GB2393260A

GB2393260A - Photoconducting terahertz transmitter/receiver system employing alternating bias voltage synchronised to the exciatation radiation pulse frequency

Info

Publication number: GB2393260A
Application number: GB0221803A
Authority: GB
Inventors: Bryan Edward Cole
Original assignee: TeraView Ltd
Current assignee: TeraView Ltd
Priority date: 2002-09-19
Filing date: 2002-09-19
Publication date: 2004-03-24
Anticipated expiration: 2022-09-19
Also published as: AU2003269156A1; WO2004027942A1; GB0221803D0; GB2393260B

Abstract

A Terahertz, THz, transmission/receiver system comprising a photoconducitve material, such as GaAs connected to first and second electrodes, a pulsed electromagnetic radiation excitation device, such as a sub picosecond mode-locked oscillator laser, and a biasing means which applies an oscillating voltage between the first and second electrodes. The oscillating bias voltage frequency is synchronised with the excitation pulse, preferably at half the pulse frequency. This serves to reverse the effect of space charge polarisation. Moreover the device uses space charge polarisation advantageously to enhance the generation efficiency. Furthermore a receiver system is provided employing high input impedance pre-amplifier suitable for coupling with a photoconductive THz receiver. The input impedance is determined by a resonant inductor-capacitor, LC, circuit whose component values are chosen such that the resonant frequency is the alternating bias frequency of the transmitter system.

Description

Method and Apparatus for Operating an Antenna The present invention

relates to emitters for the generation of radiation which can be used to image or determine compositional information from structures particularly radiation in the frequency range colloquially referred to as the TeraHertz frequency range, the range being that from 25GHz to 1 00THz, particularly that in the range of 50GHz to 84THz, more particularly that in the range from 90 GHz to 50 THz and especially that in the range from 100GHz to 20THz.

Recently there has been much interest in using Terahertz (THz) radiation to look at a wide variety of samples using various methods. For example, THz radiation can be used for both imaging samples and obtaining spectra at each pixel in an image. THz radiation penetrates most dry, nonmetallic and non-polar objects like plastic, paper, textiles, cardboard, semiconductors and non-polar organic substances. Therefore THz radiation can be used instead of x-rays to look inside boxes, cases etc. THz radiation also has medical uses.

The so-called "photoconductive" emitter is one device capable of generating high frequency radiation, particularly broadband pulsed THz radiation when used in conjunction with sub-picosecond optical/NIR laser pulses. A photoconductive emitter is one incorporating a photoconductive material between two electrodes, where exposure to light of an appropriate wavelength results in a dramatic increase in the electrical conductivity of the exposed material. The increased conductivity persists for a time corresponding to the "lifetime" of electrical charge carriers in the material.

When an electric field is applied between the electrodes this change in conductivity

results in the generation of broadband electromagnetic radiation that radiates away from the electrodes with frequencies extending into the Terahertz range. Using a photoconductive emitter in conjunction with a laser pulse of appropriate wavelength is one of the most effective methods of generating broadband pulsed radiation

In its simplest form, a photoconductive emitter comprises two electrodes provided on a surface of a photoconductive material, such as a semiconductor material. Many semiconductor materials, such as GaAs and Si are photoconductive. To operate the emitter, a bias voltage is applied between the electrodes and the photoconductive material is exposed to one or more laser sources (depending upon the generation approach being utiliscd such as pulsed laser or CW laser) of suitable wavelength This significantly increases the conductivity of the semiconductor to the extent that a current flows through the material between the electrodes due to the presence of the bias electric field. Provided the bias field is maintained, the current will persist for a time

corresponding to the lifetime of the photo-created charge carriers in the material. For example, for a GaAs photoconductive device the lifetime is typically of the order of ins.

If the laser pulse used to illuminate the photoconductive material between the electrodes is sufficient short (i.e. preferably <I picosecond) then the resulting current transient will radiate electromagnetic radiation away from the electrodes.

When a DC electrical bias is applied between the two electrodes on the surface of a photoconductive material, the electric field is found to concentrate in the regions

immediately adjacent to the electrodes, while in the remaining area midway between the electrodes the field is substantially reduced. This effect results from a build up of

"space charge polarization" in the photoconductive material.

Space charge polarization is the migration of charge carriers, which concentrates the electric field at a boundary layer adjacent to each electrode and effectively cancels the

applied electric field between the electrodes. This effect is illustrated schematically in

Figures 1 a and 1 b.

Figure I a illustrates the state of a photoconductive material I a in contact with and between a first electrode 2a and second electrode 3a before space charge polarization takes effect. The graph 4a positionally illustrates the corresponding voltage level (electric potential) throughout the electrodc/photoconductive material combination.

