Method for co-integration of III-V devices with group IV devices

Abstract

The present disclosure relates to a semi-conductor structure and method for co-integrating a III-V device with a group IV device on a Si.sub.xGe.sub.1-x(100) substrate. The method includes: (a) providing a Si.sub.xGe.sub.1-x(100) substrate, where x is from 0 to 1; (b) selecting a first region for forming therein a group IV device and a second region for forming therein a III-V device, the first and the second region each comprising a section of the Si.sub.xGe.sub.1-x(100) substrate; (c) forming a trench isolation for at least the III-V device; (d) providing a Si.sub.yGe.sub.1-y(100) surface in the first region, where y is from 0 to 1; (e) at least partially forming the group IV device on the Si.sub.yGe.sub.1-y(100) surface in the first region; (f) forming a trench in the second region which exposes the Si.sub.xGe.sub.1-x(100) substrate, the trench having a depth of at least 200 nm, at least 500 nm, at least 1 μm, usually at least 2 μm, such as 4 μm, with respect to the Si.sub.yGe.sub.1-y(100) surface in the first region; (g) growing a III-V material in the trench using aspect ratio trapping; and (h) forming the III-V device on the III-V material, the III-V device comprising at least one contact region at a height within 100 nm, 50 nm, 20 nm, usually 10 nm, of a contact region of the group IV device.

Claims

1. A method comprising: a) providing a Si.sub.xGe.sub.1-x(100) substrate, wherein x ranges from 0 to 1; b) selecting a first region for forming therein a group IV device and a second region for forming therein a III-V device, the first and the second region each comprising a section of the Si.sub.xGe.sub.1-x(100) substrate; c) forming a trench isolation in the second region of the Si.sub.xGe.sub.1-x(100) substrate for at least the III-V device; d) after step c, providing a Si.sub.yGe.sub.1-y(100) surface in the first region of the Si.sub.xGe.sub.1-x(100) substrate, wherein y ranges from 0 to 1; e) after step d but before step f, at least partially forming the group IV device on the Si.sub.yGe.sub.1-y(100) surface in the first region; f) after step e but before step g, forming a trench in the second region of the Si.sub.xGe.sub.1-x(100) substrate through the trench isolation and partially into the Si.sub.xGe.sub.1-x(100) substrate which exposes the Si.sub.xGe.sub.1-x(100) substrate, the trench having a depth of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, with respect to the Si.sub.yGe.sub.1-y(100) surface in the first region; g) after step f but before step h, growing a III-V material in the trench using aspect ratio trapping; and h) after step q, forming the III-V device on the III-V material in the trench, the III-V device comprising at least one contact region at a height within 100 nm, 50 nm, 20 nm, or 10 nm, of a contact region of the group IV device.

2. The method according to claim 1, wherein the trench has a V-shaped abutment defined by (111)-oriented surfaces of the Si.sub.xGe.sub.1-x(100) substrate.

3. The method according to claim 1, wherein the III-V device is a heterojunction bipolar transistor.

4. The method according to claim 1, wherein the Si.sub.xGe.sub.1-x(100) substrate is a Si(100) substrate and/or the Si.sub.yGe.sub.1-y(100) surface is a Si(100) surface.

5. The method according to claim 1, wherein the trench isolation formed in step c has a bottom and wherein step d comprises providing the Si.sub.yGe.sub.1-y(100) surface in the first region at a height of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, above the trench isolation bottom.

6. The method according to claim 1, wherein step d comprises depositing a dielectric layer over the Si.sub.xGe.sub.1-x(100) substrate having a thickness of at least 100 nm, at least 200 nm, at least 500 nm, or at least 1 μm.

7. The method according to claim 1, wherein step d comprises exposing the Si.sub.xGe.sub.1-x(100) substrate in the first region and growing Si thereon to form the Si.sub.yGe.sub.1-y(100) surface.

8. The method according to claim 1, wherein step d comprises transferring a Si layer comprising the Si.sub.yGe.sub.1-y(100) surface onto the Si.sub.xGe.sub.1-x(100) substrate.

9. The method according to claim 1, wherein the trench isolation provided in step c is a deep trench isolation and wherein the Si.sub.xGe.sub.1-x(100) substrate as provided in step a comprises the Si.sub.yGe.sub.1-y(100) surface of step d.

10. The method according to claim 1, wherein the trench of step f comprises a dielectric material and wherein the III-V material of step is embedded in the dielectric material.

11. The method according to claim 10, comprising a further step g′, after step g, of: g′1) at least partially removing the dielectric material adjacent to the III-V material; g′2) coating at least a lateral side of the III-V material with a passivation layer; and g′3) re-embedding the III-V material having the passivation layer in a dielectric material that is the same as or different from the dielectric material of step g′1.

12. The method according to claim 1, wherein the trench formed in step f comprises: a lower trench portion having a first trench width; and an upper trench portion having a second trench width, wherein the first trench width ranges from 0.05 to 0.75 times the second trench width.

13. The method according to claim 12, wherein the first trench width ranges from 0.1 to 0.5 times the second trench width.

14. The method according to claim 1, wherein the III-V material grown in the trench step g is a III-V nano-ridge, the III-V nano-ridge comprising a lower nano-ridge portion having a first nano-ridge width; an upper nano-ridge portion having a second nano-ridge width; and a middle nano-ridge portion between the upper and lower nano-ridge portions and tapering from the upper to the lower nano-ridge portion, wherein the first nano-ridge width ranges from 0.05 to 0.75 times the second nano-ridge width.

15. The method according to claim 14, wherein the first nano-ridge width ranges from 0.1 to 0.5 times the second nano-ridge width.

16. The method according to claim 1, wherein after step c but before step d, depositing a SiO.sub.2 dielectric layer on the trench isolation.

17. The method according to claim 1, wherein the III-V device comprising the at least one contact region has the same height of the contact region of the group IV device.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. A Si(100) wafer (10) can be provided, and a first region (21) and second region (22)—each comprising a section of the Si substrate (10)—may be selected for forming therein respectively a group IV device (50) and a III-V device (80). A shallow trench isolation (30) for the III-V device (80) may then formed with a SiO.sub.2 dielectric.

(2) FIG. 2 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. A SiO.sub.2 dielectric layer (40) can be deposited over the shallow trench isolation (30) and a window may be opened therein over the Si(100) base (11). Si (12) may then selectively regrown on the Si(100) base (11), thereby providing a Si(100) surface (15) at a height offset (h.sub.o) with respect to the bottom (32) of the trench isolation (30) (e.g. an offset of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm).

