Classifications

• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D30/00—Field-effect transistors [FET]
  • H10D30/01—Manufacture or treatment
  • H10D30/015—Manufacture or treatment of FETs having heterojunction interface channels or heterojunction gate electrodes, e.g. HEMT
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D30/00—Field-effect transistors [FET]
  • H10D30/40—FETs having zero-dimensional [0D], one-dimensional [1D] or two-dimensional [2D] charge carrier gas channels
  • H10D30/47—FETs having zero-dimensional [0D], one-dimensional [1D] or two-dimensional [2D] charge carrier gas channels having two-dimensional [2D] charge carrier gas channels, e.g. nanoribbon FETs or high electron mobility transistors [HEMT]
  • H10D30/471—High electron mobility transistors [HEMT] or high hole mobility transistors [HHMT]
  • H10D30/473—High electron mobility transistors [HEMT] or high hole mobility transistors [HHMT] having confinement of carriers by multiple heterojunctions, e.g. quantum well HEMT
  • H10D30/4732—High electron mobility transistors [HEMT] or high hole mobility transistors [HHMT] having confinement of carriers by multiple heterojunctions, e.g. quantum well HEMT using Group III-V semiconductor material
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D64/00—Electrodes of devices having potential barriers
  • H10D64/01—Manufacture or treatment
  • H10D64/012—Manufacture or treatment of electrodes comprising a Schottky barrier to a semiconductor
  • H10D64/0124—Manufacture or treatment of electrodes comprising a Schottky barrier to a semiconductor to Group III-V semiconductors
  • H10D64/0125—Manufacture or treatment of electrodes comprising a Schottky barrier to a semiconductor to Group III-V semiconductors characterised by the sectional shape, e.g. T or inverted T