More specifically, in Figure 1 a a potential difference has been applied between the two electrodes so as to create an electric field in the photoconductive material. Graph 4a

( shows that the voltage within the first electrode 2a with +Vmax applied is constant.

Similarly, the potential within the second electrode 3a is zero volts, and a constant zero i voltage value is evident on the graph 4a for the region corresponding to this electrode.

From the graph 4a it is also apparent that a linear decrease in voltage occurs from +Vmax to zero volts through the semiconductor material 1 a (which in this example is GaAs), l with the +Vmax voltage occurring in the region adjoining the first electrode and zero volts in the region adjoining the second electrode.

Figure 1 b illustrates the state of the photoconductive material of Figure 1 a once space charge polarization has taken effect. Space charge polarization generally occurs after the electric field has been applied for a period of time. In Figure lb, with reference to

the graph 4b, the constant potentials of Vmax and zero volts through the first and second electrodes continue to exist. However, now, in the photoconductive material lb, negative charge carriers have gathered in the region adjacent the first electrode 2b of potential +Vmax and positive charge carriers have gathered in the region adjacent the second electrode 3b. An electric field still exists in these regions adjacent the

electrodes, as illustrated by the sharp change in potential difference on the graph 4b in these locations. In fact, the electric field that exists in each instance is an increased

electric field, which is evident in graph 4b through the increased gradient.

However, these electric fields, in the regions of concentration of negative and positive

charge carriers adjacent the first and second electrodes respectively, serve to virtually cancel the electric field in the remaining central region of the photoconductive material.

This effect is illustrated in graph 4b by the virtually constant voltage between the two sharp voltage drops whose sum equals, or at least substantially equals, the total potential difference between the two electrodes. The polarization therefore creates an essentially constant voltage through the majority of the GaAs material so that no, or at least minimal, electric field exists in the central region.

Hence, the change in potential difference from +Vmax to zero volts, which previously occurred throughout the photoconductive material, now only occurs in the regions immediately adjacent the electrodes.

For photoconductive emitters, this polarisation problem is exacerbated since the rate of charge migration is substantially increased by exposure to light.

It has been recognised that the radiation emitted from a photoconductive emitter exhibiting space charge polarization may be increased by concentrating the optical excitation at the regions of the GaAs material adjacent to the electrodes, where the electric field is increased. This is because the power of radiation emitted by a

photoconductive device increases with the local electric field at the point of optical

excitation. In an article entitled "Trap-enhanced electric fields in semiinsulators: The role of

electrical and optical carrier injection" by S. Ralph and D Grischkowsky and published in Appl. Phys. Lett. 59 (16), 14 October 1991 it was also recognised that the electric field preferentially concentrates at the anode of such a photoconductive emitter, so that

optimum radiation emission can be achieved by focussing the optical beam directly adjacent the anode rather than the cathode.

Another approach to further enhance the local electric field was disclosed in US Patent

Number 5,729,017 and involves designing the electrodes with sharply pointed corners, in order to create one or more geometric singularities in the electric field.

While such photoconductive emitters provide radiation with acceptable power emission levels, for optimum radiation emission efficiency, it must be ensured that the near-

infrared excitation beam is absorbed within the region of GaAs experiencing the highest electric field. That is, it is necessary to maximise the overlap between the optical beam

and the high-field regions of the GaAs. However, with emitters exhibiting space charge

polarization, the highest electric field is found in a sub-micron layer adjacent to the

electrodes. This is a poor match to a broad gaussian mode of an optical excitation beam, which is typically used to excite such emitters.

Further, with these configurations, the maximum emitted energy per pulse is limited by space charge polarization. This is because the maximum energy per pulse that can be converted to radiation (and hence emitted power) is proportional to the maximum

( electrostatic energy stored in the electric field of the emitter. Hence, in order to increase

the power of the emitted radiation, the stored energy must be increased. The total electrostatic stored energy is proportional to the integral of the square of the electric field over the volume of the device. However, since the electric field is concentrated

into a small volumetric region adjacent to the electrodes in the configurations discussed above, the stored energy is limited. In other words, the emitted power with these configurations is limited by the localised electric field caused by space charge

polarization. Therefore, while existing emitter designs have aimed to utilise the space charge polarization effects in an advantageous manner, there are still inefficiencies caused by space charge polarization. There is therefore room for improvements in the design and operation of photoconductive emitters.

There is therefore a need for an emitter with enhanced radiation emission efficiency.

It is also desirable to enhance power-handling and the lifetime of photoconductive emitters. It is an aim of the present invention to overcome or alleviate at least one of the problems of the prior art.

