Intra-cavity second harmonic generation for UV laser emission

Abstract

Broadly speaking, embodiments of the present techniques provide a laser device. In particular, the present techniques provide a laser device comprising a laser cavity, a non-linear crystal for generating converted radiation from radiation at a fundamental frequency, and a first optical element configured to separate the radiation at the fundamental frequency from the converted radiation. The present techniques also provide a laser device comprising a laser cavity and an optical diode which is within the laser cavity and which is configured to ensure that radiation at a fundamental frequency is resonating in a single direction around the laser cavity. Advantageously, the present techniques provide means of intra-cavity frequency conversion which minimises damage to the optical components within the laser cavity.

Description

LASERField

[0001] The present techniques relate to a laser device, including for example an intracavity Second Harmonic Generation (SHG) laser and / or an ultraviolet laser.Background

[0002] The laser comprises an arrangement of mirrors or other optical elements that forms a cavity resonator for light waves. A cavity resonator may also be termed an optical cavity, a resonating cavity or an optical resonator. An example of an intra-cavity SHG laser device is described in US8498316. The device comprises a laser cavity formed by the first and the second laser cavity end reflectors. An active medium and at least one non-linear crystal for the second harmonic generation are provided within the laser cavity. At least one of the laser cavity end reflectors comprises an interferometric layout providing spectrally selective reflection for the radiation about the laser fundamental frequency so that the laser cavity restricts the spectrum of the laser radiation to a single longitudinal mode.

[0003] Intracavity frequency conversion to generate the second harmonic can provide high efficiency and lower complexity than employing external frequency conversion. In such devices, two wavelengths typically circulate within the cavity. In wavelength ranges below 380nm (i.e. in the ultraviolet range) laser radiation can damage optical coatings and optical materials. The applicant has recognised the need for minimising interaction between the ultraviolet light and the coatings and materials within the laser cavity (i.e. intracavity).

[0004] One example of a cavity resonator is a ring resonator, e.g. a bow-tie ring resonator comprising four mirrors defining the closed path in which the beam of light circulates. In a ring resonator, light can circulate in two different directions and thus an optical diode may be employed intracavity to ensure that the light circulates in only one direction. An optical diode may also be termed an optical isolator. The applicant has recognised the need for an intracavity optical isolator for use in lasers creating output wavelengths in ranges below 650nm, including lasers operating in the ultraviolet spectrum, e.g. below 400nm.Summary

[0005] According to the present invention there is provided an apparatus as set forth in the appended claims. Other features of the invention will be apparent from the dependent claims, and the description which follows.

[0006] According to a first aspect of the present techniques, there is provided a laser device for intra-cavity frequency conversion of a radiation with a laser fundamental frequency (ro) to a converted radiation with a second harmonic frequency (2a>) of the fundamental frequency (ro). The device (which may also be termed a laser) comprises a laser cavity formed by a plurality of reflectors which are reflective for radiation at the laser fundamental frequency; a non-linear crystal in the laser cavity, wherein the non-linear crystal is for transmitting radiation at the fundamental frequency and generating the converted radiation with the second harmonic frequency wherein the second harmonic wavelength is ultraviolet; and a first optical element which is positioned to receive the radiation at the fundamental frequency and the converted radiation with the second harmonic frequency from the non-linear crystal and which is configured to separate the radiation at the fundamental frequency from the converted radiation with the second harmonic frequency. The fundamental frequency may be the frequency which is generated by the laser crystal and may also be termed a first frequency.

[0007] The term “for” used herein means “suitable for” or as a word to specify a functional limitation. For example, “a laser device for intra-cavity frequency conversion” means “a laser device suitable for intra-cavity frequency conversion”. Similarly, a “plurality of reflectors which are reflective for radiation at the laser fundamental frequency” means a “plurality of reflectors whose function is to reflect radiation at the laser fundamental frequency”.

[0008] The term “ultraviolet” used herein means a wavelength or frequency which is in the ultraviolet range of the electromagnetic spectrum. That is, wavelengths typically in the range 10 nm to 400 nm and frequencies typically in the range 750 THz to 30,000 THz.

[0009] The term “optical element” used herein means an optic or optical component formed of any suitable optical material. For example, an optical element may be a crystal formed of fused-silica, ultraviolet fused-silica, calcium fluoride or magnesium fluoride.

[0010] The term “non-linear crystal” used herein means an optic or optical component form of any suitable optically non-linear material. For example, a non-linear crystal may be a crystal formed of lithium triborate, beta barium borate or caesium lithium borate.

[0011] The term “optical contact” used herein means a contact of mated surfaces of different optical components which results in an interface between the mated surfaces that minimizes or mitigates losses of any radiation propagating across the interface that are due to reflection, scattering and / or distortion (i.e. a seamless transition) It will be appreciated that any combination of optical elements, non-linear crystals, and / or any other optical component, may be in optical contact. When two or more components are joined together via optical contact, a monolithic structure is formed. Optical contact may be achieved in multiple different ways, such as the examples described in R. Paschotta, "Optical Contact - an encyclopedia article",RP Photonics Encyclopaedia (2023), the contents of which are incorporated herein by reference in their entirety. For example, optical contact may be achieved by a glueless process where two closely conformal surfaces are joined together, for example by intermolecular forces, such as Van der Waals forces, which occur when precisely polished surfaces are brought together. In an alternative example, optical contact may be achieved through bonding using, for example, adhesive-free chemical bonding, hydroxide catalysis or adhesive supports, which provides a seamless - reflection-free - transmission of light on the mating interface. When using hydroxide catalysis or similar techniques, there will be a thin layer of silicate or a bonding / adhesive substance between the pairs of adjacent surfaces that may result in some reflective losses. However, the connection still enables a seamless transition across the interface, and which nonetheless minimises reflective / scattering / distortive losses. It will be appreciated that optical contact may be achieved using any suitable technique.

[0012] Radiation in the ultraviolet spectrum has larger photon energy and when the radiation passes through some materials, the materials can heat up as they absorb the radiation. This can damage the material, particularly any coatings. By separating the radiation at the fundamental frequency from the converted radiation with the second harmonic frequency, it is possible to minimise the number of components through which the radiation at the second harmonic frequency is transmitted and thus reduce damage to components within the laser. The second harmonic wavelength may be below 400nm, more specifically below 380nm and above 100nm.

[0013] The second harmonic wavelength may be generated using any suitable technique. For example, the second harmonic generation may be achieved using Type 1 critical phase matching, which means the fundamental wavelength is orthogonally polarised to the second harmonic wavelength. Alternatively, non-critical Type 1 phase matching can be used. However, non-critical Type 1 typically requires the crystal needs to be heated and hence is more complex and less often used than Type 1 critical phase matching.

[0014] The first optical element may be spaced from the non-linear crystal whereby there is an air-filled gap between the two components. Alternatively, the first optical element may be in optical contact with (i.e. form an optical contact bond) the non-linear crystal to form a monolithic component. The optical contact is typically achieved by a glueless process where two closely conformal surfaces are joined together, for example by intermolecular forces such as Van der Waals forces which occur when precisely polished surfaces are brought together. Alternatively, optical contact may be achieved through bonding using, for example, adhesive- free chemical bonding, hydroxide catalysis or adhesive supports, which provides a seamless - reflection-free - transmission of light on the mating interface. When using hydroxide catalysisor similar techniques, there will be a thin layer of silicate between the pairs of adjacent surfaces that may result in some reflective losses. However, the connection still enables a seamless transition across the interface. In both arrangements, the first optical element may have a first surface on which radiation from the non-linear crystal enters and a second surface through which radiation at the fundamental frequency exits. In other words, the second surface may be considered to be the opposed or opposite surface to the first surface of the first optical element.

