Thin film semiconductor switching device
12593467 ยท 2026-03-31
Assignee
Inventors
- Douglas W. BARLAGE (Edmonton, CA)
- Lhing Gem SHOUTE (Edmonton, CA)
- Kenneth C. CADIEN (Edmonton, CA)
- Alex Munnlick MA (Edmonton, CA)
- Eric Wilson MILBURN (Edmonton, CA)
Cpc classification
H10D30/6713
ELECTRICITY
H10D89/60
ELECTRICITY
H10W42/20
ELECTRICITY
H10D99/00
ELECTRICITY
H10W90/297
ELECTRICITY
H10W20/20
ELECTRICITY
International classification
H10D89/60
ELECTRICITY
H10D99/00
ELECTRICITY
H10W20/20
ELECTRICITY
H10W42/20
ELECTRICITY
Abstract
Novel semiconductor devices are taught. The novel devices include a thin film transistor (TFT) with an n-type semiconductor layer to form a channel between a source and a drain. The TFT further includes a source-channel interfacial member adjacent to at least the source contact of the device to provide depletion layer control of the operation of the TFT.
Claims
1. A method of making a thin-film transistor, the method comprising: forming a well in a thin-film structure, wherein the thin-film structure includes a source contact and a drain contact; forming a layer of n-type semiconductor material over the thin-film structure within the well and outside the well; forming a layer of dielectric material over the layer of n-type semiconductor material within the well and outside the well; forming a layer of gate material over the layer of dielectric material within the well and outside the well; and removing portions of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material that are outside the well; wherein portions of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material remaining within the well are operable with the source contact and the drain contact as the thin-film transistor.
2. The method of claim 1, wherein removing the portions of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material that are outside the well comprises polishing.
3. The method of claim 1, wherein removing the portions of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material that are outside the well comprises wet chemical etching.
4. The method of claim 1, wherein removing the portions of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material that are outside the well comprises dry chemical etching.
5. The method of claim 1, wherein removing the portions of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material that are outside the well comprises atomic layer etching.
6. The method of claim 1, further comprising forming a source-channel interfacial member over the source contact before forming the layer of n-type semiconductor material.
7. The method of claim 6, wherein forming the source-channel interfacial member comprises oxidizing material of the source contact.
8. The method of claim 6, wherein forming the source-channel interfacial member comprises depositing the source-channel interfacial member on the source contact.
9. The method of claim 6, wherein the source-channel interfacial member comprises a p-type metal oxide and is operable to deplete the carrier channel in a region of the layer of n-type semiconductor material adjacent the source contact to reduce leakage current when the thin-film transistor is off.
10. The method of claim 1, wherein forming the layer of n-type semiconductor material comprises using atomic layer deposition.
11. The method of claim 1, wherein the n-type semiconductor material is tin oxide.
12. The method of claim 1, wherein the drain contact is positioned vertically with respect to the source contact.
13. The method of claim 1, further comprising: forming a layer of gate tuning material over the layer of dielectric material within the well and outside the well before forming the layer of gate material; and removing a portion of the layer of gate tuning material that is outside the well.
14. A thin-film transistor comprising: a thin-film structure including a source contact and a drain contact; a layer of n-type semiconductor material formed within a well in the thin-film structure; a layer of dielectric material formed over the layer of n-type semiconductor material within the well; a layer of gate material formed over the layer of dielectric material within the well; and a source-channel interfacial member formed over the source contact; wherein no portion of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material extends outside the well.
15. The thin-film transistor of claim 14, wherein the source-channel interfacial member comprises oxidized material of the source contact.
16. The thin-film transistor of claim 14, the source-channel interfacial member comprises material deposited on the source contact.
17. The thin-film transistor of claim 14, wherein the source-channel interfacial member comprises a p-type metal oxide and is operable to deplete the carrier channel in a region of the layer of n-type semiconductor material adjacent the source contact to reduce leakage current when the thin-film transistor is off.
18. The thin-film transistor of claim 14, wherein the n-type semiconductor material is tin oxide.
19. The thin-film transistor of claim 14, wherein the drain contact is positioned vertically with respect to the source contact.