According to a first aspect, the present invention provides an emission system comprising: an emitter comprising a photoconductive material and first and second electrodes; an optical excitation device for applying a pulsed radiation signal to the emitter at a defined pulse repetition frequency; and a biasing means configured to apply an alternating bias voltage signal to the first and second electrodes at a frequency synchronized with the defined pulse repetition frequency.

In deriving this aspect of the invention, it was recognised that the concentration of the electric field at the electrodes of a photoconductive emitter, rather than enhancing

device performance through the increased electric field, in fact exhibited a destructive

phenomena that inhibited the performance of the device. The present invention

( provides an approach that, by applying an alternating bias signal to the emitter at a l'requcncy synchronized with the pulse repetition frequency, or"irradiation frequency", the detrimental effects of space charge polarization may not only be reduced but reversed such that the space charge is used advantageously to increase the power of the emitted radiation.

Preferably the frequency of the alternating bias signal corresponds to half the irradiation frequency. According to a further aspect the present invention provides an emission system comprising: an emitter comprising a photoconductive material and first and second electrodes; an excitation device for applying a pulsed radiation signal to the emitter at a defined pulse repetition frequency; and a biasing means configured to apply an alternating bias signal to the first and second electrodes wherein the frequency of the alternating bias signal is set such that the polarity of an electric field between the first

and second electrodes is reversed for each pulse of the pulsed radiation signal.

When the alternating bias signal is utilised to reverse the polarity of the electric field

between the emitter's electrodes after each excitation pulse, the space charge polarization that has built up adjacent the emitter's electrodes then acts to enhance the electric field midway between the electrodes while reducing the field directly adjacent

to the electrodes.

It is preferable for the emission system to generate THz radiation.

The excitation device may be a laser such as a sub picosecond mode-locked oscillator laser or an amplified laser.

Preferably the emission system also includes a phase shifter for synchronizing the maxima and minima of the alternating bias signal with the arrival of the pulses of the pulsed radiation signal at the emitter. Further, the frequency of the alternating bias signal may be controlled using a reference output from the excitation device.

( The biasing means of the emission system preferably comprises: input means for receiving a low voltage bias signal; and a transforming element for generating a high voltage alternating bias signal from the low voltage bias signal, such that the high voltage alternating bias signal is applied to the first and second electrodes. The transforming element may be a transformer, such as a flyback transformer or part of a tuned LCresonating circuit or an inductor in a "flyback" pulse generating circuit.

Alternatively, the biasing means of the emission system may comprise a first pulse generator and a second pulse generator, such that the two pulse generators operate in anti-phase with each other and the output of the first pulse generator is applied to the first electrode and the output of the second pulse generator is applied to the second electrode. Preferably the first and second pulse generators comprise one or more high voltage transistors in conjunction with an inductor. It is also preferable that the one or more high voltage transistors are MOSFET or IGBT type devices.

According to a related aspect the present invention provides a method of operating an emitter, the emitter comprising a photoconductive material and first and second electrodes and being configured to emit radiation upon receiving a pulsed radiation signal at a defined pulse repetition frequency from an excitation device, the method comprising applying an alternating bias voltage signal to the first and second electrodes of the emitter at a frequency synchronized with the pulse repetition frequency.

Preferably the method further comprises obtaining a first and second bias signal, such that the first signal is an inversion of the second signal; applying the first bias signal to a first pulse generator; applying the second bias signal to a second pulse generator; applying the output of the first pulse generator to the first electrode of the emitter; and applying the output of the second pulse generator to the second electrode of the emitter.

It is also preferable that the first and second bias signals are derived from a reference output of the excitation device.

According to another related aspect, the present invention provides a biasing circuit for biasing an emitter, the emitter comprising a photoconductive material and first and

( second electrodes and being configured to emit radiation upon receiving a pulsed radiation signal at a donned pulse repetition frequency from an excitation device, the biasing circuit comprising: a generator configuecd to apply an alternating bias voltage to the emitter at a frequency syncllronised with the pulse repetition frequency.

Preferably the generator comprises a first pulse generator and a second pulse generator, such that the first pulse generator operates in antiphase relative to the second pulse generator and the output of the first pulse generator is applied to the first electrode and the output of the second pulse generator is applied to the second electrode.

These aspects of the present invention result in a higher modulation frequency than is typically utilised. This higher modulation frequency poses problems for the use of many receivers, in that a very high impedance preamplifier is required to achieve optimum performance.

Therefore, according to a still further aspect the present invention provides a radiation receiver system comprising: a receiver unit; a preamplifier; and a resonance circuit for increasing the input impedance of the preamplifier.

Preferably the receiver unit is a photoconductivc THz receiver. It is also preferable that the resonance circuit comprises an inductor and a capacitance and the value of the inductor is such that the resonance of the resonance circuit lies at an AC bias frequency of a transmitter system corresponding to the receiver system.

