WO1991008602A1

WO1991008602A1 - Variable frequency dye laser

Info

Publication number: WO1991008602A1
Application number: PCT/GB1990/001822
Authority: WO
Inventors: Jacqueline Staromlynska
Original assignee: The Secretary Of State For Defence In Her Britannic Majesty's Government Of The United Kingdom Of Great Britain And Northern Ireland
Priority date: 1989-11-24
Filing date: 1990-11-23
Publication date: 1991-06-13
Also published as: GB8926539D0

Abstract

The invention provides a variable frequency dye laser (1), where the tuning element within the cavity is a Fabry Perot etalon (2). The etalon cavity (12) contains a layer of nematic liquid crystal material or long pitch cholesteric material. The Fabry Perot etalon is constructed such that the liquid crystal material has a uniform tilt direction. Application of a gradually increasing voltage to the Fabry Perot etalon causes a gradual tilting of the liquid crystal director. As the liquid crystal director tilts this changes the refractive index of the liquid crystal material, and thus the selected wavelength (i.e. the wavelength allowed to traverse the dye laser cavity) changes.

Description

VARIABLE FREQUENCY DYE LASER

This invention relates to dye lasers, and more specifically to variable frequency dye laser apparatus.

A laser beam is produced by stimulated emission of photons, and provides a source of intense, coherent radiation which may be in the visible, infrared or ultraviolet regions of the spectrum. The emergent radiation, or spectral output^*, is the resultant effect of the laser cavity reflectivity characteristic on the gain curve of the lasing medium. Many lasers are capable of producing two or more wavelengths, eg, the helium-neon laser operates at wavelengths including 1.153 m, 3.391 m and 632.8 nm. The wavelengths at which the laser can operate is determined by the energy of the photons emitted.

There are many applications, such as for example spectroscopy and the study of chemical reactions, where spectral output over a significant wavelength range is required. Such a requirement can be met by a dye laser, which is capable of producing a gain curve over a wide wavelength range (eg, typically 580-640 nm for a Rhodium 6G dye laser or 420-480 nm for a carbostyril 165 laser). The active medium is usually an organic dye in a solvent, where each available atomic level is broadened into a near continuum of levels by the effects of rotation and vibration of the dye molecules and also by the effects of the solvent molecules. All dye lasers are optically pumped, which is the provision of sufficient energy for stimulation of molecules to an excited electronic level state from the ground electronic level state. The pumping is normally carried out by the use of some other type of laser radiation, although flash lamps may also be used. Pumping results in the excitation of the dye molecule from the ground state to the first excited state. This is followed by very rapid non-radiative decay processes to .lower energy levels of the first excited state. The laser transition is then from these levels to an energy level in the ground state. Since there are many such rotational/vibrational levels within both the" ground and first excited states, there are many transitions available resulting in a gain curve which is very broad.

Selection of a specific dye laser wavelength spectral output can be made within the laser cavity by using an element which is capable of introducing loss at all but the selected wavelength. The action of . changing the characteristic relied upon the element for loss is introduced at all but the selected wavelength tunes the spectral output, eg, the laser.

Varying the output of the frequency of dye lasers (or tuning) is mostly carried out by the use of device called a tuning element. At present tuning elements are mostly mechanically driven variable birefringence filters known as Lyot filters (J Evans, J Opt Soc Am 3£ 1949 p229). Such systems are susceptible to wear of mechanical parts, are slow to scan through the wavelength range and lack the ruggedness required for uses outside the laboratory. Computer controlled scanning systems are known for C0₂ TEA lasers (A Crocker et al, J Phys E JJ3 1985 p133), but although speed of scanning is increased up to about 30 Hz, they still suffer the disadvantages of other mechanically driven tuning systems. Also, the interfacing of mechanical components to electronic components can be complex.

