WO2005109083A2

WO2005109083A2 - Optical image exposing method and apparatus

Info

Publication number: WO2005109083A2
Application number: PCT/DK2005/000304
Authority: WO
Inventors: Niels-Søren BØGH; Brian Andersen; Hans Peter Ballegaard
Original assignee: Esko-Graphics A/S
Priority date: 2004-05-06
Filing date: 2005-05-04
Publication date: 2005-11-17
Also published as: WO2005109083A3

Abstract

A modulator device for modulating a pulsed laser beam in response to a modulation signal, the pulsed laser beam comprising a sequence of laser pulses at a predetermined pulse rate, the modulator device comprising a modulation medium and a driver circuit for applying a control signal to the modulation medium. The driver circuit is adapted to receive the modulation signal and a clock signal indicative of the pulse rate of the laser beam and to generate the control signal as a pulsed signal having a frequency determined from the clock signal, wherein the pulsed control signal is modulated by the modulation signal.

Description

Optical image exposing method and apparatus

The present invention relates to a modulator device and a method for modulating a pulsed laser beam.

It is known to modulate a laser beam by means of an acousto-optic modulator (AOM), i.e. by a device comprising a medium with a refractive index that can be controlled by an acoustic wave. Acousto optical modulators have numerous technical applications such as laser printers and optical spectrum analyzers. The acousto-optic effect occurs when a light beam passes through a transparent material, such as glass, in which travelling acoustic waves are also present. Acoustic waves are generated in the glass by a piezoelectric transducer that is driven by a RF signal source. The spatially periodic density variations in the glass corresponding to compressions and rarefactions of the travelling acoustic wave are accompanied by corresponding changes in the index of refraction for propagation of light in the medium. These travelling waves of index of refraction variation diffract the incident light much as the atomic planes of a crystal diffract x-rays in Bragg scattering. For acoustic waves of sufficiently high power, most of the light incident on the acousto-optic modulator can be diffracted and therefore deflected from its incident direction.

In prior art modulators, the transducer generates a carrier wave inside the material which carrier wave is turned on and off according to a modulation signal. Hence, dependant on the modulation signal, the laser beam propagating through the material experiences a different refractive index causing a diffraction of the laser beam responsive to the modulation signal.

However, it is a problem of the prior art modulators that the modulation speed of such a modulator is limited by the rise and fall times of the acoustic wave, which in turn is governed by the time required for an acoustic wave packet to propagate through the cross section of the laser beam. A reduction of the cross section of the laser beam by focussing the beam in a small spot size reduces the rise time of the modulator. However such a reduction also increases the power density of the laser beam in the medium and, therefore, limits the maximum power of the laser beam that can be modulated without damaging the AOM medium.

Furthermore, the power induced by the acoustic waves in the modulating medium causes the medium to be heated up. In the prior art systems, the amount of temperature increase depends on the bit pattern to be imaged, i.e. on the ratio of exposed and not-exposed regions.

It is therefore a problem to provide a modulator that reduces the effect of the bit pattern to be imaged on the temperature increase of the medium of the modulator. It is further desirable to provide a modulator that can modulate high power laser beams at high frequencies.

The above and other problems are solved by a modulator device for modulating a pulsed laser beam according to a modulation signal, the pulsed laser beam comprising a sequence of laser pulses at a predetermined pulse rate, the modulator device comprising a modulation medium and a driver circuit for applying a control signal to the modulation medium, wherein the driver circuit is adapted to receive the modulation signal and a clock signal indicative of the pulse rate of the laser beam and to generate the control signal comprising signal pulses at a frequency determined from the clock signal; wherein the control signal is modulated by the modulation signal.

Hence, the laser beam comprises a sequence of laser pulses and the modulator device is controlled by means of a modulation induced in a pulsed control signal such that the relative timing of the laser pulses and the signal pulses control the modulation of the laser beam, thereby reducing the variability of the power induced into the medium and, thus, reducing the temperature changes of the modulating medium due to changes in the image pattern. The modulator device described herein avoids the limitations related to finite rise and fall times of the transducer signal and allows a modulation of high-power laser beams with ultra-short pulses without damaging the modulator.

