GB2167205A - Apparatus for producing pulse trains of light pulses - Google Patents

Abstract

An apparatus for producing pulse trains of ultrashort light pulses may be used particularly for excitation- and test beam methods of ultrashort time laser spectroscopy. Parallel non- collinear beaming of the individual pulses, derived from a single ultrashort light pulse (1) is achieved with the aid of two plates (2,3) disposed parallel to one another, one of which plates (2) is partially reflective. To ensure a fixed mutual spacing b of the beams, there is provided a mechanism which, simultaneously with a change in the spacing (d) between the two plates necessary for changing the pulse repetition frequency, turns the two plates commonly through a corresponding angle with respect to the direction of the beam. <IMAGE>

Description

SPECIFICATION Apparatus for producing pulse trains of light pulses The invention relates to an apparatus for producinvg pulse trains of light pulses, which may be of selectable repetition frequency. The arrangement may be used for the time resolution of processes which take place rapidly, for example in the range of nano- or subnanoseconds. In such case, an object under test is probed strobascopically by the pulse train.

The invention may be used especially for the excitation- and test beam method of ultrashort-time laser spectroscopy.

In the so-called excitation- and test beam method, an excitation pulse puts the sample to be tested into a state of non-equilibrium whose change with respect to time is probed by means of a test pulse of selectable time delay with respect to the excitation. Various technical solution are known for this purpose.

The variable time delay of an individual test pulse with respect to the excitation pulse is used (IBEE Journal of Quantum Electronics QE3 (1967), 302). Elimination of the effect of the statistical shot-to-shot fluctuations in the properties of the excitation pulse on the measuring effect requires averaging over a large number of individual measurements for each delay time of the test pulse. The disadvantage of this is the expenditure of time needed for a time-resolved measurement. Furthermore, the high radiation load on the sample may lead to a change in its properties.

Two methods are also known for obtaining time resolution with an individual test pulse (Chem. Phys. Lett. 3 (1969), 534 and Chem.

Phys. Lett. 9 (1971), 1). For this purpose, the test beam is expanded spatially and one-dimensionally in order to be able to realise different time delays of different components of the beam with respect to the instant of excitation. In a first method, this is done by crossed propagation of the excitation and test beams, an extended area of the sample being excited along the direction of propagation of the excitation pulse. Since the test is made transversely of the excited track on the sample, the varying time delay chiefly results from the varying excitation time along the track as a result of the finite rate of propagation of the excitation pulse. In this instance, the range of time resolvable with one event is limited by the length of the sample or the depth of penetration of the excitation radiation.However, the change in extinction for the test radiation must be sufficiently large over the width of the excited track.

In a second method, a so-called echelon is introduced into the expanded test beam in order to obtain a varying time delay of the components of the expanded beam over its stepwise differing optical length of path. The disadvantage of this solution resides in the fact that the time delay of the individual components of the test beam is fixedly prescribed by the step length of the echelon, and that the latter's over-all length limits the resolvable time range.

Time-resolved fluorescence spectroscopy is also known (Appl. Phys. Lett. 15 (1969), 192 and Opt. Commun. 1 61969, 254). The fluorescent behaviour of a sample excited by a picosecond pulse is observed. By way of example, a so-called gate based on the optical Kerr effect is used for this purpose. The Kerr switch may be controlled by an optical pulse which is delayed in a defined manner with respect to the fluorescence excitation, and hence may be made transparent for a specific time range of the decaying fluorescent radiation. A disadvantage of this method is the limitation to states of excitation relaxing with adequate fluorescent quantum efficiency and, according to the construction, the necessity for a measuring statistic or limitation to a specific range of measuring time.

Time resolution by means of image-converter cameras is also possible if fluorescing samples are being used (Nuovo Cimento 63B (1981), 411). In this method, the distribution of light intensity with respect to time is converted to a corresponding time distribution of the intensity of a photoelectron beam. This distribution of intensity with respect to time is transformed to a spatial distribution of brightness on a fluorescent screen by a rapid deflection system for the electron beam. A very efficient method is available if this distribution of brightness can be recorded and evaluated by a multi-channel optical analyzer.

The expenditure on technical equipment is, of course, very high, and, in particular, rapid streak cameras are very expensive.

The aim of the invention is to provide an arangement in order to perform time-resolved measurements in the range of ultra-short time with shorter measuring times and incrased accuracy without highly time-resolving radiation detectors or image converters, to extend the field of application to samples which are less resistant to radiation and which do not necessarily fluoresce, and, in a simple manner, to be able to vary the probed time range in conformity with the behaviour of the samples.

An object of the invention is to perform time-resolved measurements of absorption and reflection in the range of ultra-short time with bombardment excitation advantageous with respect to the expenditure of time and accuracy, and thereby to extend the scanned time range relative to known methods and to make it variable by means of selectable intervals between the sampling points.

In accordance with the invention, an apparatus for producing pulse trains of ultrashort light pulses of selectable repetition frequency, with parallel non-collinear beam propagation of the individual pulses, derived from a single ultrashort light pulse, comprises two plates disposed parallel to one another, one of which plates is partially transparent, and a device to ensure a fixed mutual spacing of the beams, said device being adapted to ensure that, simultaneously with a change in the spacing between the two plates necessary for changing the pulse repetition frequency, the two plates are turned commonly through a corresponding angle with respect to the direction of the beam.