Definitions

• This invention relates generally to pseudomorphic high electron mobility transistors (PHEMTs) and more particularly to transistors of such type which are formed on III-V substrates.
• MOSFET metal electrode semiconductor field effect transistor
• HEMT high electron mobility transistor
• PHEMT pseudomorphic high electron mobility transistors
• Group III-V materials such as gallium arsenide (GaAs) or indium phosphide (InP) .
• GaAs gallium arsenide
• InP indium phosphide
• a heterojunction is formed between the doped donor layer and the undoped channel layer.
• This heterojunction provides spatial separation of electrons which are injected from the doped donor layer into the undoped channel layer.
• electrons from the large bandgap donor layer are transferred into the narrow bandgap channel layer where they are confined to move only in a plane parallel to the heterojunction. This results in the formation of a two-dimensional electron so-called "gas" .
• conduction takes place in the undoped channel, impurity scattering is reduced in this undoped layer and electron mobility is thereby enhanced compared to the doped channel structure used in MESFETs. Accordingly, HEMTs provide higher frequency operation than MESFETs.
• one type of PHEMT includes a gallium arsenide substrate having formed thereon successive layers of : an undoped InGaAs channel layer; a doped barrier (donor) AlGaAs layer; and n- GaAs and n+ GaAs ohmic contact layers, as shown in FIG. 1A.
• a layer of photoresist is then deposited over the structure and patterned to have an aperture over a portion of the structure to expose a region where the gate electrode is to be formed.
• an etch is brought into contact with the portions exposed by the aperture to successively etch through portions of the n+ GaAs and n- GaAs layers and partially into the AlGaAs layer, as shown in FIG. 1A for a wet etch and FIG. IB for a dry etch.
• a relatively wide recess is formed in the n+ GaAs and n- GaAs ohmic contact layers thereby improving the breakdown voltage of the FET.
• the dry etch has better selectivity and less undercut than a wet etch; however, the dry etch always causes some damage on the surface layer being etched which may induce more unwanted surface states .
• the photoresist is stripped and another layer of photoresist is deposited over the structure and patterned to define the narrow gate recess and gate metalization (i.e., the gate electrode) in Schottky contact with the AlGaAs channel layer as shown in FIG. 1C (when the wet etch is used to form the wide recess) , or in FIG. ID (when the dry etch is used to form the wide recess) .
• the AlGaAs PHEMT shown in either FIG.
• this narrow recess is performed with a wet chemical etch by a timed etch which is checked by measuring the open channel current between the source S and drain D electrodes.
• a gate metal is then deposited over the photoresist and through the electron beam patterned aperture formed therein onto the exposed portion of the aluminum gallium arsenide layer. After the photoresist layer and extraneous metal thereon are lifted-off, the gate electrode G is formed resulting FET - is shown in FIG. IE for the wet etch process and FIG. IF for the dry etch process.
• the use of the second photoresist layer on an InGaP surface and the use of a wet etch are not compatible. More particularly, the wet etch used for etching InGaP are solutions containing strong acids. These strong acids cause sever undercutting in the photoresist resulting in a complete loss of the InGaP surface layer. Further, it is noted from FIGS. IE and IF, that there is a significant ungated recesses in regions U which results from either of these processes.
• a method for forming a gate of a field effect transistor includes providing a structure having: a III-V substrate; a channel layer over the substrate; a doped barrier layer over the channel layer; a protective layer disposed on the donor layer; an etch stop layer disposed over the protective layer; and source and drain contact layer disposed over the etch stop layer.
• a mask is provided over the surface of the structure to expose a surface portion of the contact layer. The exposed surface portion of the contact layer is exposed to a first etch which etches through the contact layer to expose an underlying surface portion of the etch stop layer.
• the first etch etches the contact layer at a substantially greater etch rate than the etch rate of such etch to the etch stop layer.
• the exposed surface portion of the etch stop layer is exposed to a second etch which etches through the etch stop layer to expose an underlying surface portion of the protective layer.
• the second etch etches the etch stop layer at a substantially greater etch rate than the etch rate of such second etch to the protective layer.
• a metal is deposited over the mask and through etched portions of the etch stop layer onto the exposed portion of the protective layer.
• the gate recess is formed by selective wet etching.
• no dry etch is used and therefore there is no dry etching induced damage.
• the selectivity of the etching there is no need to measure the channel current to determine the etch end point resulting in better uniformity, better reproducibility and a less labor intensive process.
• a method for forming a gate of a field effect transistor includes providing a structure having: a III-V substrate; a channel layer over the substrate; a doped barrier layer over the channel layer; a protective layer disposed on the donor layer; an etch stop layer disposed over the protective layer; and source and drain contact layer disposed over the etch stop layer.
• a mask is provided over the surface of the structure, such mask having an aperture therein to expose a surface portion of the contact layer.