In a preferred embodiment, the modulator device is an acousto-optic modulator comprising a modulating medium and a wave generator adapted to generate a modulating wave in the modulating medium in response to the control signal. It is an advantage of the acousto-optic modulator that the constant relative phase offset between the control signal and the laser pulses can easily be adjusted by mechanical adjustment of the position of the modulator relative to the laser beam. In alternative embodiments, other modulator devices may be used, such as an electro-optic modulator.

In one embodiment, the pulsed control signal has a phase shift relative to the clock signal; and the driver circuit is adapted to control the relative phase shift in response to the modulation signal. It is an advantage that the modulator receives a constant average power of the modulating signal, irrespective of the content of the modulation signal. Hence, the temperature of the material remains substantially constant, irrespective of the modulation signal.

In an alternative embodiment, the pulsed control signal is pulse width modulated responsive to the modulation signal. In one embodiment, the modulator device is an acousto-optic modulator comprising a modulating medium and a wave generator adapted to generate a modulating wave in the modulating medium; wherein the driver circuit is adapted to control the wave generator to generate the modulating wave as a carrier wave modulated by the control signal to comprise a sequence of wave pulses/packets; and wherein the driver circuit is adapted to control the pulse width of the wave pulses such that, responsive to the modulation signal, the laser pulses overlap with the wave pulses in the modulating medium.

Further preferred embodiments are disclosed in the dependant claims. The present invention can be implemented in different ways including the modulator device described above and in the following, different uses of the modulator device, and a method described in the following, each yielding one or more of the benefits and advantages described in connection with the modulator device, and each having one or more preferred embodiments corresponding to the preferred embodiments disclosed in connection with the modulator device.

In particular, the invention further relates to a method of modulating a pulsed laser beam in response to a modulation signal, the pulsed laser beam comprising a sequence of laser pulses at a predetermined pulse rate, the method comprising - receiving the modulation signal and a clock signal indicative of the pulse rate of the laser beam; - generating a control signal comprising signal pulses at a frequency determined from the clock signal; wherein the control signal is modulated by the modulation signal; - applying the control signal to a modulation medium.

The present invention further relates to an optical image exposing method.

In a known type of image setters a laser beam is directed onto a printing plate, thereby causing the photosensitive layer of the printing plate to be exposed or ablated due to a thermal interaction of the laser beam with the photosensitive medium of the printing plate. In this process, the energy of the laser beam induces a collision impact ionisation process in the photosensitive medium where an electron avalanche is initiated by the thermal energy from the laser beam. This type of process is also known as Joule process or a light-to-heat conversion process, as it is determined by the amount of energy transferred by the laser beam to the photosensitive material. Different printing plates are characterised by different energies per unit area required for an exposure due to the above process.

EP 1 154 629 discloses an exposure system of the internal drum type, in which a pulsed laser beam is employed to expose a sensitised material. However, it is a problem of the above prior art system that the light-to-heat conversion process is sensitive to the wavelength of the laser light and to the properties of the sensitised material. The sensitivity of a sensitised material of a printing plate is typically expressed by a sensitivity curve which shows the energy per unit area in mJ/cm² required for an exposure of the material vs. the wavelength of the light. In the graphical industry, different types of printing plates are employed that are sensitised to different wavelengths, such as ultra-violet printing plates, printing plates for visible light, printing plates for infrared light, etc. Hence, in such systems the wavelength of the laser beam has to match the sensitivity of the sensitised material.

The above and other problems are solved by an apparatus for imaging a pixel on a photosensitive medium with a laser beam, the apparatus comprising a pulsed laser source for generating a pulsed laser beam; and

an optical system for directing the pulsed laser beam towards a predetermined pixel position on the photosensitive medium;

wherein the apparatus is adapted to direct a laser pulse towards the pixel position with a power per unit area large enough to induce a multiphoton ionisation process in the photosensitive medium..