(3) FIG. 3 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. A shallow trench isolation (35) for the group IV device (50) can be formed in the regrown Si (12) with a SiO.sub.2 dielectric, to define a group IV active region (51).

(4) FIG. 4 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. One or more dummy gates (52) may be defined over the active region (51) and these may be flanked with a gate spacer.

(5) FIG. 5 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. The partially formed group IV device (50) may be covered with a SiO.sub.2 zeroth-level interlayer dielectric (ILD0; 55). Next, a trench (60) may be formed in the second region (22) down to and partially into the Si(100) substrate 10).

(6) FIG. 6 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. A III-V buffer layer (71) (e.g. GaAs, InP or InGaAs) may be grown in the trench (60) using ART. Subsequently, III-V layers (72) (cf. supra) for the III-V device (80) (e.g. a heterojunction bipolar transistor, HBT) may be grown on the III-V buffer layer (71). The III-V material (71,72) may form a nano-ridge (75) comprising a narrower lower nano-ridge portion (76) in the lower trench portion (63), an wider upper nano-ridge portion (78) in the upper trench portion (64), and a middle nano-ridge portion (77) tapering from the upper to the lower nano-ridge portion (76). Because of the tapering of the middle nano-ridge portion (77), air gaps (78) may be formed at the bottom of the upper trench portion (64).

(7) FIG. 7 schematically depicts a semiconductor structure at different stages of a method in accordance with a first exemplary embodiment of the present disclosure. the ILD0 (55) over the dummy gates (52) in the first region (21) can be opened up and a replacement metal gate process if performed thereon, after which the group IV device (50) may be covered with a protective layer, such as a oxide- or nitride-based dielectric material (e.g. SiO.sub.2, SiN, SiCN, SiON or Al.sub.2O.sub.3). Next, the III-V device (80) may be formed, e.g. by etching an HBT emitter and exposing the HBT base layer and cutting the base layer to expose an HBT subcollector (out of plane from the drawing). Remaining gaps in the second region (22) can then be filled with an oxide- or nitride-based dielectric material (90) (e.g. SiO.sub.2) and planarized with respect to the protective layer in the first region (21).

(8) FIG. 8 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. A Si(100) wafer (10) can be provided, and a first region (21) and second region (22) each comprising a section of the Si substrate (10) can be selected for forming therein respectively a group IV device (50) and a III-V device (80). A shallow trench isolation (30) for the III-V device (80) can then be formed with a SiO.sub.2 dielectric.

(9) FIG. 9 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. A SiO.sub.2 dielectric layer (40) can be deposited over the shallow trench isolation (30) and a Si(100) layer (13) can be layer transferred thereon, thereby providing a Si(100) surface (15) at a height offset (h.sub.o) with respect to the bottom (32) of the trench isolation (30) (e.g. an offset of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm).

(10) FIG. 10 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. The Si(100) layer can be patterned to delineate a group IV active region (51) and a SiO.sub.2 dielectric shallow trench isolation (30)—or a mesa isolation without the need for SiO.sub.2 refilling (when the transferred layer is less than 100 nm thick)—for the group IV device (50) can be formed with a. One or more dummy gates (52) can then be defined over the active region (51) and these can be flanked with a gate spacer.

(11) FIG. 11 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. The partially formed group IV device (50) can be covered with a SiO.sub.2 zeroth-level interlayer dielectric (ILD0; 55). Next, a trench (60) can be formed in the second region (22) down to and partially into the Si(100) substrate (10). The trench (60) typically has a depth (d.sub.t) such that a bottom (61) thereof can be at the same height as, or somewhat higher (e.g. between 5 and 50 nm) than, the bottom (32) of the shallow trench isolation (30) for the III-V device (80).

(12) FIG. 12 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. A III-V buffer layer (71) (e.g. GaAs, InP or InGaAs) can be grown in the trench (60) using ART. The buffer forms a nano-ridge (75) comprising a narrower lower nano-ridge portion (76) in the lower trench portion (63), a wider upper nano-ridge portion (78) in the upper trench portion (64), and a middle nano-ridge portion (77) tapering from the upper to the lower nano-ridge portion (76). Because of the tapering of the middle nano-ridge portion (77), air gaps (78) can be formed at the bottom of the upper trench portion (64).

(13) FIG. 13 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. the ILD0 (55) over the dummy gates (52) in the first region (21) may be opened up and a replacement metal gate process if performed thereon, after which the group IV device (50) may be covered with a protective layer, such as a oxide- or nitride-based dielectric material (e.g. SiO.sub.2, SiN, SiCN, SiON or Al.sub.2O.sub.3). Next, the III-V device (80) may be formed, e.g. by etching an HBT emitter and exposing the HBT base layer and cutting the base layer to expose an HBT subcollector (out of plane from the drawing). Remaining gaps in the second region (22) may then filled with an oxide- or nitride-based dielectric material (90) (e.g. SiO.sub.2) and planarized with respect to the protective layer in the first region (21).

(14) FIG. 14 schematically depicts a semiconductor structure at different stages of a method in accordance with a second exemplary embodiment of the present disclosure. Contact regions (53,83) for the III-V device (80) and the group IV device (50) can be formed. As previously indicated, the III-V device (80) and the group IV device (50) can be formed such that contact regions (53,83) for each can be formed at a similar height. Subsequently, the zeroth-level (M0) and first-level (M1) metallization (91) to the contact regions (53,83) of the III-V (80) and group IV (50) devices may be performed.

(15) FIG. 15 schematically depicts a semiconductor structure at different stages of a method in accordance with a third exemplary embodiment of the present disclosure. A Si(100) wafer (10) can be provided, and a first region (21) and second region (22)—each comprising a section of the Si substrate (10)—may be selected for forming therein respectively a group IV device (50) and a III-V device (80). A deep trench isolation (30) for the III-V device (80) and the group IV device (50) may then formed with a SiO.sub.2 dielectric, thereby providing a Si(100) surface (15) in the first region (21) at a height offset (h.sub.o) with respect to the bottom (32) of the trench isolation (30) in the second region (22) (e.g. an offset of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm).

(16) FIG. 16 schematically depicts a semiconductor structure at different stages of a method in accordance with a third exemplary embodiment of the present disclosure. One or more dummy gates (52) can be defined over the active region (51) and these can be flanked with a gate spacer.