20. The thin-film transistor of claim 14, further comprising a layer of gate tuning material formed over the layer of dielectric material within the well.
21. A stack of thin-film transistors comprising: two or more thin-film transistors stacked vertically and separated by insulating and/or dielectric material, each of the two or more thin-film transistors including: a thin-film structure including a source contact and a drain contact; a layer of n-type semiconductor material formed within a well in the thin-film structure; a layer of dielectric material formed over the layer of n-type semiconductor material within the well; a layer of gate material formed over the layer of dielectric material within the well; and wherein the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material are flush with an opening of the well.
22. The stack of thin-film transistors of claim 21, wherein each of the two or more thin-film transistors further comprises a source-channel interfacial member formed over the source contact, wherein the source-channel interfacial member includes a p-type metal oxide and is operable to deplete the carrier channel in a region of the layer of n-type semiconductor material adjacent the source contact to reduce leakage current when the thin-film transistor is off.
23. The stack of thin-film transistors of claim 21, wherein the n-type semiconductor material is tin oxide.
24. The stack of thin-film transistors of claim 21, wherein the drain contact is positioned vertically with respect to the source contact.
25. A thin-film transistor comprising: a thin-film structure including a source contact and a drain contact; a layer of n-type semiconductor material formed within a well in the thin-film structure; a layer of dielectric material formed over the layer of n-type semiconductor material within the well; and a layer of gate material formed over the layer of dielectric material within the well; wherein no portion of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material extends outside the well; and wherein the drain contact is positioned vertically with respect to the source contact.
26. A thin-film transistor comprising: a thin-film structure including a source contact and a drain contact; a layer of n-type semiconductor material formed within a well in the thin-film structure; a layer of dielectric material formed over the layer of n-type semiconductor material within the well; a layer of gate material formed over the layer of dielectric material within the well; and a layer of gate tuning material formed over the layer of dielectric material within the well; wherein no portion of the layer of n-type semiconductor material, the layer of dielectric material, and the layer of gate material extends outside the well.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1) For a better understanding of the various embodiments of the present invention described herein, reference will be made to the accompanying drawings. The drawings are not intended to limit the scope of the invention described herein and are not drawn to scale, to improve clarity of understanding of this specification. In the Figures:
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36) Further aspects and features of the example embodiments described herein will appear from the following description taken together with the accompanying drawings.
DETAILED DESCRIPTION OF THE INVENTION
(37) In the following discussion, like components and elements may be indicated with like reference numbers. Further, as is understood by those of skill in the art, most semiconductor devices are manufactured in a series of steps wherein substantially planar layers of material are formed, portions of those layers subsequently being removed, treated and/or replaced with subsequent layers to obtain the desired semiconductor device. Accordingly, as used herein, the terms horizontal and vertical are used with respect to the plane of those layers, with horizontal referring to a direction largely parallel to the plane of the layers and vertical referring to a direction largely orthogonal to the plane of the layers. Similarly, the terms up, down, on, under or the like are also used with respect to those planar layers.
(38) Further, the manufacturing of those layers and/or features of semiconductor devices described herein is referred to herein as forming those layers and/or features and, as will be apparent to those of skill in the art, forming is intended to comprise all semiconductor manufacturing techniques suitable and applicable therefor including, without limitation: deposition (chemical, atomic layer, physical vapor, etc.); sputtering; PECVD (Plasma Enhanced Chemical Vapor Deposition); implantation and annealing, oxidation; etc. as will be apparent to those of skill in the art.
(39) Before describing embodiments of the present invention, examples of prior art TFTs will be described, for clarity, with respect to
(40)
(41) Gate 102 includes a gate electrode 104 and a gate contact 103, source 106 includes a source electrode 108 and a source contact 107, and drain 110 includes a drain electrode 112 and a drain contact 111. As will be apparent to those of skill in the art, source electrode 108 and source contact 107 can be the same or different materials. In particular, the material of source contact 107 can be selected to provide a desired Work Function while the material of source electrode 108 can be selected to simplify connection of source electrode 108 to other devices, etc. Gate electrode 104 and gate contact 103, and drain electrode 112 and drain contact 111 can similarly be formed of the same respective materials or different materials, as desired.