In particular, it has turned out that directing a laser pulse towards the pixel position with a power per unit area large enough to induce a multiphoton ionisation process in the photosensitive medium causes an exposure of essentially all types of printing plates, irrespective of the sensitivity curve of the printing plate. Consequently, the same light source may be used to print on all types of printing plates, because it has turned out that, in contrast to the Joule process exploited by prior art systems, the multiphoton ionisation process does not strongly depend on the wavelength.

It is often a problem for print shops to obtain the printing plates because of marked fluctuations. Further it is a problem that most image setters are limited to the use of a few types of plates. Therefore, image setters that can print on any plate are a considerable improvement in imaging technology.

It is a further advantage of the invention that the amount of heat energy transferred to the photosensitive medium is relatively small, thereby reducing the detrimental effects caused by a heating up of the printing plates, in particular cross-talk between neighbouring pixels. Hence, it is an advantage of the invention that it provides an improved sharpness of the pixel edges, thereby improving the printing quality.

It is a further advantage of the invention that the high power per unit area of the photosensitive medium only needs to be applied for a very short period of time, thereby reducing the effect of pixel deformations when the light beam scans over the photosensitive medium during the pixel exposure.

It is a further advantage of the invention that, in contrast to a Joule process, the multuiphoton ionisation process does not require a relatively high minimum amount of energy to be transferred to the sensitive medium, thereby reducing the required exposure times per pixel and, thus, increasing the printing productivity.

Multiphoton ionisation as such is a known process which has mainly been studied in connection with micro-engineering, such as drilling and cutting, as is e.g. described in X. Liu et al. "Laser ablation and micromachining with ultrashort laser pulses," IEEE Journal of quantum optics, vol. 33, no. 10, 1997. In contrast to Joule processes, the multiphoton ionisation is caused by a direct interaction of photons and atoms. In a Joule process, photons interact with free electrons in the sensitive material causing the electrons to obtain sufficient kinetic energy to ionise atoms by collision impact ionisation, thereby starting an avalanche of electrons. Multiphoton ionisation, on the other hand, is not a kinetic energy process but an interaction where photons are absorbed by atoms of the photosensitive material. If the photon density is sufficiently high, the atoms are ionised due to a multiphoton interaction where a number of photons are absorbed simultaneously. Hence, this process is not governed by the total thermal energy transferred by the laser beam, but it requires a sufficiently high power per unit area in order to ensure a sufficiently high photon density. In micromachining applications these effects have been studied at power densities of 10000 GW/cm² or more. However, it has turned out that an exposure of photosensitive materials may be achieved at considerably lower powers per unit area.

In particular, it has turned out that a power per unit area larger than 0.1 GW/cm², preferably larger than 0.5 GW/cm² is sufficient for the exposure of most types of printing plates. Furthermore, a power per unit area of 5 GW/cm² has turned out to be sufficient to selectively ablate the photosensitive layer of most printing plates without damaging the underlying substrate.

Hence, in a preferred embodiment, the power per unit area is between 0.1 GW/cm² and 10 GW/cm², preferably between 0.1 GW/cm² and 5GW/cm², more preferably between 0.1 GW/cm² and 2.5GW/cm², most preferably between 0.5 GW/cm² and 2GW/cm².

Preferably, the required power per unit area is provided by a pulsed laser with an ultra-short pulse length, wherein the pulse duration of a single pulse is smaller than 20 ps, preferably smaller than 10ps, most preferably smaller than 5ps. Consequently, due to the short pulse duration, the average power of the laser beam, the so-called quasi continuous-wave power P_qcw> may be kept small, preferably P_qCw= 1-5W or smaller, more preferably between 2W and 4W.

Since the multiphoton ionisation process is governed by the power per unit area rather than by the energy, it is sufficient to expose each pixel with a single ultrashort pulse, thereby further increasing the productivity of the apparatus.

Preferably, the apparatus comprises beam deflection means adapted to deflect the pulse laser beam in synchronism with the pulse rate of the laser source to cause each pixel position of the photosensitive medium to the exposed to only one laser pulse.