[0015] Both the first and second surfaces of the first optical element may be aligned at Brewster’s angle for the radiation at the fundamental frequency. In this case, it will be appreciated that the first and second surfaces may be parallel to each other. Brewster’s angle is also termed the polarisation angle and is the angle of incidence at which light at a particular frequency with a particular polarisation (in this case p-polarisation) is perfectly transmitted through the crystal with no reflection. In other words, radiation at the fundamental frequency which is p-polarised passes through the first surface of the optical element, through the optical element and out through the second surface without reflection. The Brewster’s angle will depend on the wavelength of the radiation and also the nature of the interface through which the radiation passes.

[0016] When there is space between the first optical element and the non-linear crystal, the first optical element may have a coating on a first surface, the first surface being the surface on which radiation from the non-linear crystal is incident, wherein the coating is anti-reflective for radiation at the fundamental frequency and reflective for the converted radiation with the second harmonic frequency. The second surface may be uncoated. In this way, further exposure of the coatings to ultraviolet radiation may be further reduced. The first optical element may be generally planar in shape. The first optical element may thus be termed a pick-off mirror because it is highly reflective for the second harmonic light and transmitting for the fundamental frequency light.

[0017] When the first optical element is in optical contact with the non-linear crystal, all or part of the first surface of the optical element may contact the non-linear crystal to form an optical contact. In both arrangements, the degree of optical contact is such that the first optical element receives the radiation at the fundamental frequency and the converted radiation with the second harmonic frequency. When forming an optical contact, the adjacent surfaces (which are typically polished) are in good contact so that radiation can pass through and minimize reflective losses at the optical contact (i.e. the interface between the optical element and the non-linear crystal). The first optical element may have a coating on the second surface, wherein the coating is anti-reflective for radiation at the fundamental frequency and reflective for the converted radiation with the second harmonic frequency.

[0018] As set out above, the radiation at the fundamental frequency enters the optical element through its first surface which may be in optical contact with the non-linear crystal and exits the optical element through a second surface. The converted radiation with the second harmonic frequency also enters through the first surface but is reflected from at least part of the second surface so that the converted radiation exits the optical element through a different or separate location to the location that the radiation at the fundamental frequency exits the optical element. In other words, the radiation at the fundamental frequency and the converted radiation exit the optical elements in non-coaxial directions. For example, the converted radiation may be reflected at the second surface so that the converted radiation does not exit the first optical element through the second surface and instead exits the first optical element through a different face to the second surface. Alternatively, the converted radiation may be reflected at the second surface and then reflected back towards the second surface, for example from a part of the first surface with a reflective coating, so that the converted radiation exits through the second surface in a different location to that which the radiation at the fundamental frequency exits the optical element.

[0019] The converted radiation with the second harmonic frequency may be reflected to exit the first optical element through a side face. This side face may be aligned at Brewster’s angle for the radiation at the second harmonic frequency. Such an arrangement may be particularly appropriate when the whole of the first surface of the optical element is in optical contact with the non-linear crystal because otherwise the converted radiation with the second harmonic frequency would pass back through the non-linear crystal but it will be appreciated that it can be used in other arrangements. When the whole of the first surface of the optical element is in optical contact with the non-linear crystal, the first optical element may be considered to be an end cap for the non-linear crystal. The first optical element may be wedge shaped.

[0020] When only a part of the first surface of the optical element optically contacts the nonlinear crystal, the converted radiation with the second harmonic frequency may be reflected back through the first surface of the optical element without entering the non-linear crystal. In other words, a first portion of the first surface of the optical element may optically contact the non-linear crystal and a second portion of the first surface of the optical element may extend beyond the non-linear crystal, and the converted radiation with the second harmonic frequency may be reflected to exit the first optical element through the second portion of the first surface. In other words, the converted radiation may exit the first optical element through a different part of the same surface through which the converted radiation entered the first optical element. In this arrangement, the first optical element may also be considered to be an end cap for the non-linear crystal, but the second portion of the first optical element is not in contact with the non-linear crystal. The first optical element may also be wedge shaped.

[0021] The coating which is applied to either the first surface of the first optical element and / or the second surface of the first optical element may be highly reflective for the second harmonic frequency, in other words over 98% of the radiation may be reflected. Similarly, an anti- reflective coating may be defined as a coating which means that less than 0.2% is reflected. The anti-reflective nature of the coating may be more important because the fundamental frequency is amplified within the laser device and thus any drop in laser power is highly nonlinear with any loss associated with the fundamental frequency. The radiation at the second harmonic frequency (i.e. the ultraviolet radiation) is not amplified within the laser device and thus any loss of second harmonic frequency is not amplified.

[0022] The laser device may further comprise a second optical element which is positioned to receive the reflected radiation with the second harmonic frequency from the first optical element and which is configured to reflect the radiation with the second harmonic frequency towards an output of the laser device. The second optical element may thus be considered to be a steering mirror. The second optical elements may be generally planar in shape.

[0023] The laser device may comprise other standard components within the laser cavity, for example any of an active medium (also termed a gain medium), tuning means for selecting the fundamental frequency within the cavity, for example an etalon, and an optical oven may be used to efficiently control the temperature of optical element, such as the etalon. There may also be a feedback mechanism to adjust a length of the optical path, e.g. using a piezoelectric element or an electro-optic modulator.

[0024] The plurality of reflectors may form a ring resonator, more specifically a bow-tie ring resonator with four reflectors. The device may further comprise an optical diode which is configured to ensure that the radiation at the fundamental frequency is resonating with unidirectional oscillation (i.e. in a single direction) around the ring resonator. The optical diode may comprise a magneto-optical element having a first surface and an opposed second surface, a waveplate having a first face and an opposed second face, wherein the first face of the waveplate faces the second surface of the magneto-optical element; and a magnet which is configured to generate a magnetic field within the magneto-optical element. When exposed to the magnetic field, the magneto-optical element is configured to rotate a polarisation direction of radiation at the fundamental frequency which is incident on the first surface of the magneto-optical element by a first rotation and rotate a polarisation direction of radiation at the fundamental frequency which is incident on the second surface of the magneto-optical element by a mirrored first rotation, wherein the mirrored first rotation is different from the first rotation; and wherein the waveplate is configured to rotate a polarisation direction of the radiation at the fundamental frequency which is transmitted in either direction through the waveplate by asecond rotation, wherein the second rotation is in the opposite direction to the first rotation and is equal in size to the first rotation. The mirrored first rotation is equal and opposite to the first rotation.

[0025] Thus, when radiation at the fundamental frequency is incident on the first surface of the magneto-optical element, its polarisation direction will be rotated by the first rotation as it travels through the magneto-optical element to exit at the second surface of the magnetooptical element. In other words, the first rotation corresponds to a rotation of the polarisation direction of the radiation at the fundamental frequency which exits the magneto-optical element at the opposed second surface with respect to the polarisation direction of the radiation at the fundamental frequency which is incident on the first surface. Here, the first rotation may also be understood as a continuous rotation of the polarisation direction which occurs within the magneto-optical element between the first surface and the opposed-second surface. The first face of the waveplate faces and / or is in contact with the second surface of the magneto-optical element; and thus radiation exiting the magneto-optical element through the second surface is incident on the first face of the waveplate and its polarisation direction will be rotated by the second rotation which is equal and opposite to the first rotation. In a similar way to the first rotation, the second rotation corresponds to a rotation of the polarisation direction of the radiation at the fundamental frequency which exits the waveplate at the opposed second face with respect to the polarisation direction of the radiation at the fundamental frequency which is incident on the first face. Here, the second rotation may also be understood as a continuous rotation of the polarisation direction which occurs within the waveplate between the first face and the opposed-second face. In other words, the rotation applied by the waveplate cancels out the rotation applied by the magneto-optical element.