(42) TFT 100a also includes a dielectric layer 114 and a semiconductor layer 116. As is understood by those of skill in the art, TFT 100a can optionally include a gate tuning layer 122.
(43) Operation of TFT 100A relies upon the use of semiconductor layer 116 as an active material in which a channel can be formed. When semiconductor 116 is in an active state, a channel is formed between source contact 107 and drain contact 111 allowing electrical current to flow therebetween. In the illustrated embodiment, an n-type semiconductor is shown.
(44) For an n-type semiconductor based TFT (or n-type TFT), such as TFT 100a, the TFT will be in an active (ON) state when a voltage applied to gate 102 (i.e. V.sub.g) is greater than the threshold voltage (i.e. V.sub.th), the threshold voltage being defined by the structure and materials of TFT 100a. In a TFT with an n-type semiconductor, the application of a positive gate voltage greater than the threshold voltage results in the injection of negatively charged electrons into the n-type semiconductor from the source. Upon application of a voltage between the source and the drain (i.e. V.sub.d), the injected electrons move through the channel formed in the semiconductor, from the source to the drain, resulting in current flow, i.e. the source-drain current (I.sub.ds) or drain current (I.sub.d).
(45) In
(46) When a low (compared to the gate voltage) drain voltage is applied, the concentration of the carriers in the channel tends to be uniform across the channel and TFT 100a is considered to be ohmic. When a high drain voltage is applied, the channel becomes pinched-off and a highly insulating region adjacent to drain contact 111 begins to form a depletion layer 135, showing a decrease in the density of the carriers, is shown in
(47) Conventional TFTs typically have two channels of current flow. A first channel is formed through an ohmic connection between source contact 107 and drain contact 111, and a second channel is formed by an accumulation layer at the interface between gate dielectric layer 114 and semiconductor layer 116.
(48) Because of the channel available underneath gate contact 103, TFT 100a has a limited OFF state due to current flow through this channel. Accordingly, conventional TFTs invariably do not pinch-off well, and subsequently can have substantial leakage problems wherein some current flow continues despite the TFT being intended to be OFF.
(49) Conventional TFTs can also have output resistance problems if source contact 107 and drain contact 111 are too close to one another. For example, as mentioned above, if source contact 107 and drain contact 111 are less than about 1000 nm apart, and particularly if they are less than 200 nm apart, significant levels of leakage occurs. Accordingly, conventional TFTs are typically in an ON state even under a zero-bias. In most cases, a negative gate voltage must be applied to conventional TFTs to create a depletion layer or region in order to limit the flow of current between source contact 107 and drain contact 111 to effectively turn the TFT OFF.
(50)
(51)
(52) However, both TFT 100d and 100e face similar challenges as those described above with respect to TFT 100a, particularly when fabricated in smaller dimensions.
(53)
(54) TFT 200a also includes a dielectric layer 214 and a semiconductor layer 216. TFT 200a can optionally include a gate tuning layer 222. These layers, and TFT 200a in general, are formed over an insulation layer 218, which is formed over a substrate 220. It is contemplated that in many circumstances substrate 220 will be a silicon wafer, as commonly used in the manufacture of semiconductors, however the present invention is not so limited and substrate 220 can, in fact, be a wide variety of materials including, without limitation: glass; ceramics; metals; flexible polymers or other plastics; etc.
(55) In the illustrated embodiment, semiconductor layer 216 is an n-type semiconductor layer. N-type semiconductor layer 216 is a semiconductor material which is selected to preferably allow it to be formed at temperatures below 300 degrees centigrade. Examples of suitable materials for semiconductor layer 216 include, without limitation, zinc oxide, tin oxide, indium gallium zinc oxide, gallium oxide, germanium oxide, etc.
(56) Unlike prior art TFTs, TFT 200a further includes a source-channel interfacial member 250 that, in this embodiment, is a layer of p-type semiconductor material extending from source contact 207 to drain contact 211 under semiconductor layer 216. Source-channel interfacial member 250 electrically connects source contact 207 to semiconductor layer 216 and source contact 207 is otherwise electrically separated from semiconductor layer 216. Further, in this example, source-channel interfacial member 250 electrically connects semiconductor layer 216 to drain contact 211.
