WO2001063248A1

WO2001063248A1 - Fluorescence measurement apparatus and method

Info

Publication number: WO2001063248A1
Application number: PCT/AU2001/000181
Authority: WO
Inventors: Michael Ron Hammer
Original assignee: Varian Australia Pty Ltd
Priority date: 2000-02-22
Filing date: 2001-02-22
Publication date: 2001-08-30
Also published as: AUPQ578300A0

Abstract

A sample cell (2) for containing a fluid sample (26) for spectrophotometric fluorescence analysis comprises a small diameter thin walled tube (22) which is mounted or housed to provide a surrounding medium (24) that has a reflective index less than the refractive index of the fluid sample (26) so that an excitation beam of light (28) will be totally internally reflected (34) to traverse the length of the tube (22) to interact with the fluid sample. Fluorescence light (36) emitted by the fluid sample (26) passes through the tube 22 for detection. The invention provides for increased interaction between excitation light and a very small sample volume thus increasing sensitivity. It also provides for physical separation of the excitation light (28, 34) from the emitted fluorescence light (36) making it easier to measure the emitted light.

Description

FLUORESCENCE MEASUREMENT APPARATUS AND METHOD

Technical Field

The present invention relates to a sample cell for containing a fluid sample for spectrophotometric fluorescence analysis and to a method and apparatus for measuring fluorescence of a fluid sample.

Background

In fluorescence spectrophotometry instruments, excitation light of an appropriate wavelength is passed through a sample to cause the sample to fluoresce. Typically the total fluoresced light which is emitted by the sample is only a small fraction of the excitation light intensity and furthermore it is emitted in all directions, which in practice means that only a fraction can be collected. This means that the performance of fluorescence spectrophotometry instruments is usually light level limited.

The fluorescence light emitted can be increased by using a brighter excitation light source but this is not always possible. The fluorescence light emitted however also depends on the degree of interaction between excitation light and sample. Thus increasing this interaction can also increase the fluoresced light emitted, thereby conferring an advantage in overall instrument performance. Such interaction can be improved by increasing the path length the light travels in the sample. In practice, available sample volumes are usually quite limited. Thus increasing the path length without increasing the sample volume implies reducing the sample diameter through which the light must travel.

The need to be able to freely select the wavelength of excitation light in general purpose fluorometers dictates the use of a non coherent excitation source such as a hot filament or arc discharge, rather than a coherent source such as a laser. The laws of geometric optics limit the amount of light that can be coupled from such a non-coherent source into a long thin sample. The light collected from the source can be focussed down to an image which will fit within the sample diameter at the focal point but this is at the expense of a greater divergence angle. Thus, for a long thin sample geometry, as one moves away from the focal point, a larger and larger fraction of the excitation light passes out of the sample volume and is lost. Conversely the angle of divergence can be made smaller but only at the expense of making the image size bigger again causing excitation light to miss the sample. The option of configuring the available sample into a shorter cell of larger diameter would reduce the interaction due to the shorter path length of light through the sample. Whichever approach is adopted, the laws of geometric optics limit the available interaction and this problem becomes progressively worse as the available sample volume is reduced.

Collecting the fluoresced light also presents a problem particularly in the case of wavelength resolved fluorescence because of very low light intensity.

Wavelength resolved fluorescence involves passing the collected fluorescence light through a monochromator or polychromator before being detected so that the fluorescence signal can be measured as a function of the wavelength at which the emission occurs. In such a case, the amount of fluorescence light ultimately collected is limited by the acceptance aperture of the wavelength resolver. This aperture is typically rectangular with a substantial aspect ratio.

Ratios of 10:1 (ratio of width to height or vice versa) or even more are quite possible. Light collection is best optimised if the shape of the emission source

(that is, a fluorescing sample in the present application) substantially matches this acceptance aperture. The situation again favours the use of a sample cell whose length is substantially greater than the diameter, that is, a long thin cell.