In a particularly preferred embodiment, the apparatus is an internal drum plate making apparatus. More preferably, the apparatus comprises: - a drum member defining a centre axis and having an internal surface for receiving a printing plate; - a pulsed laser source for injecting a pulsed laser beam into the drum member along said centre axis, the pulsed laser beam pulsing at a predetermined pulse rate ; - a deflector rotably mounted around the centre axis of the drum member and adapted to deflect the pulsed laser beam towards pixel positions along a circumferential scanning line on the internal surface of the drum member; - control means for controlling the deflector to rotate at a predetermined speed such that each pixel position along the circumferential scanning line is exposed by only one laser pulse.

It is an advantage of an internal drum image setter that the dwell time, i.e. the potential time during which a pixel can be printed due to the mechanical properties of the scanning device, is very short. Furthermore, the rotational speed of an internal drum system and the pulse duration and pulse repetition rate of available pulsed laser sources may be matched such that each pixel is exposed by a single ultrashort laser pulse with a sufficiently high power per unit area to induce a multiphoton ionisation. This results in a particularly efficient exposure system.

In another preferred embodiment, the laser source is adapted to generate an ultra-violet laser beam, preferably in the range of 177nm-390 nm, for example 355nm. As mentioned above, the multiphoton interaction is insensitive to the wavelength, thereby allowing exposure of different types of printing plates with an ultra-violet laser source, even printing plates that are sensitised to visible light, infrared light, or the like. It is an advantage that the energy per photon is high for ultra-violet light, thereby reducing the number of simultaneously absorbed photons required for a multiphoton absorption.

Hence, it is an advantage that - due to the wavelength insensitivity of the employed process - the choice of the laser wavelength can be optimised by choosing short wavelengths. Short wavelengths have a higher efficiency for the multiphoton ionization process. Furthermore, short wavelengths can be focused to smaller spot sizes, resulting in higher resolutions of the printed images. Furthermore, polygons and spinning mirror devices for the scanning system of image setters can be made smaller to obtain the same spot (pixel) sizes than for e.g. thermal wavelengths. This means that the polygons or mirror devices can spin faster due to their aerodynamic properties and, consequently, the productivity of the image setter can be raised and construction costs can be lowered.

Further preferred embodiments are disclosed in the dependant claims.

The present invention can be implemented in different ways including the apparatus described above and in the following, different uses of the apparatus, and a method described in the following, each yielding one or more of the benefits and advantages described in connection with the first- mentioned apparatus, and each having one or more preferred embodiments corresponding to the preferred embodiments disclosed in connection with the first-mentioned apparatus.

In particular, the invention further relates to a method of imaging a pixel on a photosensitive medium with a laser beam, the method comprising

directing a laser pulse of a pulsed laser beam towards a predetermined pixel position on the photosensitive medium; and

controlling the power per unit area of the laser pulse to be large enough to induce a multiphoton ionisation process in the photosensitive medium.

It has been realised by the inventor that an efficient exposure of the printing plate can be achieved by means of a multiphoton ionisation process. It has further been realised that, in order to achieve an exposure/ablation of the photosensitive medium, the relevant parameter to be controlled is the power per unit area of the laser pulse(s) on the photosensitive medium rather than the total energy per unit area transferred to the medium.

In a preferred embodiment, the method further comprises determining a lower threshold value of the power per unit area of the laser pulse, above which lower threshold value the laser pulse causes an exposure/ablation of the photosensitive medium; and controlling the power per unit area of the laser pulse to be higher than the determined lower threshold value.

It has turned out that each printing plate has a rather well-defined lower threshold value of the power per unit area where the multiphoton ionisation process sets in and results in an exposure/ablation of the photosensitive medium. Hence, this threshold value can be determined for a given type of printing plate, allowing an efficient subsequent control of the power per unit area. In another preferred embodiment the photosensitive medium has a layered structure comprising a substrate layer and a photosensitive layer; and wherein the method further comprises controlling the power per unit area of the laser pulse to be small enough to prevent a multiphoton ionisation process in the substrate layer.

It has further turned out that the multiphoton ionisation process can be accurately controlled to only affect selected layers of a layered structure. In particular, by selecting the power per unit area in a suitable interval between a lower and a higher threshold value, the photosensitive layer of a printing plate is exposed/ablated without damaging the underlying substrate, i.e. the multiphoton ionisation process is selectively induced in the photosensitive layer.

Accordingly, in a further preferred embodiment, the method further comprises determining a higher threshold value of the power per unit area of the laser pulse, above which higher threshold value the laser pulse causes an ablation of the substrate; and controlling the power per unit area of the laser pulse to be lower than the determined threshold value.

Further preferred embodiments are disclosed in the dependant claims.

The above and other aspects of the invention will be apparent and elucidated from the embodiments described in the following with reference to the drawing in which:

fig. 1 illustrates ionisation based on the Joule effect;

fig. 2 shows a sensitivity curve of a printing plate; fig. 3 illustrates the mechanism of multiphoton ionisation;

fig. 4 illustrates the exposure of a printing plate;

fig. 5 illustrates the different regimes of exposure of a printing plate;

fig. 6 illustrates the effect of the pulse duration on the print quality;

fig. 7 schematically shows a block diagram of an internal drum image setter utilising ultra short light pulses;

fig. 8 schematically shows an arrangement for generating a modulated laser beam;

fig. 9 illustrates the RF signal relative to the laser pulses; and

fig. 10 illustrates a pulse width modulation of the control signal for controlling a modulator device.

Fig. 1 illustrates ionisation based on the Joule effect. A print plate, generally designated 109, is typically built up as a layered structure comprising a substrate 106, e.g. anodised aluminium, on which one or more thin layers (typically a few nm thick) are disposed including one or more photosensitive layers 107, e.g. polymer layers. In prior art systems, an exposure of the printing plate typically is based on an ionisation of the photosensitive polymer layer caused by a Joule process. Fig. 1a schematically illustrates an atom 100 of the polymer layer 107 before interaction with the laser beam. The atom 100 includes a number of electrons, one of which is shown in fig. 1a and designated by reference numeral 103. The polymer layer 107 further includes a small number of free electrons 101 or loosely bound electrons. Upon successive interactions with photons 102 of the laser beam, the kinetic energy of these electrons 101 is gradually increased. If an electron with sufficiently high energy collides with an atom 100 of the polymer, the atom is ionised.

Fig. 1b shows the ionised atom 100 and the resulting two free electrons, i.e. the original free electron 101 and the electron 103 that was removed from the atom 100.

Hence, the Joule process is governed by the amount of energy from the laser beam that is transformed into kinetic energy of the electrons.

Fig. 2 shows a sensitivity curve of a printing plate. The curve 205 indicates the sensitivity of the printing plate for different wavelength. Hence, where the curve 205 has a maximum, the required energy fluence measured in mJ/cm² for exposing the printing plate is small and vice versa. In the present example, the printing material is sensitive mainly for light in the wavelength region 400nm to approx. 850 nm, i.e. for visible light and near-infrared light. For example, a print plate sensitised to approximately 830nm cannot be exposed by a 1064nm laser using a Joule process due to the wavelength dependency of the Joule process.

Fig. 3 illustrates the mechanism of multiphoton ionisation. In particular, fig. 3 shows a printing plate 109 with a photosensitive layer 107 on a substrate 108 as described above. An atom 300 of the photosensitive layer 107 is schematically shown with one of its bound electrons 303. The electron 303 may occupy one of a number of energy levels or valence bands 304, each with a corresponding energy. In multiphoton ionisation, the electron 303 is removed from the atom 300 by simultaneous absorption of multiple photons 302. The number m of simultaneous photons required for ionisation depends on the ionisation potential or band gap Uι and on the frequency vof the photons: m ≥ Uι/hv . Hence, multiphoton ionisation is an m-th order process requiring a simultaneous interaction of the atom with m photons. Therefore, the multiphoton ionisation has a low cross section, and multiphoton ionisation is negligible at low power densities and only contributes to the light-matter interaction at high power densities, where the photon density is sufficiently high.

It is an advantage of the multiphoton ionisation process that it allows processing of different types of printing plates. In an experiment by the inventor, the multiphoton ionization printing was performed with a 355nm, 10ps pulse laser with a quasi continuous-wave power of 3W (W_qcw). This system could print on any type printing plate, irrespective of whether the plate was optimised for UV, violet, visible or thermal light. Even some plates that should not be sensitive to 355nm at all have been exposed.

Multiphoton ionisation has previously been studied in the context of micro- machining (see e.g. X. Liu, ibid.), where metal is cut or drilled using pulsed lasers generating peak powers of 10¹³ W/cm² or higher and pulse lengths in the picosecond or femtosecond range. At these power levels, the material is ablated due to multiphoton ionisation even though only moderate energies are transferred during the short pulses.

However, in the context of print plate exposure, it is not desirable to destroy the print plate but rather to expose it, as will be illustrated with reference to fig. 4.

Fig. 4 illustrates the exposure of a printing plate. In particular, figs. 4a-c illustrate a printing plate with a substrate 106 and a photosensitive polymer layer 107 as described in connection with fig. 1. In a printing application, the photosensitive layer is locally exposed by directing a laser beam to selected pixel areas 408.

As illustrated by fig. 4a, an exposure of the printing plate is achieved by causing the polymer layer 107 to be hardened at the pixel areas 408. The hardening caused by the exposure with the laser beam has the effect that the polymer layer at the exposed pixel locations does not react with the development chemicals any longer. Hence, during subsequent development of the printing plate, the non-exposed areas of the polymer layer are removed due to a chemical reaction with the development chemical, while the exposed parts of the polymer layer remain intact, thereby resulting in an exposed image.

Another option of processing the printing plate is by selective ablation of the polymer layer, as illustrated by fig. 4b. Hence, in this mode, the polymer layer is ablated, i.e. removed, by the laser beam at the pixel areas 408. Consequently, the print plate can be used without subsequent development. It is important to notice that in this mode material is actually removed from the printing plate, thereby producing dust during the illumination by the laser beam. However, the presence of dust particles may be a problem to the application/apparatus and may also cause an environmental problem. It is an advantage of an ablation by the multiphoton process that it results in an evaporation of the material, thereby avoiding large dust particles.

Fig. 4c illustrates a situation where not only the polymer 107 but also the substrate 106 is destroyed by the laser ablation at the illuminated pixels 408 due to an illumination with too high a power density. In this regime, the material will be freed of all electrons and enter a state of plasma, causing the material to evaporate. Another effect of such an overexposure may be that the polymer is loosened form the substrate. In any case, these effects result in a destruction of the printing plate or at least in poor printing quality.

It has been realised by the inventor that the power of the laser beam may be controlled to enable an exposure of the printing plate by multiphoton ionisation without destroying the printing plate. This has been achieved by selecting a power of the laser beam per unit area that is sufficiently large to induce multiphoton ionisation of the photosensitive layer. On the other hand, it has been realised that the required power per unit area resulting in an exposure or selective ablation of the polymer layer is smaller than the power levels that would result in a destruction of the substrate. Hence, since it is the power per unit area which is decisive for the printing process by multiphoton ionisation rather than the energy which is decisive for a Joule process, it is important to accurately control the power per unit area in a printing device in order to obtain a useful result.

In particular, it has turned out that a power per unit area of between 0.1 GW/cm² and 2.5 GW/cm², preferably between 0.5 GW/cm² and 2 GW/cm², is sufficiently high for exposures of most types of printing plates by a multiphoton ionisation process, while 5 GW/cm² for ablations. It is understood that the exact value depends on the type of polymers or other like materials of the photosensitive layer.

