HEMT device

Abstract

The utility model relates to an HEMT (High Electron Mobility Transistor) device, comprising a body which is provided with a top surface and comprises a heterostructure configured to generate two-dimensional charge carrier gas; and a gate structure extending on a top surface of the body and capable of being biased to electrically control a two-dimensional charge carrier gas, the gate structure comprising: a channel modulation region of a semiconductor material, the channel modulation region having a top surface; a functional region of a semiconductor material; a first gate contact region of conductive material; a second gate contact region of an electrically conductive material, the second gate contact region extending at least partially over the functional region and in contact with the first gate contact region, where the functional region and the first gate contact region extend over a top surface of the channel modulation region, and where the first gate contact region is arranged laterally with respect to the functional region, the channel modulation region has different conduction types relative to the functional region; and an insulating region of a non-conductive material, the insulating region between the first gate contact region and the second gate contact region.

Description

Technical Field

[0001] The present disclosure relates to a High Electron Mobility Transistor (HEMT) device with an improved gate structure and a manufacturing process thereof. Background Art

[0002] HEMT devices are known in which the conducting channel is based on the formation of a high-mobility two-dimensional electron gas (2DEG) layer at a heterojunction (i.e., at the interface between semiconductor materials with different band gaps). For example, HEMT devices based on a heterojunction between an aluminum gallium nitride (AlGaN) layer and a gallium nitride (GaN) layer are known.

[0003] HEMT devices based on AlGaN / GaN heterojunctions or heterostructures offer several advantages that make them particularly suitable and widely used in different applications. For example, the high breakdown threshold of HEMT devices is utilized to achieve high-performance power switching; the high mobility of electrons in the conductive channel allows for high-frequency amplifiers; and the high concentration of electrons in the 2DEG allows for low on-state resistance (R ON ).

[0004] Furthermore, HEMT devices used in radio frequency (RF) applications generally have better RF performance than similar silicon LDMOS devices.

[0005] HEMT devices based on AlGaN / GaN heterojunction are usually depletion-type, ie, they are normally-on.

[0006] However, in various applications, enhancement mode, ie, normally-off GaN / AlGaN HEMTs are needed to improve the reliability of electronic devices, reduce the design complexity of the driving circuit of the HEMT device, and reduce its power consumption.

[0007] Various methods have been proposed for realizing a normally-off HEMT device, such as forming a recessed gate structure, performing specific fabrication steps including incorporating a fluorine plasma under the gate structure, and forming a p-GaN gate region.

[0008] Figure 1 A known HEMT device 1 is shown, which comprises a semiconductor body 2 having a substrate 3 of semiconductor material, a channel layer 4 of GaN and a barrier layer 6 of AlGaN superimposed on one another.

[0009] The channel layer 4 and the barrier layer 6 are of N-type.

[0010] A gate structure 7 extends on the barrier layer 6. The gate structure 7 comprises a channel modulation region 8 of p-GaN and a gate contact region 9 of a conductive material (eg a stack of TiN / AlCu / TiN).

[0011] A channel modulation region 8 extends on the barrier layer 6 , and a gate contact region 9 extends on the channel modulation region 8 .

[0012] A first insulating region 10 , for example made of silicon oxide, extends on the barrier layer 6 , on the side surfaces of the channel modulation region 8 and partly on the top surface of the channel modulation region 8 .

[0013] The opening 11 extends through the first insulating region 10 along the axis Z, and the gate contact region 9 extends into the opening 11 .

[0014] A second insulating region 12 made of silicon oxide, for example, extends over the first insulating region 10 and the gate contact region 9 .

[0015] A field plate region 13 of conductive material extends on the first insulating region 10 , laterally to the gate contact region 9 and below the second insulating region 12 .

[0016] A source region 15 of electrically conductive material extends on a first side of the gate structure 7 along the axis X, on the barrier layer 6 and partly on the second insulating region 12 .

[0017] A drain region 16 of electrically conductive material extends on a second side of the gate structure 7 along the axis X, on the barrier layer 6 and partly on the second insulating region 12 .

[0018] In use, the HEMT device 1 is of the normally-off type. A positive voltage applied to the gate structure 7 causes a two-dimensional electron layer (2DEG) to form at the interface between the channel layer 4 and the barrier layer 6 below the channel modulation region 8 and thus turns the HEMT device 1 on.

[0019] In the presence of a positive voltage, the HEMT device 1 is subject to a leakage current that flows from the gate contact region 9 through the gate structure 7 of the HEMT device 1 to the 2DEG.

[0020] The leakage current may cause malfunction of the HEMT device 1 .

[0021] For example, the leakage current may cause the on-threshold voltage of the HEMT device 1 , ie, the voltage at which the 2DEG is formed below the channel modulation region 8 , to shift relative to a designed value.

[0022] Leakage current has bulk-type contribution and parasitic-type contribution.

[0023] The body shape contribution is mainly affected by the concentration of doping substances (e.g., magnesium atoms) in the channel modulation region 8; and by defects present at the interface between the channel modulation region 8 and the gate contact region 9, which are caused, for example, by partial removal or high roughness of the channel modulation region 8 during the manufacture of the gate contact region 9.

[0024] The parasitic contributions originate from the top surface and the lateral sidewalls 17 of the channel modulation region 8 .

[0025] In fact, the lithography and etching processes used during the manufacturing steps for patterning the channel modulation region 8 result in the formation of defects on the lateral sidewalls 17 .

[0026] Such defects can lead to the formation of vacancies (e.g., those of nitrogen atoms) in which a high concentration of electrons are trapped.

