GB2306692A

GB2306692A - Optically activated semiconductor device

Info

Publication number: GB2306692A
Application number: GB9521999A
Authority: GB
Inventors: Andrew James Shields
Original assignee: Toshiba Cambridge Research Centre Ltd
Current assignee: Toshiba Europe Ltd
Priority date: 1995-10-27
Filing date: 1995-10-27
Publication date: 1997-05-07
Anticipated expiration: 2015-10-27
Also published as: GB9521999D0; GB2306692B

Abstract

An optically operable semiconductor device 1 comprising a modulation region 3 which comprises a quantum well layer 19 formed of a semiconductor material having a predetermined forbidden bandgap. The modulation region 3 comprises at least one barrier region 26 and preferably, at least another barrier region 28 formed of semiconductor material having a higher bandgap than the predetermined bandgap. When the barrier region 26 (28) is irradiated with an incident light beam 15 having a photon energy higher than the bandgap of the barrier region 26 (28), the excess carrier density in the latter is altered. The device may, for example, be constructed as an optical modulator or an optical detector.

Description

OPTICALLY ACTIVATED SEMICONDUCTOR DEVICES The present invention relates to optically activated semiconductor devices in which an incident beam (hereinafter called a "control beam") is either used to generate an electrical signal or else to control the intensity or switching of a beam transmitted through the device. A device of the former kind is, of course, an optical detector and a device of the latter kind is an optical modulator. In the context of the present invention, the term "optical" is used in a sense pertaining to electromagnetic radiation of infra-red or visible wavelengths and the term "light" is also used to refer to any such radiation.

The principle of controlling or switching a transmitted beam, using a control beam is disclosed, for example, in US-A-4 872 744. This reference discloses a device in which the control beam is said to change the electron density in a quantum well layer so that at zero - or low carrier density, the device is relatively opaque to a transmitted beam and at higher carrier densities, it is relatively transparent. However, no specific means of putting this into effect appears to be described.

US-A-4 626 075 describes a device in which a control beam having a photon energy within a quantum well absorption band creates excess carriers within the well and so alters the refractive index of a layer. This changes the angle of deflection of a transmitted beam having a photon energy lower than the bandgap of the quantum well. Therefore, this kind of device acts as a phase modulator.

Another form of device which has a waveguide with a P-I-N structure is disclosed in US-A-4 716 449. With this device, control beam is used to generate a photo-voltage which in turn is used to modulate a transmitted beam by changing the biasing condition of the waveguide.

A new form of optically activated device has now been devised, in accordance with the present invention which provides an optically operable semiconductor device comprising a modulation region which comprises a quantum well layer formed of a semiconductor material having a predetermined bandgap and at least one barrier region formed of semiconductor material having a higher bandgap than said predetermined bandgap such that when the at least one barrier region is irradiated with an incident light beam having a photon energy higher than the bandgap of said at least one barrier region, the excess carrier density in the quantum well layer is altered.

Hereinbelow are described, both embodiments of optical modulators and of an optical detector, all in accordance with the present invention. The optical modulators are capable of operating in the mode of operation described in US-A-4 872 744, i.e. so that the control beam changes the electrical density in the quantum well layer so that at zero-or low carrier density, the device is relatively opaque to the transmitted beam and at higher carrier densities, it is relatively transparent. However, these optical modulators according to the present invention are also capable of operating in a different mode. This different mode is described and claimed in a co-pending UK patent application number 9508466.1.It causes the modulator to be substantially transparent to transmitted light when the carrier density in the quantum well layer is zero or negligible and to be more opaque to the light when there is a finite carrier density in the quantum well layer.

This alternative mode of operation reduces absorption in the transparent state, relative to the equivalent absorption in the operation described in US-A-4 872 744. The alternative mode of operation achieves this for two reasons. First, there is a residual valence-to-conduction band absorption remaining when excess electrons are added to the quantum well. Second, the excess electrons in the well cause absorption due to intra- and inter-conduction band transitions. With the alternative method, the transparent state is formed with negligible excess electron density in the quantum well, thereby almost eliminating intra- and inter-conduction band absorption by excess electrons in the transparent state. Indeed, the intra- and inter-conduction band absorption is significant only for the opaque state, thereby enhancing the transparent/opaque contrast ratio.Furthermore, the residual valence band-to-conduction band absorption in the transparent state is also reduced since the device is operated at a photon energy below the bandedge neutral exciton line at zero electron density, where the quantum well will be nearly transparent.