In addition it is preferable to provide a variable capacitor in parallel with the preamplifier unit for fine adjustment of the resonant frequency of the resonant circuit.

The present invention will now be described with reference to the following non-

limiting embodiments in which: Figure la illustrates schematically an electrode/photoconductive material combination before the effects of space-charge polarization, together with a graphical representation of the applicable electric potential.

( Figure lb illustrates schematically the elcctrode/photoconductive material combination of F igure 1 a after polarization, together with a graphical representation of the applicable electric potential.

Figures 2a, 2b, 2c, 2d and 2e illustrate graphically the effect of reversing the sign of the bias-voltage between each laser pulse.

Figure 3 illustrates a schematic of an AC-biasing system according to one embodiment of the invention.

Figure 4 illustrates a schematic diagram of an AC-biasing system with an amplified laser according to another embodiment of the invention.

Figure 5 illustrates a schematic diagram of a pulse generator suitable for use in an AC-

biasing system with an amplified laser according to an embodiment of the present invention. Figure 6 illustrates a circuit diagram of a JFET-based pre-amplifier for a THz photoconductive receiver.

A schematic of a biasing arrangement according to one embodiment of the present invention is illustrated in Figure 3. The excitation component of this arrangement includes the laser 31 which directs a near infra-red (NIR) pulse train to a lens 32 which focuses the pulse train appropriately on the photoconductive emitter 33, such as a THz emitter. It is to be appreciated that the present invention is not specific to any particular type of emitter geometry or photoconductive material.

The laser 31 in this example is a sub-picosecond mode-locked oscillator laser (as opposed to an amplified laser), which typically has an irradiation frequency or repetition rate in the region of 50MHz to I OOMHz. For example, the Coherent "Vitesses" Ti:sapphire laser has an irradiation frequency of 80MHz.

( 1() Thereforc, a high voltage AC biasing signal at half the irradiation frequency (40MHz) needs to be generated to apply to the emitter 33.

T o achieve this, with rcEerence to Figure 3, an 80MHz clock signal is obtained from the laser 31 via a synchronous output. This is illustrated at c' in Figure 3. This clock signal is applied to a "divide-by-2" filter 34, whose output b is a 40MHz clock signal in the form of a square wave. This signal is then input to a phase shifter 35, which preferably has an external user phase control so as to allow phase adjustment with a view to aligning the signal peak with the arrival of the laser pulse on the emitter. The output c from the phase shifter 35 is then input to a filter 36 in order to create a 40MHz sinusoidal signal from the 40MHz square wave signal, as output d.

This signal d is the amplified in an RF power-amplifier 37 and the output e is fed into a tuned LC series resonant circuit 38. The resonance frequency may be adjusted by varying the value of the capacitor. On resonance, energy is stored in the LC resonator 38 and the voltagefacross the capacitor, relative to the input driving voltage e, is a much larger 40MHz AC voltage. This signalf, which is the alternating bias signal, is then applied to an electrode of the emitter 33 to provide the AC bias. The other electrode of the emitter is grounded. Where an external phase control of the phase shifter 35 is provided, the phase of the signal applied to the emitter can be independently controlled to ensure that the arrival of the laser pulse occurs at the time of maximum voltage drop across the emitter's electrodes. This allows the power of the emitted radiation to be maximised.

The voltage amplification of the Figure 3 circuit is determined by losses in the inductor of the LC series resonant circuit 38 and the emitter 33. A voltage gain of approximately 10 is typical. For example, a 10W RF power amplifier can generate a peak voltage of over 20 volts, resulting in over 200 V peak across the emitter.

It is to be appreciated that the high voltage pulsed signal may be generated by other means such as using a "flyback" inductor or transformer. High voltage pulses are accomplished with a fly-back inductor by passing a high-value saw-tooth waveform current through the inductor. The voltage across the device is the rate-of-change

( (differential) of the current, so a low-value constant voltage causes the current to linearly ramp up in the inductor. The current is then forced to zero by closing a transistor switch. The resulting fast drop in current generates the desired higl1 voltage l pulse across the inductor.

A "flyback" transformer operates in a similar manner. The device has a primary coil with a small number of turns and a secondary coil with a larger number. A low voltage is applied to ramp up the current in the primary coil then the current is rapidly forced to zero, generating a voltage pulse across it. The voltage across the secondary coil mirrors that of the primary, but is multiplied up by a factor equal to the ration of turns on the two coils.