In order to eliminate mechanically driven tuning elements with their associated disadvantages, it is possible to change the tuning element's optical characteristics by some external non-mechanical influencing means. One method is that of using a cholesteric liquid crystal tuning element (F Simoni et al, Mol Cryst Liq Cryst, 139 1986 p161). An externally applied electric field is used to induce a structure deformation in the cholesteric liquid crystal material. This results in a change in the polarisation of the incident radiation in a continuous range. The system can operate at low voltages of up to 15 Volts, but produces a small range of tuning of about only 12A. Another method is that of Umerton et al (C Umerton et al, App Opt 22 1988 p210). They also used a cholesteric liquid crystal optical device as a tuning element in a dye laser. When strain is applied to the molecular layers, through positive or negative voltages to piezo¬ electric ceramics, tuning is achieved due to the induced variation of the rotary power of each wavelength. The dye laser tuned by this method had a continuous tuning range of only 60A, which was capable of being carried out at a scanning frequency of 10Hz.

In order to achieve a greater tuning range, Telle and Lang (App Phys Lett 2_4 1974 p85) used a suitably cut and oriented electro-optical crystal of ADP within the dye laser cavity. The laser had a reported maximum tuning range of about 35 nm, but the maximum tuning rate was about 6 nm per kV applied. Thus, for a tuning range of about 35 nm a very high voltage of about 5.8 kV needs to be applied for the maximum tuning ranges. No rates of tuning or material response times were stated in the reported results. This tuning apparatus relies on high technology and expensive crystal growing, orienting and cutting facilities.

_*

According to this invention a variable frequency dye laser is provided in which the tuning element is lighter and physically smaller than the mechanical lining elements, operates at relatively low voltages of up to about 10 volts, contains less complex circuitry^' than is used for mechanical/electronic tuning elements and is relatively cheap.

According to this invention a variable frequency dye laser comprises: a dye cuvette containing a dye and solvent suitable for lasing, means for optically pumping the dye, an intracavity tuning element, a cavity between a back reflector and a partially transmitting forward reflector, characterised by the intracavity tuning element comprising a Fabry Perot etalon incorporating a layer of nematic or long pitch cholesteric liquid crystal material having a uniform tilt direction across the layer thickness.

A nematic material in a layer forms itself into very thin planes with all the molecules in that plane lying in a common direction. To be technically correct it is the director which lies in a common direction; the director being the localised average direction of molecules moving about under Brownian motion. A cholesteric material has a twisted spiral structure with the director changing direction like the steps of a spiral staircase. Pure cholesteric material has a very small pitch spiral. Adding small amounts of cholesteric material to a nematic material results in a spiral structure whose cholesteric pitch is proportional to the amount of cholesteric material added; this mixture is often called a long pitch cholesteric material.

Nematic or long pitch cholesteric liquid crystal materials exhibit very large non-linear refractive index coefficients and it is possible to induce large changes in the refractive index at low voltages and/or low optical input powers (J Staromlynska et al, Optics Comms 6 _ 1987, p229).

Typical nematic liquid crystal materials which may be used within the Fabry Perot etalon are 4-Cyano-4'-pentylbiphenyl and 4-Cyano-1-(trans- 4-propylcyclohexyl)-benzene, which may be purchased from BDH Chemicals, Poole, Dorset.

A long pitch cholesteric is typically achieved by. the addition of between about 0.1% and 3% by weight of cholesteric material to a nematic material. Suitable cholesteric material include S(+)4-Cyano- 4'-(2-Methylbutyl) biphenyl, also available from BDH Chemicals. Typically cholesteric material may be added in sufficient quantity to provide a natural pitch of multiples of π, eg π, 2τt, 3rc, ... etc. In order to achieve a uniform tilt direction across the layer thickness, the liquid crystal director (or average molecular direction) can be surface aligned. This is normally achieved by the use of an alignment layer on the walls of the liquid crystal cells. Two possible methods of surface alignment are the use of a polyimide layer which is softly rubbed unidirectionally, or oblique evaporation of silicon dioxide.

Liquid crystals can display severe distortion of the beam spatial profile due to non-linear self-focusing effects. This adds the constraint that non-linear effects must be kept to the minimum. Since the non-linearity of a liquid crystal is dominated by thermal effects, it is important to minimise any absorption in the mirrors and the material itself. Absorption in the etalon mirrors can be minimised by the use of dielectric multilayer stacks composed of appropriate material, ie materials with low absorption in range of interest. Such materials for the visible are MgF2 and Titanium dioxide.

Nematic and long pitch cholesteric liquids crystal materials have a response time which is of the order of milliseconds (J Staromlynska and R A Clay, Optics Comms 64 1987 p474).