Another problem associated with measuring the fluorescence light is achieving physical separation between the fluoresced and excitation light. While the two are typically at different wavelengths and thus in theory can be separated by a suitable monochromator, the excitation light may be several orders of magnitude more intense than the fluoresced light. Although a practical monochromator will typically pass only a very small fraction of light at wavelengths other than that for which the monochromator is set (called the stray light performance of the monochromator), because of the much greater initial intensity, the residual excitation light passed may still be large enough to seriously degrade the fluorescence measurement. In situations where the object is to measure total fluorescence (that is, without using a fluorescence monochromator) the problem of separation is worse since the separation achieved by the fluorescence monochromator is absent.

Summary Of The Invention

An object of the present invention is to provide a sample cell for achieving a large degree of interaction between a non coherent optical source and a limited sample volume to thereby improve the sensitivity with which fluorescence measurements may be made.

A further object is for the sample cell to provide for physical separation of the excitation light from the emitted light so that the latter may be more readily measured.

The concept behind the present invention is to enclose the sample in a structure (that is, a sample cell) which traps and guides the excitation light through the sample along the full length of the structure while allowing fluoresced light to escape through a side of the structure.

Accordingly, a first aspect of the invention provides a sample cell for containing a fluid sample for spectrophotomethc fluorescence analysis, including a tube defined by a wall of an optically transparent material, a medium surrounding the tube wall which has a refractive index less than the refractive index of the fluid sample thereby establishing a critical angle for total internal reflection of light at the interface of the tube wall and the surrounding medium, the tube having a length and including an end for injection of an excitation beam of light into, in use, a fluid sample contained in the tube, the beam having a predetermined included angle to provide an angle of incidence relative to the tube axis for the excitation light at said interface which is less than said critical angle, wherein the excitation light is totally internally reflected such that it traverses the length of the tube for substantially all of it to interact with the fluid sample for the sample to emit fluorescence light, a portion of which passes through the wall of the tube, the surrounding medium being optically transparent to allow for detection of the fluorescence light which passes through the wall of the tube.

According to a second aspect, the invention provides fluorescence spectrophotometry apparatus including a sample cell according to the first aspect of the invention for containing in use, a fluid sample, a light source and optical system for directing a beam of excitation light of selected wavelength into the fluid sample at an end of the sample cell, wherein the beam has a predetermined included angle such that, in use, the excitation light is totally internally reflected and traverses the length of the sample cell and fluid sample contained therein, an emission optical system for collecting and detecting fluorescent light of a selected wavelength from the fluid sample, such fluorescent light being a portion of that which passes through the wall of the sample cell.

According to a third aspect the invention provides a method for measuring a fluorescence characteristic of a fluid sample including, providing a tubular cell having a wall of an optically transparent material surrounded by an optically transparent medium having a refractive index less than the refractive index of the fluid sample, filling the tubular cell with the fluid sample, injecting a beam of excitation light of selected wavelength and predetermined included angle into the sample at an end of the cell, wherein the predetermined included angle is determined in relation to said refractive indices such that the excitation light is totally internally reflected for substantially all of it to traverse the sample within the cell to cause the sample to emit fluorescent light, collecting at least a portion of the emitted fluorescent light which passes through the wall and the surrounding medium, and measuring the intensity of a selected wavelength of the collected fluorescent light.

Thus, in using the invention, the sample is enclosed in a tube, which is ideally thin walled, made of an optically transparent material (for example, quartz) the outer surface of which is surrounded by a material of low refractive index, for example a gas such as air, nitrogen, argon etc. The excitation light is injected into the sample via one end either by way of an optical fibre or via conventional geometric optics. In either case, the excitation light typically diverges after entering the sample until it encounters the sample/tube material interface. If the refractive index of the sample is higher than that of the tube material the ray may be totally internally reflected at this boundary back into the sample. More generally, the ray will continue on into the tube material until it encounters the boundary between the tube material and the external environment. The objective is to ensure that rays of excitation light reaching this boundary are totally internally reflected and the parameters to ensure that this occurs are given below. Such reflected rays propagate back through the tube material and into the sample to the opposite wall where they are again totally internally reflected. In this way the sample plus tube effectively form a "light guide" or "composite optical fibre" thus constraining the excitation light to travel down the entire length of the "guide" or "fibre" interacting with the sample as it goes. The requirements for such propagation are calculated as shown herein below.