While the alternating bias signal in this embodiment of the invention is a sinusoidal signal, it is to be appreciated that this is not essential to the invention and that any appropriate alternating signal may be utilised. For example, the alternating signal may be an alternating "short pulse" waveform, a square wave, ramp, sawtooth or a triangular wave. A schematic of a biasing arrangement according to another embodiment of the present invention is illustrated in Figure 4. This arrangement is suitable for use with an amplified laser system used in THz generation. As with the Figure 3 arrangement, the excitation component includes a laser 41 which directs a near infra-red (NIR) pulse train to a lens 42 which focuses the pulse train appropriately on the photoconductive emitter 43, such as a THz emitter. Once again, this embodiment of the invention is not specific to any particular type of emitter geometry or photoconductive material. One suitable emitter is a conventional 1 mm gap design formed on a semi-insulating GaAs substrate with gold electrodes evaporated onto its surface using conventional lithographic techniques. The amplified laser 41 may be any suitable laser such as a Coherent RegA9000. Such lasers typically have a repetition rate of 250kHz. This repetition rate is sufficiently low for a dual pulse- generator configuration to be used. A clock signal a is obtained from the laser 41 via its synchronous output (sync out), which is applied to a "divide-by-2"

( filter 44, whose output h is a 125kHz clock signal. The circuit then branches into two, with each branch including a phase shifter (45a, 45b) and a pulse generator (46a, 46b) and the output of each branch being applied to the emitter 43. The signal b is applied to each branch.

One of the branches additionally includes an inverter 47 situated before both the phase shifter (45b) and the pulse generator (46b). By virtue of this inverter 47, the input to the I phase shifter 45a on the first branch is the 125kHz clock signal and the input to the phase shifter 45b on the second branch is an inverted version of that 125kHz reference signal: that is the signals are in anti-phase. The phase shifters provide a means of fine tuning the circuit timing. In an ideal circuit, where the signals in the first and second branch are in-phase, they would not be necessary.

The output of phase shifter 45a is then applied to pulse generator I (46a) and the output of phase shifter 45b is applied to pulse generator 2 (46b) . The nature of the two input signals serves to alternately trigger the two high voltage (positive) pulse generators.

A schematic showing the operation of a high voltage pulse generator (46a, 46b) suitable for switching voltages over 1000V is illustrated in Figure 5. The incoming clock signal 51 is input to a "one-shot" astable-multivibrator filter 52, which generates a short pulse 53 from the positive edge of the input signal. This is inverted by inverter 54 and then buffered 55 before being fed into the gate of a high voltage transistor 56 such as a MOSFET or IGBT type device. Although only one high voltage transistor is illustrated in Figure 5, more than one may be utilised.

Between pulses the transistor is held "on" such that a large current (generally >5A) may build up through the inductor 57. During this time, the current output from the inductor travels through the drain of the transistor and is effectively shorted to ground at the source of the transistor 56. To generate a high voltage pulse 59, the transistor is shut off by applying the negative pulse 58 to the gate input of the transistor. This results in a very fast decrease in current through the inductor 57 with the generation of a correspondingly large back-EMF across the inductor creating the high voltage pulse 59,

( since the voltage across the inductor is proportional to the rate of change of current through the inductor (i.e. Vat - L x (dl/dt)).

The outputs of the two pulse generators are respectively applied to the electrodes of the emitter. In this way, the two electrodes arc alternately pulsed at half the repetition rate of the laser 41. This has the effect of reversing the sign of the electric f eld across the emitter at each laser pulse. Also, the phase of the two pulsed bias signals are preferably controlled independently to ensure that the arrival of the laser pulse at the emitter occurs at the time of maximum voltage drop across the electrodes in order to maximise the emitted radiation.

Applying an AC bias to the emitter, such that the frequency is half the repetition rate of the excitation laser, serves to reverse the effects of space charge polarisation. The present invention does not prevent space charge polarisation occurring, but utilises it advantageously. That is, the photoconductive material of the emitter implementing the present invention still gains some space charge polarisation after each laser pulse.

However, when the sign of the bias signal for the next laser pulse is reversed, the electric field between the electrodes is reversed and the space charge polarisation that

has built up adjacent the electrodes then acts to enhance the electric field midway

between the electrodes while reducing the field directly adjacent to the electrodes. At

the point at which the bias signal changes sign, the applied bias signal is zero (by definition), so at this point the total field is due to the space charge polarisation. Then

as the applied bias increase from zero, it enhances the space charge polarisation field,

until it peaks and the next optical pulse arrives.

While AC biasing is a known technique, bias frequencies utilised to date are typically substantially less than half the laser repetition rate andhence do not counteract space charge polarisation. For such low laser repetition rates, the photoconductive material of the emitter gains space charge polarisation after each laser pulse, which acts to reduce the electric field midway between the electrodes without embodying the advantageous

increase in the electric field between the electrodes to boost the radiation emission, as

per the present invention. By applying the AC bias at a frequency of half the laser

( repetition rate, the sign of the electric field between the electrodes is reversed for each

laser pulse.