[0027] This results in the formation of a current path for the gate leakage current through the lateral sidewalls 17 of the channel modulation region 8 . Utility Model Content

[0028] According to one aspect of the present disclosure, a HEMT device is provided, comprising: a body having a top surface and comprising a heterostructure configured to generate a two-dimensional charge carrier gas; and a gate structure extending on the top surface of the body and capable of being biased to electrically control the two-dimensional charge carrier gas, the gate structure comprising: a channel modulation region of semiconductor material, the channel modulation region having a top surface; a functional region of semiconductor material; a first gate contact region of conductive material; a second gate contact region of conductive material, the second gate contact region at least partially extending over the functional region and contacting the first gate contact region, wherein the functional region and the first gate contact region extend on the top surface of the channel modulation region, and wherein the first gate contact region is laterally arranged relative to the functional region, the channel modulation region having a different conductivity type relative to the functional region; and an insulating region of non-conductive material, the insulating region being between the first gate contact region and the second gate contact region.

[0029] In some embodiments, the channel modulation region has a peripheral portion forming a lateral sidewall of the channel modulation region, and the functional region extends at least partially over the peripheral portion.

[0030] In some embodiments, the functional region has outer walls that abut lateral sidewalls of the channel modulation region.

[0031] In some embodiments, the functional region is intrinsic.

[0032] In some embodiments, the channel modulation region has a first conductivity type and the functional region has a second conductivity type opposite to the first conductivity type.

[0033] In some embodiments, the concentration of the dopant in the functional region is less than 1015 Atom / cm 3 .

[0034] In some embodiments, the functional region has an inner wall in contact with the first gate contact region.

[0035] In some embodiments, the functional region has a width along the first direction, and a width of the first gate contact region along the first direction is greater than a width of the functional region.

[0036] In some embodiments, a width of the second gate contact region along the first direction is smaller than a width of the functional region along the first direction.

[0037] In some embodiments, a material of the second gate contact region is selected from Ti, Ta, TiN and TaN.

[0038] In some embodiments, the insulating region extends on lateral sidewalls of the channel modulation region and on outer walls of the functional region and extends at least partially on and at a distance from a top surface of the channel modulation region.

[0039] According to one aspect of the present disclosure, a device is provided, comprising: a heterostructure having a surface and configured to generate a two-dimensional charge carrier gas; a channel modulation region of semiconductor material on the surface; a functional region of semiconductor material on the channel modulation region; a first gate contact region on the functional region; an opening passing through the first gate contact region and the functional region, the opening exposing a top surface of the channel modulation region; a passivation layer on the surface of the heterostructure, the functional region, and the first gate contact region, the passivation layer not in the opening and not on the top surface of the channel modulation region; and a second gate contact region in the opening, contacting the top surface of the channel modulation region and overlapping with the passivation layer.

[0040] In some embodiments, the functional area includes two separate parts, and the opening is between the two separate parts.

[0041] In some embodiments, the shape of the functional area is circular or polygonal.

[0042] In some embodiments, the first gate region has a first width and the functional region has a second width greater than the first width.

[0043] In some embodiments, the passivation layer contacts sidewalls of the functional region and a top region of the functional region, the top region is coplanar with a first plane where the functional region contacts the first gate contact region, and the sidewalls of the functional region are on a second plane transverse to the first plane.

[0044] Therefore, according to the present disclosure, a HEMT device and a process for manufacturing the same are provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] For a better understanding of the present disclosure, embodiments of the present disclosure will now be described purely by way of non-limiting examples with reference to the accompanying drawings, in which:

[0046] Figure 1 shows a cross section of a known HEMT device;

[0047] Figure 2 shows a cross section of the present HEMT device according to an embodiment;

[0048] Figure 3 Shows Figure 2 A magnified portion of the HEMT device;

[0049] Figure 4 shows an enlarged portion of the present HEMT device according to various embodiments;

[0050] Figure 5A-5H Shows Figure 3 A cross section of a HEMT device in a subsequent manufacturing step; and

[0051] Figure 6 A cross section of the present HEMT device according to another embodiment is shown. DETAILED DESCRIPTION

[0052] The following description refers to the arrangements shown in the accompanying drawings; therefore, expressions such as "above", "below", "lower", "upper", "right", "left", "top", "bottom", etc. are related to the accompanying drawings and should not be interpreted in a limiting manner.

[0053] Figure 2 The HEMT device 20 is shown in a Cartesian reference frame XYZ comprising a first axis X, a second axis Y and a third axis Z.

[0054] The HEMT device 20 is formed in a body 22 having a top surface 22A and including a substrate 24 and a heterostructure 25 extending on the substrate 24 .

[0055] Substrate 24 may be formed of one or more layers of silicon, silicon carbide, gallium nitride (GaN), sapphire (Al 2 O 3 ), or other materials.

[0056] exist Figure 2 In the embodiment of FIG. 2 , the substrate 24 is made of a semiconductor material.

[0057] In detail, although Figure 2 25, but the substrate 24 may include a substrate layer, for example made of silicon or silicon carbide, and a buffer layer, for example made of GaN, extending on the substrate layer. In this case, the buffer layer extends between the substrate layer and the heterostructure 25 to allow the heterostructure to grow during the manufacture of the HMET device 20.

[0058] The heterostructure 25 includes a compound semiconductor material including elements of Group III and Group V of the periodic table, extends on the substrate 24 and forms a top surface 22A of the body 22 .

[0059] The heterostructure 25 includes a channel layer 28 and a barrier layer 29 that overlap each other.

[0060] The channel layer 28 is made of a first semiconductor material, for example gallium nitride (GaN) or an alloy including gallium nitride, such as InGaN. Here, the channel layer 28 is made of gallium nitride (GaN), extends on the substrate 24 and has a top surface 28A.

[0061] The barrier layer 29 is made of a second semiconductor material, such as a compound based on a ternary or quaternary gallium nitride alloy, such as Al x Ga 1-x N, AlInGaN, In x Ga 1-x N、Al x In 1-x Al, AlScN, where barrier layer 29 is made of aluminum gallium nitride (AlGaN), and barrier layer 29 extends between a top surface 28A of channel layer 28 and a top surface 22A of body 22 .