The following detailed description with reference to the accompanying drawings is provided to give a better understanding of the invention in all its aspects and to show how it may carried into effect. This description and drawings are given by way of non-limiting example only and are not to be interpreted as limiting the generality of the preceding description. Brief Description Of Drawings

Fig. 1 schematically illustrates total internal reflection in a cell according to the invention to assist in explaining the principle on which the invention is based.

Fig. 2 schematically illustrates the invention,

Fig. 3 shows an experimental cell according to an embodiment of the invention,

Fig. 4 is a graph of fluorescence measurements for a Rhodamine sample using the experimental cell of Fig. 3, and

Fig. 5 schematically illustrates fluorescence spectrophotometry apparatus including a sample cell according to the invention.

Detailed Description With reference to Fig. 1 consider a ray of light 98 entering a sample volume through one end 102 of a cell 100 (that is, a thin walled tube) at an angle A to the normal (equivalent to an angle (90-A) relative to the tube axis 96).

For the purposes of the calculation define Ns = refractive index of the sample

Nt = refractive index of tube material Next = refractive index of medium outside the tube

When the ray 98 encounters the sample/tube interface at 104 it will be refracted, according to Snell's law, to angle B such that

Ns *sin(A) = Nt ^*sin(B) or sin(B) = sin(A) ^*Ns/Nt If sin(A) *Ns/Nt is >1 the ray 98 will be totally internally reflected at this interface 104 (this requires Ns > Nt); otherwise the ray will proceed into the tube material. When the ray encounters the tube/external medium interface at 106 it will be again refracted to angle C such that

Nt * sin(B) = Next *sin(C), or substituting for sin(B); Nt ^* sin(A) ^* Ns/Nt = Next ^* sin(C) hence sin(C) = sin(A) ^* Ns/Next

Total internal reflection will occur at this interface if sin (A) * Ns/Next is > 1 , that is, A > asin (Next/Ns) or the angle to the tube axis is < (90-A).

Note that total internal reflection at this interface 106 does not depend on the refractive index of the tube material. It depends only on the refractive index of the sample and external medium.

By reversing the light path it can also be shown that any ray emanating from the sample and totally internally reflected from the tube/external medium interface must pass back through the tube/sample interface into the sample, re- entering the sample at angle A and proceeding to the opposite boundary where it will be again totally internally reflected. Thus any ray of light injected into the sample at an angle of less than plus or minus (90-A) degrees relative to the tube axis 96 will be trapped within the structure by repeated total internal reflection and will pass down the length of the tube interacting with the sample as it does so. As an example, if the external medium is air (Next = 1 ) and the sample is aqueous (Ns = 1.33) the critical value for A is 48.75 degrees. Thus any ray at an angle of plus or minus 41.2 degrees or less relative to the tube axis will be captured. Thus a beam with a 41.2 degree half angle or 82.4 degree included angle will be fully captured. This corresponds to an F 0.57 beam.

Quite small refractive index differences will still allow a beam of substantial included angle to be trapped. For example, if the refractive index of the sample is only 1.1 , the above calculation yields a critical half angle of the beam of 24.6 degrees corresponding to an F1.1 beam. Even such a beam is very broad by optical standards.

It is possible to further increase interaction between excitation light and sample by making the far end of the cell reflective. In this case the excitation light will be reflected at this boundary and guided back through the sample to emerge at the sample injection port. This refinement effectively doubles the path length of the excitation light in the sample.

The fluorescence light is generated within the sample at all angles. Those rays that are emitted at an angle of less than (90-A) degrees relative to the tube axis 96 will be trapped and lost. Rays emitted at an angle greater than this critical angle however will emerge through the sides of the tube 100 and may be thus collected.