(17) FIG. 17 schematically depicts a semiconductor structure at different stages of a method in accordance with a third exemplary embodiment of the present disclosure. The partially formed group IV device (50) can be covered with a SiO.sub.2 zeroth-level interlayer dielectric (ILD0; 55). Next, a trench (60) may be formed in the second region (22) down to and partially into the Si(100) substrate (10). The trench (60) can typically have a depth (d.sub.t) such that a bottom (61) thereof may be at the same height as, or somewhat higher (e.g. between 5 and 50 nm) than the bottom (32) of the shallow trench isolation (30) for the III-V device (80).

(18) FIG. 18 schematically depicts a semiconductor structure at different stages of a method in accordance with a third exemplary embodiment of the present disclosure. A III-V buffer layer (71) (e.g. GaAs, InP or InGaAs) can be grown in the trench (60) using ART. The buffer may form a nano-ridge (75) comprising a narrower lower nano-ridge portion (76) in the lower trench portion (63), a wider upper nano-ridge portion (78) in the upper trench portion (64), and a middle nano-ridge portion (77) tapering from the upper to the lower nano-ridge portion (76). Because of the tapering of the middle nano-ridge portion (77), air gaps (78) can be formed at the bottom of the upper trench portion (64).

(19) FIG. 19 schematically depicts a semiconductor structure at different stages of a method in accordance with a third exemplary embodiment of the present disclosure. The ILD0 (55) over the dummy gates (52) in the first region (21) can be opened up and a replacement metal gate process if performed thereon, after which the group IV device (50) may be covered with a protective layer, such as a oxide- or nitride-based dielectric material (e.g. SiO.sub.2, SiN, SiCN, SiON or Al.sub.2O.sub.3). Next, the III-V device (80) can be formed, e.g. by etching an HBT emitter and exposing the HBT base layer and cutting the base layer to expose an HBT subcollector (out of plane from the drawing). Remaining gaps in the second region (22) can then be filled with an oxide- or nitride-based dielectric material (90) (e.g. SiO.sub.2) and planarized with respect to the protective layer in the first region (21).

(20) FIG. 20 depicts electron microscope images of semiconductor structures in accordance with the present disclosure. Part a shows the quality of the interface (marked by an arrow) of the III-V material with the dielectric material can be typically relatively bad. Part b shows that the situation can be further aggravated when partially recessing the dielectric material (e.g. a partial wet SiO.sub.2 removal); and Part c shows that by partially removing the dielectric next to the III-V material, depositing a passivation layer (marked by an arrow) around the III-V material and subsequently filling up the remaining gap, the interface with the III-V material (i.e. between the III-V material and the passivation layer) can be considerably improved.

(21) In the different figures, the same reference signs refer to the same or analogous elements.

DETAILED DESCRIPTION OF THE DISCLOSURE

(22) The present disclosure will be described with respect to particular embodiments and with reference to certain drawings, but the disclosure is not limited thereto but only by the claims. The drawings described are only schematic and are non-limiting. In the drawings, the size of some of the elements may be exaggerated and not drawn on scale for illustrative purposes. The dimensions and the relative dimensions do not correspond to actual reductions to practice of the disclosure.

(23) Furthermore, the terms first, second, third and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequence, either temporally, spatially, in ranking or in any other manner. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the disclosure described herein are capable of operation in other sequences than described or illustrated herein.

(24) Moreover, the terms top, bottom, over, under, upper, lower and the like in the description and the claims are used for descriptive purposes and not necessarily for describing relative positions. It is to be understood that the terms so used are interchangeable with their antonyms under appropriate circumstances and that the embodiments of the disclosure described herein are capable of operation in other orientations than described or illustrated herein.

(25) The term “comprising”, used in the claims, should not be interpreted as being restricted to the means listed thereafter; it does not exclude other elements or steps. It is thus to be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. The term “comprising” therefore covers the situation where only the stated features are present and the situation where these features and one or more other features are present. Thus, the scope of the expression “a device comprising means A and B” should not be interpreted as being limited to devices consisting only of components A and B. It means that with respect to the present disclosure, the only relevant components of the device are A and B.

(26) Similarly, the term “coupled” should not be interpreted as being restricted to direct connections only. The terms “coupled” and “connected”, along with their derivatives, may be used. It should be understood that these terms are not intended as synonyms for each other. Thus, the scope of the expression “a device A coupled to a device B” should not be limited to devices or systems wherein an output of device A is directly connected to an input of device B. It means that there exists a path between an output of A and an input of B which may be a path including other devices or means. “Coupled” may mean that two or more elements are either in direct physical or electrical contact, or that two or more elements are not in direct contact with each other but yet still co-operate or interact with each other.

(27) Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments.

(28) Similarly, it should be appreciated that in the description of exemplary embodiments of the disclosure, various features of the disclosure are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various disclosed aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed disclosure requires more features than are expressly recited in each claim. Rather, as the following claims reflect, disclosed aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this disclosure.

(29) Furthermore, while some embodiments described herein include some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the disclosure, and form different embodiments, as would be understood by those in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination.

(30) In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the disclosure may be practised without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

(31) Reference will be made to transistors. These are devices having a first main electrode such as a drain or emitter, a second main electrode such as a source or collector and a control electrode such as a gate or base for controlling the flow of electrical charges between the first and second main electrodes.

(32) The following terms are provided solely to aid in the understanding of the disclosure.

(33) As used herein, and unless otherwise specified, a III-V device can be a device comprising a III-V material (e.g. GaAs, InGaAs or InP) as an active material. Likewise, a group IV device can be a device comprising a group IV material (e.g. Si, Ge or SiGe) as an active material. In particular embodiments, the group IV device may comprise a group IV and a III-V active material. For example, a group IV complementary metal-oxide-semiconductor device (CMOS) could comprise a Ge p-channel metal-oxide-semiconductor field-effect transistor (p-channel MOSFET or PMOS) and a III-V n-channel MOSFET (NMOS).