While it is preferable that the AC bias frequency is applied at half the pulse repetition rate, this bias frequency is not essential to the invention. When it is considered that nearly all the space charge polarization occurs during and immediately after the arrival of the optical pulse, it is to be appreciated that it is the sign of the applied bias at the instant of optical excitation that is important. Therefore, it is unimportant what the bias does in between optical pulses, so a bias frequency of 1.5 times the laser repetition rate would also embody the principles of the present invention.

However, it is generally true that high voltage signals at higher frequencies are harder to generate. Thus while any odd-integer harmonic of half the laser repetition frequency could be used to improve emitter performance, the half repetition rate frequency is the easiest to implement. Other high frequency AC blase signals, such as a quarter of the repetition rate would show a small improvemenet over low frequency AC biasing, since a little cancellation of the space charge polarization could occur, but not to any great degree. With reference to Figure 2a to 2e the effect of reversing the sign of the bias-voltage between laser pulses is shown schematically. The phase of the AC-hias is adjusted such that the peak voltage coincides with the arrival of the laser pulse at the device. Figure 2a illustrates an essentially sinusoidal wave that corresponds to the bias voltage. Point ( I), is a point in time which corresponds to the positive rising edge of the bias voltage, just before the arrival of the first laser pulse and just before the bias reaches its peak positive value. At this point, the potential gradient through the photoconductive material is approximately uniform. This is evident in Figure 2b, which illustrates the voltage through the photoconductive material at the time corresponding to point ( I). The voltage drop through the photoconductive material, which also indicates the electric field through the material, is substantially linear. The position of the two electrodes is

evident by the two constant voltage regions of Vn,ax and zero volts, at the two extremes.

( Point (2) in Figure 2a is the point in time shortly after the first laser pulse. As the peak bias voltage also occurred in the region of the first laser pulse, the bias voltage at point (2) lies on the positive trailing edge of the bias voltage wave. The optical excitation provided by the first laser pulse served to trigger the motion of charge within the photoconductive material such that a space charge polarization built up immediately after the first laser pulse. RcEerring to Figure 2c, which illustrates the voltage through the photoconductive material at point (2), it is evident that this space charge polarization results in the concentration of the potential gradient close to the electrodes as illustrated by the two sharp voltage drops immediately adjacent the electrode positions evident at the constant voltages of Vmax and zero volts. Midway between the electrodes, a reduced and negligible electric field is apparent by virtue of the almost uniform potential

between the two sharp potential drops.

At point (3) on Figure 2a the bias voltage has reversed its sign, so that at this point the bias voltage is on a negative leading edge, just before a peak negative bias voltage and the next laser pulse. As the bias voltage reverses sign for the second laser pulse, the space charge polarization acts to enhance the potential gradient midway between the electrodes. This is illustrated in Figure 2d, which shows the voltage through the photoconductive material at point (3). In other words, the total electric field between

the electrodes is the sum the contribution from the applied field due to charge on the

electrodes and the space charge polarization field.

Under the prior art arrangements, these two fields acted to cancel one another out, such

that the total field was somewhat less than the applied field. However, by reversing the

sign of the applied field, the two contributions act together to give a total field that is

greater than either the applied field or space-charge field alone.

At point (3) the polarity of the electrodes have been reversed, so that the electrode that was previously Vmax volts is now zero volts and vice versa. Space charge polarization still occurs in the regions adjacent the two electrodes in the form of a voltage drop as shown in Figure 2d. The gradient of the voltage drop adjacent the electrodes is essentially unchanged from that which existed before the change in voltage polarity as the space charge polarization is retained. However, in view of the electrode polarity

( change caused by the sign reversal for the bias voltage, and combined with the space charge polarization effect, an electric field is generated between the two electrodes of

opposite polarity to the previous field. Further, the electric field generated is enhanced

by the space charge polarization and causes the potential gradient between the two electrodes to exceed that expected due to the applied bias alone.

Point (4) in Figure 2a is the point in time shortly after the second laser pulse. As the peak negative bias voltage also occurred in the region of the second laser pulse, the bias voltage at point (4) lies on the negative trailing edge of the bias voltage sinusoidal wave. From Figure 2e, which illustrates the voltage across the photoconductive material at point (4), it is apparent that charge migration again occurs, but in the opposite direct to before, so that a potential difference occurs in the regions immediately adjacent the two electrodes. That is, in view of the reversed polarity between the two electrodes, the potential difference appears as a positive potential difference rather than a negative potential difference, as occurred at point (2) where the reverse polarity was applied to the electrodes. As with the potential difference midway between the electrodes, the space charge polarization that occurs results in minimal electric field, as

is evident in Figure 2e by the essentially constant voltage midway between the two electrodes. Hence it is apparent that the voltage across the photoconductive material is virtually identical to that at point (2), except of reverse polarity.