For an aqueous sample with a critical angle of 48 degrees this suggests that approximately half the fluoresced light is lost. In practice however the light emitted at such angles would anyway not be capable of collection by conventional optics. Thus, while in absolute terms this light is lost, relative to conventional approaches there is little if any difference. Since the excitation light is constrained to travel inside the composite tube while the fluoresced light being measured is emitted through the side wall of the tube there is effective separation between the excitation light and fluoresced light as desired.

In cases where sample volume is limited it is preferred if the tube has a relatively small diameter in the order for example of 0.5-3.0 mm. A small bore cell of this order of size filled with a fluid sample of higher refractive index than the surrounding medium acts as a composite optical fibre, as desired. This embodiment meets the increasing demand for smaller and smaller sample volumes while ensuring there is ample interaction of the excitation light with the sample and ample fluorescent light collected, as shown by the comparative example calculations given hereinafter.

The preferred choice of tube material and surrounding medium depends on the wavelength range of interest. For operation in the visible only, suitable tube materials could be for example quartz, glass or a transparent plastic such as acrylic with the surrounding medium air. For operation into the ultraviolet down to about 200 nm the preferred tube material would be quartz again with air as the surrounding medium. For operation at shorter wavelengths the preferred tube material would be magnesium fluoride or calcium fluoride with the surrounding medium nitrogen or argon. For operation into the infrared the preferred tube material would be water-free quartz with dry nitrogen or dry air as the surrounding medium. The surrounding medium may alternatively be a thin coating of a light transparent material applied to the outside of the tube. Such a coating may be designed to be selectively transparent to a relatively narrow range of wavelengths to separate fluorescence occurring at one wavelength from other possible fluorescences occurring at other wavelengths.

The emission of fluorescence light from a small diameter elongated source, (that is, from a sample cell according to an embodiment of the invention), also allows advantageous coupling into a monochromator or polychromator for wavelength resolved fluorescence applications. In some cases, the cell may become the entrance aperture into such a wavelength resolving device.

A cell according to the present invention may also be placed inside a reflective cavity such as an integrating sphere. Such a cavity structure is known and used in spectroscopic applications. With this feature, a larger fraction of the emitted fluorescence light can be collected.

Because the light is trapped within the composite fibre there is no need for the cell to be straight. Indeed, where very long interaction paths are desired it is possible to coil the cell into a multi-turn spiral. Such coiling allows very long path cells to be accommodated in a small compact space such as within the integrating sphere referred to above.

The injection of the beam of excitation light into the sample at the end of the tube may be done via appropriate imaging or via an optical fibre.

As the critical angle of the cell plus sample is increased, the included angle of the cone of light that may be injected is increased allowing more light to interact with the sample. Conversely however, more of the fluorescence signal will be lost through total internal reflection trapping it inside the cell. There may be advantages to be gained by not making the critical angle too large (for example, the incident cone included angle is limited by other means such as the numerical aperture of the injecting fibre). This critical angle can be reduced by increasing the refractive index of the medium outside the tubing. Nonetheless, the refractive index of this medium must always be less than that of the sample and must be substantially transparent to the fluoresced light so as to allow the sensing of such light.

Comparative Calculations

Consider a fluorometer using an excitation monochromator with a 2 mm exit slit and an F4 beam. That means the exit optical beam at a distance X from the focal point will have a width Y such that

Y = X/4 + 2 mm

Further assume a sample volume of 10 microlitres which is typical of the sample volume from a liquid chromatograph.

Optical re-imaging using a magnification of K can give a beam width of

Y = X/4K + 2^*K mm

The longest cell that will accept all the light while having a volume of 10 microlitres is 3.6 mm long using a K of 0.5. This means the longest sample path length for a 10 microlitre sample that will accommodate all the excitation light is 3.6 mm.

By contrast, assuming an aqueous sample, a cell according to the current invention can accept a cone of light down to F0.6. Allowing some margin a K of 0.2 could be used giving an F number of 4^*0.2 = 0.8. This gives an image size of 2*0.2 or 0.4 mm. Again allowing some margin an 0.5 mm diameter cell could be used giving a path length of 10/(0.25^*3.1415^*.5^*.5) = 51 mm. This represents 14 times greater interaction than for the conventional case above. Using a mirror at the far end of the cell to cause the excitation beam to traverse the sample twice raises the gain to 28 times. These are substantial gains over a conventional approach.