[0062] Depending on the specific application, the channel layer 28 and the barrier layer 29 may be of intrinsic type, P-type, or N-type; in particular, both the channel layer 28 and the barrier layer 29 may be of N-type.

[0063] For example, when barrier layer 29 is made of AlGaN, the presence of aluminum atoms may cause barrier layer 29 to be N-type.

[0064] The heterostructure 25 is configured to accommodate a two-dimensional gas (2DEG) of (mobile) charge carriers (especially electrons here), Figure 2 It is indicated by a dashed rectangle 30 in FIG.

[0065] In detail, Figure 2 In the embodiment of FIG. 5 , the two-dimensional gas 30 is disposed at the interface between the channel layer 28 and the barrier layer 29 , that is, disposed at the top surface 28A of the channel layer 28 .

[0066] The HEMT device 20 includes a gate structure 33 (in Figure 3 ), source region 34 and drain region 35.

[0067] The gate structure 33 extends over the top surface 22A of the body 22 , in particular here in contact with the heterostructure 25 , and can be biased to electrically control the formation of the two-dimensional gas 30 .

[0068] The gate structure 33 includes a channel modulation region 40 of semiconductor material having a top surface 40A; a functional region 41 of semiconductor material; and a gate contact region 42 of conductive material, which are in electrical and physical contact with each other.

[0069] The channel modulation region 40 is made of a compound material containing elements of Group III and Group V of the periodic table. For example, the channel modulation region 40 can be made of gallium nitride (GaN) or an alloy including gallium nitride. In particular, the channel modulation region 40 is made of GaN here, and the channel modulation region 40 extends on the top surface 22A of the body 22.

[0070] In particular, the modulation zone 40 is in direct contact with the barrier layer 29 .

[0071] The thickness T of the channel modulation region 40 along the third axis Z is g For example comprised between 30 nm and 150 nm, in particular approximately 100 nm, and a width W along the first axis X g For example, it is comprised between 0.5 μm and 2 μm.

[0072] For example, during the design phase, the width W g The selection can be made according to the maximum voltage that can be applied to the gate terminal G in use. For example, the higher the maximum voltage, the smaller the width W can be selected. g The bigger.

[0073] The channel modulation region 40 extends in a manner not shown here, for example in the form of a strip elongated along the second axis Y.

[0074] In detail, the channel modulation region 40 includes a central portion 45 arranged parallel to the first axis X between two peripheral portions 46 .

[0075] The channel modulation region 40 may comprise further peripheral portions in a manner not shown here, which extend at a distance from each other parallel to the second axis Y. The central portion 45 is arranged between two further peripheral portions parallel to the second axis Y. The peripheral portions 46 may be joined together and form a single peripheral portion defining the outer perimeter of the channel modulation region 40 .

[0076] The peripheral portion 46 forms lateral sidewalls 47 of the channel modulation region 40 .

[0077] The central portion 45 and the peripheral portion 46 form a central portion and a peripheral portion of the top surface 40A of the channel modulation region 40 , respectively.

[0078] The channel modulation region 40 has a different conductivity type relative to the barrier layer 29 , so that the channel modulation region 40 and the barrier layer 29 form a junction, in particular a pn junction here.

[0079] The channel modulation region 40 is of P type and the concentration of the doping material (eg, magnesium atoms) is, for example, within 10 18 with 10 20 Atom / cm 3 between 2 and 10 19 Atom / cm 3 .

[0080] The channel modulation region 40 is designed to modulate the formation of the two-dimensional gas 30 based on the voltage applied to the gate electrode G.

[0081] In detail, the channel modulation zone 40 controls the formation of the two-dimensional gas 30 in a portion 49 of the heterostructure 25 which is arranged parallel to the third axis Z at the channel modulation zone 40 , in particular vertically aligned with the channel modulation zone 40 .

[0082] In the embodiment shown, in the absence of voltage applied to the gate contact region 42 , or in the case of zero voltage applied, the channel modulation region 40 causes a portion 49 of the heterostructure 25 to be depleted of charge carriers, ie, such that the two-dimensional layer 30 is interrupted at the portion 49 .

[0083] The HEMT device 20 is of a normally-off type.

[0084] The functional region 41 is made of a compound material containing elements of Group III and Group V of the periodic table, for example, the functional region 41 can be made of gallium nitride (GaN) or an alloy including gallium nitride, and extends on the top surface 40A of the channel modulation region 40, and in particular is in direct contact with the top surface 40A.

[0085] The functional region 41 may be made of the same material as the channel modulation region 40 , or made of a different material; in this embodiment, the functional region 41 is made of the same material, specifically GaN.

[0086] The functional region 41 is designed to form a junction with the channel modulation region 40 .

[0087] The functional region 41 has a different conductivity type relative to the channel modulation region 40 .

[0088] In detail, according to one embodiment, the functional region 41 may be of an intrinsic type, that is, the functional region 41 has a zero concentration of doping substances, for example, is made of intrinsic GaN.

[0089] According to different embodiments, the functional region 41 may have a conductivity type opposite to that of the channel modulation region 40, that is, the functional region 41 may be N-type. For example, the functional region 41 may be N-type GaN. For example, the concentration of doping atoms (e.g., silicon atoms) in the functional region 41 may be included in 10 12 with 10 15 Atom / cm 3 between.

[0090] The thickness T of the functional area 41 parallel to the third axis Z f For example, it is comprised between 5 nm and 50 nm, in particular approximately 30 nm.

[0091] In detail, the functional region 41 is formed by portions 50 , each portion 50 having an inner wall 51 and an outer wall 52 , each portion 50 extending over a respective peripheral portion 46 of the channel modulation region 40 .

[0092] Portions 50 of the functional region 41 are joined together in a manner not shown here so as to form an annular structure, for example a circle, a polygon or any other shape, for example depending on the shape of the underlying channel modulation region 40 .