In order that the invention by more fully understood the apparatus will be described, by way of example only, with reference to the accompanying figures in which:

Figure 1 is a schematic representation of a dye laser utilising a Fabry Perot etalon as a tuning element.

Figure 2 is a schematic representation of the construction of a Fabry Perot etalon tuning element.

Figure 3(a), (b) , and (c) are general wavelength characteristics of a Fabry Perot etalon and illustrate these characteristics for a refractive index change of 0.02 for etalon cavity lengths of 2, 5 and 10 microns respectively.

Figure 4(a), (b) , (c) are general wavelength characteristics as for Figure 3, except that they represent a refractive change of 0.2 in each case.

Figure 5(a), (b) and (c) show change of refractive index approximately required to map superimposed adjacent fringe maxima for etalon cavity lengths of 2, 5 and 10 microns respectively.

Figure 6 demonstrates the resolution of an etalon cavity of 2 microns length.

Figure 1 is a schematic representation of a standard dye laser 1 containing intracavity Fabry Perot etalon device 2 which acts as a tuning element. A dye cuvette 3 contains mixture 4 of dye (eg Rhodamine-G for the green/red part of the visible spectrum with a spectral output between 560 nm and 640 nm) and solvent. The mixture is pumped by Xenon flash lamps 5. Stimulated emission is resonated between a back mirror 6 of reflectivity 99% and an output coupler (or front reflector) 7 which has a reflectivity of 85%. Determination of allowed wavelength of resonance between mirror 6 and the output coupler is carried out by the Fabry Perot etalon 2 with a frequency of 1 KHz. Where radiation emitted from the dye during lasing is unpolarised, then it is advisable to place a polariser 9 between the cuvette 3 and the Fabry Perot etalon 2.

The structure of the Fabry Perot etalon 2 may be seen in Figure 2 to comprise a liquid crystal cell with integral mirrors forming an etalon 2. Two cell walls 10 are separated by a spacer ring 11 to define the etalon cavity 12. The spacer ring 11 is normally of thickness between about 5 and 15 microns. The cavity -12 is filled with a layer of nematic liquid crystal material or long pitch cholesteric material.

Prior to assembly the cell walls are constructed. A coating 13 of about 100 nm Indium Tin Oxide (ITO) on glass slides 14 provides a rugged, optically transparent electrode. Etalon mirrors 15 are partially reflecting, and formed by vacuum evaporation of BaF₂ and ZnSe, each of λ/4 thickness. Each mirror consists of two layers of each of BaF2 and ZnSe (16(a)). (b) and 17(a), (b) respectively). A liquid crystal alignment layer 18 is provided on each mirror. Such alignment may be formed by spinning a layer of polyimide (about 20- 40 nm thick) on to the mirror surface. The polyimide is then unidirectionally rubbed with cotton wool or a soft tissue. This forms microscratches in the polyimide surface which align the liquid crystal director along the rubbing directions D₁ , D₂ and at an angle of about 2° to the surface. Alternatively surface alignment may be provided by the known technique of oblique evaporation of silicon dioxide. This technique is described in GB1454296. Depending on the angle, or angles, of evaporation, a surface tilt angle of 5° to 30° or more may be produced.

The cell is assembled with alignment directions anti-parallel, D₁ and D₂ as seen in figure 2. This ensures a uniform tilt direction of the liquid crystal. The cell walls 10 and spacer ring 11 are held together with epoxy resin glue, with two small holes left at the corners. These holes allow the liquid crystal material to be drawn by capillary action into the cell; the holes are then sealed.

Application of a gradually increasing voltage causes a gradual tilting of the liquid crystal director. The tilting starts at the centre of the layer and continues outwards with increasing voltage. The limit occurs when substantially all the molecules align along the direction of the applied electric field, ie, perpendicular to the plane of the layer. Where Fabry Perot etalon 2 has a 10 micron thick cavity 12 filled with liquid crystal material K15, and BaF₂/ZnSe mirrors (16(a), (b)/17(a), (b)), then the measured free spectral range is 13 nm. The voltage required to scan through one free spectral range is 1.3 volts giving a frequency scan measurement of 7nm. The frequency scan range achieved by applying voltage to this Fabry Perot etalon can be seen in table 1 whilst table 2 gives the refractive index change with changing voltage.

TABLE 2

Table 2 shows that a large refractive index change is.achieved with the above described Fabry Perot etalon. Thus, a reduction in cavity thickness to 2 microns would increase the scanning range to approximately 35 nm.

The transmission of a Fabry Perot etalon is given by the equation

_Ψ . exp (-od) ( -R)² 1 ₍₁₎

(1 - R exp(-αd))"^ 1 + F sin^ (6/2)

where α is the absorption of the etalon cavity, d is the etalon cavity length, R is the etalon mirror reflectivity (assumed to be the same for back and front mirrors), F is the cavity finesse and δ is the etalon cavity round trip phase change.

A maximum etalon transmission occurs when the following condition is satisfied:

δ/2 = mπ where m = 0, 1, 2, (2)

This may also be expressed as

2 n d = mλ (3)

where n is the refractive index of the etalon material and λ is the wavelength of the incident radiation.

Equation (3) shows that a change in the refractive index of the etalon material will result in a shift in the wavelength of the maximum transmission. Scanning of the transmission maximum through a spectral range may Δλ be achieved by changing the refractive index of the etalon cavity material through an electro-optic mechanism. Use of a Fabry Perot (in the transmission mode) as a laser intra-cavity element will allow scanning of the peak gain of a dye laser system. The change in refractive index, Δn, required to shift the transmission maximum of mode m by an amount Δλ is given by the equation: Δn/n = Δλ/λ ( 4 )

There is no etalon cavity length -dependence in this equation. Equation (4) is an over simplification as it ignores the possible • scanning through of higher orders at the initial wavelength. For a particular wavelength, λ, the number of the modes scanned through for n, is

Δ = 2dΔn/λ (5)

This illustrates that shifting the mth order fringe by an amount from the initial wavelength may, (depending on the etalon cavity parameters), result in the transmission at the initial wavelength going through several maxima. The larger the etalon cavity length, d, the greater the number of modes scanned through. Figures 3(a), (b) and (c) illustrate some of the general features of the wavelength response of a Fabry Perot etalon and the effect of changing the refractive index. In the case given, the front and back mirror reflectivities are 32%, the absorption of the etalon material is 10cm^~' and the etalon cavity lengths are 2, 5 and 10 microns respectively giving etalon cavity finesses of 2.757, 2.741 and 2.715. In each figure the response is shown for two values of refractive index, 1.5 and 1.52.

There are four main points to be noted from the figures:

1. Within a given wavelength range, the number of transmission maxima increases with etalon cavity length.

2. The shorter the etalon cavity, the poorer the chromatic resolving power of the etalon.

3. Figures 3(b) and (c) show that in agreement with equation (4), the shift in wavelength mode m from the initial wavelength, λ, remains the same regardless of etalon cavity length (eg, approximately 9 nm from λ = 655 nm) .

4. Examination of figure 3(c) reveals that although the m = 46 mode is shifted by the predicted 9 nm for a Δn of 0.02, this Δn also causes the a = 47 to shift such that a transmission maximum occurs at only 5 nm from the initial wavelength. This is an important point since it. illustrates that an injudicious choice of etalon cavity parameters could result in the spectral range of the laser being clipped.

Figures 4(a) (b) and (c) highlight this even more dramatically. In this case the change in refractive index change is 0.2, and hence from equation (4) Δλ should be 90 nm. However, for the 2 micron etalon cavity case the minimum wavelength separation between two transmission maxima is only 14 nm, for the 5 micron etalon cavity it is 2 nm, and finally for the 10 micron etalon cavity it is 0 nm.

A definitive way of describing the wavelength response of an etalon when the refractive index of the etalon material is changed is that the transmission maxima will map back onto themselves whenever the refractive index is great enough to induce a Δλ equal to the free spectral range. The free spectral range is defined as the range between two transmission peaks, and is given by the expression:

(Δλ)_fsr = λ²/2 n d (6)

The free spectral range can thus be seen to be inversely proportional to the etalon cavity length. The change in refractive index which is needed to map a maxima onto another maxima is given by:

Δn = λ/2 d (7) Calculations using equations (6) and (7) predict that the Δn required for this mapping to occur for a 2, 5 and 10 microns etalon cavity is. 0.16, 0.012 and 0.032 respectively. Figure 5(a), (b) and (c) confirm these values.

The theoretical points discussed above define the main specifications for the use of a Fabry Perot etalon as a tuning element for a dye ^• laser. They may be summarised as:

1. The lower limit on the necessary refractive index change is given by Δn = nΔλ/λ.

2. The maximum achievable spectral range is half -the free spectral range of the cavity, is Δλ_maχ = λ^/4 n d. for a given etalon cavity material of refractive index n, this relationship defines the maximum acceptable etalon cavity length.

3. The specific value of etalon cavity length is determined by the starting value of the wavelength scan, ie, d = mλ/2 n. This along with the criterion set out in point 2 above defines the mode numbers of the etalon fringes. If the cavity length, as determined by the free spectral range, means that the transmission maximum falls at too large a wavelength, this may be overcome by merely biassing the etalon cavity material with a small voltage which will shift the maximum down in wavelength to the desired value.

4. The final specification to be set is the cavity finesse. The minimum acceptable finesse will be influenced by the desired spectral resolution and the etalon cavity length. Where the etalon cavity is reduced, then the necessary finesse also becomes reduced. Figures 3(a) and 6 compare two etalons with a finesse of 2.757 and 14.88, and highlight the influence on the spectral resolution of the etalon. The spectral resolution is proportional to finesse: Res = F 2 n d/λ (8)

where Res = spectral resolution or the chromatic resolving power. The finesse of an etalon cavity is thus defined as:

_F 4Rexp (-αd) ₍₉₎

(1 - R exp (-αd))²

An additional advantage of a Fabry-Perot etalon geometry is that the Finnesse (and hence the spectral resolution) may be easily tailored by changing the mirror reflectivity, R.

Applying a voltage across a nematic or long pitch cholesteric liquid crystal material results in a change in the^• material's refractive index. This is due to an induced change of tilt of the liquid crystal molecules or more correctly the director. The relationship between applied voltage and resultant refractive index change is not linear.

Where molecular orientation is substantially perpendicular to the incident laser radiation and an applied voltage across the liquid crystal molecules results in a change of orientation of angle A away from the perpendicular, then the change in n with angle A is given by:

where n₀ is the ordinary refractive index of the liquid crystal material, n_e is the extraordinary refractive index of the liquid crystal material and n(A) is the effective refractive index in the perpendicular direction. At low voltages of about 2-5 volts the molecules in the centre of the etalon cavity reorientate, at higher voltages (about 8-1OV) the molecules close to the etalon cavity walls also re-align.

Claims

1. Variable frequency dye laser comprising: a dye cuvette containing a dye and solvent suitable for lasing, means for optically pumping the dye, an intracavity tuning element, a cavity between a back reflector and a partially transmitting forward reflector, characterised by the intracavity tuning element comprising a

Fabry Perot etalon incorporating a layer of nematic or long pitch cholesteric liquid crystal material having a uniform tilt direction across the layer thickness.

2. A variable frequency dye laser according to Claim 1 where the nematic liquid crystal is selected from a list including K15, E64, 3/5/7 PCH and 16522.

3. A variable frequency dye laser according to Claim 1 where the long pitch cholesteric liquid crystal material consists of a mixture of between 0.1% and 3% by weight of a cholesteric liquid crystal material added to a nematic liquid crystal material.

4. A variable frequency dye laser according to Claim 1 where the layer of nematic or long pitch cholesteric liquid crystal material is in antiparallel alignment

5. A variable frequency dye laser according to Claim 1 here the uniform tilt direction across the layer thickness is achieved by the use of an alignment layer.

6. A variable frequency dye laser according to Claim 4 where the alignment layer is polyimide which is rubbed unidirectionally.

A variable frequency dye laser according to Claim 4 where the alignment layer is produced by vacuum evaporation.

8. A variable frequency dye laser according to Claim 6 where the alignment layer produced by vacuum evaporation is silicon dioxide.

9. A variable frequency dye laser according to Claim 1 where the layer of long pitch cholesteric liquid crystal has a twist of π.

10. A variable frequency dye laser according to Claim 1 where the layer of long pitch cholesteric liquid crystal has a twist of 2π.