The parallel paths of the individual beams enable all of the beams to be combined by a focussing optical system in a small volume to be investigated. The non-collinear propagation makes it possible to dispense with a highresolution detection of radiation with respect to time, since the spatially separated beams can be recorded independently of one another on appropriately positioned receivers. The fact that the direction of propagation and the propagation intervals of the individual pulses are independent of their time delay obviates readjustment of the measuring apparatus in the event of a variation of the same.

The invention will be further described, by way of example, with reference to the accompanying drawings, in which: Figure 1 its a diagrammatic illustration for clarifying the optical mode of operation of an arrangement in accordance with the invention, Figure 2 shows a technical embodiment, and Figure 3 shows an application in test beam spectroscopy.

An incident light pulse i (Fig. 1) encounters at an angle a of incidence, a partially reflective plate 2, whose back is non-reflective, and the part reflected by the plate 2 is reflected from a plate 3 which is disposed parallel to the plate 2 and at a distance d thereform and whose surface facing the plate 2 is fully reflective. After reflection from the said surface, the beam is transmitted partially through the plate 2 and leaves the latter parallel to, and at a distance b from, the adjacent beam components transmitted by the plate 2.The following geometrical relationships apply: AC=d AB=d cos a - b BC=- b=2d sina 2 Hence, the path difference of adjacent beams passing through the plate 2 is D=2d cost4 AB, and the corresponding time delay of adjacent pulses is D 2 AB iSt== c c wherein c represents the speed of light.

Upon a variation of D or At, the positions of the beams emerging from the plate 2 do not change when, by turning the pair of plates about the point of incidence C of the incident beam (axis of rotation at right angles to the plane of incidence), the angle of incidence is at the same time changed in such a way that the product b=2d.sin a, corresponding to the beam spacing, remains constant. The arrangement has this property if D or At is varied by displacing the point A along the straight lines through A and B, and the triangle ABC fulfils the following conditions: a) the straight line through A and C passes through the plates 2 and 3 at right angles thereto at precisely these fixed points (i.e. AC is normal to the reflecting surface) b) path BC is fixed with respect to length and position) c) ABC=90" d) AB reamins parallel to incident beam.

Fig. 2 shows, diagrammatially, the mechanical construction for realising the triangle ABC of Fig. 1 having the above-mentioned properties. The designations of the points A, B, C of the triangle correspond to one another in Fig.

1 and Fig. 2.

Furthermore, Fig. 2 shows a fixed toothed rack 4 having a scale for reading off the distance AB which is calibrated to 2 AB At= c A tigid tube (or rod) 5 is connected to the toothed rack 4 by a spacer 6 which is normally fixed but which may be adjustable for adjusting the distance b BC=-.

2 The tube 5 is rotatable about the point C and is longitudinally movable in a guide sleeve 8 which is in turn secured to a sliding carriage 7, displaceable along the toothed rack 4, and which is rotatable about the point A. The two plates 2 and 3 are secured parallel to one another and at right angles to the plane of the drawing in such a way that their mutually facing surfaces contain the point C and A respectively. Point A must be sufficiently close to that edge of the plate 3 which faces the incident beam to ensure that, with the given distance BC the plate 3 does not shade the incident beam. On the other hand, the beam component reflected at C muct impinge entirely on the plate 3.

An arrangement for test beam spectroscopy is illustrated diagrammatically in Fig. 3. The test beams are focussed in the excitation region of the sample 10 by means of an optical system 9 (without aperture- and chromatic-aberrations). The distance b must be sufficiently large to ensure that the individual bundles of rays do not overlap, and the latter must be parallel bundles to a good approximation (only slight beam divergence). As a result of the self-diffraction of the light bundle, this requirement constitutes a lower limit to the crosssection of the beam and an upper limit to the maximum time delay. More accurate quantitative data also depend upon, for example, the quality of the beam, the required number of beams and the aperture of the focussing system 9.

An optical system 11 makes the adjacent bundles once more parallel. The lenses L, to L5 for focussing the individual beams onto the receivers E1 to E5 may be integrated in the system 11. Alternatively, the receiver line of an optical multi-channel analyzer may be used instead of the individual receivers. The optical system 11 must then be replaced by a focussing objective. In this case, the receiver line must be positioned sufficiently far outside the objective focal plane to ensure that it is illuminated to an optimum extent by the test bundles still substantially free from overlap.

Claims (3)

1. An apparatus for producing pulse trains of ultrashort light pulses of selectable repetition frequency, with parallel non-collinear beaming of individual pulses derived from an ultrashort light pulse, comprising two plates disposed parallel to one another, one of which plates is partially reflective, and a mechanism for maintaining a fixed mutual spacing of the beams, said mechanism being such that, simultaneously with a change in the spacing between the two plates necessary for changing the pulse repetition frequency, the two plates are turned commonly through a corresponding angle with respect to the direction of the beam.

2. Apparatus as claimed in claim 1, in which said mechanism comprises a rod member pivoted at a first pivotal axis about a fixed support and having one of said plates secured thereto so that the first pivotal axis lies in the plane of said one plate, a sleeve slidable on said rod and having the other plate attached thereto so that the two plates remain parallel to one another, and a slider slidable along a fixed slideway along a line spaced from said first pivotal axis, said sleeve being pivoted to said slider at a second pivotal axis which lies in the plane of said other plate.

3. Apparatus for producing pulse trains, constructed and adapted to operate substantially as heein described with reference to and as illustrated in the accompanying drawings.