In Fig. 2 a sample cell 20 according to the invention comprises a quartz wall 22 defining a cylindrical tube which is mounted or housed (not shown) such that it is surrounded by air 24. The tube 22 is filled with an aqueous sample 26. A beam of excitation light 28 is injected into the sample 26 at an end 30 of the tube 22. The included angle 32 of the beam 28 is determined in relation to the ratio of the refractive index of air 24 to the refractive index of the aqueous sample 26 such that the excitation light 28 will be totally internally reflected, as described hereinabove This is shown by the rays 34. Thus the excitation light 28 is confined within and travels the length of the cell 20, with substantially all of it interacting with the sample 26 to cause the sample to emit fluorescent light 36. The fluorescent light 36 is emitted in all directions and that which is incident at the quartz 22 - air 24 interface at an angle greater than the critical angle passes through the quartz wall and is thus able to be detected.

Experimental Results

To test the invention an experimental cell was built using readily available materials. This comprised a piece of quartz tubing 40 (see Fig. 3) mounted by and within a stainless steel housing 42 via an O ring seal 44 in a shoulder formation 47 of a recess 46 at an end structure 48 of the housing 42. An opposite end of the tubing 40 was passed through an aperture 50 in an opposite end 52 of the housing 42, and held and sealed therein via an O ring 54 and clamp plate 56. This end of the tubing 40 is accessible for a sample to be fed into the cell as indicated by arrow 57. An optical fibre 58, which is mounted in a passage 60 in the end 48 of the housing 42, enters the tubing 40 a very short distance for injection of excitation light into a sample in the tubular cell. Sample passes out of the tubing 40 via a passageway 62 which connects with recess 46, in the housing end 48. The cell was completed by surrounding the tubing 40, leaving an air space 64, in a pressed halon tube 66 mounted by housing 42. A window 68 was provided in the halon tube 66 and the housing 42 for a detector for detecting fluoresced light from the sample. The quartz tubing 40 was 1.6 mm internal diameter, 4 mm outside diameter. In practice the wall thickness of this tubing is excessive and will result in lower efficiency since a significant fraction of the light will be travelling in the quartz walls and not the sample. It was however readily to hand and is able to demonstrate the principle. Light was injected into this cell via the optical fibre 58, which was 0.6 mm diameter, 0.22 NA (numerical aperture). The fluorescence light emitted from a 10 mm path length of tube 40 was collected by a standard Cary 50 detector and preamplifier (the Cary 50 is a spectrophotometer available from Varian, Australia Pty Ltd, 679 Springvale Road, Mulgrave, Victoria, Australia). The tubing 40 was enclosed in the pressed halon tube 66, which was 8 mm internal diameter, to make the efficiency as high as possible.

The light throughput of a coupler and the optical fibre 58 for the excitation light, measured by placing a detector at the far end of the fibre, was 31 %.

A solution of rhodamine having a concentration of 1 milligram per litre was used as the sample material and the absorbance of this sample was first measured using a 10 mm path prior art cuvette. At 600 nm, at which virtually no fluorescence occurs, the cuvette had a transmittance of 88%. At 554 nm, the wavelength exciting maximum fluorescence, it passed 67%. Overall, the shape of the plot of transmission as a function of wavelength closely mirrored the published fluorescence curve with a constant background. It was assumed from this that (88-67) or 21 % of the incident light is absorbed and converted to fluorescence at 554 nm. Further, rhodamine is known to have a very high quantum efficiency, close to 100%. This means that the total fluorescence signal generated over the 10 mm path length should be about 21 % of the incident signal.