A similar pattern as per points (3) and (4) will continue to occur as the bias voltage polarity changes between each laser pulse.

The present invention was tested using a 250KHz repetition rate Coherent RegA9OOO laser system and the dual-pulser antenna bias system shown in Figure 4. The THz antenna was a conventional Imm-gap design formed on a semi-insulating GaAs substrate, with gold electrodes evaporated onto its surface. The averaged THz power generated was measured using a Golay type detector (a Bolometric detector). It was found that by biasing the antenna at half the laser repetition rate, four times the power could be obtained as compared with unipolar biasing. This is because with unipolar pulsed biasing, the bias pulse is always of the same sign, and detrimental charge build up continues to occur, which reduces the performance of the emitter. This fourfold

( increase is an unexpected additional benefit as conventional expectations would predict a doubling of power rather than a quadrupling.. This four-fold increase in power also corresponds to a two-fold increase in emission efficiency. That is, for a constant emission efficiency, conventional expectations would predict a two-fold increase in power when operating both pulse generators compared to a single pulse generator operating alone, since we are only biasing the emitter for 50 X, of the laser pulses.

However, since a four-fold increase in power is possible when going from a unipolar biasing system to a bipolar AC bias system of the present invention, this implies a two-

fold increase in emission efficiency when AC biasing at half the laser repetition rate, is possible. Also, by biasing the emitter with an AC bias at half the laser repetition rate, radiation pulses are generated with alternating signs. There are various approaches for detecting such a train of pulses. For example, conventional bolometric detection (direct power measurement) is possible, as this approach is unaffected by the sign of the pulses. Also, the Electro-Optic Sampling (EOS) method of detection may be used. With the EOS approach, however, in view of the higher modulation frequency, specialized electronics such as a Radio Frequency Lockin Amplifier may be required.

The higher modulation frequency utilised in the present invention (for example 40MHz for many Ti:sapphire lasers) poses problems for the use of photoconductive receivers for detection, as these devices must feed a very high impedance pre-amplifier (typically over 10,000 kQ) to achieve optimum performance. However, preamplifiers of sufficiently high input impedance at 40MHz are not available due to the finite input capacitance of these devices. For example, for optimum performance, an input impedance of approximately I MQ is required. However, an input capacitance of 5pF (which is a very conservative value, as most preamplifiers are far in excess of this) will only yield an input impedance of 800Q.

To address the high frequency performance problem, the present invention proposes an approach utilising JFET transistors. At low frequencies the input impedance of JFET transistors is sufficient and these have been used successfully in the optimization of

photoconductive devices. Figure 6 illustrates a schematic diagram of an arrangement that the present invention proposes for increasing the input impedance of a JFET pre-

amplifier. To increase the input impedance of a JFET pre-amplifer 61 the present invention proposes placing a low loss inductor 62 in series with a JFET input capacitance 63 to create a resonance circuit. The preamplifier 61 and the resonance circuit are in turn in series with the receiver 64, which for THz generation is a THz receiver. The value of the inductor 62 should be such that the LC resonance formed by the inductor 62 and the input capacitance 63a of the JFET lie at the AC bias frequency, which for a typical oscillator based system is 40MHz. Therefore, L is determined by the formula: L = 1/ ((27rf)2C) Where C is the capacitance of the LC resonance circuit and f is the AC-bias frequency.

Therefore, for a 1 SpF circuit capacitance (5pF JFET input capacitance pluse I OpF variable capacitance) and a 40MHz frequency, the inductance required would be 1.061lH. Typically, the inductance could be expected to lie approximately in the range of O.l to IOOpH.

For fine adjustment of the resonant frequency, a variable capacitor 65 in parallel with the JFET input can be added.

Variations and additions are possible within the general inventive concept as will be apparent to those skilled in the art. It will be appreciated that the broad inventive concept of the present invention may be applied to any conventional type of emitter and that the exact embodiment show is intended to be merely illustrative and not limitative.

For example, the present invention may be readily applied to a number of different photoconductive emitter geometries such as the so-called coplanar strip line.

Furthermore, the emitter of the present invention may be used in various forms, such as in a combined TH% generation and detection system.

Claims

( CLAIMS:

1. An emission system comprising: an emitter comprising a photoconductive material and first and second electrodes; an excitation device for applying a pulsed radiation signal to the emitter at a defined pulse repetition frequency; a biasing means configured to apply an alternating bias signal to the first and second electrodes at a frequency synchronised with the pulse repetition frequency.

2. An emission system comprising: an emitter comprising a photoconductive material and first and second electrodes; an excitation device for applying a pulsed radiation signal to the emitter at a defined pulse repetition frequency; a biasing means configured to apply an alternating bias signal to the first and second electrodes wherein the frequency of the alternating bias signal is defined such that the polarity of an electric field between the first and second electrodes is reversed

for each pulse of the pulsed radiation signal.

3. The emission system of claim I or 2 wherein the frequency of the alternating bias signal corresponds to half the pulse repetition frequency.

4 The emission system of claim 1 or 2 wherein the frequency of the alternating bias signal corresponds to odd integers of half the pulse repetition frequency.

5. The emission system according to any one of claims I to 4 wherein THz radiation is generated.

6. The emission system according to any one of claims I to 5 wherein the excitation device is a sub pico-second mode-locked oscillator laser or an amplified laser.

( 2()

7. The emission system according to any one of claims 1 to f;rthcr including a phase shifter for aligning peaks of the alternating bias signal with the pulses of the pulsed radiation signal.

8. The emission system according to any one of claims I to 7 wherein the frequency of the alternating bias signal is controlled using a reference output from the excitation device.

9. The emission system according to any one of claims I to 8 wherein the biasing means comprises: input means for receiving a low voltage bias signal; a transforming element for generating a high voltage alternating bias signal from the low voltage bias signal, such that the high voltage alternating bias signal is applied to the first and second electrodes.

10. The emission system of claim 8 wherein the transforming element comprises a transformer, such as a flyback transformer, a part of a tuned LC-resonating circuit or an inductor in a flyback pulse generating circuit.

I I. The emission system according to any one of claims I to 10 wherein the biasing means comprises a first pulse generator and a second pulse generator, such that the first pulse generator operates in anti-phase relative to the second pulse generator and the output of the first pulse generator is applied to the first electrode and the output of the second pulse generator is applied to the second electrode.

12. The emission system according to claim I I wherein the first and second pulse generators comprise one or more high voltage transistors in conjunction with an inductor.

13. The emission system according to claim 12 wherein the one or more high voltage transistors are MOSFET or IGBT.

(

14. A method of operating an emitter, the emitter comprising a photoconductive material and first and second electrodes and being configured to emit radiation upon receiving a pulsed radiation signal at a defined pulse repetition frequency from an excitation device, the method comprising: applying an alternating bias voltage signal to the first and second electrodes of the emitter at a frequency synchronized with the pulse repetition frequency.

15. The method of claim 14 further comprising: obtaining a first and second bias signal, such that the first signal is an inversion ofthe second signal; applying the first bias signal to a first pulse generator; applying the second bias signal to a second pulse generator; applying the output of the first pulse generator to the first electrode of the emitter; and applying the output of the second pulse generator to the second electrode of the emitter.

16. The method of claim 15 wherein the first and second bias signals are derived from a reference output of the excitation device.

17 A radiation receiver system comprising: a receiver unit; a preamplifier; and a resonance circuit for increasing the input impedance of the preamplifier.

18. The receiver system of claim 17 wherein the receiver unit is a photoconductive THz receiver.

19. The receiver system of claim 17 or 18 wherein the resonance circuit comprises an inductor and a capacitance and the value of the inductor is such that the resonance of the resonance circuit lies at an AC bias frequency of a transmitter system corresponding to the receiver system.

(

20. The receiver system according to any one of claims 17 to 19 further comprising a variable capacitor in parallel with the preamplifier unit for fine adjustment ofthe resonant frequency of the resonant circuit.

21. The receiver system according to any one of claims 17 to 19 further comprising a variable inductor for fine adjustment of the resonant frequency of the resonant circuit.

22. A biasing circuit for biasing an emitter, the emitter comprising a photoconductive material and first and second electrodes and being configured to emit radiation upon receiving a pulsed radiation signal at a pulse repetition frequency from an excitation device, the biasing circuit comprising: a generator configured to apply an alternating bias voltage to the emitter at a frequency synchronized with the pulse repetition frequency.

23. The biasing circuit of claim 22 wherein the alternating bias voltage signal corresponds to an odd integer harmonic of half the pulse repetition frequency.

24. The biasing circuit of claim 22 or 23 further comprising a reference signal input for receiving a signal indicative of the pulse repetition frequency.

25. The biasing circuit according to any one of claim 22 to 24 wherein the generator comprises: a first pulse generator and a second pulse generator, such that each pulse generator operates in anti-phase and the output of the first pulse generator is applied to the first electrode and the output of the second pulse generator is applied to the second electrode.

26. An emitter substantially as herein described with reference to the accompanying drawings.

27. A receiver substantially as herein described with reference to the accompanying drawings.

(

28. A biasing circuit substantially as herein described with refcrcncc to the accompanying drawings.