(57) Gate contact 203 can comprise any elemental material or compound of materials with conducting properties. Examples of materials that can be used for gate contact 203 include titanium, titanium nitride, chrome, hafnium, tantalum nitride or any other single element or bi-metal element or compound. Gate electrode 204 can be formed of the same material as gate contact 203 or from another material which provides desired properties for connecting gate electrode 204 to other components.
(58) Source 206 and drain 210 can comprise any elemental material or compound of materials with conducting properties. Specifically, source contact 207 and drain contact 211 can be a metal or a degenerate (highly doped) semiconductor. Examples of suitable materials that can be used in source contact 207 and/or drain contact 211 include: nickel, tungsten, molybdenum, aluminum; gold; copper; cobalt; ruthenium; titanium nitride; tantalum nitride; silicon; and/or any III-V compound semiconductor with a high conductivity. In some cases, drain contact 211 consists of the same material as source contact 207. In other cases, drain contact 211 consists of a material with a higher conductivity that is capable of providing a reduced barrier to electron transport than the material used in source contact 207.
(59) Source-channel interfacial member 250 is provided to create a repository of complementary excess negative charge that functions to deplete the channel in at least the region of semiconductor layer 216 adjacent source contact 207. In this manner, source-channel interfacial member 250 serves as an electron transport barrier, resulting in substantially no current flow through semiconductor layer 216 when TFT 200a is in a OFF state.
(60) The operation of source-channel interfacial member 250 can be achieved in a variety of configurations, including a p-type semiconductor, an induced piezoelectric dipole, a controllable tunneling barrier, combinations thereof or other mechanisms to modulate injected current by an external applied field.
(61) For example, when implemented as a p-type semiconductor, source-channel interfacial member 250 can consist of elemental germanium or source-channel interfacial member 250 can be formed by using a p-type dopant in a relevant portion of semiconductor layer 216. In other cases, source-channel interfacial member 250 can consist of an oxide or a sulfide, or another element corresponding to group VI(A) of the periodic table, or the chalcogens, such as oxygen, sulfur, selenium, tellurium or polonium.
(62) Source-channel interfacial member 250 can also be formed by a catalytic growth of the material that forms source contact 207. In such cases, the oxygen can preferentially traverse to the source-channel interface, creating the source-channel interfacial member. In other cases, source-channel interfacial member 250 can be formed by depositing a p-type dopant, such as a p-type metal oxide or another semiconductor, by way of a deposition technique, such as atomic layer deposition, sputtering physical vapor deposition or chemical vapor deposition.
(63) In the embodiment of TFT 200a, source-channel interfacial member 250 has been formed as a continuous member extending from source contact 207 to drain contact 211. While source-channel interfacial member 250 need only be located adjacent source contact 207 to electrically connect source contact 207 and semiconductor layer 216, depending upon the manufacturing processes used to fabricate TFT 200a, source-channel interfacial member 250 can be a formed as a layer extending between source contact 207 and drain contact 211, or adjacent to just source contact 207, or adjacent to each of source contact 207 and drain contact 211.
(64) It can be desired that semiconductor layer 216 is formed with compositional control such that semiconductor layer 216 is heavily n-type near source contact 207 and drain contact 211 to prevent carrier starvation during the ON state of TFT 200a. In other embodiments, semiconductor layer 216 is formed to be heavily n-type near source contact 207, drain contact 211 and gate contact 203 which can allow tuning of the threshold voltage of TFT 200a. Accordingly, the properties of TFT 200a can be controlled based on the location and extent of n-type doping in semiconductor layer 216.
(65) As illustrated in
(66) TFT 200a can also include an optional gate tuning layer 222. Gate tuning layer 222 can be an atomic layer of metal that is used to adjust the effective barrier height and/or Work function of the gate contact 203, as desired, as will be apparent to those of skill in the art.
(67) In TFT 200a, typically each of gate electrode 204, source electrode 208 and drain electrode 212 are formed of highly conducting metals, such as copper or aluminum, that are used to electrically connect the corresponding electrode to the appropriate elements of the rest of the integrated circuit in which TFT 200a is formed.