[0093] However, the portions 50 may be separate from each other.

[0094] The width W of each of the portions 50 parallel to the first axis X is f Smaller than width W g For example, the width W f It can be comprised between 20nm and 200nm.

[0095] Here, the portion 50 extends across the width parallel to the first axis X of the peripheral portion 46 of the channel modulation region 40 .

[0096] In other words, in this embodiment, the outer wall 52 of the portion 50 forms an extension of the lateral sidewall 47 of the channel modulation region 40 , in particular is vertically aligned with the lateral sidewall 47 .

[0097] A gate contact region 42 (eg, multi-layer Ti / AlCu / TiN) extends over the top surface 40A of the channel modulation region 40 , and in particular is in direct contact with the top surface 40A.

[0098] The gate contact region 42 forms a first Schottky contact with the channel modulation region 40 .

[0099] A first layer (not explicitly shown here) of the gate contact region 42 in contact with the channel modulation region 40, e.g. Ti, TiN, Ta, TiW, TaW, may have a thickness e.g. comprised between 10 nm and 50 nm, in particular greater than 30 nm, to facilitate the formation of electrical contact.

[0100] The gate contact region 42 extends and is also in direct contact with the functional region 41 .

[0101] In detail, the gate contact region 42 extends over a central portion 45 of the channel modulation region 40 , contacting an inner wall 51 of a portion 50 of the functional region 41 .

[0102] In other words, the gate contact region 42 is arranged in parallel with the functional region 41. Figure 2 and Figure 3 Laterally adjacent in the direction of the first axis X.

[0103] Depending on the specific doping level of the functional region 41 and the specific material of the gate contact region 42 contacting the functional region 41 , the gate contact region 42 may form an ohmic contact or a second Schottky contact with the functional region 41 .

[0104] In the case of a Schottky contact, due to the different conductivity types between the functional region 41 and the channel modulation region 40 , the Schottky barrier of the second Schottky contact in an equilibrium state has a different height, in particular a smaller height, relative to the Schottky barrier of the first Schottky contact.

[0105] The width W of the gate contact region 42 parallel to the first axis X is m Smaller than width W g .

[0106] According to one embodiment, the width W can be selected at the design stage. m and W f , to increase the ratio W m / W f , that is, to increase the contact surface between the gate contact region 42 and the channel modulation region 40, for example, so that each width W f In this way, a homogeneous biasing of the entire channel modulation region 40 can be obtained.

[0107] A passivation layer 58 of insulating material (e.g., oxide, particularly aluminum oxide (Al2O3)) extends laterally toward the channel modulation region 40 on a portion of the top surface 22A of the body 22 between the source region 34 and the channel modulation region 40 and between the drain region 35 and the channel modulation region 40.

[0108] The passivation layer 58 also extends on the lateral sidewalls 47 of the channel modulation region 40 and on the outer walls 52 of the functional region 41 .

[0109] In this embodiment, the passivation layer 58 also extends over and contacts the functional region 41 .

[0110] The thickness of the passivation layer 58 is, for example, comprised between 2 nm and 10 nm, in particular approximately 5 nm.

[0111] A first insulating layer 60 , for example made of silicon oxide or silicon nitride, extends over the passivation layer 58 .

[0112] The gate contact region 42 extends through the passivation layer 58 and the first insulating layer 60 and extends partially on the first insulating layer 60 .

[0113] A second insulating layer 62 made of, for example, silicon oxide or silicon nitride extends over the first insulating layer 60 and over the gate contact region 42 .

[0114] The passivation layer 58 and the first and second insulating layers 60 and 62 form two openings 64 , 65 extending on two opposite sides of the gate structure 33 parallel to the first axis X, which expose the top surface 22A of the body 22 .

[0115] The source region 34 is made of a conductive material, for example, Ti / AlCu / TiN, and extends on the top surface 22A of the body 22 within the opening 64 , contacting the barrier layer 29 .

[0116] exist Figure 2 In the embodiment of FIG. 5 , the source region 34 also includes a portion extending on the second insulating layer 62 above the gate contact region 42 and acting as an electromagnetic shield.

[0117] The HEMT device 20 may further include a field plate region 69 extending between the first insulating layer 60 and the second insulating layer 62 laterally and at a distance relative to the gate contact region 42 along the first axis X. In this case, the source region 34 may also extend over the field plate region 69, where the source region 34 again acts as a shield.

[0118] The drain region 35 is made of a conductive material, for example, Ti / AlCu / TiN, and extends on the top surface 22A of the body 22 within the opening 65 , contacting the barrier layer 29 .

[0119] The source region 34 and the drain region 35 form an electrical contact, in particular an ohmic type electrical contact, with the heterostructure 25 , in particular with the two-dimensional gas 30 . The source region 34 and the drain region 35 are current conducting regions of the HEMT device 20 .

[0120] Depending on the specific application of HEMT device 20 and the specific manufacturing process used to obtain source region 34 and drain region 35 , source region 34 and drain region 35 may be in direct ohmic contact with two-dimensional gas 30 , or may be in electrical contact with two-dimensional gas 30 through barrier layer 29 .

[0121] For example, according to one embodiment, the source region 34 and the drain region 35 may be of a recessed type, i.e., they may extend completely or partially through the barrier layer 29 parallel to the third axis Z. In this way, the source region 34 and the drain region 35 may be in direct electrical contact with the two-dimensional gas 30, thereby obtaining a low contact resistance.

[0122] The HEMT device 20 allows to obtain improved performance with respect to gate leakage current that may originate from the lateral sidewalls 47 of the channel modulation region 40 .

[0123] The presence of the functional region 41 on the top surface 40A of the channel modulation region 40 actually results in the formation of a depletion region at the interface between the functional region 41 and the channel modulation region 40 .