[0026] There are thus two rotations applied to the radiation, a first applied by the magnetooptical element (the first rotation) and a subsequent rotation applied by the waveplate (the second rotation). When radiation is travelling in the opposite direction, the radiation is first incident on the wave plate and is rotated by the second rotation. That is, in contrast to the magneto-optical element, the rotation of the polarisation direction applied by the waveplate does not depend on the direction of transmission of the radiation through the waveplate. That is, the second rotation also corresponds to a rotation of the polarisation direction of the radiation at the fundamental frequency which exits the waveplate at the first face with respect to the polarisation direction of the radiation at the fundamental frequency which is incident on the opposed second face. The rotated radiation is then incident on the second surface of the magneto-optical element and is subject to the mirrored first rotation. In a similar way to the first rotation, the mirrored first rotation corresponds to a rotation of the polarisation direction of the radiation at the fundamental frequency which exits the magneto-optical element at the first surface with respect to the polarisation direction of the radiation at the fundamental frequencywhich is incident on the second surface. Thus again, thus two rotations applied to the radiation but in this arrangement the first rotation is applied by the waveplate and the subsequent rotation by the magneto-optical element.

[0027] The optical diode may be used independently from the second harmonic generation and thus according to a second aspect, there is a laser device comprising: a laser cavity in the form of a ring resonator formed by four reflectors which are reflective for radiation at a laser fundamental frequency; and an optical diode which is configured to ensure that the radiation at the fundamental frequency is resonating in a single direction around the ring resonator. The optical diode may have the features described above. The further details described below may be applied to the combination of the optical diode within an SHG laser device or in the separate case.

[0028] Unidirectional oscillation is achieved by the two rotations which are separately applied by the magneto-optical element and the waveplate. As explained above, radiation at the fundamental frequency which is at the correct polarisation and which passes first through the magneto-optical element is rotated along its length by approximately 4 to 5 degrees and then rotated by the waveplate in the opposite direction by the same amount so that the waveplate rotation compensates for the rotation of the magneto-optical element. Effectively, radiation in this direction is unchanged by the optical diode. By contrast, radiation which passes first through the waveplate will first be rotated by the second rotation (i.e. 4 to 5 degrees) and then further rotated by the mirrored first rotation (i.e. by 4 to 5 degrees in the same direction as the second rotation) as is passes through the magneto-optical elements. These rotations do not balance each other out and thus the radiation which passes in the wrong direction, i.e. through the waveplate first followed by the magneto-optical element, has a different polarisation which results in a reflective loss at each component within the laser, e.g. the reflectors, the non-linear crystal and / or the optical elements when used.

[0029] Each of the first, mirrored first and second rotations may be a rotation of approximately 3 to 5 degrees, more specifically 4 to 5 degrees. Thus a relatively small amount of rotation is required. The cumulative rotation when radiation is travelling in the wrong direction may be 8 to 10 degrees and this may result in a reflective loss on one round trip around the device which is effective to prevent the laser from oscillating in this direction.

[0030] The magneto-optical element is an optical element which exhibits birefringent properties when exposed to a magnetic field. A suitable example is a Faraday crystal. Suitable materials include potassium terbium fluoride (KTF) or terbium gallium garnet (TGG). The magnet may be a permanent magnet or an electro-magnet with a radial magnetic field distribution, i.e. ring magnet.

[0031] The waveplate may be made from crystalline quartz. Crystalline quartz has known birefringent and optical activity which can be exploited separately in the different arrangements to achieve the desired rotation.

[0032] The magneto-optical element may optically contact or be bonded to the waveplate or may be spaced apart from the waveplate. When the magneto-optical element is bonded to or optically contacts the waveplate, there is a monolithic optical component. As described above, the optical contact may be formed by precise polishing of the surfaces. As an alternative to or in addition to an optical contact, the connection may be formed by using a bonding substance applied to one or both of the adjacent surfaces. For example, by using hydroxide catalysis bonding or a silicate solution, it is possible to form a thin silicate layer between the components (i.e. true covalent bonds) which connects the surfaces.

[0033] Each of the first and second surfaces of the magneto-optical element may be aligned at Brewster’s angle for the radiation at the fundamental frequency (which as explained above is p-polarised). In a similar way to the first and second surfaces of the first optical element above, it will be appreciated that the first and second surfaces of the magneto-optical element may be parallel to each when aligned at Brewster’s angle. This ensures that radiation at the fundamental frequency is transmitted through the magneto-optical element without reflection. Alternatively, each of the first and second surfaces may be planar, i.e. each of the first and second surfaces of the magneto-optical element may be aligned perpendicular to an axis of the magneto-optical element. In both cases, the axis of the magneto-optical element may be aligned with the intended direction of radiation travel around the ring resonator. In the first arrangement, radiation at the fundamental frequency is refracted when incident upon the first and second surfaces. However, the radiation is not reflected at the first and second surfaces because of the alignment of the surfaces at Brewster’s angle. In the second arrangement, to ensure the correct refraction through the magneto-optical element, each of the first and second surfaces are coated whereby the refractive index of the coated magneto-optical element is such that radiation at the fundamental frequency which is parallel to the axis of the magnetooptical element is not refracted as it passes through the magneto-optical element. The coating may also be anti-reflective to the fundamental frequency to encourage transmission through the magneto-optical element. Thus, there are two possible options for the magneto-optical element: no coating and wedged surfaces or flat-flat with coatings.

[0034] Similarly, there are two arrangements for the waveplate. In one arrangement, each of the first and second faces of the waveplate may be aligned perpendicular to an axis of the waveplate (i.e. flat-flat). In other words, the waveplate may be cut so that an optic axis (c-axis) is in the plane of the waveplate surface. The waveplate may be supported within a mount andthe second rotation may be achieved using birefringent properties of the material of the waveplate. Thus, the waveplate can be rotated within the mount to a fixed setting so that the rotation applied by the waveplate cancels out the rotation applied by the magneto-optical element when the radiation is travelling in the correct direction. In a second arrangement, each face of the waveplate may be cut at Brewster’s angle. The waveplate has a thickness which achieves the second rotation. The waveplate may be any suitable shape, e.g. circular or rectangular.

[0035] The waveplate may be coated with an anti-reflective coating whereby the refractive index of the coated waveplate is such that radiation at the fundamental frequency which is orthogonal to each of the first and second faces is not refracted as it passes through the magneto-optical element. The coating is particularly applied when a waveplate which is cut so that the optic axis (c-axis) is in the plane of the waveplate surface is used.

[0036] In all arrangements described above, the fundamental frequency may have a wavelength of 650nm or below. The fundamental frequency may thus be in the green, red, blue or possibly ultraviolet regions.Brief Description of the Drawings

[0037] For a better understanding of the invention, and to show how embodiments of the same may be carried into effect, reference will now be made, by way of example only, to the accompanying diagrammatic drawings in which:

[0038] Fig. 1 is a drawing showing the components of the laser and the beam path within the optical cavity;

[0039] Fig. 2a shows more detail of one second harmonic generator which is used in the laser of Fig. 1 ;

[0040] Fig. 2b is a schematic sketch showing details of an alternative second harmonic generator which is used in the laser of Fig. 1 ;

[0041] Fig. 2c is a schematic sketch showing details of another alternative second harmonic generator which is used in the laser of Fig. 1 ;

[0042] Figs. 3 and 4 are two perspective views of an optical diode which may be used in the laser of Fig. 1 to provide unidirectional oscillation in the direction of the dashed line;

[0043] Fig. 5a is a half waveplate which is used in the optical diode of Fig. 3;

[0044] Fig. 5b is a side view of the half waveplate of Fig. 5a showing radiation passing therethrough;

[0045] Figs. 6a and 6b show components of a mount which is used to support the half waveplate of Fig. 5a;

[0046] Fig. 7a is a perspective view of a first Faraday crystal which may be used in the optical diode of Fig.3;

[0047] Fig. 7b is a perspective view of a mount for the Faraday crystal of Fig. 7a;

[0048] Figs. 7c and 7d are alternative Faraday crystals which may be used;

[0049] Figs. 8a to 8d show beam paths through three different arrangements of Faraday crystal and waveplate which are used in the optical diode of Fig. 3;

[0050] Fig. 9a plots refractive index against wavelength for light having p-plane polarisation or s-plane polarisation respectively as it is incident on an KTF crystal; and

[0051] Fig. 9b plots reflection against incident angle for light having p-plane polarisation or s- plane polarisation respectively.Detailed Description of the Drawings

[0052] Figure 1 shows a laser comprising a laser cavity in the form of a bow-tie ring resonator. The ring resonator is formed by four reflectors 104a, 104b, 104c, 104d which are highly reflective for radiation (or light, the terms are used interchangeably) at the laser fundamental frequency ro. Due to a nearly 100% reflectivity of the cavity mirrors about the laser fundamental frequency, the cavity resonates and enhances the circulating laser power at frequency ro. Each reflector 104a, 104b, 104c, 104d comprises a mirror held in a mount. An active (laser gain) medium 106 is held in a mount within the cavity. The active medium 106 can be any suitable solid state laser material, for instance, a Neodymium doped laser crystal, or a Praseodymium doped crystal. Laser light is output from the laser device through output 150.

[0053] In this example, the laser is an intracavity Second Harmonic Generation (SHG) laser. Thus, the laser further comprises a SHG nonlinear crystal 140 which generates second harmonic light, i.e. light at frequency 2a. As explained in more detail below, some light at the laser fundamental frequency is converted to light at the second harmonic frequency and thus radiation at the second harmonic frequency may be termed converted radiation. The generated (i.e. converted) light is directed to a first optical element 142 which separates the fundamental wavelength from the second harmonic wavelength by reflecting the secondharmonic wavelength (in this example ultraviolet light). In other words, the fundamental wavelength radiation is spatially separated from the second harmonic wavelength radiation. The spatial separation is achieved by a reflection of the second harmonic wavelength radiation, wherein the fundamental wavelength radiation and the second harmonic wavelength radiation follow spatially separated paths as a result of the reflection. As shown in Fig. 1 , the first optical element 142 is positioned within the laser cavity to receive light at the laser fundamental frequency. In other words, the first optical element 142 is positioned along the beam path of the laser cavity. The reflected light is directed towards a second optical element 144 which then directs the second harmonic light towards the output 150. In other words, power at the second harmonic frequency 2a is output from the laser.

[0054] There may be an etalon 130 within the beam path of the ring resonator. The etalon 130 may be an optical cavity through which only optical waves which are in resonance with the optical cavity can pass. In other words, the etalon acts to tune the laser to the desired fundamental frequency. The temperature of the etalon 130 is carefully controlled and thus the etalon 130 may be termed an etalon oven.

[0055] As illustrated by the dotted line, there may be leakage of radiation 204 at the fundamental wavelength. This leaked radiation 204 may be reflected to an optional feedback mechanism using a reflector 110, particularly when using a high finesse broadband coated etalon 130 in which the fundamental frequency has been locked using fringe locking. The leaked radiation 204 is then split using a beam splitter 136 to generate two separate signals, each of which are received separately at a photodiode 134. A first signal is an intensity reference signal, and a second signal is spectrally selected by a second etalon 132. These signals may be used to measure the light leakage. One of the reflectors, in this example the first reflector 104a may be attached to a piezoelectric element 214 which can be used to change the optical length within the laser cavity based on the measured light leakage. As an alternative to using a piezoelectric element, a birefringent optical element (such as an electro optic modulator) may be used.

[0056] Figure 1 shows the beam path within the ring resonator. A ring resonator has a travelling wave which requires unidirectional operation. Accordingly, an optical diode 120 is preferably located within the cavity (i.e. intracavity). The optical diode 120 prevents the laser from lasing in one direction but allows it to laser in the other. The optical diode is described in more detail below.

[0057] Light from the active medium 106 is incident on a first reflector 104a which reflects the incident light at the fundamental frequency a through the optical diode 120 towards a second reflector 104b. The second reflector reflects the light at the fundamental frequency a so that it passes through the SHG crystal 140 to generate the second harmonic wavelength. Twowavelengths are now circulating intracavity: the fundamental wavelength 200 which is shown in red and the second harmonic wavelength 202 which is shown in blue. The second harmonic wavelength 202 is discussed in more detail below. The fundamental wavelength 200 is incident on the third reflector 104c and is reflected towards the fourth reflector 104d. As the wave travels between the third and fourth reflectors, the wave passes through the etalon oven 130. The fourth reflector 104d then reflects the wave back towards the first reflector 104a. The first, second, third and fourth reflectors thus form a bow-tie resonator in which the beam path for the light at the fundamental frequency resembles a bowtie.

[0058] It will be appreciated that a bow-tie resonator can be used without the SHG crystal. Similarly, it will be appreciated that this is just one suitable arrangement of resonator. In this arrangement, both components of the optical diode are placed in a diagonal arm of the bowtie resonator and the SHG crystal is in one of the horizontal arms (horizontal as viewed in Figure 1). However, the components of the optical diode may be separated into different arms, for example the Faraday crystal may be in the diagonal arm (as in Figure 1) but the waveplate may be after the second reflector in the horizontal arm (e.g. in the arm in which the SHG crystal is shown in Figure 1 . In general, we aim to position the crystal close to the middle of an arm to minimise beam divergence through the crystal. When separating the components into two different arms, there can only be a reflector between the waveplate and the Faraday crystal. In other words, no other components of the laser device such as the SHG crystal may be between the optical diode components.

[0059] Figure 2a shows more detail of the components of one arrangement for generating the second harmonic. The SHG crystal 140 is supported in a mount 340. The SHG crystal 140 is a non-linear crystal which is generally wedge shaped. The SHG crystal 140 may be formed of any suitable non-linear optical material, for example, lithium triborate (LBO). Alternatively, the SHG crystal 140 may be formed of beta barium borate (BBO) or caesium lithium borate (CLBO). The front face 352 of the crystal 140 is cut at Brewster’s angle for the fundamental frequency light which is p-polarised. Brewster’s angle is also termed the polarisation angle and is the angle of incidence at which light with a particular polarisation (in this case p-polarisation) is perfectly transmitted through the crystal with no reflection. Brewster’s angle 6Bis defined as:6B= arctan where is the refractive index of the initial medium through which the light propagates (i.e. the air within the cavity) and n2 is the refractive index of the crystal. Since the refractive index for a given medium changes depending on the wavelength of light, Brewster’s angle will also vary with wavelength. The rear face of the crystal may also be cut at Brewster’s angle to minimise any loss of light at the fundamental wavelength.

[0060] The SHG crystal also generates the second harmonic light, i.e. light having twice the frequency and half the wavelength). Second harmonic generation is a non-linear optical process in which two photons with the same frequency interact with a non-linear material and generate a new photon with twice the energy. In one example, the second harmonic light is generated as type 1. In other words, two photons having ordinary polarization with respect to the crystal will combine to form one photon with double the frequency and extraordinary polarization. In other words, the second harmonic has s-polarisation while the fundamental has the opposite polarization.

[0061] As shown in Figure 2a, the fundamental wavelength 200 which is shown in red and the second harmonic wavelength 202 which is shown in blue both exit the SHG crystal 140. Both wavelengths are incident on a first optical element 142 which is in the form of a rectangular cuboid, and which is supported in a support or mount 342. The first optical element 142 has a first or front face on which the fundamental and second harmonic wavelengths are incident. This first face is set at Brewster’s angle for the p-polarised fundamental frequency so that the light at the fundamental frequency passes through the optical element 142 without reflective losses and out through a second or rear face. The first face has a coating which is highly reflective at the second harmonic wavelength so that the light at the second harmonic wavelength is reflected. The coating is anti-reflective for the fundamental wavelength so that it can pass through the optical element. There is no coating on the second face, but the second face may also be set at Brewster’s angle to minimise any reflective loss of light at the fundamental wavelength as it travels through the optical element. Returning to Figure 1 , the light at the fundamental frequency which travels through the first optical element is incident on one of the four reflectors of the ring resonator. The first optical element 142 is typically formed of any suitable optical material, such as fused-silica or UV fused-silica (UVFS). Alternatively, the first optical element may be formed of calcium fluoride (CaF2) or magnesium fluoride (MgF2). It will be appreciated that any suitable optical material may be used to form the first optical element.

[0062] As shown in Figure 2a, the light at the second harmonic wavelength is reflected onto a second optical element 144. The second optical element 144 may have the same shape as the first optical element 142 and may be mounted in a similar support 344. The second optical element 144 reflects the light at the second harmonic wavelength towards the output. The second optical element 144 may be termed a steering mirror for the second harmonic wavelength. At this point, theoretically there is no fundamental light, because the first optical element 142 has transmitted it all to the reflector. Similarly to the first optical element 142 above, the second optical element 144 may be formed of any suitable optical material, such as UVFS, CaF2or MgF2.

[0063] When the fundamental wavelength is in the red spectrum (i.e. between 630nm to 700nm) or in the lower end of the infrared spectrum (i.e. between 700nm and 800nm), the second harmonic wavelength is typically in the ultraviolet spectrum (i.e. below 400nm). It will be appreciated that as the fundamental wavelength is shortened, e.g. below 650nm, the wavelength of the second harmonic also shortens. The shorter wavelengths such as those in the ultraviolet spectrum have larger photon energy and when they pass through some materials can heat up the materials as they are absorbed by them. In particular, this can destroy coatings. The second harmonic wavelength is separated from the fundamental wavelength using the first optical element. Both the first and second optical elements only have a single coating and thus heating of coatings within the laser is minimised. It is also noted that, when using type 1 second harmonic generation, the fundamental and the second harmonic wavelengths are orthogonal in polarisation, and this helps with the beam separation.

[0064] As an alternative to using the first and second optical elements, an output coupler which is reflective for the fundamental and transmitting for the second harmonic wavelength could be used. However, in this case, the second harmonic wavelength will pass through the output coupler and may be absorbed by the material of the output coupler, including, any coatings on one or both of the surfaces. This may lead to heating of the material and / or the coatings.

[0065] Figure 2b shows an alternative arrangement for generating the second harmonic. In this arrangement, the SHG crystal 240 and the first optical element 242 form a monolithic structure. As in the previous arrangement, the SHG crystal 240 is a non-linear crystal which is generally wedge shaped. The SHG crystal 240 and the first optical element 242 may similarly be formed of any of the materials described above in relation to Figure 2a. The front face 252 of the crystal 240 is cut at Brewster’s angle for the fundamental frequency light which is p- polarised. The rear face 254 of the crystal 240 is also cut at Brewster’s angle to minimise any reflective loss of light at the fundamental wavelength. That is, the front face 252 and rear face 254 may be parallel to each other when cut at Brewster’s angle. In both cases, the Brewster’s angle is calculated for the fundamental frequency. For the first face 252, the interface is air to the material of the crystal and for the second face 254, the interface is the material of the crystal to the material of the optical element.

[0066] As shown in Figure 2b, the fundamental wavelength which is shown in red enters the SHG crystal 240 through the first face 252 and exits the SHG crystal through the second face 254. The light is refracted upon entering the SHG crystal 240 at the first face. It will be appreciated that refraction normally occurs when the radiation passes from one component to another. When the refractive indices of the materials are similar, the angle is very small and may be approximated by a straight line. The SHG crystal has a higher refractive index than the adjacent component, so the refracted angle is visible in Figure 2b. No reflection or reflectiveloss occurs because the first face 252 is cut at Brewster’s angle for p-polarised radiation at the fundamental wavelength. The SHG crystal 240 also generates the second harmonic wavelength and this also exits the SHG crystal through the second face 254. No reflective losses occur at the second face because it is also cut at Brewster’s angle. A first optical element 242 has a first or front face 244 on which the fundamental and second harmonic wavelengths are incident.

[0067] In this example, the first face 244 and the second face 254 of the SHG crystal are optically contacting. Furthermore, the entire first face (surface) 244 of the first optical element 242 optically contacts the non-linear crystal 240 to form a monolithic component. A connection or bond is formed between the SHG crystal 240 and the first optical element 242 using any suitable mechanism. For example, by a glueless optical contact such as intermolecular forces acting across the first 244 and second 254 faces or by using any suitable bonding processes, such as hydroxide catalysis techniques, which provide a seamless transmission of light across the first 244 and second 254 faces. Optically contacting the SHG crystal 240 to the first optical element 242 ensures that reflective losses at the interface between the SHG crystal 240 and the first optical element 242 are minimized. In other words, any light propagating across the optically contacting interface of the crystal 240 and the optical element 242 is not affected by the interface.

[0068] When precisely polished faces are brought together, attractive Van der Waals forces induce a bond between the faces. The precision may be defined by standard specifications. For example, the specification typically includes one of more of a surface quality greater than scratch / dig (S / D) = 10 / 5, surface flatness of less than lambda / 10 (633 nm / 10, i.e. less than 63nm across the rms of the surface) and surface roughness of less than 5 Armstrong. The forces are typically sufficient to maintain the connection between each of the adjacent faces. As an alternative to or in addition to an optical contact, the connection may be formed by using a bonding substance applied to one or both of the adjacent faces. For example, by using hydroxide catalysis bonding or a silicate solution, it is possible to form a thin silicate layer between the components (i.e. true covalent bonds) which connects the surfaces.

[0069] The first face 244 is set at Brewster’s angle for the p-polarised fundamental frequency so that the light at the fundamental frequency is not reflected and passes through the optical element 242 and out through a second or rear face 246 which is also set at Brewster’s angle. In this case, the first face 244 and 246 may be parallel to each other when set (i.e. aligned) at Brewster’s angle. The Brewster’s angle for the first face 244 is calculated for the interface between the material of the non-linear crystal to the material of the optical element which may be for example fused silica. The Brewster’s angle for the second face 246 is calculated for the interface between the material of the optical element and air.

[0070] The light at the second harmonic frequency also passes through the optical element. The second face 246 of the optical element has a coating which is applied on its outer surface (i.e. on the surface which contacts the air). The coating is anti-reflective for the fundamental wavelength so that it passes through the second face 246 without any change in direction. The coating is highly reflective at the second harmonic wavelength (i.e. over 98% is reflected) so that the light at the second harmonic wavelength is reflected and does not exit the optical element through the second face 246 but instead exits the optical element through a third or side face 248. This side face 248 is cut at Brewster’s angle for the second harmonic wavelength based on an interface between the material of the optical element and air.

[0071] Figure 2c shows another alternative arrangement for generating the second harmonic. The features relating to the SHG crystal 240 described in relation to Figure 2b above apply equally to the example shown in Figure 2c, and are not repeated for conciseness. In addition, light at the second harmonic frequency also passes through the optical element and the second face 246 of the optical element has a coating which is applied on its outer surface. However, rather than the entire first face 244 of the first optical element 242 optically contacting the non-linear crystal 240 as shown in Figure 2b, only a first portion 244-1 of the first face 244 of the optical element optically contacts the SHG crystal 240.

[0072] In this arrangement, the light at the second harmonic frequency is reflected and exits the optical element through a second portion 244-2 of the first surface which extends beyond and does not contact the SHG crystal. In other words, the second harmonic wavelength exits the optical element through a different part of the same surface through which it entered the optical element. The first face 244 and the second face 246 are also set at their respective Brewster angles, as described above in relation to Figure 2b. However, unlike the arrangement in Figure 2b, the first face 244 of the first optical element is not coterminous with corresponding face of the SHG crystal 240. It will be appreciated that the arrangement of Figure 2c allows the light at the second harmonic frequency to be reflected back through the first surface of the optical element, but it is also possible to adapt the arrangement to reflect the second harmonic frequency through the side face as described in Figure 2b. It is also possible to adapt the arrangement to reflect the second harmonic frequency back through the second surface in a different location to the location that the light at the fundamental frequency exits the optical element.

[0073] Returning to Figure 1 , the light at the fundamental frequency which travels through the first optical element is incident on one of the four reflectors of the ring resonator. The light at the second harmonic wavelength may then be directed towards the exit, e.g. using a second optical element or other suitable mechanism.

[0074] In this arrangement, as in the first arrangement, the second harmonic wavelength is separated from the fundamental wavelength using the first optical element. By contrast, in this example, the second harmonic wavelength passes through the first optical element. This reduces the potential interaction of the ultraviolet light with air (oxygen in particular) that can catalyse some of the degradation processes.

[0075] As shown in Figure 1 , an optical diode may be used in a ring resonator to ensure that the light at the fundamental frequency is resonating in a single direction around the cavity. Figures 3 and 4 show more detail for the optical diode which comprises two key components: a Faraday crystal (also termed a magneto-optical element and the terms can be used interchangeably) mounted in a first holder 410 and a half waveplate (also termed a waveplate or plate rotator and the terms are used interchangeably) mounted in a second holder 420. As shown in Figure 1 , the light at the fundamental frequency which is going in the correct direction around the cavity passes first through the Faraday crystal and then through the waveplate. A waveplate may also be termed a retarder and is an optical device which alters the polarisation state of a light wave travelling through it. More specifically a half-wave plate shifts the polarisation direction of linearly polarised light.

[0076] An example of a half waveplate 610 is shown in Figures 5a and 5b. The half waveplate 610 is birefringent, and thus its refractive index depends on the polarization and propagation direction of light which is incident on the half waveplate 610. In other words, the half waveplate may be considered to be an optically anisotropic material or birefractive. The waveplate 610 is also a non-reciprocal optical element. In other words, light propagating in both directions through the waveplate 610 has the same polarisation rotation. The waveplate may be made of any suitable material, for example crystalline quartz or calcite and has a thickness t.

[0077] The waveplate can be designed using either of the following properties, which are described in detail for Crystalline Quartz:

[0078] Plate rotator - Optical activity: In this case the quartz crystal is cut so that the beam propagation is parallel to the crystal optical axis. Optical activity occurs naturally in some materials such as quartz, where it is possible to define a rotatory power (p), described in “The dispersion, birefringence and optical activity of quartz” by Radhakrishnan published in January 1947, which specified the rotation in degrees per mm of propagation of travelling through the material. As an example, for the 640 nm wavelength this is approximately 18.251 deg / mm, and as such for a 4-degree rotation, the thickness t of the quartz should be 0.2191 mm.4.948 4.617 2.311 4.617P~ A2- 0.014161+A2- 0.011236 " A2- 0.00974 " (A2- 0.0195)2" °'1905

[0079] Waveplate - Birefringence properties: In this case the quartz crystal is cut so that the beam propagating is perpendicular to the crystal optical axis. Light polarised along the optic axis (fast axis) travels faster than light polarised perpendicular to the optic axis (slow axis) and, in this way, the birefringence property of quartz gives rise to a relative phase retardation between components of polarisation along the fast and slow axes. A half waveplate provides a phase retardation of half the wavelength (or pi retardation), hence a linear polarisation input remains linear on exiting the waveplate. The thickness d of the waveplate is proportional to the birefringence at a specific wavelength , and M is an integer specifying the order number.

[0080] As such a 5thorder half waveplate made of quartz at 640 nm has a thickness of 0.389 mm. In order to achieve a rotation of 4 degrees, the waveplate is rotated, so that the incoming linear polarisation is separated by 2 degrees relative to the waveplate’s fast axis (or half of the rotation required), noting that the propagation direction remains orthogonal to the optical axis.

[0081] Returning to Figures 3 and 4, the half waveplate is mounted in a generally circular mount 430 which sits within a corresponding generally circular aperture on a support 432. As shown in Figures 6a and 6b, the mount comprises an outer ring 630 and a central body 632 having an outer rim 634 which sits within and is secured to the outer ring 630. The central body 632 also has a second part 636 which projects from the outer rim 632. As shown in Figure 3, the outer ring 630 protrudes from a rear face of the support 432. As shown in Figure 4, the second part 636 of the central body sits within the aperture on the support 432.

[0082] The mount 430 may be rotatable so that the birefringent properties of the material of the waveplate can be used to create retardation. Alternatively, the waveplate may be cut as a plate rotator relying on the optical activity concept to apply rotation.

[0083] A Faraday crystal is a crystal which exhibits birefringent properties when exposed to a magnetic field. Thus, the Faraday crystal may also be termed a magneto-optical element and the terms may be used interchangeably. Examples of suitable materials of crystal include potassium terbium fluoride (KTF) and terbium gallium garnet (TGG). The Faraday crystal is a reciprocal element and thus any rotation which is applied to radiation passing through the Faraday crystal is dependent on the magnetic field and propagation direction of the radiation. Thus, radiation which is incident on a first face of the Faraday crystal and then is transmitted through the Faraday crystal may be subject to a first rotation which is different to a second rotation which is applied to radiation which is incident on a second face of and then transmitted through the Faraday crystal.

[0084] In a first arrangement shown in Figure 7a, the Faraday crystal 710 is cylindrical in shape and has planar end faces 712, 714 which are at right angles to the long axis 720 of the crystal. The Faraday crystal is mounted in a crystal mount 750 which is shown in more detail in Figure 7b. The crystal mount 750 has a cylindrical (or rod-shaped) cavity to receive the crystal and a circular flange 752 around which a ring magnet 450 is attached. The ring magnet 450 is used to expose the crystal to a magnetic field.

[0085] In the second arrangement of Figure 7c, the Faraday crystal 730 is generally cylindrical in shape but has angled end faces 732, 734 which are both at Brewster’s angle, and the end faces 732, 734 may be parallel to each other. In both Figures 7a and 7b, the length of the crystal is similar to that of the diameter (e.g. a ratio of 4:3 or 3.5:3). In Figure 7d, the Faraday crystal 740 is an elongate version of the crystal in Figure 7c. Thus, in Figure 7d, the end faces are also angled at Brewster’s angle, but the length of the crystal is approximately twice as long as the diameter (e.g. a ratio of 8:3).

[0086] There are various arrangements of the optical diode depending on the shape of the crystal and / or the orientation of the wave plate as well as any coatings applied to some or all surfaces of the crystal and / or the wave plate. A first arrangement as shown in Figure 8a in which a crystal 810 such as that shown in Figures 7c and 7d is used. The crystal 810 is thus cut or subject to a wedged polish so that the first surface is at Brewster’s angle and is uncoated. The waveplate 820 is also uncoated. The waveplate 820 is a generally flat disc of a certain thickness and is set at Brewster’s angle. Light 812 at the frequency which corresponds to the Brewster’s angle is not reflected by the crystal. As explained above, refraction occurs because the materials are different and the refraction angle is exaggerated to show the radiation travelling through the crystal in a direction which is not parallel to the axis of the crystal. The direction of the light 814 exiting the crystal is parallel to the direction of light 812 which entered the crystal. Similarly, the light 814 which is incident on the waveplate 820 is refracted through the waveplate and exits in a parallel direction 816. In this arrangement, both the Faraday and waveplate are at Brewster cut, so the beam propagates without a reflective loss, the Brewster angle might be different because the materials are different. There is air between the two components.

[0087] In Figure 8b, the same crystal 810 is used as in Figure 8a. However, in this arrangement, the waveplate 830 is coated with an anti-reflective coating and uses the birefringent properties of the material of the waveplate 830. In this arrangement, the waveplate 830 can be set perpendicular to the direction of the light 814 which is exiting the crystal. When the light path is orthogonal to the incidence plane of the waveplate, the light passes through the waveplate without changing its direction 836. There is no refraction of the light through the waveplate. In this arrangement, the Faraday crystal is Brewster cut and the waveplate iscoated with antireflective coating on both faces at the fundamental wavelength to minimise reflective losses.

[0088] In Figure 8c, the same waveplate 830 is used as in Figure 8b. However, in this arrangement, the crystal 840 is also coated and has end faces which are perpendicular to its main axis as shown in Figure 7a. Light 812 at the frequency which corresponds to the refractive index of the coated crystal and which has a light path orthogonal to the incidence plane of the crystal 830 passes through the crystal without any change in direction 844. This unchanged light is then incident on the waveplate and passes through the waveplate without changing its direction 836. There is no refraction of the light through either the waveplate or the crystal. In this arrangement, both the Faraday crystal and the waveplate are antireflective coated on both surfaces.

[0089] As an alternative to the arrangement in which the crystal and waveplate are spaced apart as shown in Figure 8a, it is possible to create a monolithic optical diode by bonding the exit surface of the crystal 810 to the entry surface of the waveplate 820 (or vice versa depending on the direction of travel). This arrangement is shown in Figure 8d. Each surface which is bonded together should be uncoated. When two optical surfaces are optically contacted in this way, they become one material and thus there is no reflective loss between. Thus, such an arrangement can be beneficial. Light 812 at the frequency which corresponds to the refractive index of the combined crystal and waveplate passes through without any change in direction 846. In this arrangement, both the Faraday crystal and waveplate are wedged so that beam incident at any surface experiences Brewster reflection. This potentially requires the waveplate to have non-parallel faces, i.e. refraction between the Faraday crystal and the waveplate is different to refraction between the waveplate and air. Thus, the second face of the Faraday crystal is optically contacted with the first face of waveplate, i.e. refraction is gradual and there are no refractive losses.

[0090] The options shown in Figures 8a to 8d, show two different configurations for the waveplate. In a first configuration shown in Figures 8b and 8c, the waveplate is cut so that the optic axis (c-axis) is in the plane of the waveplate surface. In this arrangement, a generally round waveplate (such as shown in Figure 5a) can be utilised and the waveplate can be rotated to match the retardation of the Faraday crystal. Thus, the waveplate is held in a round mount as shown in Figure 6a. In this arrangement, the crystalline birefringence properties are being used to achieve the desired rotation (or retardation).

[0091] In a second configuration shown in Figures 8a and 8d, the waveplate may be termed a Brewster cut waveplate. In this case, the waveplate may be generally rectangular and is cut at a certain angle, for example as illustrated in Figure 5b. The thickness is calculated to achieve the rotation which matches the rotation applied by the Faraday crystal. Using a rectangularwaveplate facilitates alignment but is noted that a rectangular waveplate can be rotated as described above. Moreover, as shown in the arrangement in Figure 8d, the waveplate can be optically contacted to the Faraday crystal and the sides are parallel. In this case, there is no rotation of the waveplate, so there is less freedom / flexibility but a more compact design. In this arrangement, we are using the optical activity property of the waveplate, and these are well known for quartz as described in “The dispersion, birefringence and optical activity of quartz” by Radhakrishnan published in January 1947. Thus, the two different arrangements can exploit two different properties of the waveplate.

[0092] For each of the options shown in Figures 8a to 8d, it is important that the material of the crystal is cut differently. The crystal has different optical axes sometimes referred to as an a-cut and a c-cut. For each arrangement, the refractive index of the material at the given wavelength is needed. KTF crystals are known for the telecommunication industry but are normally used in applications in the near infrared and infrared ranges (e.g. from 980 to 1550nm). Similarly, TGG has been previously used in ranges covering 650nm to 1500nm. There was thus no data available on this crystal below these wavelengths.

[0093] Figure 9a plots the refractive index against wavelength for a crystal formed from KTF. Laser light at three different wavelengths was passed through and the refractive index was measured. With three points, it is possible to estimate the refractive index change over all wavelengths. The change was plotted for both p-plane and s-plane polarisation so that both the change for both the ordinary and extraordinary refractive indices over the visible wavelength spectrum can be understood.

[0094] Figure 9b plots the Fresnel curves for the selected material. The Fresnel curves plot the reflection at different angles of incidence. The reflection is measured at three different wavelengths for the p-plane and s-plane polarisation for each wavelength to give six plots in total. Each plot gives one refractive index estimate which is plotted in Figure 9b. The refractive index number is calculated or estimated from the line fit.

[0095] The data from Figures 9a and 9b can be used to calculate the Brewster’s angle and also to design an anti-reflective coating. The anti-reflective coating may be less than 0.02% reflective. Knowing the reflection and incident intensity, it is possible to measure the absorption of the crystal from the difference between the incident power and the reflection loss. For example, the absorption for a crystal made from TGG was measured at about 2% over 10mm at a wavelength of 650nm. The absorption for a crystal made from KGF was measured at less than 0.1% over 10mm at a wavelength of 650nm.

[0096] As mentioned above, the crystal exhibits birefringence properties when exposed to a magnetic field. It is known to use a KTF or TGG crystal as an optical isolator or optical togetherwith a polariser, for example as described in US10,120,213. The KTF crystal applies a 45 degree rotation on the polarised light which is incident on the crystal in the forward direction. In the reverse direction, back-reflected light is again polarised and is rotated again by 45 degrees as it passes through the crystal. In this way, a full 90 degrees is applied by the Faraday rotator and the light can be fully attenuated.

[0097] Considering the beam path shown in Figure 2, in the present case, when an electric field is applied to the Faraday crystal, the polarised light which is incident on the crystal from the first reflector 104a is rotated by a first rotation of a few degrees, e.g. by 3 to 5 degrees. The rotated light is then incident on the waveplate which then counters the rotation by applying a second rotation so that the light passes through the optical diode without any rotation. If light is passing through the optical diode 120 in the wrong direction, e.g. from the second reflector 104b towards the first reflector 104a, light is rotated by the second rotation by the waveplate by a few degrees (e.g. by 3 to 5 degrees) and is further rotated by a mirrored first rotation which is a few degrees (e.g. by 3 to 5 degrees) by the Faraday crystal. In other words, there is an overall rotation of approximately 8 to 10 degrees. There are several optical surfaces within the cavity including reflectors, specific coatings on optical elements and optical elements cut at or set at Brewster’s angle. This relatively small rotation results in a reflective loss over one round trip which is effective to restrict the laser from oscillating in this direction because the laser is a light amplifier and thus small reflective losses amplify into huge losses on the output power. In other words, a little bit of loss is enough to suppress bi-directional lasing and for the optical diode to perform its function of generating unidirectional oscillation in the ring resonator.