Within the tube 40 of Fig. 3 this means the total amount of fluorescence light generated should be about 0.31 x 21 % or 6.5%T. Of this the light emitted within 41 degrees (for an aqueous sample RI = 1.33) will be trapped by total internal reflection. This component represents 0.245 of the generated fluorescence. The theoretical fluorescence signal is thus 6.5 x 0.755 = 4.9%T. The cell of Fig. 3 was filled first with distilled water and the %T signal versus wavelength measured. It was then filled with 1 milligram per litre rhodamine and the signal again measured as a function of wavelength. The measured signal with distilled water was constant with wavelength and close to zero (see Fig. 4 - trace 70). The signal with rhodamine (see Fig. 4 - trace 72) conformed to the published fluorescence versus wavelength data. The rhodamine signal minus distilled water signal at 554nm was 3.3%T. This corresponds to 67% of the theoretical signal. The discrepancy is believed to be mainly due to the excessively thick walls of the quartz tube 40 resulting in less than ideal interaction between the excitation light and the sample.

Figure 5 shows fluorescence spectrophotometry apparatus comprising a light source 81 which emits light 82 that passes into an excitation monochromator 83, which transmits first essentially monochromatic light 84 of a first wavelength, said first wavelength having been chosen to ensure that said light 84 can interact with a particular chemical species of interest and cause said chemical species to emit light of a known second wavelength. On emerging from excitation monochromator 83, light 84 passes to a first focussing means 85 that focuses light 84 into a sample 86 contained in a cell 87 constructed according to the teachings of the present invention. Light 84 interacts with said chemical species of interest in sample 86, causing it to emit light 88 of said second wavelength in all directions. Some of said light 88 passes into a second focussing means 89 that focuses said light 88 as a focussed beam 90 into an emission monochromator 91 that allows second essentially monochromatic light 92 of said second wavelength to pass to a first light detecting means 93. Said first light detecting means 93 produces a first electrical signal 94 in proportion to the intensity of said second essentially monochromatic light 92. First electrical signal 94 passes to amplifying and processing circuitry 95 and therein is converted to an amplified and processed electrical signal 96 that in turn passes to a computing means 97 wherein it is further processed and is then displayed by a display means 98. The mathematical relationship between the concentration of said chemical species of interest in sample 86 and the signal displayed at 98 is established by consecutively placing samples having known but different concentrations of said chemical species in cell 87 and noting the corresponding signal displayed at 98. The mathematical relationship between the known concentration of said chemical species and the corresponding displayed signal is then established, and said mathematical relationship is then used to compute the concentration of said chemical species in samples of unknown concentration consecutively placed in cell 87.

Thus the apparatus of Fig. 5 includes a sample cell 87 according to the invention, a light source 81 and an optical system 83-85 for directing a beam of excitation light 84 of selected wavelength into a fluid sample 86 at an end of the cell 87 such that the light 84 is totally internally reflected and traverses the length of cell 87 and sample 86 therein. An emission optical system 89-91-93 collects and detects fluorescent light 88 of a second wavelength from the fluid sample 86.

Excitation monochromator 83 and emission monochromator 91 may each or both be a grating monochromator, a prism monochromator, an optical filter or any other optical device adapted to selectively pass light of a desired wavelength and prevent the passage of light of other wavelengths. The first light detecting means 93 can be a photomultiplier tube, a photodiode, a charge- coupled device, or any other means of converting incident light into an electrical signal.

An optional device 99 may be added to the Fig. 5 apparatus, however its presence is not essential to the operation of the apparatus. If present, device 99 intercepts light 100 emerging from cell 87. In a first variation, device 99 may be a second light detecting means, in which case device 99 produces a second electrical signal (not shown) proportional to the intensity of light 100 emerging from cell 87. Said second electrical signal can then be amplified and processed by known means (not shown) and can serve as a reference for said first electrical signal 94. In a second variation device 99 may be a mirror adapted to reflect and focus light 100 back into cell 87, wherein said reflected and focussed light (not shown) interacts with sample 86 thus increasing the emission of light 88. In a third variation, device 99 may be adapted to absorb light 100 and thus prevent light 100 being reflected or scattered in an uncontrolled fashion and possibly entering first light detecting means 93 and thereby producing an erroneous or inaccurate measurement.