[0124] The depletion region, which is devoid of (mobile) charge carriers (ie, electrons and holes), is the region that electrically insulates the gate contact region 42 from the lateral sidewalls 47 of the channel modulation region 40 .

[0125] Therefore, the gate contact region 42 is parallel to the functional region 41. Figure 2 and Figure 3 The fact that the first axis X in FIG. 4 is arranged laterally allows obtaining a high electrical insulation between the gate contact region 42 and the lateral sidewalls 47 .

[0126] Therefore, in the on-state (the voltage applied to the gate electrode G is greater than zero), the HEMT device 20 has a low leakage current.

[0127] In particular, when the channel modulation region 40 and the functional region 41 have opposite conductivity types, for example, P conductivity type and N conductivity type, respectively, electrical insulation and leakage current reduction effects may be better than the case where the functional region 41 is of intrinsic type.

[0128] The fact that the gate contact region 42 is in contact with the functional region 41, and in particular can be in ohmic contact with the functional region 41, makes the junction formed by the functional region 41 and the channel modulation region 40 in a reverse biased state when the HEMT device 20 is turned on (the voltage applied to the gate electrode G is greater than zero). In this way, the depletion region generated prevents the flow of lateral leakage current.

[0129] Additionally, in the present embodiment, the presence of the passivation layer 58 extending on the lateral sidewalls 47 of the channel modulation region 40 allows for reduced formation of surface electronic states on the lateral sidewalls 47, further contributing to reduced leakage current.

[0130] In addition, the fact that the gate contact region 42 extends directly on the top surface 40A allows obtaining a Schottky junction with the channel modulation region 40 and obtaining a more effective biasing of the channel modulation region 40 relative to, for example, a situation in which the functional region 41 extends parallel to the third axis Z between the channel modulation region 40 and the gate contact region 42.

[0131] Figure 4 A portion of a HEMT device, indicated here by 120 , is shown according to various embodiments.

[0132] The HEMT device 120 has Figure 2 The HEMT device 20 has a similar overall structure; therefore, Figure 4 Only a portion of the HEMT device 120 is shown, which portion highlights the differences of the HEMT device 120 relative to the HEMT device 20 .

[0133] Figure 4 Elements of the HEMT device 120 visible in FIG. 1 that are common to the HEMT device 20 are indicated by the same reference numerals and are not described in further detail.

[0134] Furthermore, although not shown, the HEMT device 120 also includes a source region, a drain region, and a substrate, as discussed with respect to the HEMT device 20 .

[0135] The HEMT device 120 has a gate structure, indicated here by 133 , which here also includes a channel modulation region 40 with corresponding portions 45 , 46 , a functional region 41 with corresponding portion 50 , and a gate contact region 42 .

[0136] In this embodiment, the gate contact region 42 is a first gate contact region and the HEMT device 120 further includes a second gate contact region 144 in direct electrical contact with the first gate contact region 42 .

[0137] The second gate contact region 144 is made of a conductive material, for example a single layer of Ti, TiN, Ta or TaN, having a thickness comprised between 5 nm and 50 nm, and extends over and contacts the functional region 41. The second gate contact region 144 creates a barrier that prevents contaminants from diffusing.

[0138] The second gate contact region 144 is arranged adjacent to the first gate contact region 42 along the first axis X.

[0139] Depending on the conductivity type and doping level of the functional region 41 , the second gate contact region 144 makes ohmic or quasi-ohmic contact with the functional region 41 .

[0140] In particular, when the functional region 41 is of N-type, the second gate contact region 144 may be in ohmic contact with the functional region 41 .

[0141] In detail, second gate contact region 144 is formed of two parts, each part extending over a corresponding part 50 of functional region 41. These parts can be separated from each other or joined together in a manner not shown here, for example based on a specific design layout and based on the shape of functional region 41.

[0142] The width of each portion of the second gate contact region 144 parallel to the first axis X may be smaller than or equal to (here, smaller than) the width of the portion 50 of the functional region 41 .

[0143] The passivation layer indicated here by 158 is made of the same material as the passivation layer 58 of the HEMT device 20 and extends over the top surface 22A, the lateral sidewalls 47 of the channel modulation region 40 and the outer walls 52 of the functional region 41 .

[0144] Furthermore, the passivation layer 158 extends partially over the portion 50 of the functional region 41 , on the outer walls of the second gate contact region 144 and on the top of the second gate contact region 144 .

[0145] A second insulating layer, indicated here by 160 , made of silicon oxide or silicon nitride, for example, extends over the passivation layer 158 and is in contact therewith.

[0146] HEMT device 120 has the same advantages discussed with reference to HEMT device 20 .

[0147] Furthermore, the presence of the second gate contact region 144 in ohmic or quasi-ohmic contact (here ohmic contact) with the functional region 41 allows for more accurate control of the behavior of the electric field in the peripheral portion 46 of the gate modulation region 40 .

[0148] Therefore, the HEMT device 120 allows the entire corresponding width W g The formation of the 2DEG 30 below the channel modulation region 40 is accurately controlled.

[0149] In the following, reference Figure 5A-5H , describing the manufacturing steps of the HEMT device 120.

[0150] Figure 5A 2 shows a cross section of a working body 200 having a top surface 200A. Figure 2 and Figure 3 Elements common to the described elements are denoted by the same reference numerals and are not described in further detail.

[0151] In the working body 200 , a substrate 24 and a heterostructure 25 including a channel layer 28 and a barrier layer 29 have been formed.

[0152] In this embodiment, the barrier layer 29 forms the top surface 200A of the working body 200 .

[0153] In addition, a channel modulation layer 202 (for example, P-type GaN) of a semiconductor material has been formed on the top surface 200A and is used to form a channel modulation region 40; and a functional layer (for example, intrinsic or N-type GaN) of a semiconductor material has been formed on the channel modulation layer 202 and is used to form a functional region 41.