• the exposed surface portion of the contact layer is subjected to a first etch to etch through the contact layer to expose an underlying surface portion of the etch stop layer, such first etch etching the contact layer at a substantially greater etch rate than the etch rate of such etch to the etch stop layer.
• the aperture in the mask is enlarged.
• the exposed surface portion of the etch stop layer exposed by the enlarged aperture is subjected to a second etch to etch through the etch stop layer to expose an underlying surface portion of the protective layer, such second etch etching the etch stop layer at a substantially greater etch rate than the etch rate of such second etch to the protective layer.
• a metal is deposited over the mask and through etched portions of the etch stop layer onto the exposed portion of the protective layer.
• the substrate is gallium arsenide
• the channel layer is indium gallium arsenide
• the protective layer is a material different from the etch stop layer.
• the protective layer is the same material as the source and drain contact layer.
• the protective layer has a thickness in the order of 10 to 20 Angstroms.
• the protective layer is a III-V material.
• the protective layer is gallium arsenide.
• the etch stop layer is indium gallium phosphide.
• a field effect transistor in accordance with another feature of the invention.
• the transistor includes: a III-V substrate; a channel layer over the substrate; a doped barrier layer over the channel layer; an additional layer disposed over the barrier layer, such additional layer having an aperture therein; source and drain contact layer disposed over the laterally spaced regions of the additional layer; and a gate electrode having side portions thereof disposed in the aperture, walls of such aperture being in contact with such side portions of the gate electrode.
• a field effect transistor in accordance with still another feature of the invention, includes a gallium arsenide substrate.
• An indium gallium arsenide channel layer is disposed over the substrate.
• a doped, aluminum gallium arsenide barrier layer is disposed over the channel layer.
• a protective layer is disposed on the barrier layer.
• An indium gallium phosphide layer is disposed over the protective layer, such indium gallium phosphide layer having an aperture therein.
• a source and drain contact layer is disposed over the laterally spaced regions of the indium gallium phosphide layer, the aperture being disposed between the laterally spaced regions of the indium gallium phosphide layer.
• a gate electrode is provided having side portions thereof disposed in the aperture in the indium gallium phosphide layer, walls of such aperture in the indium gallium phosphide layer being in contact with such side portions of the gate electrode and a bottom portion of such gate being in Schottky contact with the gallium arsenide layer.
• FIGS. 1A-1F are diagrammatical sketches of a FET at various stages in the fabrication thereof in accordance with the PRIOR ART;
• FIG. 2 is a diagrammatical sketch of a cross-section of a field effect transistor according to the invention.
• FIGS 3A-3E are diagrammatical sketches of a cross-section of a field effect transistor of FIG. 2 at various stages in the fabrication thereof according to the invention.
• a field effect transistor 10 here a PHEMT, is shown to include a gallium arsenide (GaAs) substrate 12, an indium gallium arsenide (InGaAs) channel layer 14 over the substrate 12, a doped, aluminum gallium arsenide (AlGaAs) barrier (donor) layer 16 over the channel layer 14, a protective layer 18 disposed on the barrier layer 16, an indium gallium phosphide (InGaP) layer 20 disposed over the protective layer, gallium arsenide (GaAs) source and drain contact layers 22, 24 disposed over the laterally spaced regions of the indium gallium phosphide layer 20.
• GaAs gallium arsenide
• InGaAs indium gallium arsenide
• AlGaAs aluminum gallium arsenide
• InGaP indium gallium phosphide
• GaAs gallium arsenide
• the indium gallium phosphide layer 20 has an aperture 26 therein.
• a gate G electrode 30 is provided having side portions thereof disposed in the aperture 26 in the indium gallium phosphide layer 20.
• the sidewalls 32 of such aperture 26 in the indium gallium phosphide layer 20 are in contact with such side portions of the gate electrode 30 and a bottom portion 34 of such gate electrode 30 is in Schottky contact with the gallium arsenide layer 18.
• Source S and drain D electrodes 36, 38, respectively, are in ohmic contact with the gallium arsenide layer 24, as shown.
• the protective layer 18 here has a thickness in the order of 10 to 20 Angstroms and a doping concentration of less than 5x10 17 cm "3 .
• the indium gallium arsenide (InGaAs) channel layer 14 has a thickness of 100 to 200 A and an electronic sheet concentration of lxlO 12 cm “2 to 3xl0 12 cm “2 .
• the doped, aluminum gallium arsenide (AlGaAs) barrier (donor) layer 16 has a thickness of 100 to 300 A and an electronic sheet concentration of 2xl0 12 cm “2 to 5xl0 12 cm “2 .
• the indium gallium phosphide (InGaP) layer 20 has a thickness of 100 to 300 A and a doping concentration of less than 5xl0 17 cm “3 .
• the gallium arsenide (GaAs) source and drain contact layer 22 has a thickness of less than 400 A and has a doping concentration of less than 5xl0 17 cm “3 .
• the contact layer 24 has a thickness of 400 to 600 A and doping concentration of 2xl0 18 cm “3 to 6xl0 18 cm “3 .