A cell according to the invention has potential use in all fluorescence work and is most advantageous when working with very small sample volumes. In this application the cell offers substantially higher sensitivity than existing approaches. Small volumes have always been an issue for liquid chromatography but are also becoming increasingly an issue for bioscience applications.

The invention described herein is susceptible to variations, modifications and/or additions other than those specifically described and it is to be understood that the invention includes all such variations, modifications and/or additions which fall within scope of the following claims.

Claims

1. A sample cell for containing a fluid sample for spectrophotometric fluorescence analysis, including a tube defined by a wall of an optically transparent material, a medium surrounding the tube wall which has a refractive index less than the refractive index of the fluid sample thereby establishing a critical angle for total internal reflection of light at the interface of the tube wall and the surrounding medium, the tube having a length and including an end for injection of an excitation beam of light into, in use, a fluid sample contained in the tube, the beam having a predetermined included angle to provide an angle of incidence relative to the tube axis for the excitation light at said interface which is less than said critical angle, wherein the excitation light is totally internally reflected such that it traverses the length of the tube for substantially all of it to interact with the fluid sample for the sample to emit fluorescence light, a portion of which passes through the wall of the tube, the surrounding medium being optically transparent to allow for detection of the fluorescence light which passes through the wall of the tube.

2. A sample cell as claimed in claim 1 wherein the tube has a closed end opposite the end for injection of an excitation beam into the fluid sample, and the closed end is reflective for reflecting the excitation light for return through the sample.

3. A sample cell as claimed in claim 1 or 2 wherein the medium surrounding the tube wall is a gas such as air, nitrogen or argon.

4. A sample cell as claimed in claim 1 or 2 wherein the medium surrounding the tube is a coating of a light transparent material.

5. A sample cell as claimed in claim 4 wherein the coating of a light transparent material is selected which is transparent to a predetermined wavelength to separate fluorescence occurring at that wavelength from fluorescences occurring at other wavelengths.

6. A sample cell as claimed in any one of claims 1 to 5 wherein the optically transparent material of the tube is selected from the group comprising quartz, glass, a transparent plastic, magnesium fluoride or calcium fluoride.

7. A sample cell as claimed in claim 1 including a housing which includes spaced supports for mounting the tube such that the tube extends between the supports.

8. A sample cell as claimed in claim 7 wherein one support allows for the fluid sample to be fed into an open end of the tube and the other support provides a mounting for an optical fibre for injection of the excitation beam of light into the fluid sample.

9. A sample cell as claimed in claim 7 or 8 wherein the housing encloses the tube and contains said medium surrounding the tube, the housing including a window for passage of the fluorescence light to allow for its detection.

10. Fluorescence spectrophotometry apparatus including a sample cell as claimed in any one of claims 1 to 9 for containing, in use, a fluid sample, a light source and optical system for directing a beam of excitation light of selected wavelength into the fluid sample at an end of the sample cell, wherein the beam has a predetermined included angle such that, in use, the excitation light is totally internally reflected and traverses the length of the sample cell and fluid sample contained therein, an emission optical system for collecting and detecting fluorescent light of a selected wavelength from the fluid sample, such fluorescent light being a portion of that which passes through the wall of the sample cell.

11. Fluorescence spectrophotometry apparatus as claimed in claim 10 wherein the optical system for directing a beam of excitation light of selected wavelength into the fluid sample includes an optical fibre having an end from which the beam of excitation light exits located adjacent said end of the sample cell.

12. A method for measuπng a fluorescence characteristic of a fluid sample including, providing a tubular cell having a wall of an optically transparent material surrounded by an optically transparent medium having a refractive index less than the refractive index of the fluid sample, filling the tubular cell with the fluid sample, injecting a beam of excitation light of selected wavelength and predetermined included angle into the sample at an end of the cell, wherein the predetermined included angle is determined in relation to said refractive indices such that the excitation light is totally internally reflected for substantially all of it to traverse the sample within the cell to cause the sample to emit fluorescent light, collecting at least a portion of the emitted fluorescent light which passes through the wall and the surrounding medium, and measuring the intensity of a selected wavelength of the collected fluorescent light.