Optical excitation of electrons in semiconductors from the valence to conduction band results in a continuum of optical transitions, with a threshold photon energy equal to the bandgap of the material. It is known, that the electron in the conduction band and the "hole" it leaves behind in the valence band feel a mutual Coulombic attraction due to their opposite charge. The interaction leads to an electron-hole bound state being formed (consisting of one electron and one hole), referred to as a (neutral) exciton. These neutral excitons produce sharp resonances in the absorption spectrum at a photon energy below the threshold energy of the valence band-to-conduction band continuum.The difference in the continuum threshold and lowest-energy, discrete excitonic transition energy is approximately equal to the binding energy of the electron and hole due to their mutual attraction. For a semiconductor quantum well the interband absorption threshold (at zero electron density) is due to the (neutral) exciton formed between the first electron and hole subbands.

Instead of modulating the intensity of the transmitted beam, the device can also be fabricated so that the control beam alters the refractive index of the device at the photon energy of the transmitted beam. In this way, the device acts as a phase modulator.

In another variant of the invention, a photovoltage can be generated by the control to change the gate bias of a device with gate and ohmic contacts and hence the electron density in the quantum well and hence the transmission of a beam travelling through the waveguide. This is analogous to the operation described in US-A4 716 449, except in the latter device the photovoltage changes the electric field across the quantum well, whilst with the present invention it changes the electron density in the quantum well.

Optical modulators according to the present invention which are intensity (amplitude) modulators fall into two distinct classes, in terms of their mechanism of operation for modulating the intensity of a light beam transmitted through the waveguide region. For one of these classes, the incident light beam (i.e. the control beam) functions to decrease the excess carrier density in the quantum well layer. In the other class of modulator, the control beam functions to increase the excess carrier density in the quantum well layer.

In the embodiments of optical modulator described hereinbelow, the optical path through the modulation region of the device is defined by a stripe waveguide structure, although in principle, any known optical waveguide structure may be employed in a detector as according to the present invention.

Alternatively, waveguide structure can be dispensed with altogether and the transmitted beam can impinge at a normal or oblique angle to the upper surface of the device.

In the first class of optical modulator according to the present invention, i.e. that in which the control beam functions to decrease excess carrier density in the quantum well layer, the modulation region further comprises a first doped barrier layer. However, preferably, the modulation region also comprises a second doped barrier layer and the quantum well layer is situated between the first and second doped barrier layers. In the embodiment described to exemplify this class of device, the quantum well layer is separated from the first and second doped barrier layers by respective spacer layers. In this device, the first and second doped barrier layers are doped so as to have the same conductivity type.

In the second class of optical modulator according to the invention, i.e.

that in which the control beam functions to increase the excess carrier density in the quantum well layer, the layer structure of the device is such as to develop an electric field across the modulation region. This can be achieved by the modulation region being situated between first and second doped layers, respectively of different conductivity types. In a described embodiment, the modulation region is separated from these first and second doped layers by respective first and second cladding layers. However, the same effect can be achieved by eg. a p-i-n, n-i-p, (Schottky metal)-i-p, (Schottky metal)-i-n layer structures or any other means well known to those skilled in the art.

In the embodiment described to exemplify this second class of optical modulator according to the invention, one of the first and second doped layers forms part of a stripe waveguide structure.

Optionally, respective ohmic contacts can be made to the first and second doped layers so that a voltage applied between them may be used to increase the electric field across the waveguide region. This can have the benefit of reducing switching time, as will be explained in more detail hereinbelow. The doped layer which is on the substrate side of the waveguide region can conveniently be contacted by an ohmic contact located in a recessed portion of the waveguide region. It is also possible to make ohmic contact via the back of the substrate if the lower doped region is grown on buffer and substrate layers of same doping type.