(34) In a first aspect, the present disclosure relates to a method comprising: (a) providing a Si.sub.xGe.sub.1-x(100) substrate, where x is from 0 to 1; (b) selecting a first region for forming therein a group IV device and a second region for forming therein a III-V device, the first and the second region each comprising a section of the Si.sub.xGe.sub.1-x(100) substrate; (c) forming a trench isolation for at least the III-V device; (d) providing a Si.sub.yGe.sub.1-y(100) surface in the first region, where y is from 0 to 1; (e) at least partially forming the group IV device on the Si.sub.yGe.sub.1-y(100) surface in the first region; (f) forming a trench in the second region which exposes the Si.sub.xGe.sub.1-x(100) substrate, the trench having a depth of at least 200 nm, at least 500 nm, at least 1 μm, and at least 2 μm, such as 4 μm, with respect to the Si.sub.yGe.sub.1-y(100) surface in the first region; (g) growing a III-V material in the trench using aspect ratio trapping; and (h) forming the III-V device on the III-V material, the III-V device comprising at least one contact region at a height within 100 nm, 50 nm, 20 nm, or 10 nm, of a contact region of the group IV device.

(35) In embodiments, the Si.sub.xGe.sub.1-y(100) substrate may be a Si(100) substrate (i.e. x may be 1), Ge substrate (i.e. x may be 0) or a mixture thereof (i.e. x may be between 0 and 1); usually a Si(100) substrate. In embodiments, the Si.sub.yGe.sub.1-y(100) surface may be Si(100) surface (i.e. y may be 1), Ge surface (i.e. y may be 0) or a mixture thereof (i.e. y may be between 0 and 1); usually a Si(100) surface. Si(100) substrates and surfaces—compared to SiGe or Ge substrates and surfaces—can be typically more easily and/or cheaply provided (such acquired, e.g. as Si(100) wafers, or produced, e.g. grown on top of a substrate) and otherwise worked with (e.g. processed); hence they can often be the usual substrates and/or surfaces within the present disclosure.

(36) In embodiments, the III-V device may be a vertical or a planar device; usually vertical. Indeed, the present disclosure can be beneficial to both vertical and planar III-V devices, but the problem of the height difference (cf. infra) can be typically larger for vertical than for planar devices. In embodiments, the III-V device may be a heterojunction bipolar transistor (HBT) or a high-electron-mobility transistor (HEMT), usually HBT. In embodiments, the group IV device may be a complementary metal-oxide-semiconductor device (CMOS). In embodiments, the CMOS device may be a fin field-effect transistor (FinFET).

(37) Selecting a first and second region in step b typically involves choosing locations where the group IV and the III-V device are to be made. These regions are not necessarily delimited by physically distinguishable boundaries, but they each include a—typically different—section of the Si.sub.xGe.sub.1-x(100) substrate and a region above said substrate encompassing one of the aforementioned locations; thereby each forming a 3D space in which the respective devices will be made. The first and second region may typically be selected in such a way that they do not overlap and are located next to (e.g. adjacent to) one another—at the same height with respect to the substrate.

(38) In embodiments, the trench isolation may be a shallow trench isolation (e.g. with a depth up to 1 μm) and/or a deep trench isolation (e.g. with a depth of from 1 μm to 4 μm). The depth may for example be measured as a height difference between a bottom (e.g. lowest surface) and a top (e.g. highest surface) of the trench isolation. In order to achieve good aspect ratio trapping in step g, it can be typically beneficial to have a depth which is at least two times the size (e.g. width or length) of the active area of the III-V material in the final III-V device. For a typical active area size of from 20 to 200 nm, this translates to a minimum depth of from 40 to 400 nm. The larger the ratio between the trench isolation depth and active area size, the more defects can typically be trapped in step g. Optionally, step c may further comprise forming a trench isolation for the group IV device. Such a trench isolation could however also be provided at a later stage (e.g. as part of step d or e).

(39) In order to achieve a good co-integration of the III-V device and the group IV device, it can be beneficial to match (e.g. within 100 nm, 50 nm, 20 nm, or 10 nm from one another) the top level—such as the contact region(s)—of the III-V and group IV devices. However, III-V devices—be they planar or vertical—typically need considerably more vertical space (e.g. 500 nm to 2 μm) to be formed than group IV devices. For example, they may typically require a buffer layer (e.g. for reducing therein the number of growth defects), below the actual III-V active layer(s). This buffer layer may typically already require a thickness of several tens or hundreds of nanometres (e.g. before the number of growth defects is reduced to a desired level). Such a buffer layer may typically be unnecessary—or at least a same thickness thereof is not required—for group IV devices, since the group IV active layer can be part of the Si.sub.xGe.sub.1-x(100) substrate as such or can be grown directly thereon with minimal growth defects. In order to overcome this difference in vertical space then, the present disclosure realizes a height offset between the III-V and group IV device; i.e. a height difference between the starting level for III-V material growth on the one hand and the Si.sub.yGe.sub.1-y(100) surface on the other hand.

(40) In embodiments, the trench isolation formed in step c may have a bottom and step d may comprise providing the Si.sub.yGe.sub.1-y(100) surface in the first region at a height of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm, above the trench isolation bottom.

(41) In embodiments, step d may comprise depositing a dielectric layer (e.g. SiO.sub.2) over the Si.sub.xGe.sub.1-x(100) substrate. In embodiments, the dielectric layer may have a thickness of at least 100 nm, at least 200 nm, at least 500 nm, or at least 1 μm, such as 2 μm or 4 μm. The dielectric layer may be typically used in combination with embodiments of the first or second type, cf. infra. Where used, the dielectric layer advantageously allows to overcome the vertical distance between the top of the trench isolation and the level where the Si.sub.yGe.sub.1-y(100) surface is to be realized. Furthermore, it plays a role in achieving a degree of planarization between the first and second regions.

(42) Providing the Si.sub.yGe.sub.1-y(100) surface with the desired height offset can be achieved in multiple ways. In a first type of embodiments, step d may comprise exposing the Si.sub.xGe.sub.1-x(100) substrate in the first region and growing (e.g. regrowing) Si.sub.yGe.sub.1-y thereon to form the Si.sub.yGe.sub.1-y(100) surface. In embodiments, the Si.sub.yGe.sub.1-y may be lattice matched to Si.sub.xGe.sub.1-x(100) substrate. In embodiments, the part of the Si.sub.xGe.sub.1-x(100) substrate which is exposed may be a Si.sub.xGe.sub.1-x(100) base for growing Si.sub.yGe.sub.1-y thereon. In embodiments, the Si.sub.xGe.sub.1-x(100) base may have been defined while forming the trench isolation in step c.