(68) TFT 200a also includes an insulation layer 218 formed over substrate 220. Substrate 220 serves as the foundation for the construction of components and devices, such as transistors and integrated circuits, and insulation layer 218 is a dielectric separating substrate 220 from source contact 207 and drain contact 211. Examples of insulation layer 218 include materials such as: silicon dioxide (SiO.sub.2); silicon nitride (Si.sub.3N.sub.4); aluminum oxide (Al.sub.2O.sub.3); etc. and examples of substrate 220 include silicon, glass, plastic materials and/or flexible polymers, printed circuit boards, etc. In some circumstances, such as the case illustrated in
(69) As discussed above, TFT 200a is intended to perform, when compared to prior art TFTs, with improved operating characteristics, and in particular with a higher threshold voltage, especially when fabricated with channel length dimensions of 1000 nm or less.
(70) As described above, source-channel interfacial member 250 serves to create a repository of complementary excess negative charge that functions to deplete the channel in at least the region of semiconductor layer 216 adjacent source contact 207. In TFTs fabricated in accordance with embodiments of the present invention, such as TFT 200a, the distance between gate contact 203 and source contact 207, through gate dielectric layer 214, semiconductor layer 216 and source-channel interfacial member 250, is selected such that the electric field induced by a voltage applied to gate contact 203 will lower the blocking potential of the repository of complementary excess negative charge introduced by source-channel interfacial member 250 to place the TFT, such as TFT 200a, into an ON state. In the absence of such an applied gate voltage, the blocking potential results in the TFT being in an OFF state.
(71) TFT 200a can be fabricated with its various elements formed in ranges of thicknesses. As will be apparent to those of skill in the art, the primary limiting factor is that gate contact 203 must be within sufficient electrical proximity to source contact 207 such that the threshold voltage applied to gate 202 serves to effectively reduce the barrier to electron flow into semiconductor layer 216 induced by source-channel interfacial member 250. Failure to induce a sufficient electric field reduces the ability to turn TFT 200a on. This is very similar to the well known criteria for MOSFETs and the like, typically referred to as Ada () factor.
(72) In a first example of TFT 200a, semiconductor layer 216 was formed with a thickness of about 20 nm (dielectric constant of about 8) and source-channel interfacial member 250 with a thickness of about 2 nm (dielectric constant of about 10) and dielectric layer 214 with a thickness of about 10 nm (dielectric constant of about 24).
(73) In another example of TFT 200a, semiconductor layer 216 was formed with a thickness of about 5 nm (17 dielectric constant), source-channel interfacial member 250 formed was with a thickness of about 0.5 nm (dielectric constant of about 10) and dielectric layer 214 was formed with a thickness of about 5 nm (dielectric constant about 24).
(74) The actual selection of the vertical thickness of each of semiconductor layer 216, dielectric layer 214 and source-channel interfacial member 250 will be, to some extent, dependent upon properties of the materials selected and the fabrication techniques employed. However, in view of the disclosure herein, these specific selections will now be well within the capabilities of those of ordinary skill in the art,
(75)
(76)
(77) The presence or absence of source-channel interfacial member 250 on drain contact 211 does alter the electrical properties of TFT 200c. In TFT 200d, the absence of source-channel interfacial member 250 on drain contact 211 raises the threshold voltage of TFT 200d between approximately 0.1V to 2V, compared to TFT 200a and TFT 200c. In cases where source-channel interfacial member 250 is formed by catalytic growth of an oxide or sulfide, the resulting formation of source-channel interfacial member 250 on drain contact 211 in addition to the formation of source-channel interfacial member 250 on source contact 207 can be readily tolerated and no additional processing is required to remove source-channel interfacial member 250 from drain contact 211, unless it is desired to reduce the threshold voltage.
(78) Reference is next made to
(79) In the various embodiments illustrated herein, with source-channel interfacial member 250 adjacent source contact 207 a structure which is functionally similar to a bipolar junction transistor (BJT) is introduced. The TFTs of the embodiments disclosed herein effectively combine a field-effect transistor (FET) architecture with a structure analogous to the emitter of a BJT at the source contact.