In optical detectors according to the present invention, the incident light beam functions to generate an electrical signal from output terminals which are in electrical contact with the quantum well layer (effectively like source and drain ohmic contacts to a quantum well layer in a high electron mobility transistor (HEMT)). In a described embodiment of such a detector, the light beam functions to decrease the excess carrier density in the quantum well layer.

This device further comprises a first doped barrier layer, although preferably, it also comprises a second doped barrier layer and the quantum well layer is situated between these'two doped barrier layers. The two doped barrier layers are doped to have the same conductivity type. As described, the quantum well layer is separated from the two doped barrier layers by respective spacer layers.

Optionally, a gate electrode overlying the structure (between the source and drain-like electrical contacts) to alter or enhance the electrical field in the upper barrier region. Additionally or alternatively, a back gate (i.e. substrate side) could be formed between the front of the lower barrier region in a similar manner. This is also optionally applicable to modulation according to the present ivention.

The aforementioned kind of optical detector may be re-configured to detect light using a second optical beam by incorporating a waveguide structure so that it provides an optical rather than electrical output signal. Furthermore, the structure may be utilised as a discrete device or integrated into an array for image detection purposes.

All described embodiments of the present invention are based on a GaAs/AIxGal xAs heterostructure. However, the layers could be formed from a number of different materials. Nevertheless, those with strong bandedge excitonic resonances are preferred, e.g.

InP/InxGa 1 xAs, InxGa1-xAs/InyAl1-yAs, InPllnxAll xAs, InxGa1 xAs/GaAs, GaInP/A1GaInP, CdTh/CdxZnlxTe, CdTe/CdxMn 1 xTe, ZnSe/Zn 1 xMnxSe, Znl xCdxSe/ZnSe, ZnSySe1 y/Znl xCdxSe, CdTe/Cd 1 xZnxTe, GaN/AlN.

GaN/AlxGal-xN. tnxGal-xN/GaN InxGal-xN/AlyGal-yN etc.

The present invention will now be explained in more detail by way of the following description of preferred embodiments and with reference to the accompanying drawings in which: Figure 1 shows a sectional perspective view of a first embodiment of an optical modulator according to the present invention; Figure 2 shows details of the waveguide region of the device shown in Figure 1; Figure 3 shows details of the complete layer structure of the device shown in Figures 1 and 2; Figure 4 shows the conduction and valence band profiles for the quantum wellfbarrier layer structure for explaining operation of the device shown in Figures 1-3 and also the device shown in Figures 12 and 13; Figure 5 shows absorption spectra for neutral and charged excitons; Figure 6 shows operation in neutral and charged exciton modes for the device shown in Figures 1-3;; Figure 7 shows a sectional perspective view of a second embodiment of an optical modulator according to the present invention; Figure 8 shows details of the waveguide region of the device shown in Figure 7; Figure 9 shows details of the complete layer structure of the device shown in Figures 7 and 8; Figure 10 shows the conduction and valence band profile for the quantum well/barrier layer structure for explaining operation of the device shown in Figures 8 and 9; Figure 11 shows operation in neutral and charged exciton modes for the device shown in Figures 7-9; Figure 12 shows a sectional perspective view of an embodiment of an optical detector according to the present invention; Figure 13 shows details of the complete layer structure of the device shown in Figure 12; and Figure 14 shows the source-drain current as a function of incident light intensity for the device shown in Figures 12 and 13.

The structure of a first embodiment of an optical modulator according to the present invention is shown in Figures 1-3. The optical modulator 1 comprises a waveguide region 3 situated between a lower cladding layer 5 and an upper cladding layer 7. This waveguide region 3 constitutes a "modulation region" in the terminology of the claims. The upper cladding layer is etched to form a ridge 9 which functions as a strip waveguide defining an optical path 11 therebelow, in the waveguide region 3. A beam 13 to be modulated is transmitted through the optical path 11. Modulation is effected by means of an applied incident control beam 15 used to irradiate the top of the device, and in particular, the top surface 17 of the ridge 9.Optionally the area of the device outside of the ridge can be covered with a suitable material opaque to the control light, so as to prevent penetration of the control beam in these regions.

Details of the waveguide region 3 are shown in more detail in Figure 2.

