface may be overexposed. The pressureat which this critical transition occurs is between 0.5 and 1.5 Torr and the pressure recommended is above 1.0 Torr. Thus according to the invention surface passivation is optimized and is unexpectedly made independent of exposure time through the use of pressure levels above those used in the prior art.
For plasma treatments under the conditions of FIGS. 2 and 3 - high pressure, short times, low temperature only the surface should be passivated (i.e. ~ 1000 A). Bulk passivation is not likely to play an importnnt role in enhancing the PL yield. For a peak to peak voltage of 380V, typical oE our operating conditions, the time-averaged sheath voltage that ions can respond to in being accelerated to the surface is only #120V. At 4 Torr, ions experience a signiEicant number oE chargeexchange collisions as they traverse the ~s0.5mm thick sheath and their average energy is reduced further to ~50V. With such low ion energies, the penetration of ionichydrogen into the bulk beyond a few hundred A can be neglected. Similarly, at the low temperature employed here (< 50C), diffusion of atomic hydrogen beyond 100 ~ can be neglected. These conditions are quite different Erom those typically employed for bulk dopant or defect passivation - long exposures, high temperature, ~ s _ 4 _ 1,., . .
.. .
... .
\
~24~3~
energetic ion bombardment.
To further evaluate the consequences of low-temperature (less than 100C) H2 plasma passivation, we treated n-type GaAs with an ohmic contact on the back side of the wafer. On the front side, Schottky diodes were fabricated both prior 5 to and after plasma treatment. No change was observed in the chracteristics of the diodes forrned prior to plasma treatment. However, for diodes fabricated on passivated areas of the same wafer (i.e. after plasma treatment), the barrier heights showed small but significant increases (Table 1). Consistent with a reduction insurface state density, we found barrier heights to depend slightly on the metal used to 10 make the contact. T~e barrier heights are also close to those reported in the literature for clean GaAs prepared under UHV conditions. Note that such clean surfaces are still thought to be "pinned".
Schottky Barrier Heights for n-GaAs 15 Metal n ~ (eV) ~ (eV) (cleanGaAs)b Au 1.06 0.89 0.88 Ti 1.01 0.84 0.83 Pt 1.02 0.85 a For untreated n-GaAs: n=1.02 and ~ = 0.75 eV, where n = ideality factor, ~ = barrier height.
b E. H. Rhoderick and R. H. Williams, Metal-Semiconductor ContactsLClaredon 25 Press (Oxford 1988), p. 70.
The increases in Schottky barrier height shown in Table I are significant and useful for reducing leakage current in MES devices. The increases are also consistent with an increase in band-bending as the Fermi level shifts to a lowerpinning position closer to the Ga antisite defect. Evidently, the increase in band 30 bending is more than compensated for by the reduction in the near mid-gap As antisite defect state density.
With few exceptions, most plasma passivation treatments reported in the literature do not exhibit good long-term stability. In most cases, PL intensity degrades within hours after treatment. Seven samples were treated in accordance 35 with the invention and mentioned for long term stability. The results appear in FIG.
4. The seven different symbols represent data points for the seven samples. As the data in FIG. 4 show, the long-terrn stability of the treatment of this invention here is excellent. Within the first few days, the PL intensity drops between 10 and 20% of its peak value (which is usually more than 10 times the initial value). After this 2 ~
period, the PL intensity remains constant within our experimental precision for at least 40 days.
period, the PL intensity remains constant within our experimental precision for at least 40 days.
. . .
:~ .
. .
:~ .
. .
Claims (5)
1. A method for passivating the surface of a III-V semiconductor comprising the steps of exposing the surface of a gas plasma having a major constituent of hydrogen for a period sufficient to passivate the surface, the invention characterized in that said plasma has a pressure greater than 1.5 Torr, and the temperature of the surface is less than 100 degrees centigrade for the duration of said exposure to hydrogen plasma.
2. The method of claim 1 in which the III-V semiconductor comprises gallium arsenide.
3. The method of claim 2 in which the semiconductor is gallium arsenide.
4. The method according to claim 1 wherein said pressure is at least 4 Torr and said substrate comprises gallium arsenide.
