WAVEGUIDE DEVICE AND METHOD OF DOPING A WAVEGUIDE DEVICE

Abstract

A waveguide device and method of doping a waveguide device, the waveguide device comprising a rib waveguide region, the rib waveguide region having: a base, and a ridge extending from the base, wherein: the base includes a first slab region at a first side of the ridge and a second slab region at a second side of the ridge; a first doped slab region extends along the first slab region; a second doped slab region extends along the second slab region; a first doped sidewall region extends along a first sidewall of the ridge and along a portion of the first slab, the first doped sidewall region being in contact with the first doped slab region at a first slab interface; and a second doped sidewall region extends along a second sidewall of the ridge and along a portion of the second slab, the second doped sidewall region being in contact with the second doped slab region at a second slab interface; and wherein the separation between the first sidewall of the ridge and the first slab interface is no more than 10 m; and wherein the separation between the second sidewall of the ridge and the second slab interface is no more than 10 m.

Claims

1. A waveguide device comprising a rib waveguide region, the rib waveguide region having: a base, and a ridge extending from the base, wherein: the base includes a first slab region at a first side of the ridge and a second slab region at a second side of the ridge; a first doped slab region that extends along the first slab region; a second doped slab region that extends along the second slab region; a first doped sidewall region that extends along a first sidewall of the ridge and along a portion of the first slab, the first doped sidewall region being in contact with the first doped slab region at a first slab interface; and a second doped sidewall region that extends along a second sidewall of the ridge and along a portion of the second slab, the second doped sidewall region being in contact with the second doped slab region at a second slab interface; wherein the separation between the first sidewall of the ridge and the first slab interface is no more than 10 m; and wherein the separation between the second sidewall of the ridge and the second slab interface is no more than 10 m.

2. The waveguide device of claim 1, wherein the separation between the first sidewall of the ridge and the first slab interface is no more than 5 m; and wherein the separation between the second sidewall of the ridge and the second slab interface is no more than 5 m.

3. The waveguide device of claim 1 or claim 2, further comprising a first electrical contact located on the first doped slab region and a second electrical contact located on the second doped region; wherein the separation distance between the first electrical contact and the first sidewall of the ridge is no more than 10 m; and wherein the separation between the second electrical contact and the second sidewall of the ridge is no more than 10 m.

4. The waveguide device of claim 3, wherein the separation distance between the first electrical contact and the first sidewall of the ridge is no more than 5 m; and wherein the separation between the second electrical contact and the second sidewall of the ridge is no more than 5 m.

5. The waveguide device of any preceding claim, further comprising an epitaxial crystalline cladding layer located between the base of the waveguide device and a silicon substrate.

6. The waveguide device of claim 5, further comprising a buried oxide layer, disposed on opposing horizontal sides of the epitaxial crystalline cladding layer, and wherein the epitaxial crystalline cladding layer is formed of a material which is different from the buried oxide layer.

7. A method of fabricating a waveguide device comprising: providing a rib waveguide, the rib waveguide comprising: a base, and a ridge extending from the base; wherein: the base includes a first slab region at a first side of the ridge and a second slab region at a second side of the ridge; and creating a first doped slab region which extends along the first slab region; the step of creating the first doped slab region comprising: providing a photoresist over at least a portion of the second slab region, the photoresist extending further from the base than the ridge extends from the base; implanting the first slab region with a dopant at an angle to the first sidewall of the waveguide, using the photoresist as a mask to cast a shadow over regions not to be doped including the second sidewall of the ridge.

8. The method of fabricating a waveguide device according to claim 7; further comprising the step of: implanting a first sidewall of the ridge and a portion of the first slab region with the dopant, at an angle to the first sidewall of the ridge, to create a first doped sidewall region which extends along a first sidewall of the ridge and along a portion of the first slab, the first doped sidewall region being in contact with the first doped slab region at a first slab interface.

9. The method of fabricating a waveguide device according to claim 8, wherein the separation between the first sidewall of the ridge and the first slab interface is no more than 10 m.

10. The method of fabricating a waveguide device according to claim 9, wherein the separation between the first sidewall of the ridge and the first slab interface is no more than 5 m.

11. The method of fabricating a waveguide device according to any one of claims 7 to 8, wherein the dopant for doping the first slab region and the first sidewall region is an N-type dopant.

12. The method of fabricating a waveguide device according to claim 11, wherein the dopant of the first slab is the same material as the dopant of the first sidewall region.

13. The method of fabricating a waveguide device according to any one of claims 7 to 12, further comprising the steps of: removing the photoresist from over the second slab region; creating a second doped slab region which extends along the second slab region; the step of creating the second doped slab region comprising: providing a photoresist over at least a portion of the first slab region, the photoresist extending further from the base than the ridge extends from the base; implanting the second slab region with a second dopant at an angle to the second sidewall of the waveguide, using the photoresist as a mask to cast a shadow over regions not to be doped, including the first sidewall of the ridge.

14. The method of fabricating a waveguide device according to claim 13; further comprising the step of: implanting a second sidewall of the ridge and a portion of the second slab region with the dopant, at an angle to the second sidewall of the ridge, to create a second doped sidewall region which extends along a second sidewall of the ridge and along a portion of the second slab, the second doped sidewall region being in contact with the second doped slab region at a second slab interface.

15. The method of fabricating a waveguide device according to any one of claims 7 to 14, wherein the separation between the second sidewall of the ridge and the second slab interface is no more than 10 m.

16. The method of fabricating a waveguide device according to claim 15, wherein the separation between the second sidewall of the ridge and the second slab interface is no more than 5 m.

17. The method of fabricating a waveguide device according to any one of claims 14 to 16, wherein the dopant for doping the second slab region and the second sidewall region is a P-type dopant.

18. The method of fabricating a waveguide device according to claim 17, wherein the dopant of the second slab is the same material as the dopant of the second sidewall region.

19. The method of any one of claims 7 to 18, wherein the device is fabricated on a silicon wafer and wherein the step of implanting the first slab region with a dopant at an angle to the first sidewall includes the step of tilting the wafer relative to the direction of implantation.

20. The method of any one of claims 13 to 19, wherein the device is fabricated on a silicon wafer and wherein the step of implanting the second slab region with a dopant at an angle to the second sidewall includes the step of tilting the wafer relative to the direction of implantation.

21. A waveguide device comprising a rib waveguide region, the waveguide device being fabricated by the method of any one of claims 7 to 18.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0034] These and other features and advantages of the present invention will be appreciated and understood with reference to the specification, claims, and appended drawings wherein:

[0035] FIG. 1 depicts a schematic diagram of a waveguide device and method of fabricating the waveguide device according to the present invention;

[0036] FIG. 2 depicts a schematic diagram of a semiconductor wafer bearing the regions to be doped being rotated to expose the surface at the correct angel for doping.

[0037] FIGS. 3a, 3b and 3c illustrate examples of the photoresist applied during the fabrication process to enable shadow doping of the waveguide device;

[0038] FIG. 4 shows a representative circuit for a waveguide device according to the present invention, the waveguide device taking the form of an electro absorption modulator;

[0039] FIG. 5 shows a representative circuit for a waveguide device of the present invention, the waveguide device taking the form of a waveguide photodiode;

[0040] FIG. 6 shows a schematic diagram of a variant waveguide device; and

[0041] FIG. 7 shows a schematic diagram of a further variant waveguide device.

DETAILED DESCRIPTION

[0042] The detailed description set forth below in connection with the appended drawings is intended as a description of exemplary embodiments of an active waveguide device and a method of fabrication of a waveguide device provided in accordance with the present invention and is not intended to represent the only forms in which the present invention may be constructed or utilized. The description sets forth the features of the present invention in connection with the illustrated embodiments. It is to be understood, however, that the same or equivalent functions and structures may be accomplished by different embodiments that are also intended to be encompassed within the spirit and scope of the invention. As denoted elsewhere herein, like element numbers are intended to indicate like elements or features.

[0043] A waveguide device and method of fabricating the device is described below in relation to FIG. 1. Initially the basic active waveguide structure is formed as an upstanding waveguide fabricated on a silicon platform, for example a silicon on insulator (SOI) platform using standard techniques, the structure including a waveguide, the waveguide comprising: a base, and a ridge 11 extending from the base. The base includes a first slab region 12 at a first side of the ridge and a second slab region 13 at a second side of the ridge 11. The ridge 11 includes a first sidewall and a second sidewall.

[0044] The basic active waveguide structure may comprise materials such as silicon germanium (in particular, for example when the waveguide device is an optical modulator) or germanium (in particular, for example, when the waveguide device is a photodiode). The techniques for fabrication of such waveguide structures are known and will not be described here in more detail. Instead, this application focusses on the novel doping structures and techniques which form part of the fabrication of the final device, whether that be a modulator, a photodiode, or another waveguide-based device. A first doped slab region 22 is created which extends along the first slab region 12; the step of creating the first doped slab region comprising: providing a photoresist (not shown in FIG. 1) over at least a portion of the second slab region 13, the photoresist extending further upwards (i.e. further in a direction parallel to that in which the sidewalls of the ridge extend) from the base than the ridge extends from the base and then implanting the first slab region 12 with a dopant N++ at an angle .sub.1 to the first sidewall of the waveguide, and thereby using the photoresist as a mask to cast a shadow over regions not to be doped including the second sidewall 21b of the ridge. Importantly, the lateral distance covered by the shadow on the first slab dictates where the slab doped region 22 will terminate (i.e. the position of a first slab interface). For fast devices, the goal is to get the first slab interface as close to the first sidewall as possible. For example, a separation of no more than 10 m is desirable and a separation of no more than 5 m is even more desirable.

[0045] In a separate doping step, a first sidewall of the ridge and a portion of the first slab region is implanted with a dopant at an angle .sub.1 to the first sidewall of the ridge, to create a first doped sidewall region 21a which extends along a first sidewall of the ridge and along a portion of the first slab, the first doped sidewall region 21a therefore contacting the first doped slab region physically and electrically at a first slab interface 42 which is laterally offset in a first direction from the first sidewall of the ridge. The fabrication process is repeated on the second side of the waveguide, starting initially with the step of removing the photoresist from over the second slab region 13 and instead providing a photoresist over at least a portion of the first slab region 22, the photoresist extending further from the base than the ridge extends from the base. A second doped slab region 23 which extends along the first slab region can then be created using the shadow doping method by implanting the second slab region with a second dopant at an angle .sub.2 to the second sidewall of the waveguide, using the photoresist (not shown) as a mask to cast a shadow over regions of the second slab and second sidewall of the ridge that are not to be doped.

[0046] In a final doping step, the second sidewall of the ridge and a portion of the second slab region is doped by implanting them with a dopant at an angle .sub.2 to the second sidewall of the ridge, to create a second doped sidewall region 21b which extends along a second sidewall of the ridge and along a portion of the second slab, the second doped sidewall region being in contact physically and electrically with the second doped slab region 23 at a second slab interface 43 which is laterally offset in a second direction from the second sidewall of the ridge.

[0047] A first electrical contact 32, typically a metal layer, is located on top of the first doped slab region 22 in electrical contact with the first doped slab region and a second electrical contact 33, typically a metal layer, is located on top of and in electrical contact with the second doped slab region. In this way, an electrical bias applied between the two electrical contacts 32, 33 will provide a corresponding bias across the waveguide. For optimal working speeds of the device, it is desirable to locate the electrical contacts as close to the waveguides as possible. Again, a separation of no more than 10 m is desirable and a separation of no more than 5 m is even more desirable.

[0048] Where the electro optical waveguide device takes the form of an electro absorption modulator (EAM), the waveguide is formed of an electro-absorption material. By applying a bias across the first and second contacts, an electrical field is generated in the electro-absorption material in which the Franz Keldysh effect occurs, the presence of an electrical field thereby giving rise to an increase in the absorption of light within the electro-absorption material.

[0049] Where the electro optical waveguide device is a photodiode, the ridge will comprise an optically active material. Upon application of a reverse bias across the first and second contacts, an electrical field will be created between the doped regions. Absorption of light within the waveguide will cause a change in the electrical current between the contacts 32, 33, the magnitude of which indicates the intensity of the light detected.

[0050] FIG. 2 shows examples in which a wafer can be placed relative to the angle at which the ion beam of the dopant is applied. In this way, the angle of the wafer itself during the dopant process facilitates the shadow doping procedure. In the first example (a), the wafer is angled so that its face is orthogonal (i.e. at 90 degrees or substantially 90 degrees) to the ion beam of dopant. In other words, no tilt is applied to the wafer. In the second example (b), a tilt is applied to the wafer so that the face of the wafer is at a non-orthogonal (or substantially non-orthogonal) angle to the ion beam. For example, a wafer tilt of 30 degrees may be applied, in which case, the smallest angle between the face of the wafer and the ion beam would be 60 degrees or substantially 60 degrees. Clearly, for the embodiment depicted in FIG. 1, during the deposition of the dopant for the first slab doped region and the second doped region, the smallest angle between the face of the wafer and the ion beam will have an angle of 1, or 2 respectively.

[0051] Suitable dopants could include phosphorus for N-type doping and boron for P-type doping.

[0052] By utilizing the methods described above, particularly the shadow masking layer or layers, it is possible to create a shadow precise enough produce a highly doped region very close to the active waveguide. If such high doping were to extend into the waveguide region, the device would not perform as required. Neither would it perform as effectively if the doping were too far from the waveguide. The ability to tilt accurately the face of the wafer at an angle to direction of a doping beam further facilitates the control of the angle and in combination with the shadow doping, therefore creates an improved method by which a highly doped region can be applied in close proximity to the waveguide.

[0053] The shadow doping mechanism is explained in more detail below with reference to FIGS. 3a, 3b and 3c. The embodiment described in relation to FIGS. 3a, 3b and 3c show one example of suitable dimensions that would result in a desirable sub-Sum measurement between the doped slab region and the adjacent sidewall of the ridge.

[0054] FIG. 3a shows an example of desirable dimensions for a particular instance of doping of the second slab 13. In this example, a photoresist mask is applied over the first slab 12, leaving a clearance distance of 0.3 m between the photoresist and the first sidewall of the ridge. The photoresist has a height which is greater than that of the ridge. In the embodiment shown, a layer of cladding is applied to the top of the ridge, in this example with a thickness of 0.5 m. The height of the photoresist is therefore chosen so that it is greater than the sum of the ridge and the cladding layer. In the embodiment shown, the ridge has a height of 2.6 m and a thickness of 0.8 m, and the photoresist mask 50 applied has a height of 5.6 m.

[0055] FIGS. 3b and 3c depict the tolerance and design considerations that must be taken when determining the optimum angle at which the ion dopant beam should be set in order to ensure that the opposite sidewall (in this case the first sidewall) lies entirely within the shadow of the photoresist mask but that the doped region of the slab (in this case the second slab) lies as close as possible to the second sidewall of the ridge, thereby reducing the series resistance R of the device as much as possible.

[0056] The minimum value for the implant angle of FIG. 3a which ensures that it does not dope the first sidewall can be calculated by the following equation:

tan =maximum separation/(height of photoresistheight of ridge)(1.1)

[0057] where the height of the ridge includes the thickness of any cladding layer if a cladding layer is present.

[0058] For the example dimensions depicted in FIG. 3b, this gives the following minimum implant angle:

[00001]$\begin{array}{cc}\tan ()=\frac{0.8.Math..Math.\mathrm{um}}{5.6.Math..Math.\mathrm{um}-\left(2.6.Math..Math.\mathrm{um}+0.5.Math..Math.\mathrm{um}\right)}& \left(1.2\right)\\ =17.7.Math.& \left(1.3\right)\end{array}$

[0059] The calculation has assumed a 0.5 m error in the separation distance between the first sidewall and the photoresist mask. For the calculation of the smallest possible angle , the maximum possible separation for this example is used (i.e. 0.3 m+0.5 m=0.8 m).

[0060] FIG. 3c illustrates the other extreme within the tolerance of the system (i.e. that the photoresist ends up deposited 5 m in the other direction). In this case, there is no separation between the photoresist and the first sidewall. Some of the photoresist is deposited on top of the ridge. Using equation 1.1 above, it is therefore possible to calculate the worst case scenario in terms of the minimum separation between the doped (second) slab and the second sidewall that can be achieved when doping at the minimum implant angle. Using the dimensions of FIG. 3c, and the angle calculated in 1.3 above, this gives:

[00002]$\begin{array}{cc}\tan \left(17.7.Math.\right)=\frac{x}{5.6.Math..Math.\mathrm{um}+\left(2.6.Math..Math.\mathrm{um}+0.5.Math..Math.\mathrm{um}\right)}& \left(1.4\right)\end{array}$

[0061] Note that it is assumed that the resist which is located on top of the cladding on the ridge will have a thickness corresponding to the height of the photoresist on the first slab (in this case 5.6 m).

x=2.78 um(1.5)

[0062] So, for the dimensions shown in FIG. 3c, an implantation angle of 17.7 degrees will lead to a worst case scenario of 2.78 m for the separation between the second sidewall and the second doped slab region.

Doping is achieved in a standard implanter with capability to tilt and rotate the substrate holder. For a given dopant the dose of dopant received by the regions of semiconductor to be doped depends upon the energy of the ion beam and the time of exposure.
The slab doped regions (typically heavily doped) and sidewall doped regions (typically lightly doped) may use different dopants. That is to say, it would be possible to have 2 different P dopants and 2 different N dopants. One example of a benefit provided by different dopants could be the ability to more easily obtain different depths of penetration that may be required for performance optimization.

[0063] FIG. 4 shows the representative circuit of a waveguide EA modulator where the EAM modulator shown is a top view and the fabrication of the device corresponds to that of the waveguide device 1 shown in FIG. 1. A bias V.sub.j is applied across the contacts 32, 33, giving rise to a resulting capacitance and Resistance R.sub.5.

[0064] It is established that the performance (especially speed) of the device improves as C.sub.j.Math.R.sub.5 gets smaller. The capacitance C is reduced significantly as the sidewall doping concentration can be engineered to increase effective intrinsic region width. Series resistance is reduced since the highly doped region can be deeply doped by multiple implantations with different energies without affecting the junction capacitance. Hence the product C.sub.j.Math.R.sub.5 can be reduced significantly.

[0065] The embodiment of FIG. 4 shows an example of a large signal lumped circuit model of an EAM.

[0066] FIG. 5 shows the representative circuit of a waveguide photodiode modulator where the fabrication of the photodiode will again correspond to that of the waveguide device 1 shown in FIG. 1. In this case, a current I.sub.PD is generated by putting optical power inside photodiode waveguide for a given reverse bias Vj across the contacts 32, 33. As with the modulator example, the device, in use, will therefore have an inherent capacitance Cj and Resistance R.sub.5. Again, the performance of the device improves as C.sub.j.Math.R.sub.5 gets smaller.

As will be appreciated, the method of fabrication described above can be used to fabricate many variants of electro-absorption modulator or photodetector. For example: proud waveguide, single silicon sidewall, and BOX-less devices i.e. those with a epitaxial crystalline layer. FIGS. 6 and 7 show examples of these devices made using the same method of fabrication described above and can generally be referred to as proud waveguide devices. In the devices shown there is no buried oxide below the ridge of the waveguide, instead there may be a regrown or epitaxial crystalline cladding layer. The slab and a portion of the ridge is formed of a first material M.sub.1, and a region beneath the first slab region and the second slab region is formed of a second material M.sub.2. The second material M.sub.2 may be buried oxide (BOX) e.g. silicon oxide. The remaining part of the ridge not formed of material M.sub.1 may be formed of a different material e.g. Si or SiGe. Of course, in other examples, buried oxide may be present below the ridge such that the second material M.sub.2 forms a substantially continuous layer. In FIG. 6, at least one sidewall 601 of the ridge is formed of doped silicon. Generally the sidewall doping concentration is smaller than the slab doping. This can be achieved by separating the implantation processes. The structures shown are similar to those disclosed in U.S. 62/429,701, the entire contents of which is incorporated herein by reference.

[0067] Although exemplary embodiments of an active waveguide device have been specifically described and illustrated herein, many modifications and variations will be apparent to those skilled in the art. Accordingly, it is to be understood that an active waveguide device constructed according to principles of this invention may be embodied other than as specifically described herein. The invention is also defined in the following claims, and equivalents thereof.