[0098] Various combinations of optional features have been described herein, and it will be appreciated that described features may be combined in any suitable combination. In particular, the features of any one example embodiment may be combined with features of any other embodiment, as appropriate, except where such combinations are mutually exclusive. Throughout this specification, the term “comprising” or “comprises” means including the component(s) specified but not to the exclusion of the presence of others.

[0099] Attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

[0100] All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternativefeatures serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

[0101] The invention is not restricted to the details of the foregoing embodiment(s). The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed. Although a few preferred embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes and modifications might be made without departing from the scope of the invention, as defined in the appended claims.

Claims

CLAIMS1 . A laser device for intra-cavity frequency conversion of radiation with a laser fundamental frequency (ro) to a converted radiation with a second harmonic frequency (2a>) of the fundamental frequency (ro), the device comprising: a laser cavity formed by a plurality of reflectors which are reflective for radiation at the laser fundamental frequency; a non-linear crystal which is within the laser cavity and which is for transmitting the laser fundamental frequency and for generating the converted radiation with the second harmonic frequency wherein the second harmonic wavelength is ultraviolet; and a first optical element which is positioned to receive the radiation at the fundamental frequency and the converted radiation with the second harmonic frequency from the non-linear crystal and which is configured to separate the radiation at the fundamental frequency from the converted radiation with the second harmonic frequency.

2. The laser device of claim 1 , wherein the first optical element is spaced apart from the non-linear crystal and wherein the first optical element has a coating on a first surface, the first surface being the surface on which radiation from the non-linear crystal is incident, wherein the coating is anti-reflective for radiation at the fundamental frequency and reflective for the converted radiation with the second harmonic frequency.

3. The laser device of claim 1 , wherein a first surface of the first optical element optically contacts the non-linear crystal, the first surface being the surface on which radiation from the non-linear crystal is incident.

4. The laser device of any one of the preceding claims, wherein the first optical element has a second surface through which the radiation at the fundamental frequency exits the first optical element and wherein the second surface is uncoated.

5. The laser device of claim 4, wherein the first optical element has a coating on at least part of the second surface, wherein the coating is anti-reflective for radiation at the fundamental frequency and reflective for the converted radiation with the second harmonic frequency.6 The laser device of claim 5, wherein the converted radiation with the second harmonic frequency and the radiation at the fundamental frequency exit the first optical element at different locations.

7. The laser device of claim 6, wherein the converted radiation with the second harmonic frequency exits the first optical element through a side face which is aligned at Brewster’s angle for the radiation at the second harmonic frequency.

8. The laser device of claim 6, wherein a first portion of the first surface of the optical element optically contacts the non-linear crystal and a second portion of the first surface of the optical element extends beyond the non-linear crystal and the converted radiation with the second harmonic frequency exits the first optical element through the second portion of the first surface.

9. The laser device of any one of claims 4 to 8, wherein both the first and second surfaces of the first optical element are aligned at Brewster’s angle for the radiation at the fundamental frequency.

10. The laser device of any one of the preceding claims wherein the second harmonic wavelength is below 380nm.11 . The laser device of any one of the preceding claims, wherein the plurality of reflectors form a ring resonator, and the device further comprises an optical diode which is configured to ensure that the radiation at the fundamental frequency is resonating in a single direction around the ring resonator, wherein the optical diode comprises a magneto-optical element having a first surface and an opposed second surface, a waveplate having a first face and an opposed second face, wherein the first face of the waveplate faces the second surface of the magneto-optical element; and a magnet which is configured to generate a magnetic field within the magneto-optical element whereby the magneto-optical element is configured to, rotate a polarisation direction of radiation at the fundamental frequency which is incident on the first surface of the magneto-optical element by a first rotation and rotate a polarisation direction of radiation at the fundamental frequency which is incident on the second surface of the magneto-optical element by a mirrored first rotation, wherein the mirrored first rotation is in a different direction to the first rotation; and wherein the waveplate is configured to rotate a polarisation direction of the radiation at the fundamental frequency which is transmitted in either direction through the waveplate by a second rotation, wherein the second rotation is in the opposite direction to the first rotation and is equal in size to the first rotation.

12. A laser device comprising:a laser cavity in the form of a ring resonator formed by four reflectors which are reflective for radiation at a laser fundamental frequency; and an optical diode which is within the laser cavity and which is configured to ensure that the radiation at the fundamental frequency is resonating in a single direction around the ring resonator, wherein the optical diode comprises a magneto-optical element having a first surface and an opposed second surface, a waveplate having a first face and an opposed second face, wherein the first face of the waveplate faces the second surface of the magneto-optical element; and a magnet which is configured to generate a magnetic field within the magneto-optical element wherein the magneto-optical element is configured to: rotate a polarisation direction of radiation at the fundamental frequency which is incident on the first surface of the magneto-optical element by a first rotation and rotate a polarisation direction of radiation at the fundamental frequency which is incident on the second surface of the magneto-optical element by a mirrored first rotation, wherein the mirrored first rotation is in a different direction to the first rotation; and wherein the waveplate is configured to rotate a polarisation direction of the radiation at the fundamental frequency which is transmitted in either direction through the waveplate by a second rotation, wherein the second rotation is in the opposite direction to the first rotation and is equal in size to the first rotation.

13. A laser device of claim 11 or claim 12, wherein each of the first, mirrored first and second rotations is a rotation of approximately 4 to 5 degrees.

14. A laser device of any one of claims 11 to 13, wherein the magneto-optical element is a Faraday crystal made from potassium terbium fluoride (KTF).

15. A laser device of any one of claims 11 to 14, wherein the waveplate is made from crystalline quartz.

16. A laser device of any one of claims 11 to 15, wherein the magneto-optical element is bonded to or optically contacts the waveplate.

17. A laser device of any one of claims 11 to 16, wherein each of the first and second surfaces of the magneto-optical element are aligned at Brewster’s angle for the radiation at the fundamental frequency.

18. A laser device of any one of claims 11 to 16, wherein each of the first and second surfaces of the magneto-optical element are aligned perpendicular to an axis of the magneto-optical element and each of the first and second surfaces are coated with an anti-reflective coating whereby the refractive index of the coated magneto-optical element is such that radiation at the fundamental frequency which is parallel to the axis of the magneto-optical element is not refracted as it passes through the magneto-optical element.

19. A laser device of claim 17 or claim 18, wherein each of the first and second faces of the waveplate are aligned perpendicular to an axis of the waveplate.

20. A laser device of claim 19, wherein the waveplate is rotatable within a mount whereby the waveplate is configured to rotate a polarisation direction of the radiation at the fundamental frequency by the second rotation21. A laser device of claim 19 or claim 20, wherein the waveplate is coated with an anti- reflective coating whereby the refractive index of the coated waveplate is such that radiation at the fundamental frequency which is orthogonal to each of the first and second faces is not refracted as it passes through the magneto-optical element.

22. A laser device of claim 11 to 18, wherein each of the first and second faces of the waveplate are aligned at Brewster’s angle for the radiation at the fundamental frequency.

23. A laser device according to any one of the preceding claims, wherein the fundamental frequency has a wavelength which is at or below 650nm.

Images