It comprises a quantum well layer 19 situated between an upper undoped spacer layer 21 and a lower undoped spacer layer 23. An upper doped barrier layer 25 is situated above the upper spacer layer 21. A lower doped barrier layer 27 is situated below the lower spacer layer 23. Optionally a lower undoped barrier layer 31 is formed below the lower doped barrier layer 27 and an upper undoped barrier layer 33 is formed above the upper doped barrier layer 25. The upper spacer, upper doped barrier and upper undoped barrier layers 21, 25, 23 constitute an "upper barrier region" 26 and the lower spacer, lower doped barrier and lower undoped barrier layers 23, 27, 31 constitute a "lower barrier region" 28.

In use, the control beam 15 is chosen to have a photon energy larger than the bandgap of the material of the barrier layers 25, 27 to the quantum well 19.

The waveguided beam 13 has a lower photon energy around the bandgap of the well material.

Details of the complete layer structure, in order of growth, can be seen in Figure 3; 1 Rm GaAs buffer layer 29; 1 Rm Al0.5Gao.sAs waveguide cladding 5; 100 nm Alo 33Ga0 67As lower barrier 31; 100 nm A10.33Ga0.67As lower barrier (p-type 5 x 1017 cm-3) 27; 100 nm Al0.33Ga0.67As lower barrier (undoped spacer) 23; 20 nm GaAs quantum well layer 19; 100 nm Alo 33Ga0 67As upper barrier (undoped spacer) 21; 100 run Alo 33Ga0 67As upper barrier (p-type 5 x 1017 cm-3) 25; 100 nm A10.33Ga0.67AS upper barrier 33; 1 ,um AIo.5Gao.5As waveguide cladding 7; 10 nm GaAs cap 35.

The ridge 9 is 1 -Spunwide.

Figure 4 explains how the control beam 15 is used to vary the excess carrier density in the quantum well layer 19 and thereby change the absorption of the waveguided beam. The energy of the control beam 15 is chosen to have an energy greater than the bandgap of the barrier material, but smaller than that of the waveguide cladding 5. Hence most of control light is absorbed in the relatively thick barrier layers above and below the quantum well.

Figure 4 shows how the internal electric field produced by the ionised acceptors 37 in the barriers tends to spatially separate the photo-excited electrons 39 and holes 40. Electrons photo-excited in the spacer regions 21, 23 in the barrier either side of the well 19 are swept by the internal electric field into the quantum well 19, where they recombine with the excess holes, thereby lowering their density. On the other hand, the photo-excited holes are swept toward the maxima of the valence band, as shown in Figure 4. Some of these photo-excited holes recombine with ionised acceptors. The effect of the control light, therefore, is to reduce the excess hole density in the quantum well.Since the absorption of the quantum well (at energies around the quantum well bandgap) is very sensitive to the excess hole density, the transmission of the waveguided beam can be controlled and switched between high and low values.

The device of Figures 1-3 has a p-type remotely doped quantum well, although this principle can also be applied to a n-type remotely doped well.

However, it has been found that the depletion of the quantum well can be achieved for lower intensities of the control light beam for the case of a p-type remotely-doped quantum well. This may be due to photo-excited electrons being more efficiently swept into the well for remote p-type doping, than photo-excited holes for n-type structures.

Although layers 25 and 27 of the upper barrier region 26 and lower barrier region 28 have been doped for the device shown in Figures 1-3, the device will also operate if only the upper barrier layer 25 is doped. In that case, operation relies upon most of the light being absorbed by the upper barrier.

However, the double-sided doping arrangement is preferred, since this will be more sensitive to the control light.

The device could also be made with a stack of several quantum wells each remotely doped in its upper and lower barrier regions.

The excess hole density in the quantum well layer 19 is varied with the control optical beam 15, rather than an electrical voltage as with some conventional modulators. Consequently gate and ohmic contacts are not required for this device but can be optionally included. The ridge 9 can be etched to a depth within either the upper cladding layer 7 as shown in Figure 1, or the waveguiding layer 3. For the latter, the ridge 9 is etched to the depth of the dopant atoms, so that there are no excess carriers in the area of the quantum well layer 19, outside the ridge 9. This has the advantage of reducing absorption losses due to absorption outside the ridge in the transparent state when operating at a relatively lower photon energy A, corresponding to a charged exciton.When operating at a relatively higher photon energy B, corresponding to a neutral exciton (Figure 5) it may be better to etch only as far as the upper waveguide layer.