(43) In a second type of embodiments, step d may comprise transferring a Si.sub.yGe.sub.1-y layer comprising the Si.sub.yGe.sub.1-y(100) surface onto the Si.sub.xGe.sub.1-x(100) substrate. In embodiments, the Si layer may be transferred onto the dielectric layer.

(44) Embodiments of the first or second type thus advantageously allow to realize the Si.sub.yGe.sub.1-y(100) surface at a certain height above the top of the trench isolation (i.e. the top of the Si.sub.xGe.sub.1-x(100) substrate).

(45) In a third type embodiments, the trench isolation provided in step c may be a deep trench isolation (cf. supra) and the Si.sub.xGe.sub.1-x(100) substrate as provided in step a may comprise the Si.sub.yGe.sub.1-y(100) surface of step d. Rather than providing the Si.sub.yGe.sub.1-y(100) surface at a certain height above the top of the trench isolation, embodiments of the third type realize the desired height offset by lowering the level from which the III-V material will be grown.

(46) Since a deep trench isolation can be typically more difficult to form than a shallow trench isolation, the trench isolation used in embodiments of the first or second type may be typically shallow. Nevertheless, it will be clear that—in embodiments—the different approaches can be combined, e.g. by achieving a first part of the desired height offset through a deeper trench isolation and achieving the rest of the height offset as described for the first or second type of embodiments.

(47) At least partially forming the group IV device in step e may comprise performing at least those steps of the group IV device formation where a desired temperature to be used exceeds the thermal budget of the group III-V material. In embodiments, step e may comprise forming functional gates or forming dummy gates for the group IV device. In a gate-first process flow, the functional gates may typically be formed already during step e. Conversely, in a gate-last process flow, dummy gates may be formed during step e and may be replaced in a replacement metal gate (RMG) module after step g (e.g. before, during or after step h). In embodiments, step e may further comprise forming a protective layer (e.g. a zeroth-level dielectric, ILD0) over the at least partially formed group IV device.

(48) In embodiments, forming the trench in the second region in step f may comprise etching opening through the dielectric layer—if present—and through the trench isolation, down to the Si.sub.xGe.sub.1-x(100) substrate. In embodiments, forming the trench may further comprise etching partially into the Si.sub.xGe.sub.1-x(100) substrate. Note that the trench in the second region may be distinct from the trench isolation formed in step c.

(49) In embodiments, the trench formed in step f may comprise: a lower trench portion having a first trench width, and an upper trench portion having a second trench width. In embodiments, the first trench width may be from 0.05 to 0.75 times, usually from 0.1 to 0.5 times, the second trench width. The trench may thus be shaped such as to allow formation of a nano-ridge therein (cf. infra). In embodiments, the lower trench portion may be suitable for aspect ratio trapping of the III-V material, while the upper trench portion may be suitable for growth confinement of the III-V material. As described previously in the context of the trench isolation, a minimum depth of from 40 to 400 nm may be typically needed to achieve good aspect ratio trapping. As such, the bottom trench portion may have a trench depth of at least 40 nm, at least 100 nm, at least 200 nm, or at least 400 nm. In embodiments, a bottom of the trench (e.g. of the lower portion) and the bottom of the trench isolation may be at a comparable height (e.g. within 5 to 50 nm from one another, such as within 5 to 20 nm or 5 to 10 nm). Usually, the bottom of the trench may substantially align with or be above the bottom of the trench isolation. When the trench has a V-shaped abutment (cf. infra), at least the ends of the V may align with or be above the bottom of the trench isolation; the point of the V may optionally be above, aligned with or below the bottom of the trench isolation.

(50) The trench typically has a depth such that a bottom thereof may be at the same height as, or somewhat higher (e.g. between 5 and 50 nm) than, the bottom of the shallow trench isolation for the III-V device.

(51) In embodiments, the trench formed in step f may have a V-shaped abutment defined by (111)-oriented surfaces of the Si.sub.xGe.sub.1-x(100) substrate. In embodiments, the lower trench portion may be in contact with the V-shaped abutment. In embodiments, the V-shaped abutment may have an internal angle of about 70.6°. Usually, the V-shape may be oriented with point of the V facing ‘downwards’ (i.e. towards the Si.sub.xGe.sub.1-x(100) substrate) and the ends of the V facing ‘upwards’ (i.e. away from the Si.sub.xGe.sub.1-x(100) substrate); i.e. oriented as a ‘V’ rather than a ‘’. The (111)-oriented surfaces advantageously facilitate III-V material growth thereon with less defects, thereby allowing to achieve—after ART—a sufficiently defect-free III-V material already within a smaller growth thickness (as compared to growing the III-V material—with ART—on a flat bottom). The above notwithstanding, the V-shaped bottom profile may not be strictly necessary and e.g. a flat bottom can be used as well, provided that the depth and/or aspect ratio of the trench—and particularly the lower portion thereof—are chosen accordingly (e.g. sufficiently high to achieve the desired low level of defects).

(52) In step g, aspect ratio trapping—as is known to the skilled person—can be used to reduce a number of growth defects (e.g. threading dislocations) in the III-V material. It will be clear that an acceptable number of remaining defects typically depends on the integrity and/or reliability that is sought for the final device. This number also depends on the type of III-V device that is being made, where e.g. an HBT device is typically more sensitive to defects than a HEMT device. Broadly speaking however, the number of threading dislocations in a top portion (e.g. in an active layer) of the III-V material may be lower than 1×10.sup.9 defects/cm.sup.2, usually lower than 1×10.sup.6 defects/cm.sup.2.

(53) In embodiments, growing the III-V material in step g may comprise growing a layer stack comprising at least one III-V material (e.g. at least one layer may consist of a III-V material). In embodiments, the layer stack may consist of III-V materials. In embodiments, the layer stack may comprise a buffer layer; such as a buffer layer made of a III-V material (e.g. GaAs, InGaAs or InP). In embodiments, the layer stack may comprise one or more further III-V layers; such as at least one III-V active layer (e.g. an emitter layer). In embodiments, the III-V layer may comprise (e.g. consist of) a material selected from the list of GaAs, InAs, GaSb, InP, InSb, GaP, InGaAs, InAlAs, InGaP, InGaSb, AlGaAs, AlGaP, InAlP, AlGaSb and InAlP. For example, the III-V layers may comprise InGaP emitter layer on a GaAs collector or a InP emitter layer on an InGaAs buffer. The buffer layer advantageously functions to trap growth defects therein and to provide a suitable—relatively—defect-free surface on which to grow the one or more further III-V layers. The above notwithstanding, the III-V material could also consist of a single III-V material, a bottom portion of which could take up to of the buffer layer and a top portion of which could take up the role of the active layer.

(54) In some embodiments, the III-V material grown in the trench step g may be a III-V nano-ridge. In embodiments, the III-V nano-ridge may comprise a lower nano-ridge portion having a first nano-ridge width, an upper nano-ridge portion having a second nano-ridge width, and a middle nano-ridge portion between the upper and lower nano-ridge portions and tapering from the upper to the lower nano-ridge portion. In embodiments, the first nano-ridge width may be from 0.05 to 0.75 times, usually from 0.1 to 0.5 times, the second nano-ridge width. In embodiments, air gaps may be present next to the tapering middle nano-ridge portion (e.g. in the upper trench portion). Methods for forming a III-V nano-ridge form the subject-matter of co-pending U.S. patent application Ser. No. 16/996,146, filed concurrently herewith, and European patent application no. 19195256.3, which are incorporated herein by reference; the III-V nano-ridge may thus for example be formed as described therein. A practical example can be for instance described on p. 13 line 4 p. 14 line 4 in European patent application no. 19195256.3. Note that the ‘trench’ mentioned in EP application 19195256.3 corresponds to the present ‘lower trench portion’ and ‘out of the trench’ corresponds to the present ‘upper trench portion’. In embodiments, forming the III-V nano-ridge may comprise initiating growth of the III-V nano-ridge in the lower trench portion, thereby forming a filling layer of the nano-ridge inside the lower trench portion, and continuing growth in the upper trench portion on top of the filling layer, thereby forming the middle nano-ridge portion and upper nano-ridge portion, wherein at least one surfactant may be added in the chamber when the nano-ridge is growing in the upper trench portion. Suitable surfactants are described, for instance, in EP application 19195256.3.

(55) In embodiments, after step g the III-V material may be embedded in a dielectric material (e.g. SiO.sub.2). In embodiments, the method may comprise a further step g′, after step g, of: (g′1) at least partially removing the dielectric material adjacent to (e.g. in contact with) the III-V material; (g′2) coating at least a lateral side (and a optionally a top side) of the III-V material with a passivation layer; and (g′3) re-embedding the III-V material in a dielectric material (e.g. SiO.sub.2). In embodiments, step g′3 may comprise refilling an opening formed by step g′1 with the same dielectric material as removed in step g′1 or with a further dielectric material. Step g′ could—for example—be performed before or concurrently with step h, but typically before making metal contacts in step i. In embodiments, the passivation layer may be made of a dielectric (e.g. SiN or Al.sub.2O.sub.3) or a high-bandgap undoped III-V material (e.g. InGaP or InP). The passivation layer may for instance have a bandgap larger than or equal to an emitter layer—if present—of the III-V device. In embodiments, the passivation layer may have a thickness of from 2 to 20 nm. With reference to FIG. 20, part a, the quality of the interface (marked by an arrow) of the III-V material with the dielectric material can be typically relatively bad. Moreover, FIG. 20, part b, shows that the situation can be further aggravated when partially recessing the dielectric material (e.g. a partial wet SiO.sub.2 removal); where due to the weak III-V/dielectric interface, the dielectric etches from that interface as well and a gap (marked by an arrow) may be left next to the III-V material. However, by partially removing the dielectric next to the III-V material, depositing a passivation layer (marked by an arrow) around the III-V material and subsequently filling up the remaining gap—the result of which is shown in FIG. 20, part c—the interface with the III-V material (i.e. between the III-V material and the passivation layer) can be considerably improved. The improved interface may for example achieve better recombination, thereby reducing the number of surface defects.

(56) In embodiments, forming the III-V device on the III-V material in step h may comprise etching an emitter for an HBT device in the III-V material. In embodiments, step h may comprise forming at least one contact region to the III-V device for contacting the III-V device to a metal contact (cf. infra).

(57) In embodiments wherein the group IV device was not fully formed in step e, the method may further comprise fully forming the group IV device before step i (e.g. before, during or after step h). This may—for example—comprise performing a replacement metal gate process.

(58) In embodiments, the method may further comprise a step i, after step h, of: (i) forming metal contact to the contact region of the III-V device and the contact region of the group IV device.

(59) In embodiments, any feature of any embodiment of the first aspect may independently be as correspondingly described for any embodiment of any of the other aspects.

(60) In a second aspect, the present disclosure relates to a semiconductor structure comprising: (i) a Si.sub.xGe.sub.1-x(100) substrate, where x is from 0 to 1; (ii) a trench isolation on the Si.sub.xGe.sub.1-x(100) substrate; (iii) a first region comprising a first section of the Si.sub.xGe.sub.1-x(100) substrate, and an at least partially formed group IV device on a Si.sub.yGe.sub.1-y(100) surface, where y is from 0 to 1; and (iv) a second region comprising a second section of the Si.sub.xGe.sub.1-x(100) substrate, a trench which exposes the Si.sub.xGe.sub.1-x(100) substrate and filled with a III-V material, the trench having a depth of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, with respect to the Si.sub.yGe.sub.1-y(100) surface in the first region, and a III-V device on the III-V material, the III-V device comprising at least one contact region at a height within 100 nm, 50 nm, 20 nm, or 10 nm, of a contact region of the group IV device. Such a semiconductor structure may be obtainable through a method in accordance with the first aspect.

(61) In embodiments, the III-V material may be a III-V nano-ridge. In embodiments, the III-V nano-ridge may comprise a lower nano-ridge portion having a first nano-ridge width, an upper nano-ridge portion having a second nano-ridge width, and a middle nano-ridge portion between the upper and lower nano-ridge portions and tapering from the upper to the lower nano-ridge portion. In embodiments, the first nano-ridge width may be from 0.05 to 0.75 times, usually from 0.1 to 0.5 times, the second nano-ridge width.

(62) In embodiments, the trench having a V-shaped abutment defined by (111)-oriented surfaces of the Si.sub.xGe.sub.1-x(100) substrate.

(63) In embodiments, any feature of any embodiment of the second aspect may independently be as correspondingly described for any embodiment of any of the other aspects.

(64) The disclosure will now be described by a detailed description of several embodiments of the disclosure. Other embodiments of the disclosure can be configured according to the knowledge of the person skilled in the art without departing from the true technical teaching of the disclosure, the disclosure being limited only by the terms of the appended claims.

EXAMPLE 1

Approach Based on Group IV Regrowth

(65) Referring to FIG. 1, a Si(100) wafer (10) can be provided, and a first region (21) and second region (22)—each comprising a section of the Si substrate (10)—may be selected for forming therein respectively a group IV device (50) and a III-V device (80). A shallow trench isolation (30) for the III-V device (80) may then formed with a SiO.sub.2 dielectric. The shallow trench isolation (30) can also define a Si(100) base (11) in the first region (21) for (re)growing Si thereon.

(66) Referring to FIG. 2, a SiO.sub.2 dielectric layer (40) can be deposited over the shallow trench isolation (30) and a window may be opened therein over the Si(100) base (11). Si (12) may then selectively regrown on the Si(100) base (11), thereby providing a Si(100) surface (15) at a height offset (h.sub.o) with respect to the bottom (32) of the trench isolation (30) (e.g. an offset of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm). This height offset (h.sub.o) can allow to have the contact regions (53,83) of the final III-V (60) and group IV (50) devices at a similar height (e.g. within 100 nm, 50 nm, 20 nm, or 10 nm from one another); despite the relatively big difference in vertical space typically needed for the III-V device (80), compared to the group IV device (50).

(67) Referring to FIG. 3, a shallow trench isolation (35) for the group IV device (50) can be formed in the regrown Si (12) with a SiO.sub.2 dielectric, to define a group IV active region (51).

(68) Referring to FIG. 4, one or more dummy gates (52) may be defined over the active region (51) and these may be flanked with a gate spacer.

(69) Referring to FIG. 5, the partially formed group IV device (50) may be covered with a SiO.sub.2 zeroth-level interlayer dielectric (ILD0; 55). Next, a trench (60) may be formed in the second region (22) down to and partially into the Si(100) substrate (10). The trench (60) can typically have a depth (d.sub.t) such that a bottom (61) thereof may be at the same height as, or somewhat higher than (e.g. between 5 and 50 nm), the bottom (32) of the shallow trench isolation (30) for the III-V device (80). The trench (60) may have a shape suitable for aspect ratio trapping (ART). For example, the trench (60) may have a narrower lower trench portion (63) for ART and a wider upper trench portion (64) for III-V growth confinement. Usually, the lower trench portion (63) may be in contact with a V-shaped abutment (62) defined by (111)-oriented surfaces of the Si(100) substrate (10), which promotes low-defect growth of III-V materials thereon.

(70) Referring to FIG. 6, a III-V buffer layer (71) (e.g. GaAs, InP or InGaAs) may be grown in the trench (60) using ART. Subsequently, III-V layers (72) (cf. supra) for the III-V device (80) (e.g. a heterojunction bipolar transistor, HBT) may grown on the III-V buffer layer (71). The III-V material (71,72) may form a nano-ridge (75) comprising a narrower lower nano-ridge portion (76) in the lower trench portion (63), an wider upper nano-ridge portion (78) in the upper trench portion (64), and a middle nano-ridge portion (77) tapering from the upper to the lower nano-ridge portion (76). Because of the tapering of the middle nano-ridge portion (77), air gaps (78) may be formed at the bottom of the upper trench portion (64).

(71) Referring to FIG. 7, the ILD0 (55) over the dummy gates (52) in the first region (21) can be opened up and a replacement metal gate process if performed thereon, after which the group IV device (50) may be covered with a protective layer, such as a oxide- or nitride-based dielectric material (e.g. SiO.sub.2, SiN, SiCN, SiON or Al.sub.2O.sub.3). Next, the III-V device (80) may be formed, e.g. by etching an HBT emitter and exposing the HBT base layer and cutting the base layer to expose an HBT subcollector (out of plane from the drawing). Remaining gaps in the second region (22) can then be filled with an oxide- or nitride-based dielectric material (90) (e.g. SiO.sub.2) and planarized with respect to the protective layer in the first region (21). After which an HBT emitter cap may be optionally formed. As previously indicated, the III-V device (80) and the group IV device (50) can be formed such that contact regions (53,83) for each are formed at a similar height. Finally, the zeroth-level (M0) and first-level (M1) metallization (91) to the contact regions (53,83) of the III-V (80) and group IV (50) devices can be performed.

EXAMPLE 2

Approach Based on Group IV Layer Transfer

(72) Referring to FIG. 8, a Si(100) wafer (10) can be provided, and a first region (21) and second region (22)—each comprising a section of the Si substrate (10)—can be selected for forming therein respectively a group IV device (50) and a III-V device (80). A shallow trench isolation (30) for the III-V device (80) can then be formed with a SiO.sub.2 dielectric.

(73) Referring to FIG. 9, a SiO.sub.2 dielectric layer (40) can be deposited over the shallow trench isolation (30) and a Si(100) layer (13) can be layer transferred thereon, thereby providing a Si(100) surface (15) at a height offset (h.sub.o) with respect to the bottom (32) of the trench isolation (30) (e.g. an offset of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm). This height offset (h.sub.o) can allow to have the contact regions (53,83) of the final III-V (80) and group IV (50) devices at a similar height (e.g. within 100 nm, 50 nm, or 20 nm, usually 10 nm from one another); despite the relatively big difference in vertical space typically needed for the III-V device (80), compared to the group IV device (50).

(74) Referring to FIG. 10, the Si(100) layer can be patterned to delineate a group IV active region (51) and a SiO.sub.2 dielectric shallow trench isolation (30)—or a mesa isolation without the ned for SiO.sub.2 refilling (when the transferred layer is less than 100 nm thick)—for the group IV device (50) can be formed with a. One or more dummy gates (52) can then be defined over the active region (51) and these can be flanked with a gate spacer.

(75) Referring to FIG. 11, the partially formed group IV device (50) can be covered with a SiO.sub.2 zeroth-level interlayer dielectric (ILD0; 55). Next, a trench (60) can be formed in the second region (22) down to and partially into the Si(100) substrate (10). The trench (60) typically has a depth (d.sub.t) such that a bottom (61) thereof can be at the same height as, or somewhat higher (e.g. between 5 and 50 nm) than, the bottom (32) of the shallow trench isolation (30) for the III-V device (80). The trench (60) may have a shape suitable for aspect ratio trapping (ART). For example, the trench (60) may have a narrower lower trench portion (63) for ART and a wider upper trench portion (64) for III-V growth confinement. Usually, the lower trench portion (63) can be in contact with a V-shaped abutment (62) defined by (111)-oriented surfaces of the Si(100) substrate (10), which promotes low-defect growth of III-V materials thereon.

(76) Referring to FIG. 12, a III-V buffer layer (71) (e.g. GaAs, InP or InGaAs) can be grown in the trench (60) using ART. The buffer forms a nano-ridge (75) comprising a narrower lower nano-ridge portion (76) in the lower trench portion (63), a wider upper nano-ridge portion (78) in the upper trench portion (64), and a middle nano-ridge portion (77) tapering from the upper to the lower nano-ridge portion (76). Because of the tapering of the middle nano-ridge portion (77), air gaps (78) can be formed at the bottom of the upper trench portion (64). Subsequently, III-V layers (72) (cf. supra) for the III-V device (80) (e.g. a heterojunction bipolar transistor, HBT) can be grown on the III-V buffer layer (71).

(77) Referring to FIG. 13, the ILD0 (55) over the dummy gates (52) in the first region (21) may be opened up and a replacement metal gate process if performed thereon, after which the group IV device (50) may be covered with a protective layer, such as a oxide- or nitride-based dielectric material (e.g. SiO.sub.2, SiN, SiCN, SiON or Al.sub.2O.sub.3). Next, the III-V device (80) may be formed, e.g. by etching an HBT emitter and exposing the HBT base layer and cutting the base layer to expose an HBT subcollector (out of plane from the drawing). Remaining gaps in the second region (22) may then filled with an oxide- or nitride-based dielectric material (90) (e.g. SiO.sub.2) and planarized with respect to the protective layer in the first region (21). After which an HBT emitter cap (94) may be optionally formed.

(78) Referring to FIG. 14, contact regions (53,83) for the III-V device (80) and the group IV device (50) can be formed. As previously indicated, the III-V device (80) and the group IV device (50) can be formed such that contact regions (53,83) for each can be formed at a similar height. Subsequently, the zeroth-level (M0) and first-level (M1) metallization (91) to the contact regions (53,83) of the III-V (80) and group IV (50) devices may be performed.

EXAMPLE 3

Approach Based on Deep Trench Isolation (30) for the III-V Device (80)

(79) Referring to FIG. 15, a Si(100) wafer (10) can be provided, and a first region (21) and second region (22)—each comprising a section of the Si substrate (10)—may be selected for forming therein respectively a group IV device (50) and a III-V device (80). A deep trench isolation (30) for the III-V device (80) and the group IV device (50) may then formed with a SiO.sub.2 dielectric, thereby providing a Si(100) surface (15) in the first region (21) at a height offset (h.sub.o) with respect to the bottom (32) of the trench isolation (30) in the second region (22) (e.g. an offset of at least 200 nm, at least 500 nm, at least 1 μm, or at least 2 μm, such as 4 μm). This height offset (h.sub.o) can allow to have the contact regions (53,83) of the final III-V (80) and group IV (50) devices at a similar height (e.g. within 100 nm, 50 nm, 20 nm, usually 10 nm from one another); despite the relatively big difference in vertical space typically needed for the III-V device (80), compared to the group IV device (50).

(80) Referring to FIG. 16, one or more dummy gates (52) can be defined over the active region (51) and these can be flanked with a gate spacer.

(81) Referring to FIG. 17, the partially formed group IV device (50) can be covered with a SiO.sub.2 zeroth-level interlayer dielectric (ILD0; 55). Next, a trench (60) may be formed in the second region (22) down to and partially into the Si(100) substrate (10). The trench (60) can typically have a depth (d.sub.t) such that a bottom (61) thereof may be at the same height as, or somewhat higher (e.g. between 5 and 50 nm) than the bottom (32) of the shallow trench isolation (30) for the III-V device (80). The trench (60) can have a shape suitable for aspect ratio trapping (ART). For example, the trench (60) may have a narrower lower trench portion (63) for ART and a wider upper trench portion (64) for III-V growth confinement. Usually, the lower trench portion (63) can be in contact with a V-shaped abutment (62) defined by (111)-oriented surfaces of the Si(100) substrate (10), which can promote low-defect growth of III-V materials thereon.

(82) Referring to FIG. 18, a III-V buffer layer (71) (e.g. GaAs, InP or InGaAs) can be grown in the trench (60) using ART. The buffer may form a nano-ridge (75) comprising a narrower lower nano-ridge portion (76) in the lower trench portion (63), a wider upper nano-ridge portion (78) in the upper trench portion (64), and a middle nano-ridge portion (77) tapering from the upper to the lower nano-ridge portion (76). Because of the tapering of the middle nano-ridge portion (77), air gaps (78) can be formed at the bottom of the upper trench portion (64). Subsequently, III-V layers (72) (cf. supra) for the III-V device (80) (e.g. a heterojunction bipolar transistor, HBT) can be grown on the III-V buffer layer (71).

(83) Referring to FIG. 19, the ILD0 (55) over the dummy gates (52) in the first region (21) can be opened up and a replacement metal gate process if performed thereon, after which the group IV device (50) may be covered with a protective layer, such as a oxide- or nitride-based dielectric material (e.g. SiO.sub.2, SiN, SiCN, SiON or Al.sub.2O.sub.3). Next, the III-V device (80) can be formed, e.g. by etching an HBT emitter and exposing the HBT base layer and cutting the base layer to expose an HBT subcollector (out of plane from the drawing). Remaining gaps in the second region (22) can then be filled with an oxide- or nitride-based dielectric material (90) (e.g. SiO.sub.2) and planarized with respect to the protective layer in the first region (21). After which an HBT emitter cap can be optionally formed. As previously indicated, the III-V device (80) and the group IV device (50) can be formed such that contact regions (53,83) for each can be formed at a similar height. Finally, the zeroth-level (M0) and first-level (M1) metallization (91) to the contact regions (53,83) of the III-V (80) and group IV (50) devices can be performed.

(84) It is to be understood that although various embodiments, specific constructions and configurations, as well as materials, have been discussed herein for devices according to the present disclosure, various changes or modifications in form and detail may be made without departing from the scope and technical teachings of this disclosure. For example, any formulas given above are merely representative of procedures that may be used. Functionality may be added or deleted from the block diagrams and operations may be interchanged among functional blocks. Steps may be added or deleted to methods described within the scope of the present disclosure.