[0154] For example, the layers 202 and 203 are epitaxially grown by metal organic chemical vapor deposition (MOCVD).

[0155] Later, in Figure 5B In the embodiment, a contact layer 204 (for forming the second gate contact region 144 ) is formed on the functional layer 203 , such as a single layer or multiple layers of Ti, TiN, Ta, TaN, etc.

[0156] Layers 202 , 203 , 204 may be deposited or grown without a mask (ie, in a blanket mode).

[0157] The working body 200 may be subjected to annealing to promote the formation of an ohmic contact between the contact layer 204 and the functional layer 203 .

[0158] exist Figure 5C In the embodiment, the layer 202, the layer 203 and the layer 204 are defined by photolithography and etching steps. The channel modulation layer 202 is patterned to form the channel modulation region 40.

[0159] A remaining portion 206 extending over the channel modulation region 40 , in particular over the corresponding top surface, remains from the functional layer 203 .

[0160] A remaining portion 208 extending over portion 206 remains from contact layer 204 .

[0161] Then, in Figure 5D In the embodiment, a working passivation layer 212 for forming the passivation layer 158 , such as an oxide, in particular Al 2 O 3 , is formed on the front side of the working body 200 without a mask.

[0162] The working passivation layer 212 covers the exposed portions of the top surface 200A, the exposed portions of the modulation region 40 , and the exposed portions of the portions 206 , 208 .

[0163] For example, the working passivation layer 212 is formed by thermal or plasma atomic layer deposition (ALD) in a water-based atmosphere at 300°C.

[0164] A first working insulating layer 213 of, for example, silicon oxide or silicon nitride for forming the first insulating layer 160 is formed on the working passivation layer 212 without a mask.

[0165] Then, in Figure 5E A window 215 is opened through the first working insulating layer 213 , the working passivation layer 212 and the portions 206 , 208 to expose a portion of the top surface 40A of the channel modulation region 40 intended to form the first gate contact region 42 .

[0166] The second contact region 144 is formed by a remaining portion 208 of the contact layer 204 .

[0167] The functional region 41 of the gate structure 133 is formed by the remaining portion 206 of the functional layer 203 .

[0168] exist Fig. 5F In the process, a working conductive region is deposited and patterned to form a first gate contact region 42 within the window 215 and partially on the first working insulating layer 213 .

[0169] In addition, a field plate region 69 may also be formed from the active conductive region.

[0170] Later, in Figure 5G In the process, the second working insulating layer 218 is deposited on the first working insulating layer 213, the first gate contact region 42 and the field plate region 69 (if present).

[0171] exist Figure 5H In the embodiment, the working passivation layer 212 and the first working insulating layer 213 and the second working insulating layer 218 are patterned by photolithography and etching steps, thereby forming the first insulating layer 160 and the second insulating layer 62 and the passivation layer 158 having corresponding openings 64 and 65.

[0172] The source region 34 and the drain region 35 are formed in the opening 64 and the opening 65 , respectively.

[0173] After final manufacturing steps (not shown here but known per se, such as cutting the working body 200 and forming electrical connections), the HEMT device 120 is obtained.

[0174] It will be clear to those skilled in the art that reference Figure 5A-5H The described manufacturing process can also be adapted to manufacture Figure 1 The HEMT device 20 is formed by, for example, omitting steps associated with forming the second gate contact region 144 .

[0175] and Figure 5A-5HUnlike the HEMT device manufactured according to the gate-first method (forming the gate structure 33, 133 before the source region 34 and the drain region 35) shown in , the HEMT device can be manufactured according to the ohmic-first method, in which the source region and the drain region are formed before the gate structure. In this way, a heat treatment can be used while forming the source region and the drain region, which reduces the contact resistance between the source region and the drain region and the two-dimensional gas. Figure 6 An embodiment of the present HEMT device, indicated herein by 320, fabricated according to an Ohm-first approach is shown. HEMT device 320 has a similar overall structure as HEMT devices 20, 120; therefore, common elements are identified by like reference numerals and are not described again in detail.

[0176] In the HEMT device 320 , a second insulating layer, indicated here by 321 , extends over source and drain regions, indicated here by 324 and 325 , respectively.

[0177] The first gate contact region, indicated here by 342 , also extends through the second insulating layer 321 .

[0178] The field plate region, indicated here by 329 , extends laterally to the gate structure 133 between the first insulating layer 160 and the second insulating layer 321 and may be formed together with the source region 324 and the drain region 325 .

[0179] The HEMT device 320 further includes a shielding conductive region 330 that partially extends over the second insulating layer 321 and serves as an electromagnetic shield. The shielding conductive region 330 partially or completely, here partially, extends at a distance above the field plate region 329 .

[0180] Finally, it is evident that modifications and variations may be made to the HEMT devices 20 , 120 and to the processes for their manufacture described and illustrated herein, without departing from the scope of the present disclosure as defined in the appended claims.

[0181] For example, referring to HEMT device 20 , functional region 41 may extend only on one side of gate contact region 42 , such as on the source side (left side) or the drain side (right side) of gate contact region 42 . Functional region 41 may have only one of portions 50 .

[0182] Similar considerations may also apply to the HEMT devices 120 , 220 .

[0183] For example, the portion 50 of the functional region 41 may extend over only a portion of the peripheral portion 46 of the channel modulation region 40 .

[0184] For example, the outer wall 52 of the functional region 41 may not be aligned along the third axis Z with the lateral sidewall 47 of the channel modulation region 40 .

[0185] For example, the value W m +2 W f It can be smaller than the width W of the channel modulation region 40 g , for example if a further region, for example of insulating material, is arranged directly on the top surface 40A of the channel modulation region 40 .

[0186] Depending on the specific application, the source region 34, 324, the drain region 35, 325 and the gate structure 33, 133 can extend along the second axis Y according to different shapes and configurations in a manner known per se, so they will not be discussed in detail. For example, in a top view (not shown here), they can have the shape of strips elongated along the second axis Y, or can have a circular shape or any other regular or irregular shape.