The control beam 15 should preferably impinge uniformly on the whole of the upper surface 17 of the ridge so as to achieve uniform depletion of the quantum well layer 19. The intensity of the control beam 15 should be switched between two finite values which achieve maximum change in the absorption of the waveguided beam 13. Alternatively, the spacer layer thicknesses, dopant concentrations etc., can be chosen so that either the optimal transparent or opaque state is achieved with no light incident on the structure.

This allows the intensity of the control light to be varied between zero and a finite value to achieve maximum modulation of the waveguided beam 13.

Figure 6 shows a schematic of the variation in the intensity of the beam transmitted through the waveguide (Pout) as a function of the intensity of the control beam (PControl). When intensity of the control beam is low the excess carrier density in the well is relatively large and hence the absorption profile resembles the dashed line shown in Figure 5. The absorption at photon energy A is relatively large and that at photon energy B relatively small.

As the intensity of the control beam increases, the excess carrier density in the quantum well decreases, as discussed above. For some control beam intensity the excess carrier density is minimal and the absorption profile then resembles the solid line in Figure 5. The absorption at photon energy A is reduced, while that at photon energy B is increased. Hence, increasing the control beam intensity has the effect of increasing the transmission of the waveguide at photon energy A, while decreasing its transmission at photon energy B, as shown schematically in Figure 6.

Thus, the intensity of the control beam can be varied so as to switch the transmission of the waveguide at either photon energy A or photon energy B.

A second embodiment of optical modulator according to the present invention is shown in Figures 7-9. Like with the first embodiment a control optical beam is used to switch the transmission of a second beam propagating through the waveguide between low and high values. The modulator of the second embodiment operates in a similar manner to the first embodiment, except that the photo-excited electrons (or holes) increase the excess carrier density in the quantum well, rather than decrease it as with the first embodiment Referring now to Figures 7 and 8, there is shown an optical modulator 41, comprising a waveguide region 43 disposed between a lower cladding layer 45 and an upper cladding layer 47. The upper cladding layer 47 is etched to form a ridge 49 on which is formed a p-type doped layer 51. An n-type doped layer 53 is formed below the lower cladding layer 45.The ridge 49 defines an optical path 55 thereunder, through the waveguide region 43, for transmitting a beam 57.

Optionally, electrical contacts can be provided to the p-type layer 51 and n-type layer 53. In the case of the latter, this is preferably affected by means of an ohmic contact 59 with a contact pad 61, disposed in a recess 63 through the waveguide region 43 and upper cladding layer 47. The ohmic contact to the lower doped region can also be made to the back face of the substrate, if the lower doped region is grown on similarly doped buffer and substrate layers.

The transmitted beam 57 is modulated by means of an applied control beam 65 which is used to irradiate the upper surface of the p-type layer 51 on top of the ridge 49.

Details of the waveguide region 43 are shown in Figure 8. It comprises a quantum well layer 67, disposed between a lower undoped barrier layer 69 and an upper undoped barrier layer 71. Details of the complete layer structure of this device are shown in Figure 9. In order of growth, these are as follows:1 um GaAs buffer and substrate layers 73; 1 Rm GaAs back gate (doped n-type 1018 cm-3) 53; 0.5 pLm AIo.5Gao.5As waveguide cladding layer 45; 300 nm Alo 33GaO 67As lower barrier layer (undoped)69; 20 nm GaAs quantum well layer 67; 300 nm Alo 33Ga0 67As upper barrier (undoped) layer 71; 0.3 ,um mi SGao gAS waveguide cladding layer 47; 0.2 Rm Alo sGa05As waveguide cladding (doped p-type 1018 cm-3) layer 51; 10 nm GaAs cap layer 75.