5. The method according to claim 1 wherein said temperature is less than 50 degrees centigrade and said substrate comprises gallium arsenide.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US40703089A | 1989-09-01 | 1989-09-01 | |
| US407,030 | 1989-09-01 | ||
| US47701290A | 1990-02-07 | 1990-02-07 | |
| US477,012 | 1990-02-07 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2024134A1 CA2024134A1 (en) | 1991-03-02 |
| CA2024134C true CA2024134C (en) | 1993-10-05 |
Family
ID=27019735
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA 2024134 Expired - Fee Related CA2024134C (en) | 1989-09-01 | 1990-08-28 | Hydrogen plasma passivation of gaas |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA2024134C (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2164094A1 (en) * | 2008-09-16 | 2010-03-17 | Imec | A method for reducing fermi-level-pinning in a non-silicon channel mos device |
| WO2012013965A1 (en) * | 2010-07-26 | 2012-02-02 | Seren Photonics Limited | Light emitting diodes |
-
1990
- 1990-08-28 CA CA 2024134 patent/CA2024134C/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2164094A1 (en) * | 2008-09-16 | 2010-03-17 | Imec | A method for reducing fermi-level-pinning in a non-silicon channel mos device |
| WO2012013965A1 (en) * | 2010-07-26 | 2012-02-02 | Seren Photonics Limited | Light emitting diodes |
Also Published As
| Publication number | Publication date |
|---|---|
| CA2024134A1 (en) | 1991-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Meneghesso et al. | Anomalous kink effect in GaN high electron mobility transistors | |
| Gottscho et al. | Real‐time monitoring of low‐temperature hydrogen plasma passivation of GaAs | |
| US20080105880A1 (en) | SILICON NITRIDE PASSIVATION WITH AMMONIA PLASMA PRETREAMENT FOR IMPROVING RELIABILITY OF AlGaN/GaN HEMTs | |
| US5179029A (en) | Hydrogen plasma passivation of GaAs | |
| Huh et al. | Effective sulfur passivation of an n-type GaN surface by an alcohol-based sulfide solution | |
| Zolper et al. | Implantation and dry etching of group-III-nitride semiconductors | |
| Ren et al. | Effect of dry etching on surface properties of III-nitrides | |
| Cao et al. | Plasma damage in p-GaN | |
| Mehdi et al. | Investigation of N2 plasma GaAs surface passivation efficiency against air exposure: Towards an enhanced diode | |
| EP0441024B1 (en) | Hydrogen plasma passivation of GaAs | |
| CA2024134C (en) | Hydrogen plasma passivation of gaas | |
| Lootens et al. | Study of electrical damage in GaAs induced by SiCl4 reactive ion etching | |
| Sugino et al. | Phosphidization of GaAs by a remote phosphine plasma process and its application to surface passivation of GaAs metal–semiconductor field-effect transistors | |
| US5880483A (en) | Semiconductor devices | |
| KR950007955B1 (en) | Electrode forming method of mes structure | |
| Kretschmer et al. | Interelectrode metal migration on GaAs | |
| Zhang et al. | Cl2/Ar High‐Density‐Plasma Damage in GaN Schottky Diodes | |
| Beister et al. | Stability of sulfur-passivated facets of InGaAs-AlGaAs laser diodes | |
| Lau et al. | Study of electron traps in semi‐insulating gallium‐arsenide buffer layers for the suppression of backgating by the zero‐bias thermally stimulated current technique | |
| Aydil et al. | Plasma passivation of III-V semiconductor surfaces | |
| Smith et al. | The effects of plasma and ion beam processing on the properties of n-GaAs Schottky diodes | |
| Xu et al. | Se chemical passivation and annealing treatment for GaAs Schottky diode | |
| Meyer et al. | Plasma surface pretreatment effects on silicon nitride passivation of AlGaN/GaN HEMTs | |
| Kim et al. | Reduction in leakage current of InGaN-based light-emitting diodes by N2O plasma passivation | |
| US6228672B1 (en) | Stable surface passivation process for compound semiconductors |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed |