DE4343345A1 - Optical transmission or reflection characteristics measuring system - Google Patents

Abstract

The measuring system has a beam from a light source (20) directed onto the sample (22), the resulting optical reflection or transmission signal provided by a detector (24) in the path of the reflected or transmitted beam after reflection by a retro-reflector (23). In the case of a reflected measuring beam, the retro-reflector lies at a different angle to the required position of the sample to the angle of incidence of the light from the source with the required sample position.

Description

The invention relates to methods and devices for measurement the reflective or transmitting optical properties a sample in which the light from a radiation source is the sample falls and a reflected or transmitted signal is measured.

The measurement of the transmission or the reflection of a sample is one of the oldest measuring tasks in optics. That puts Transmission behavior of an optical component Radiation amount fixed, which from the radiation entry to Radiation emerges. To the transmission behavior too improve, z. B. lenses with an anti-reflective layer Mistake. With mirrors, on the other hand, you want a high level of reflection receive. This high reflection is e.g. B. important for the Mirroring in astronomical telescopes, but also with mirrors for laser arrangements.

So far, the exact measurement of the transmission or Reflection of a sample is very difficult because of the measurement result is extremely sensitive to shifting and / or twisting the sample. This problem becomes particularly serious if the measurement in an evaporation plant for the production of Layers should be used. It's about the Production of thin layers or layer systems for optical or other applications e.g. B. antireflective Layers.

It is known in the art that the reflection of a Surface is measured using the following methods:

• 1. Simple reflection measurement, whereby light falls on the surface to be measured, is reflected and measured. The reflectivity of the measuring surface is determined by a zero or reference measurement on a known sample (see Fig. 1a).
• 2. In the R-square measurement, the light is directed onto the surface a second time and reflected by it. The reflectivity of the measurement surface is included twice in the measurement result, whereby the reflectivity of the mirror must be taken into account (see Fig. 1b). The reflectivity of the reflection mirror is determined by rotating the same mirror by 180 ° around the sample plane and the reflection measurement without a measurement sample is set to 100%.

The common disadvantage of both according to the prior art Known methods can be seen primarily in the fact that they are extremely sensitive to shifts (changes in Sample distance between radiation source and detector) and against twisting (changes in the optical axis of the sample with respect to the optical axis of the measurement setup). The One can be sensitive to changes in distance reduce some samples by light that is as parallel as possible, but not against twisting.

It is the object of the invention to obtain a method which at least one known from the prior art Does not have a disadvantage as well as a by this method working device.

According to the invention, this object is achieved by a method the characterizing part of the first claim and by a device according to the characterizing part of the fifth claim solved.  

By using a retroreflector according to the invention Measurement of the reflecting or the transmitting optical properties are very simple Measuring devices which are very resistant to movements of the sample are not susceptible to interference. The retroreflector can be opened be constructed in a wide variety of ways; but should in terms of its design, the one to be measured Spectral range to be adjusted.

It is advantageous if a reflection of the Back of sample not done.

Advantageously, one derives from the radiation source and radiation coming back from the sample through a Beam guide to between the sample and the radiation source or to be able to choose a greater distance from the detector. Here should be at the sample end of the beam guide Optics are located, which come from the beam guide Radiation collimates and which comes from the sample Radiation as completely as possible into the beam conductor feeds.

The divergence of the measuring radiation is advantageous as small as possible so that the angle of incidence for each beam is equal to. The spectrum is then not distorted, i. H. the Then there is a disturbance due to the angular dependence of the spectra minimal.

Retroreflectors are the easiest to plate-like manufacture, with the construction of the retroreflector but always make sure that the beam offset is chosen as small as possible.

The invention is illustrated below by way of example of drawings explained in more detail, with other essential Features and explanations for better understanding  and design options for the inventive concept are described.

Show:

FIG. 1a is a first schematic measuring arrangement according to the prior art;

FIG. 1b shows a second schematic measuring arrangement according to the prior art;

Fig. 1c third schematic measuring arrangement according to the known prior art;

Fig. 2 shows a schematic measuring arrangement according to the invention for reflection measurement;

Fig. 3 shows a schematic measuring arrangement according to the invention for transmission measurement;

Fig. 4 shows a schematic measuring arrangement according to the invention to the optical surface inspection;

Figure 5 is a schematic measuring arrangement according to the invention for permanent transmission measurement in a coating installation. and

Fig. 6 shows a schematic measuring arrangement according to the invention for permanent reflection measurement in a coating installation.

The known prior art is shown in FIGS . 1a-c with the aid of schematic measuring arrangements.

In Fig. 1a, an optical fiber ( 1 ) is shown, which has a beam splitter ( 2 ) at its first end. (This beam splitter ( 2 ) is not necessary if the optical fiber ( 1 ) consists of at least two discrete fiber bundles, each with at least one fiber. A bundle is assigned to the light source ( 3 ) or the detector ( 6 )). The light from a light source ( 3 ) is coupled into the optical fiber ( 1 ) via the beam splitter ( 2 ) and leaves the optical fiber ( 1 ) at its second end ( 4 ). Then the measuring light falls on the sample ( 5 ), is reflected back to the second end ( 4 ) of the optical fiber ( 1 ), travels through the optical fiber ( 1 ) to the beam splitter ( 2 ), which emerges from the first end of the optical fiber ( 1 ) Feeds light to a detector ( 6 ) which carries out a spectral measurement of the reflected amount of light. The reflectivity of the sample ( 5 ) can be determined using this simple reflection measurement by comparing the emitted light quantity to the reflected light if a zero measurement or a reference measurement is available on a known sample ( 5 ).

In the measuring device according to the known prior art shown in FIG. 1b, the light from a light source ( 7 ) falls obliquely onto the surface ( 8 ) of a sample ( 9 ), is reflected from there onto a mirror ( 10 ), from this mirror ( 10 ) reflected back onto the surface ( 8 ) of the sample ( 9 ), which then ultimately reflects the light onto a detector ( 11 ). In this measuring device working according to the R-square method, the light is reflected twice from the sample surface ( 8 ). If one now knows the reflectivity of the mirror ( 10 ), one can determine the reflectivity of the sample ( 9 ) (see also Optical thin films user's handbook, James D. Rancourt, McGraw-Hill Publishing Company, JSBN 0-07-052299 -3, page 144, Fig. 5).

The common disadvantage of both methods from FIGS. 1a and 1b is primarily that the measuring methods are extremely sensitive to shifting (ie changing the sample spacing) and / or twisting (ie changing the angle of the sample surface). The sensitivity to a change in distance can e.g. B. reduce by parallel light as possible. However, this increases the sensitivity to twisting. The sensitivity of the methods to twisting can be reduced by using imaging optics (e.g. 2: 1). Here, the sensitivity to tilting is reduced, but at the same time the sensitivity to distance changes is increased. (Extreme case is the measurement with a microscope as a measuring head!)

In Fig. 1c is a measuring arrangement, there is shown, in which the light from a light source (12) in a measuring head (14) inclined by a diaphragm (13) at a certain angle to a relation to a housing (14) by means of a holder (17) fixed test glass ( 15 ) falls. The measuring light is reflected on the back and front of this test glass ( 15 ) and thus reaches the detector ( 16 ). Although this measuring head excludes errors due to changes in the position of the sample plate ( 15 ), it has the disadvantage that the shape of the sample ( 15 ) cannot be arbitrary, that the exchange of the sample ( 15 ) is very time-consuming and that the sample glass ( 15 ) ambient light (e.g. from evaporation sources in coating systems) can also be measured, even if a lock-in method is used. The measuring device is suitable for photonics purposes, but not e.g. B. for a spectral measurement with the current spectrophotometers.

In Fig. 2, a measuring arrangement according to the invention for reflection measurement is now shown schematically. The radiation from a radiation source ( 20 ) is emitted onto the front surface ( 21 ) of the sample ( 22 ) at the smallest possible divergence angle. The radiation reflected by the front surface ( 21 ) of the sample ( 22 ) then falls on a retroreflector ( 23 ) and is sent back by the same in the same way. The retroreflector ( 23 ) has the property of reflecting radiation incident on it in a limited angular range parallel to the incident radiation. Depending on the design of the retroreflector ( 23 ), there is a more or less large small beam offset.

Such a retroreflector ( 23 ) is, for. B. is known from US-PS 4 202 600. An arrangement for measuring retroreflectors is known from US Pat. No. 4,721,389. The retroreflector ( 23 ) shown in FIG. 2 reflects the radiation in the direction from which the radiation comes, namely back onto the front surface ( 21 ) of the sample ( 22 ). If the sample ( 22 ) has not moved or only moved slightly during the time between the two reflections (which can certainly be assumed given the short distances and the very high speed of the electromagnetic radiation), then each beam hits at at least almost the same point and at the same angle the reflecting front surface ( 21 ) of the sample ( 22 ) and is therefore always thrown back towards the beam exit opening. In this case, changes in position of the sample (22) are automatically corrected by this double reflection on the front surface (21) of the sample (22).

The radiation reflected back onto the radiation exit opening from the front surface ( 21 ) of the sample ( 22 ) is then directed to a detector ( 24 ) which generates an electrical output signal in accordance with the intensity of the reflected radiation.

The radiation from the radiation source ( 20 ) is conducted via a beam splitter ( 25 ) into a Y waveguide ( 26 ) (optical waveguide) to the sample ( 22 ), an optic ( 27 ) consisting of at least one radiation-shaping one at the exit of the waveguide ( 26 ) on the sample side Element (z. B. lens) is arranged, which largely collimates the light.

In principle, this measuring device would also work if the retroreflector ( 23 ) was replaced by an ordinary mirror. This would then have to be precisely adjusted in the case of parallel light irradiation of the front surface ( 21 ) of the sample ( 22 ). This solution would fail with the mirror, however, if the sample ( 22 ) to be measured is not exactly adjusted, cannot be adjusted exactly or moves due to vibrations during the measurements. In addition, a normal mirror would increase beam deflection by rotating the sample ( 22 ). This is not the case if the retroreflector ( 23 ) shown is used instead of the mirror.

In order to explain this in more detail, the sample ( 22 ) is shown in two other positions in FIG. 2:

• a) sample ( 22 ') with a changed distance to the measuring head ( 28 );
• b) sample ( 22 '') in the tilted state.

In addition, the respective marginal and central rays shown.

If the distance of the sample ( 22 ') to the position of the sample ( 22 ) changes in the desired position, then the rays deflected by the sample at the first reflection hit the retroreflector ( 23.1 ). However, since the rays are essentially reflected in themselves, these rays return to the radiation entry or exit of the measuring head ( 28 ). The same applies accordingly if the sample ( 22 '') is tilted to the target position. It can be assumed that there was no or only insignificant movement of the sample ( 22 ) between the two reflections on the front surface ( 21 ) of the sample ( 22 ) (speed of light c - 3 · 10 ⁺⁸ m / s).

The surface shape of the retroreflector ( 23 ) should be matched to the surface shape of the sample ( 22 ). This means that a retroreflective plate ( 23 ) of appropriate size should be used for a plate-shaped sample with a flat reflection surface arranged in one plane.

The exact surface shape of the retroreflector ( 23 ) thus results in connection with the corresponding surface shape of the sample ( 22 ). In addition, when selecting the retroreflector ( 23 ), care must be taken that the radiation offset on the retroreflector ( 23 ) is not so great that the staggered reflected rays are no longer reflected back to the measuring head ( 28 ).

A corresponding selection with regard to the measurement problem must also be made for the radiation source ( 20 ) and the detector ( 24 ). The radiation source ( 20 ) should be optimized both in terms of intensity and in terms of the frequency spectrum. Correspondingly, the detector ( 24 ) should be optimized for the current measurement problem with regard to sensitivity as well as with regard to its frequency behavior (e.g. mono- or multispectral resolution).

A problem can arise in the measurement setup shown in FIG. 2 if reflection on the sample ( 21 ) can occur not only on the front surface ( 21 ) but also on the rear surface ( 29 ) of the sample ( 22 ). Appropriate measures can be taken to prevent a reflex from the rear surface ( 29 ) striking the retroreflector ( 23 ):

• 1) blackening the back surface ( 29 ) of the sample ( 22 );
• 2) anti-reflective treatment of the back surface ( 29 ) of the sample ( 22 ); or
• 3) Use of a wedge-shaped sample, or cementing a Wedge on the back surface of the sample.

Without these measures, ie also with the conditions shown in FIG. 2, the reflectivity of the front surface ( 21 ) can also be determined in a transparent sample ( 22 ) if one knows the reflectivity of the rear surface ( 29 ). The same applies if you want to determine the reflectivity of the back surface ( 29 ) of the sample ( 22 ). Then you have to know the reflectivity of the front surface ( 21 ) of the sample ( 22 ). Otherwise, a measured value is obtained which is composed of the reflectivity of the front surface ( 21 ) and the reflectivity of the sample ( 22 ).

The measuring process for determining the reflectivity can be like follows:

• 1) Zero measurement without plate - first measurement value A;
• 2) Optional reference measurement
• 2.1 Measurement of the retroreflector - second measured value B; or
• 2.2 measurement of a known glass - second measured value B '; and
• 3) Measurement of the sample - third measured value.

The result can then be determined from these three measured values according to the known prior art. When calculating the reflectivity of the sample ( 22 ), one must take into account that the radiation was reflected twice on the sample ( 22 ), so that the measured intensity is a quadratic function of the reflectivity R.

The great advantage of the measuring device shown in FIG. 2 can be seen in the fact that the device is self-adjusting due to the use of the retroreflector. This means that neither changes in distance nor changes in angle of the sample ( 22 ) from the desired target position lead to a change in the measured value.

The inventive measuring device can also be used for a transmission measurement without problems. Such a measuring device is shown in FIG. 3.

In this schematically illustrated measuring device, the light from a light source ( 30 ) is coupled into a Y light guide ( 32 ) via a beam splitter ( 31 ). At the other end there is an optical arrangement ( 33 ) which largely collimates the light emerging from the light guide ( 32 ). The light emerging from the measuring head ( 34 ) then penetrates the sample ( 35 ) and falls on a retroreflector ( 36 ). This retroreflector ( 36 ) reflects the light falling on it with a minimal, construction-related offset parallel to the incident light rays. The light reflected by the retroreflector ( 36 ) then penetrates the sample ( 35 ) a second time and then reaches the light guide ( 32 ) again. At the sensor-time output of the light guide, this light is coupled out at the beam splitter ( 31 ) and directed to a detector ( 37 ), which measures the intensity of the light.

The light from the light source ( 30 ) thus penetrates the sample ( 35 ) at essentially the same places, but there is no multiple reflection between the rays going back and forth. The difference in the transmission between the points of the returning and returning beams is negligible due to the low location dependence (e.g. the layer thickness distribution). As a result, a T² measurement takes place. The transmission can be determined from the light arriving at the detector ( 37 ), a zero measurement also being carried out here beforehand.

The advantage of the measuring head ( 34 ), in which the light guide ( 32 ) ends, can be seen in the fact that the position of the light source ( 30 ) and the detector ( 37 ) can be freely selected (this also applies to the one shown in FIG. 2 measuring head ( 28 ) shown).

A major advantage of the invention is also in it see that a measuring device with retroreflector is independent the beam-shaping properties of the sample (e.g. lens, Prism, wedge, etc.). It just has to take care of it that everything transmitted through the sample if possible or light of the light source reflected from the sample onto the extensive retroreflector falls.

In FIG. 4 is a schematic representation of a measuring device is shown to the optical surface examination. The surface ( 40 ) to be tested is illuminated as homogeneously as possible by the radiation source ( 41 ). The surface ( 40 ) scatters the light in accordance with the surface conditions into typical solid angles.

A retroreflector ( 42 ) with focusing optics ( 43 ) in a housing ( 44 ) is placed in a position above the surface to be tested ( 40 ), which corresponds to a certain solid angle and at which the highest possible measurement signal can be expected (depending on the expected surface defects).

The light reflected from the surface ( 40 ) onto the retroreflector ( 42 ) is reflected back (with a minimal parallel offset, due to the specific design of the retroreflector ( 42 ), coupled out by a decoupling device ( 45 ) and onto a detector ( 46 ) which allows an evaluation of the light reflected twice on the surface to be tested ( 40 ), the returning light signal (monochrome or spectral) being a measure of the reflection of the surface at a certain solid angle.

To determine the reflectivity, a zero measurement without reflector ( 42 ) may also be necessary here. A scan (ie a movement of the housing ( 44 ) with retroreflector ( 42 ) and optics ( 43 ) over many solid angles produces a reflection profile of the surface ( 40 ) over a swept solid angle range.

If you want to determine surface properties on the surface ( 40 ), you can optimize the shape of the retroreflective surface of the retroreflector ( 42 ) and, if necessary, the optics ( 43 ) accordingly. B. a line on the surface ( 40 ) at a large angular range in the measurement result of the detector ( 46 ).

With this measuring device, too, the double reflection on the measuring surface ( 40 ) must be taken into account. Since the surface ( 40 ) of this measuring device is to be illuminated as homogeneously as possible, the light source ( 41 ) may have to be equipped accordingly with optical components known from the known prior art. The angle of incidence (range) of light must be optimized according to the measuring task.

The advantages of using the retroreflector for Transmission measurement or reflection measurement particularly clear if you have a measuring device for spectral layer thickness dimensioning with the retroreflector equips.

This is explained in more detail in the following FIGS. 5-9, in particular FIG. 5 representing a measuring arrangement for a permanent transmission measurement and FIG. 6 a measuring arrangement for a permanent reflection measurement.

In the measuring device shown in FIG. 5, the light from the light source ( 50 ) falls through a beam splitter ( 51 ) and through a window in the coating apparatus ( 56 ) onto a deflecting mirror ( 52 ) which is in the axis of rotation ( 53 ) of a receiving device ( 54 ) for optically transparent samples ( 55 ), in contrast to the light source ( 50 ), is arranged within a coating apparatus ( 56 ).

A further radiation deflecting mirror ( 57 ) and an opening in the receiving device ( 54 ) are arranged in front of the sample ( 55 ), the light passes through the sample ( 55 ) and through a second window in the coating apparatus ( 56 ) onto an outside of the coating apparatus ( 56 ) arranged retroreflector ( 58 ) deflects. At this retroreflector ( 58 ) the light is reflected back into itself and the light which is then reflected out of the coating apparatus ( 56 ) is decoupled from the illumination beam path at the beam splitter ( 51 ) and directed onto a detector ( 59 ).

If the sample ( 55 ) is now rotated in the holding device ( 54 ) (as is done in coating apparatus ( 56 )), the two deflecting mirrors ( 52 , 57 ) fixed to the holding device ( 54 ) are also rotated. As a result, a permanent transmission measurement on the sample ( 55 ) can take place during the entire coating process.

By moving the second deflecting mirror ( 57 ) or varying the angle of incidence on the first deflecting mirror ( 52 ), one can also measure the sample ( 55 ) with regard to its transmission behavior at various points during the measuring process.

In Fig. 6 measuring device, the light of the light source ( 60 ) falls through a beam splitter ( 61 ) even after passing through a window ( 66 a) in the coating apparatus ( 66 ) on one, in the axis of rotation ( 63 ) of a receiving device ( 64 ) for the first deflecting mirror ( 62 ) arranged for the sample ( 65 ) to be measured. This first deflecting mirror ( 62 ) directs the light radiated into the coating apparatus ( 66 ) onto a second deflecting mirror ( 67 ), which is also firmly connected to the receiving device ( 64 ).

From this second deflecting mirror ( 67 ) (which is used here only for a steeper light incidence angle on the sample ( 65 )), the light reaches the sample ( 65 ) and is reflected by it to a retroreflector ( 68 ). This retroreflector ( 68 ) ensures a 180 ° rotation of the light propagation axis, as a result of which the light is radiated out of the coating apparatus ( 66 ) again in the same way.

Outside the coating apparatus ( 66 ), the light reflected in this way strikes the beam splitter ( 61 ), which directs this reflected light to a detector ( 69 ) for determining a measured value.

By means of the measuring device shown in FIG. 6, a permanent reflection measurement on the sample ( 65 ) can be carried out during the entire coating process.

By moving the second deflecting mirror ( 67 ) or varying the angle of incidence on the first deflecting mirror ( 62 ), the specimen ( 65 ) can also be measured at various points during the measuring process with regard to its reflection behavior.

The second deflecting mirror ( 67 ) in this measuring device (like the deflecting mirror ( 57 ) in FIG. 5) only serves to ensure that the measuring light falls on the sample ( 65 ) at the steepest possible angle. If one accepts a flatter angle of incidence of the light on the sample ( 65 ), then one can do without this second deflecting mirror ( 67 ) and only has to adjust the position of the retroreflector ( 68 ) accordingly. With this measuring device, as already mentioned, care must be taken that the reflection of the light can take place both on the front side and on the rear side of the sample ( 65 ).

It is essential in this measuring device from FIG. 6 as well as in the measuring device from FIG. 7 that the irradiation device with the light source ( 60 ) and also the detector or the detection device can be arranged outside the actual coating apparatus ( 66 ).

In addition, the two measuring devices shown in FIGS . 5 and 6 have the significant advantage that minor adjustment errors (e.g. eccentricities, vibrations of the sample receiving device, etc.) are compensated for by using the retroreflector.

The measuring devices shown in FIGS . 2 and 4 can also be used very well as lighting systems with contrast enhancement.

Is such a measuring device, for example Illumination of the fundus is used Light from the radiation source from the back of the eye to one Retroreflector reflects. This sends the light again back to the back of the eye, which then shines the light in Direction of the radiation source is reflected. A beam splitter in front of the radiation source it couples from the back of the eye coming rays of light, which then z. B. from as Detector-serving eye of an observer can. The fundus represents the reflective one Sample. This gives a square here Enhancing the contrast of the fundus.