[0187] For example, channel layer 28 and barrier layer 29 may each be formed of a properly doped or intrinsic GaN-based alloy or multiple layers (eg, one or more GaN layers) stacked on top of each other, depending on the specific application.

[0188] For example, depending on the specific application or the specific manufacturing process used, the body 22 of the HEMT device may also include one or more insulating layers, for example made of oxide or nitride, extending between the substrate 24 and the heterostructure 25, above the heterostructure 25, within the substrate 24 and / or within the heterostructure 25.

[0189] For example, body 22 may include a cap layer, for example made of aluminum nitride (AlN), which extends over and is in contact with barrier layer 29 , ie forms surface 22A.

[0190] For example, the body 22 may include a spacer layer, for example made of AlN, between the channel layer 28 and the barrier layer 29 .

[0191] For example, in a manner known per se, the present HEMT device may comprise a stack of overlying layers extending between the substrate 24 and the heterostructure 25, the stack of overlying layers comprising, for example, a buffer layer and a hole supply layer.

[0192] For example, the HEMT device may be of a normally-on type.

[0193] Finally, the different embodiments described above may be combined to provide further solutions.

[0194] The HEMT device (20; 120; 320) can be summarized as comprising: a body (22) having a top surface (22A) and comprising a heterostructure (25) configured to generate a two-dimensional charge carrier gas (30, 49); and a gate structure (33; 133) extending over the top surface of the body and capable of being biased to electrically control the two-dimensional charge carrier gas, the gate structure comprising a channel modulation region (40) of semiconductor material and having a top surface (40A); a functional region (41) of semiconductor material; and a gate contact region (42; 342) of conductive material, wherein the functional region (41) and the gate contact region (42; 342) extend over the top surface (40A) of the channel modulation region, and wherein the gate contact region is arranged laterally relative to the functional region, and the channel modulation region (40) has a different conductivity type relative to the functional region (41).

[0195] The channel modulation region (40) may have a peripheral portion (46) forming a lateral sidewall (47) of the channel modulation region, and the functional region (41) extends at least partially over the peripheral portion.

[0196] The functional region (41) may have an outer wall (52) adjacent to the lateral sidewalls of the channel modulation region.

[0197] The functional area (41) may be of intrinsic type.

[0198] The channel modulation region (40) may have a first conductivity type (P), and the functional region (41) may have a second conductivity type (N) opposite to the first conductivity type.

[0199] The concentration of doping substances in the functional area can be less than 10 15 Atom / cm 3 .

[0200] The functional region (41) may have an inner wall (51) in contact with the gate contact region (42; 342).

[0201] The functional area (41) may have a width (W) along the first direction (X). f ), and the gate contact region (42; 342) may have a width (W) along the first direction. m ), which can be larger than the width of the functional area.

[0202] The gate contact region may be a first gate contact region (42; 342), and the gate structure (133) may further include a second gate contact region (144) of conductive material, which extends at least partially over the functional region (41) and contacts the first gate contact region.

[0203] The material of the second gate contact region (144) may be selected from Ti, Ta, TiN and TaN.

[0204] The HEMT device may further include an insulating region (58, 60; 158, 160) of non-conductive material extending on lateral sidewalls (47) of the channel modulation region (40) and on outer walls (52) of the functional region (41), and extending at least partially on a top surface of the channel modulation region (40) and at a distance relative to the top surface of the channel modulation region (40).

[0205] A process for manufacturing a HEMT device can be summarized as comprising: forming a gate structure (33; 133) on a top surface (200A) of a working body (200) comprising a heterostructure (25), wherein the heterostructure is configured to generate a two-dimensional charge carrier gas (30, 49), and the gate structure is capable of biasing to electrically control the two-dimensional charge carrier gas, wherein forming the gate structure comprises: starting from a first semiconductor layer (202), forming a channel modulation region (40) having a top surface (40A); starting from a second semiconductor layer (203), forming a functional region (41) on the top surface of the channel modulation region; and forming a gate contact region (42; 342) of conductive material, the gate contact region extending laterally on the top surface of the channel modulation region relative to the functional region, wherein the channel modulation region (40) has a different conductivity type relative to the functional region (41).

[0206] The manufacturing process may include: depositing a first semiconductor layer (202) on the top surface of the working body; depositing a second semiconductor layer (203) on the first semiconductor layer; patterning the first semiconductor layer and the second semiconductor layer; depositing at least one insulating layer (58, 60; 158, 160) of non-conductive material on the lateral sidewalls (47) of the patterned first semiconductor layer, on the outer wall (52) of the patterned second semiconductor layer, and at least partially above the top surface of the first semiconductor layer and at a certain distance relative to the top surface of the first semiconductor layer; wherein forming a gate contact area may include: forming an opening (215) exposing a portion of the patterned first semiconductor layer, the opening extending through the at least one insulating layer and the second semiconductor layer; and depositing a conductive material in the opening.

[0207] The gate contact region may be a first gate contact region, wherein forming the gate structure may also include forming a second gate contact region (144) extending at least partially over the functional region (41) and contacting the first gate region, wherein forming the second gate contact region may include: forming at least one contact layer of conductive material on the second semiconductor material layer (203); performing annealing configured to promote formation of an ohmic contact between the at least one contact layer and the second semiconductor layer; and removing a portion of the at least one contact layer above the second semiconductor layer intended to form the first gate contact region.

[0208] The manufacturing process may also include forming at least one current conducting region (34, 35; 324, 325) of conductive material in contact with the heterostructure (25), the step of forming the gate structure (33; 133) being performed before or after forming the at least one current conducting region.