The ridge 49 is 1 - 5 im wide and etched to leave 0.2 Cun of the upper waveguide cladding layer. The back gate could also be n-type Alo sGa0 sAs so as to produce extra optical confinement. The upper cladding layer could also be made from a dielectric material with a refractive index smaller than that of the Alo 33Ga0 67As guided region and be transparent to the control beam.

Optionally the area of the device outside of the ridge can be covered with a suitable material opaque to the control light, so as to prevent penetration of the control beam in these regions.

Doped n-type 53 and p-type 51 regions are arranged relatively far from either side of the quantum well layer 67, as shown in Figure 7, so that there is an (approximately) uniform electric field across the guided region 43. As mentioned above, optionally, ohmic contacts can be made to the n-type and p-type regions 51, 53 and a voltage applied between them so as to increase this electric field and thus enhance the switching speed of the device.

The upper doped region 51 can be replaced with a semi-transparent Au layer, which will act as a Schottky contact, while retaining the lower doped region 53. The layer should be sufficiently thin to allow transmission of a significant fraction of the control light. The Schottky contact will develop an electric field across the guided region in a similar manner to the p-i-n structure of US-A-4 716 449. Again electrical contacts can be made to the Schottky layer and doped lower region 53, so that a voltage can be applied between them in order to enhance the electric field across the guided region.

The control beam 65 is chosen to have a photon energy larger than the bandgap of the material of the barrier layers 69, 71 but smaller than that of the waveguide cladding layers 45, 47. Its photon energy is chosen to be sufficiently large that most of the control beam energy is absorbed in the upper barrier region 71, creating photo-excited electron-hole pairs. The electric field across the guided region separates these photo-excited carriers, with the electrons being accelerated in one direction and the holes in the other. For the arrangement of p-type and n-type layers shown in Figure 7, electrons photoexcited in the upper barrier are swept into the quantum well, while the holes are swept toward the p-type layer. This can be seen from Figure 10, which shows the spatial variation of the conduction 77 and valence 79 band edges within the guided region 43.Electrons 81 which are photo-excited in the upper barrier layer 71 are swept into the quantum well 67 by the gradient. Photo-excited holes 83 are swept towards the front of the device.

Hence the control light beam 65 causes an increase in the excess electron density within the quantum well 67 and thereby modifies its bandedge absorption spectrum. This change in the bandedge absorption profile of the quantum well 67 is used to switch the transmission of the second optical beam 57 propagating through the waveguide with a photon energy close to the quantum well bandedge.

Figure 11 shows a schematic of the variation in the intensity of the waveguided beam 57 with the intensity of the control light beam 65. It can be seen that for operation at photon energies A and B (Figure 5), the operation is reversed, relative to the situation shown in Figure 6.

Figure 7 shows the upper doped region being p-type and the lower one n-type, so that electrons photo-excited in the upper barrier 71 are swept into the quantum well 67 (Figure 10). However, interchanging the positions of the ntype and p-type layers reverses the direction of the electric field, so that holes photo-excited in the upper barrier layer 71 are swept into the well 67. The former arrangement produces faster switching times, since the electrons are more efficiently collected by the electric field.

An embodiment which is an optical detector 91 in accordance with the present invention, is shown in Figures 12 and 13. A quantum well layer 93 is disposed between an undoped lower spacer layer 95 and an upper undoped spacer layer 97. Above the upper undoped spacer layer 97 is disposed a doped upper barrier layer 99. A lower doped barrier layer 101 is disposed beneath the lower undoped spacer layer 95. Beneath the entire above-mentioned structure is situated a bottom undoped barrier layer 103 and above the aforementioned structure is a top undoped barrier layer 105, covered with a capping layer 107.

A source ohmic contact 109 contacts the quantum well layer 93 and spaced apart therefrom, a drain ohmic contact 111 also contacts the quantum well layer. The device is operated by illumination with an incident light beam 113 which illuminates the upper surface 115 of the capping layer 107.

Details of the entire layer structure are shown in Figure 13, in order of growth:1 ,um GaAs buffer layer and substrate 117; 0.5 Rm Alo 33Ga0 67As bottom barrier 103; 100 nm Al33Ga0.67As lower barrier (doped 5 x 1017 cm-3) layer 101; 100 nm Alo 33Ga0 67As lower barrier (undoped spacer) layer 95; 30 nm GaAs quantum well layer 93; 100 nm Ab 33Ga0 67As upper barrier (undoped spacer) layer 97; 100 nm Alo 33Ga0 67As upper barrier (doped 5 x 1017 cm-3) 99; 100 nm run 33Gao.67As top barrier (undoped) layer 105; and 170 A GaAs cap layer 107.

These layers are etched to form a mesa of finite area, with the etch depth extending below the quantum well. The ohmic contacts 109, 111 to the quantum well layer 93 are formed on top of, and at opposite ends of, the mesa, using conventional methods.

Si can be used as a p-type dopant in GaAs/Al0.33Ga0.67AS heterostructures grown on a (311) oriented GaAs substrate. Alternatively, other p-type dopants are available on (100) oriented GaAs substrates, as known to those skilled in the art.

In use, a voltage is applied between the source and drain ohmic contacts 109, 111, causing a current (I) to flow through the quantum well layer 93 containing excess carriers. The light beam 113 incident on the structure, of intensity P' which has a photon energy larger than the bandgap of the barrier layers 103, 101, 95, 97, 99, 105 in the structure, causes a decrease in the excess carrier density in the quantum well layer 95, hence an increase in its resistance and a decrease in the current between the source and the drain, as shown in Figure 14. Hence, the level of light incident on the top surface 115 of the structure can be determined from the measured current.

Optionally the structure can be doped solely in its upper barrier region 99. However, it is desirable to remotely dope the quantum well region in both the upper and lower barrier regions 99, 101, since in this case the device will also be sensitive to photons absorbed in the lower barrier region 101. For the asymmetrically-doped GaAs/A10 33GaO 67As heterostructure, the undoped lower AIo.33Gao.67As barrier can be replaced by a superlattice of alternating layers of GaAs and Alo 33Ga0 67As layers, each with a thicknesses of 2.5 nm (for example). This should improve the quality of the quantum well layer.

Optionally the lower barrier and its dopants can be omitted altogether.

In this case the layer structure consists of a thick (e.g. 2 micron) GaAs layer, followed by the upper barrier layers 97, 99, 105 and capping layer 107 shown in Figure 13.

Several quantum wells may also be arranged on top of one another separated by barriers, a central portion of each barrier being doped.

The area around the source and drain ohmic contacts 109, 111 can optionally be covered by an opaque material using standard techniques. This has the advantage of preventing the light from effecting the operation of the contacts and, secondly, breaks the path for parallel conduction between the source and drain by photo-excited carriers in the barriers.

The current can be determined by means of an integrated or external circuit (not shown). For instance, the voltage can be measured across a resistor placed in series with the device and the voltage source.

The light to be detected should have a photon energy larger than the bandgap of the barrier material. The device relies upon the internal electric fields of the structure disassociating electron-hole pairs photo-excited in the barrier regions. The electric field is such that the photo-excited carriers of opposite sign to the majority carriers in the well are swept into the well, thereby reducing the excess carrier density in the well and increasing its resistance.

Figure 4 illustrates the internal electric fields present in a p-type remotely doped quantum well. The field in the undoped spacer region 97 of the upper barrier disassociates photo-excited electron-hole pairs. The photoexcited electrons 39 are swept into the quantum well 93, where they recombine with the excess holes, thereby reducing their density and increasing the resistance to a current flowing in the quantum well plane. The photo-excited holes 40 collect in the maxima of the valence band in the barrier layers, where some of them will recombine with ionised acceptors. The effect of the photoexcited holes, like that of the photo-excited electrons, is to reduce the excess hole density in the quantum well and increase its resistance.

Although Figure 4 illustrates operation of the device with p-type doped layers near the quantum well, clearly n-type layers could also be used. In this case holes photo-excited in the barrier regions are swept into the quantum well, where they recombine with, and lower the density of, the excess electrons. Our investigations have shown the structure with p-type doped layers near the well is more sensitive to the incident light.

A semi-transparent Schottky gate could optionally be added to the top of the structure, and a voltage applied between it and the drain ohmic contact 111, so as to alter or enhance the electric field in the upper barrier region. Similarly the structure could be grown on a doped region to form a back gate, and a voltage applied between the drain and an ohmic contact to this back gate, so as to modify and enhance the electric field in the lower barrier region.