The possible uses shown in the examples represent only a fraction of the possible stakes of the  Invention represents and therefore they are not considered to understand the final list. The samples can consist of both living and dead matter and under a detector are all radiation sensitive elements to understand which for those from the radiation source or from outgoing rays of the sample to be measured are sensitive.

Claims (11)

1. A method for measuring the reflective or transmittable optical properties of a sample ( 22 , 22 ', 22 '', 35 , 40 , 55 , 65 ), in which the radiation from a radiation source ( 20 , 30 , 41 , 50 , 60 ) falls on the sample ( 22 , 22 ′, 22 ′ ′, 35 , 40 , 55 , 65 ) and from a detector ( 24 , 37 , 46 , 59 , 69 ) one of the sample ( 22 , 22 ′, 22 '', 35 , 40 , 55 , 65 ) incoming signal is measured, characterized in that the measuring radiation after contact with the sample ( 22 , 22 ', 22 '', 35 , 40 , 55 , 65 ) by a retroreflector ( 23 , 36 , 42 , 58 , 68 ) is essentially reflected back into itself and directed to the detector ( 24 , 37 , 46 , 59 , 69 ). 2. The method according to claim 1, characterized in that during a reflection measurement, an oblique radiation incidence at an angle alpha to the target position of the sample ( 22 , 65 ) and that the retroreflector ( 23,68 ) at an angle beta to the target position of the sample ( 22 , 65 ) is aligned, the amounts of the angles alpha and beta being at least approximately the same size. 3. The method according to any one of claims 1 or 2, characterized characterized in that the reflection from the back of the sample is suppressed. 4. Device for measuring the reflective or the transmitting optical properties of a sample ( 22 , 22 ', 22 '', 35 , 40 , 55 , 65 ), in which the radiation from a radiation source ( 20 , 30 , 41 , 50 , 60 ) falls on the sample ( 22 , 22 ′, 22 ′ ′, 35 , 40 , 55 , 65 ) and from a detector ( 24 , 37 , 46 , 59 , 69 ) one of the sample ( 22 , 22 ′, 22 ′ ′, 35 , 40 , 55 , 65 ) incoming signal is measured, characterized in that a retroreflector ( 23 , 36 , 42 , 58 , 68 ) in the sample ( 22 , 22 ′, 22 ′ ′, 35 , 40 , 55 , 65 ) coming beam path of the radiation from the radiation source ( 20 , 30 , 41 , 50 , 60 ) is arranged. 5. The device according to claim 4, characterized in that the divergence angle of the measuring radiation is as small as possible. 6. Device according to one of claims 4 or 5, characterized in that the retroreflector ( 23 , 36 , 42 , 58 , 68 ) is flat. 7. Device according to one of claims 4-6, characterized in that the beam offset of the retroreflector ( 23 , 36 , 42 , 58 , 68 ) is as small as possible. 8. Device according to one of claims 4-7, characterized in that between the sample ( 22 , 22 ', 22 '', 35 , 40 , 55 , 65 ) and radiation source ( 3 , 7 , 12 , 20 , 30 , 41 , 50 , 60 ) or detector ( 24 , 37 , 46 , 59 , 69 ) an arrangement ( 27 , 26 , 25 ; 33 , 32 , 31 ; 45 ; 51 , 61 ) is arranged which detects the sample ( 22 , 22 ', 22 '', 35 , 40 , 55 , 65 ) coming radiation on the detector ( 24 , 37 , 46 , 59 , 69 ) conducts. 9. The device according to claims 8, characterized in that as an arrangement ( 32 ) at least two fiber bundles ( 32 ) each with at least one fiber ( 32 a, 32 b) between the sample ( 35 ) and / or radiation source ( 30 ) and detector ( 37 ) is arranged, at least one fiber bundle ( 32 a, 32 b) being assigned to at least one radiation source ( 30 ) or at least one detector ( 37 ). 10. The device according to claim 8, characterized in that between the radiation source ( 20 ) or detector ( 24 ) and sample ( 22 , 22 ', 22 '') is arranged as an arrangement ( 26 ) at least one radiation guide. 11. Device according to one of claims 8-10, characterized in that an optics ( 27 ; 33 ) is arranged at the sample end of the arrangement ( 26 ; 32 ).