(80) These analogous BJT-like structures have their collector and their base tied together either through a direct ionic short or a collection of recombination centers, and the electrons are injected from the emitter-like structure into source contact 207. A schematic representation of a TFT 300 which has source-channel interfacial member 350 at each of a source contact 307 and a drain contact 311, is illustrated schematically in
(81)
(82) TFT 300 is analogous to TFT 200a of
(83) In the embodiment of TFT 300, similar to TFT 200a, source-channel interfacial member 350 extends between source contact 307 and drain contact 311, thereby forming a structure resembling a gate limited junction field effect transistor (JFET), or accumulation MOSFET, with a BJT equivalent at source contact 307 and drain contact 311.
(84)
(85)
(86)
(87) Reference is next made to
(88)
(89)
(90)
(91)
(92) The differences in the gradients and configurations (e.g. thickness) of the accumulation layer and depletion region between TFT 300 and TFT 400 result from the differences in the extent of the presence of the source-channel interfacial member in the devices. With a higher concentration and distributed presence of the source-channel interfacial member, better depletion layer control of the n-type semiconductor channel is attained.
(93) Reference is next made to
(94)
(95)
(96) Accumulation layer 530 and depletion region 535, 540 characteristics of TFT 500 are analogous to TFT 300 of
(97) Reference is next made to
(98) As used herein, the terms form or forming are intended to encompass any suitable manner for manufacturing the structures recited in this disclosure. As such, form and forming can include processes such as: atomic layer deposition; chemical vapor deposition; plasma enhanced chemical vapor deposition, sputtering; ion implantation; oxidation; electrochemical deposition; molecular beam epitaxy; etc. The selection of specific appropriate processes is well within the purview of those of skill in the art.
(99) Method 600 begins at step 605 where source contact 707 and drain contact 711 are formed on top of an insulation layer 718, which is, in turn formed on top of a substrate 720, as shown in
(100) As previously mentioned, insulation layer 718 can be formed from materials such as: silicon dioxide (SiO.sub.2); silicon nitride (Si.sub.3N.sub.4); aluminum oxide (Al.sub.2O.sub.3); etc. Substrate 720 can be any suitable substrate suitable for serving as the foundation for the construction of components and devices, such as transistors and integrated circuits, image sensors and displays and examples of which include, without limitation, silicon, flexible polymers and other plastics, ceramic materials, optical glass, metals, etc.
(101) In some cases, source contact 707 and drain contact 711 are formed by deposition via lithography process. However, as will be apparent to those of skill in the art, the methods of forming source contact 707 and drain contact 711 are not particularly limited and they can be formed through a variety of other processes as will be apparent to those of skill in the art.
(102) In cases where source contact 707 and drain contact 711 are formed from the same material, then at step 605, both source contact 707 and drain contact 711 can be formed simultaneously in the same step. In cases where source contact 707 and drain contact 711 consist of different materials, then source contact 707 and drain contact 711 can be formed in separate steps.
(103) At step 610, source-channel interfacial member 750 is formed. In the example illustrated in
(104) However, source-channel interfacial member 750 can also be formed in other ways and reference is made to
(105)
(106) Returning again to
(107) At step 620, a dielectric layer 714 is formed. As shown in the cross section of
(108) At step 625, gate contact 703 is formed. As shown in
(109) At step 630, the unwanted portions of dielectric layer 714 and underlying semiconductor layer 716 and source-channel interfacial member 750 are removed.
(110) At step 635, gate electrode 704, source electrode 708 and drain electrode 712 are formed as shown in
(111) In some circumstances, it may be desired to fabricate TFTs in accordance with embodiments of the present invention in a vertical configuration. Accordingly,
(112)
(113) In TFT 900a, source 906, including source contact 907 and source electrode 908, are at the bottom (with respect to the orientation of the Figure) of TFT 900a, having been formed on top of first insulation layer 918. Second insulation layer 925 is formed over a portion of source contact 907 leaving a portion of source contact 907 exposed in the middle of TFT 900a. Drain contact 911, with drain electrode 912 formed on drain contact 911, is formed on either side of the exposed portion of source contact 907. Source contact 907 is insulated from drain contact 911 by second insulation layer 925.