The structure of Figures 12 and 13 uses an electric means to detect the light intensity, however, this could also be achieved with a second optical beam. For this purpose, the structure could be arranged within a waveguide structure, as in the device of Figures 1 and 2.

In the light of this disclosure, modifications of the described embodiments, as well as other embodiments, or within the scope of the present invention as defined by the appended claims, will now become apparent to persons skilled in this art.

Claims

1. An optically operable semiconductor device comprising a modulation region which comprises a quantum well layer formed of a semiconductor material having a predetermined bandgap and at least one barrier region formed of semiconductor material having a higher bandgap than said predetermined bandgap such that when the at least one barrier region is irradiated with an incident light beam having a photon energy higher than the bandgap of the at least one barrier region, the excess carrier density in the quantum well layer is altered.

2. A device according to claim 1, in which the modulation region is adapted to transmit a transmitted light beam having a photon energy close to or less than the bandgap of the quantum well layer so that the incident light beam modulates the intensity of the transmitted light beam.

3. A device according to claim 2, wherein the incident light beam functions to decrease the excess carrier density in the quantum well layer.

4. A device according to claim 3, wherein the at least one barrier region comprises a first doped barrier layer.

5. A device according to claim 4, wherein the at least one barrier region further comprises a second doped barrier layer and the quantum well layer is situated between the first and second doped barrier layers.

6. A device according to claim 5, wherein the quantum well layer is separated from the first and second doped barrier layers by respective spacer layers.

7. A device according to any preceding claim, further comprising one or more gate electrodes.

8. A device according to any preceding claim, further comprising source and drain contacts to the modulation region.

9. A device according to claim 2, wherein the incident light beam functions to increase the excess carrier density in the quantum well layer.

10. A device according to claim 9, having a layer structure for developing an electric field across the modulation region.

11. A device according to claim 10, wherein the modulation region is disposed between first and second doped layers.

12. A device according to any preceding claim, further comprising upper and lower cladding layers for the modulation region.

13. A device according to any preceding claim, wherein the modulation region comprises an optical path defined by a stripe waveguide structure.

14. A device according to claim 12, when dependent upon claim 8 or claim 9, wherein one of said first and second doped layers forms part of said stripe waveguide structure.

15. A device according to any of claims 10, 11 or 14, wherein respective electrical contacts are provided for said first and second doped layers.

16. A device according to claim 15, wherein one of said first and second doped layers is situated on the substrate side of the modulation region and the electrical contact to that layer is located in a recessed portion of the waveguide region.

17. A device according to any of claims 1-13, wherein the device is adapted so that a transmitted beam incident on an upper surface of the device can transit the modulation region.

18. A device according to claim 1, in which the incident light beam functions to change the resistance of the modulation region.

19. A device according to claim 18, wherein the incident light beam functions to decrease the excess carriers density in the quantum well layer.

20. A device according to claim 18 or claim 19, wherein the at least one barrier region comprises a first doped barrier layer.

21. A device according to claim 20, wherein the at least one barrier region further comprises a second doped barrier layer and the quantum well layer is situated between the first and second doped barrier layers.

22. A device according to claim 21, wherein the quantum well layer is separated from the first and second doped barrier layers by respective spacer layers.

23. A device according to any of claims 18-22, further comprising a gate electrode.

24. A device according to any of claims 18-21, further comprising source and drain contacts to the modulation region.

25. A device according to any preceding claim, wherein the modulation region comprises a plurality of quantum well layers.

26. An array of devices according to any preceding claim.

27. A device according to claim 1, further comprising a waveguide structure so that the incident light beam modifies a second beam transmitted through the modulation region and so that the emergent second light beam acts as an optical detection output signal, the device having a structure as defined in any of claims 20-25.

28. An optical modulator substantially as hereinbefore described with reference to any of Figures 1 to 3 and 7 to 9 of the accompanying drawings.

29. An optical detector substantially as hereinbefore described with reference to any of Figures 12 and 13 of the accompanying drawings.