• a structure 40 having: the gallium arsenide substrate 12; the indium gallium arsenide channel layer 14 over the substrate 12; the doped aluminum gallium arsenide barrier (donor) layer 16 over the channel layer 14; the gallium arsenide protective layer 18 disposed on the donor layer 16; the indium gallium phosphide layer 20, here an etch stop layer, disposed over the protective layer 18; gallium arsenide source and drain contact layers 22, 24 disposed over the etch stop layer 20; and the source S and drain D contacts 36, 38, respectively, as shown.
• a mask here a layer 40 of photoresist patterned by an electron beam, not shown, with an aperture 42 having shape shown is formed over the surface of the structure 40.
• the aperture 42 exposes a surface portion of the contact layers 24, 22.
• the aperture 42 in the masking layer 40 has a width W, here 0.15 to 0.25 micrometers.
• a wet chemical solution is brought into contact with the patterned photoresist layer 40, a portion of such wet chemical passing through the aperture 40 onto the exposed surface portion of layer 24. The wet chemical is used to selectively etch the exposed portions - of the gallium arsenide source and drain contact layers 22, 24.
• the etch rate of such chemical to gallium arsenide is at least two orders of magnitude greater than the etch rate of such chemical to indium gallium phosphide.
• the indium gallium phosphide layer 20 acts as an etch stop layer to the wet chemical.
• the wet chemical can be a mixture of citric acid, hydrogen peroxide, and water, or a mixture of sulfuric acid, hydrogen peroxide and water, or a mixture of ammonium hydroxide, hydrogen peroxide and water.
• the ratios are 1 NH 2 OH - 1 H 2 0 2 - 250 H 2 0.
• an oxygen plasma is used to remove a thin layer from the photoresist layer.
• the purpose of this is to widen the photoresist layer 40 aperture 42 from width W to larger width W' so that the resist layer 40' (FIG. 3B) has an aperture 42' which is larger than, or equal to, the recess which became formed in the gallium arsenide layers 22, 24 due to undercutting of the photoresist layer 40 (FIG. 3A) by the wet chemical etch.
• the gallium arsenide protective layer 18 acts as an etch stop layer to the hydrochloric acid.
• the hydrochloric acid etch is selective; i.e., it only removes the indium gallium phosphide and leaves the gallium arsenide layer 18 intact.
• the masking material is, as noted above, now the gallium arsenide layers 22, 24 that was epitaxially grown on top of the indium gallium phosphide layer 20 and not the photoresist layer 40'.
• the gallium arsenide layers 22, 24 that was epitaxially grown on top of the indium gallium phosphide layer 20 and not the photoresist layer 40'.
• a gate metal 46 here titanium/platinum/gold, is deposited over the photoresist layer 40' and through the apertures 42' formed in the photoresist layer 40', the gallium arsenide layers 22, 24 and the aperture 26 (FIG. 3B) in the indium gallium phosphide layer 20, as shown. It is noted that the gate metal 46 fills the aperture 26 in layer 20 because the width W (FIG. 3A) of layer 40 had been widen to width W' (FIG. 3B) .
• the photoresist layer 40' is then lifted-off removing the extraneous portions of the gate metal deposited thereon, to thereby form the gate G electrode 30, as shown in FIG. 3D.
• a photoresist layer 50 is deposited over the surface of the structure shown in FIG. 3D, and patterned as shown. More ⁇ particularly, the photoresist layer 50 is patterned with a wider aperture 52 between the source and drain electrodes 36, 38 than the aperture 42, 42' (FIGS. 3A and 3B) formed in the photoresist layer 40, 40' shown. It is noted that the gate metal 46 forming the gate electrode 30 is in this wider aperture 52.
• a wet etch here a mixture of sulfuric acid, hydrogen peroxide and water or a mixture of ammonium hydroxide, hydrogen peroxide and water is brought into contact with the photoresist layer 50, the exposed gate metal 30, and the exposed surface portions of the gallium arsenide source and drain contact layers 22, 24 to remove, selectively, the exposed portions of such gallium arsenide contact layers 22, 24 while leaving substantially unetched the indium gallium phosphide etch stop layer 20.
• the photoresist layer 50 is then removed and the resulting field effect transistor 10 shown in FIG. 2.
• the method described above in connection with FIGS. 3A-3E allows the formation of a gate recess by selective wet etching. Therefore, with such method, no damage to the structure from a dry etch results. Further, because of the wet etching selectivity, which is nearly infinite, there is no need, as mentioned above, to measure the channel current between the source and drain to determine the etching end point. Thus, the method yield greater uniformity, better reproducibility and is less labor intensive. Further, there is no ungated recess area thereby eliminating most significant problems associated with ungated recesses, i.e., uncontrolled reverse breakdown voltage, transient effects such as gate and drain lag, and decreased extrinsic transconductance.

Landscapes

• Junction Field-Effect Transistors (AREA)

Priority Applications (4)

Applications Claiming Priority (2)

Publications (2)

Family

ID=22588591

Family Applications (1)

Country Status (6)

Cited By (4)

Families Citing this family (13)

Citations (11)

Family Cites Families (19)

• 1998
  • 1998-09-29 US US09/163,124 patent/US6087207A/en not_active Expired - Lifetime
• 1999
  • 1999-09-15 AU AU59238/99A patent/AU5923899A/en not_active Abandoned
  • 1999-09-15 EP EP99946936A patent/EP1131849B1/en not_active Expired - Lifetime
  • 1999-09-15 DE DE69930135T patent/DE69930135T2/de not_active Expired - Lifetime
  • 1999-09-15 WO PCT/US1999/021135 patent/WO2000019512A1/en not_active Ceased
  • 1999-09-15 JP JP2000572921A patent/JP4874461B2/ja not_active Expired - Lifetime
  • 1999-09-15 EP EP05077484A patent/EP1630860A3/en not_active Ceased

Patent Citations (11)

Also Published As

Similar Documents

Legal Events