(114) In TFT 900a, source-channel interfacial member 950 is formed on the above-mentioned exposed portion of source contact 907, the inner surface of second insulation layer 925 and on drain contact 911. Gate contact 903 is surrounded by (optional) gate tuning layer 922, which is in turn surrounded by dielectric layer 914. Semiconductor layer 916 is formed between the dielectric layer 914 and source-channel interfacial member 950.
(115)
(116) Specifically, in TFT 900b source-channel interfacial member 950 is formed as a layer over the entire upper surface of source contact 907. As shown in
(117)
(118) However, TFT 900c differs from the above-described TFTs 900a and 900b in that the bottom contact of TFT 900c is drain contact 911. Conversely, the top contact of TFT 900c is source contact 907. As shown in
(119) While the embodiments of
(120) As will now be apparent to those of skill in the art, TFT 900a, TFT 900b, TFT 900c and TFT 900d function in manners similar to the other embodiments of the present invention disclosed herein. As mentioned above, TFTs 900a, 900b, 900c and 900d can be formed as vertical cylinders, rectangular parallelepipeds, hexagonal prisms, etc. and this variety of possible shapes allows the footprint of TFTs 900a, 900b, 900c and 900d to be selected to optimize the density at which TFTs 900a, 900b, 900c and 900d can be formed in a device. This provides several advantages, especially in applications such as DRAM memory wherein increasing memory cell density is an important goal.
(121) Further, as also mentioned above, TFTs 900a, 900b, and 900c can be stacked to increase circuit element densities. Specifically, an array of TFTs 900a, 900b, or 900c can be formed upon a substrate and/o insulator covering a lower layer of circuit elements (including TFTs 900a, 900b and 900c) and, in turn covered by a substrate and/or insulator and another array of TFTs 900a, 900b, or 900c formed atop that substrate and/or insulator-resulting in true 3D integrated circuits being formed.
(122) Reference is now made to
(123) Method 1000 begins at step 1005 where source electrode 908 and source contact 907 are formed on first insulation layer 918 which is, in turn, formed on substrate 920, as shown in
(124) At step 1010, a second insulation layer 925 is formed over source contact 907, as shown in
(125) At step 1015, drain contact 911 is formed as shown in
(126) At 1020, the undesired material of drain contact 911 is removed. The undesired material can be removed in any suitable manner as will occur to those of skill in the art, such as by patterning and etching.
(127) At step 1025, undesired portions of second insulation layer 925 are removed down to source contact 907, as shown in
(128) At step 1030, a source-channel interfacial member 950 is formed on the result of step 1025. As shown in
(129) At step 1035, an n-type semiconductor layer 916 is formed over source-channel interfacial member 950, as shown in
(130) At step 1040, dielectric layer 914 is formed as shown in
(131) At step 1045, (optional) gate tuning layer 922 can be formed over dielectric layer 914, and gate contact 903 is formed over gate tuning layer 922. Gate tuning layer 922 and gate contact 903 can be formed in a variety of manners, as will be apparent to those of skill in the art, including via an atomic layer deposition technique. If gate tuning layer 922 is not present, gate contact 903 is formed on dielectric layer 914 directly. The results of step 1045 are illustrated in
(132) At step 1050, unwanted material from the various formed layers is removed to expose the underlying layers/features as shown in
(133) At step 1055, drain electrode 912 and gate electrode 904 are formed. In the embodiment shown in
(134)
(135) Method 1100 begins with a substrate 920 on which an insulation layer 918 has been formed. At step 1105 source electrode 908 and source contact 907 are formed on insulation layer 918, as shown in
(136) At step 1110, p-type source-channel interfacial member 950 is formed on source contact 907. As shown in
(137) At step 1115, a second insulation layer 925 is formed on source-channel interfacial member 950 as shown in
(138) At step 1120, drain contact 911 is formed. As shown in
(139) At step 1125, unwanted material is removed from drain contact 911, as shown in
(140) At step 1130, unwanted material is removed from second insulation layer 925, as shown in
(141) Steps 1135 to 1155 of method 1100 are analogous to steps 1035 to 1055 of method 1000, described above. At step 1135, as shown in
(142) At step 1140, a dielectric layer 914 is formed over semiconductor layer 916. This is shown in
(143) At step 1145, gate contact 903 is formed over dielectric layer 914, as shown in
(144) At step 1150, TFT 1300 is processed to remove unwanted material to expose underlying layers and features as shown in
(145) At step 1155, drain electrode 912 and gate electrode 904 are formed. As shown in
(146) As will now be apparent, the present invention comprises a variety of embodiments of novel semiconductor devices. In particular, the present invention discloses TFTs with improved performance, even when fabricated at sub 200 nm channel dimensions.