[0209] In an embodiment, a process for manufacturing a HEMT device includes: forming a gate structure on a top surface of a working body including a heterostructure, wherein the heterostructure is configured to generate a two-dimensional charge carrier gas, and the gate structure can be biased to electrically control the two-dimensional charge carrier gas, wherein forming the gate structure includes: starting from a first semiconductor layer, forming a channel modulation region having a top surface; starting from a second semiconductor layer, forming a functional region on the top surface of the channel modulation region; forming a first gate contact region of conductive material, the first gate contact region extending laterally on the top surface of the channel modulation region relative to the functional region, wherein the channel modulation region has a different conductivity type relative to the functional region; forming a second gate contact region extending at least partially on the functional region and contacting the first gate contact region; and forming at least one insulating layer of non-conductive material on the second gate contact region, the at least one insulating layer of non-conductive material being partially below the first gate region.

[0210] In some embodiments, the manufacturing process also includes: forming a first semiconductor layer on the top surface of the working body; forming a second semiconductor layer on the first semiconductor layer; patterning the first semiconductor layer and the second semiconductor layer; wherein forming at least one insulating layer of non-conductive material also includes depositing at least one insulating layer on the lateral sidewalls of the patterned first semiconductor layer, on the outer wall of the patterned second semiconductor layer, and at least partially above the top surface of the first semiconductor layer and at a certain distance relative to the top surface of the first semiconductor layer; wherein forming a first gate contact area includes: forming an opening exposing a portion of the patterned first semiconductor layer, the opening extending through at least one insulating layer and the second semiconductor layer; and forming a conductive material within the opening.

[0211] In some embodiments, forming the second gate contact region also includes: forming at least one contact layer of conductive material on the second semiconductor material layer; performing annealing configured to promote formation of ohmic contact between the at least one contact layer and the second semiconductor layer; and removing a portion of the at least one contact layer above the second semiconductor layer intended to form the first gate contact region.

[0212] In some embodiments, the manufacturing process further includes forming at least one current conducting region of conductive material in contact with the heterostructure, and the step of forming the gate structure is performed before or after forming the at least one current conducting region.

[0213] The various embodiments described above can be combined to provide further embodiments. Aspects of the embodiments can be modified, if necessary to employ concepts of the various patents, applications, and publications to provide further embodiments.

[0214] These and other changes can be made to the embodiments in light of the above detailed description. In general, in the appended claims, the terms used should not be interpreted as limiting the claims to the specific embodiments disclosed in the specification and claims, but should be interpreted as including all possible embodiments and the full range of equivalent forms given to these embodiments. Therefore, the claims are not limited by the present disclosure.

Claims

1. A HEMT device, characterized in that: include: a body having a top surface and comprising a heterostructure configured to generate a two-dimensional charge carrier gas; as well as a gate structure extending over the top surface of the body and capable of being biased to electrically control the two-dimensional charge carrier gas, The gate structure comprises: a channel modulation region of semiconductor material, the channel modulation region having a top surface; Functional regions of semiconductor materials; a first gate contact region of conductive material; a second gate contact region of conductive material, the second gate contact region extending at least partially over the functional region and in contact with the first gate contact region, wherein the functional region and the first gate contact region extend on the top surface of the channel modulation region, and wherein the first gate contact region is arranged laterally relative to the functional region, The channel modulation region has a different conductivity type than the functional region; and An insulating region of non-conductive material is between the first gate contact region and the second gate contact region. 2 . The HEMT device according to claim 1 , wherein the channel modulation region has a peripheral portion forming a lateral sidewall of the channel modulation region, and the functional region extends at least partially over the peripheral portion. 3 . The HEMT device of claim 2 , wherein the functional region has an outer wall adjacent to the lateral sidewall of the channel modulation region. The HEMT device according to claim 1 , wherein the functional region is of intrinsic type. 5 . The HEMT device of claim 1 , wherein the channel modulation region has a first conductivity type, and the functional region has a second conductivity type opposite to the first conductivity type.

6. The HEMT device according to claim 5, wherein the concentration of the doping substance in the functional region is less than 10 15 Atom / cm 3 . 7 . The HEMT device according to claim 1 , wherein the functional region has an inner wall in contact with the first gate contact region. 8 . The HEMT device of claim 1 , wherein the functional region has a width along a first direction, and a width of the first gate contact region along the first direction is greater than a width of the functional region. 9 . The HEMT device according to claim 8 , wherein a width of the second gate contact region along the first direction is smaller than a width of the functional region along the first direction. 10 . The HEMT device according to claim 9 , wherein a material of the second gate contact region is selected from the group consisting of Ti, Ta, TiN and TaN.

11. The HEMT device of claim 1 , wherein the insulating region extends on lateral sidewalls of the channel modulation region and on outer walls of the functional region, and extends at least partially on the top surface of the channel modulation region and at a distance relative to the top surface of the channel modulation region.

12. A device, characterized in that include: a heterostructure having a surface and configured to generate a two-dimensional charge carrier gas; a channel modulation region of semiconductor material on said surface; a functional region of semiconductor material on the channel modulation region; a first gate contact region on the functional region; an opening passing through the first gate contact region and the functional region, the opening exposing a top surface of the channel modulation region; a passivation layer on surfaces of the heterostructure, the functional region, and the first gate contact region, the passivation layer not being in the opening and not being on the top surface of the channel modulation region; as well as A second gate contact region in the opening contacts the top surface of the channel modulation region and overlaps the passivation layer.

13. The device of claim 12, wherein the functional area comprises two separate parts, the opening being between the two separate parts. The device according to claim 12 , wherein the functional area is circular or polygonal in shape. 15 . The device of claim 12 , wherein the first gate contact region has a first width and the functional region has a second width greater than the first width.

16. The device of claim 15, wherein the passivation layer contacts the sidewalls of the functional region and a top region of the functional region, the top region being coplanar with a first plane in which the functional region contacts the first gate contact region, and the sidewalls of the functional region being on a second plane transverse to the first plane.

Images