(147) Another contemplated advantage of semiconductor devices in accordance with embodiments of the present invention is their ability to be manufactured in Back End of Line (BEOL) semiconductor fabrication processes.
(148) As is known to those of skill in the art, semiconductor fabrication processes are often split into Front End of Line (FEOL) and Back End of Line processes. Conventionally, the transistors, capacitors, resistors and inductors of an integrated circuit are first formed on a wafer by FEOL processes. Once these devices have been fabricated on a wafer, it is then processed with BEOL processes where metalization layers and bonding sites are formed.
(149) Conventionally, once a wafer (or other substrate) has completed FEOL processing, it has generally not been possible to add additional devices to the wafer during BEOL processes, as subjecting the wafer to the temperatures which are required for conventional semiconductor device fabrication would destroy the semiconductor structures and devices already on the wafer which were formed by the FEOL processes.
(150) Recently, attempts have been made to fabricate TFTs as a BEOL process using Indium Gallium Zinc Oxide (IGZO) as the semiconductor/channel material. While some success has been achieved with such devices, they tend to not perform as well as desired, generally having low threshold voltages, leading to high leakage levels/poor OFF characteristics. Further, the processes used to fabricate the IGZO devices have proven to be difficult to control and the resulting devices are fragile and it can be difficult to have them survive other BEOL processes, such as annealing.
(151) In contrast, TFTs in accordance with embodiments of the present invention can also be fabricated with techniques and processes which will not damage previously fabricated FEOL devices or structures but such TFTs have been shown to perform well, even at semiconductor/channel lengths of less than 200 nm, and to not be damaged by other BEOL processes.
(152) As described in detail above, TFTs in accordance with embodiments of the present invention employ six basic fabrication processes: Source Formation; Drain Formation; Gate Formation; Semiconductor Formation; Dielectric/Insulator Formation; and Source-Channel Interfacial Member Formationeach of which typically can be achieved in several possible ways. Selection of one or more appropriate alternatives for each of these fabrication processes, for use to fabricate BEOL TFTs or TFTs created in FEOL or alternative processes, is within the normal skills of those of skill in the art.
(153) As an example, in deciding on how to fabricate a TFT in accordance with the present invention and wherein it is desired to not exceed a temperature of three hundred degrees centigrade during the fabrication process, it can be decided that: Source formation, Drain Formation and Gate formation will be achieved by sputtering and patterning; Semiconductor layer formation will be achieved by atomic layer deposition; Dielectric/Insulator formation will be achieved by atomic layer deposition; and Source-Channel Interfacial Member formation can be achieved by oxidation of the source contact (if the material is an oxidizable metal) or by sputtering a degeneratively doped monoatomic semiconductor layer, such as Germanium or Silicon. As will be apparent, each of these selected fabrication technologies can be performed at temperatures below three hundred degrees centigrade and will therefore not affect devices and structures which were previously fabricated in FEOL processes.
(154) As will also now be apparent, the present invention provides novel thin film semiconductor switching devices which have good performance, even when manufactured with semiconductor/channel lengths of less than 1000 nm, and in particular with semiconductor/channel lengths of less than 200 nm. Manufacturing techniques and methods for the novel devices are also taught and these techniques and methods also allow the novel devices to be manufactured as BEOL devices.
(155) The above-described embodiments of the invention are intended to be examples of the present invention and alterations and modifications may be effected thereto, by those of skill in the art, without departing from the scope of the invention which is defined solely by the claims appended hereto.