In one embodiment, a pulsed laser is used with a quasi continuous-wave power (qcw) of 2-4 W_qcw, i.e. an average power over time of 2-4 W. The laser is pulsed with a repetition rate of 80-160MHz, depending on the length of the inside cavity crystal. Such a laser has a pulse peak power of 2-4kW.

Experiments have shown that peak powers in the range of 0.5kW to 1.5kW focussed in a spot having a diameter of 10μm (corresponding to 2540dpi) are sufficient to expose any type of printing plate by multiphoton ionisation process. Furthermore, a peak power of 4kW can selectively ablate the photosensitive layer of most plates.

Fig. 5 illustrates the different regimes of exposure illustrated in connection with fig. 4. In particular fig. 5 shows the different ranges of power per unit area P/A where the multiphoton ionisation can advantageously be utilised for printing applications. For small powers per unit area, i.e. in the range below a lower threshold value Thι_ indicated by reference numeral 540, multiphoton ionisation does not significantly contribute to the interaction of the laser beam with the printing plate. In this regime, the Joule process dominates and multiphoton ionisation can be neglected. In one embodiment, a pulsed laser is used with a quasi continuous-wave (qcw) power of 3 W_qcw, focussed on a 10 μm spot and with a pulse repetition rate of 160MHz. This corresponds to a pulse duration of 8-10 ps. In fig. 5, the approximate pulse durations t_p corresponding to the powers per unit area for a laser with the above specifications are included for comparison.

It is understood that, in order to obtain a sufficiently high photon density at moderate laser powers, the laser beam should be focussed on a small spot size, e.g. having a diameter no larger than 50μm, preferably no larger than

20μm, even more preferably no larger than 10μm.

It is further noted that the effective spot size of the exposed pixel may be controlled to be smaller than the spot size of the focal spot of the laser beam. Typically, the intensity of the laser beam is not uniformly distributed across the cross section of the laser beam but has a higher intensity in the centre. Hence, by controlling the intensity of the laser beam such that only the power per unit area in the centre of the laser beam is sufficiently high to induce multiphoton ionisation, the effective exposed area is smaller than the cross section of the laser beam. Consequently, the printing resolution may be further increased.

For powers per unit area between 0.1 GW/cm² and a higher threshold value for exposure, the laser beam causes an exposure of the printing plate by multiphoton ionisation as described in connection with fig. 4a. In fig. 5, this range is indicated by reference numeral 541. For a number of different types of printing plates the above interval has been found to be between

0.1 GW/cm² 2.5GW/cm², preferably between 0.5GW/cm² and 2GW/cm².

In a range 542, i.e. above 2 GW/cm² and below a higher threshold value Th_H for ablation of approximately 5 GW/cm², the photosensitive layer is ablated as described in connection with fig. 4b. It has been found by the inventors that it is possible to control the multiphoton ionisation process to cause a selective ablation of one or more predetermined layers of a layered structure, e.g. of a printing plate that consists of e.g. a water carrier improved aluminium (anodised) substrate on which at least a thin (e.g. some nm thin) polymer layer is provided. During such a selective ablation, only the polymer layer is ablated while leaving the substrate intact and still capable of charring water.

It is an advantage of the exposure/ablation by multiphoton ionisation that the total energy which is transferred to the printing plate may be kept comparatively low, thereby avoiding cross talk between neighbouring pixels due the thermal diffusion of the energy as well as a destruction of the edges of the individual pixels. These problems are solved by application of the multiphoton ionization printing process described herein, because the energy

- and thus thermal diffusion- is very low.

At higher powers per unit area, the printing plate is destroyed, as illustrated in connection with fig. 4c. This range is illustrated by reference numeral 543 in fig. 5.

As is illustrated by fig. 5, the Joule process and the multiphoton ionization process have an overlap zone between a lower and a higher threshold value corresponding to ranges 541 and 542, where the multiphoton ionization process starts to be dominating over the Joule process. The exact values of the threshold values depend on the photosensitive material and on the type of printing plate. However, for a number of frequently used types of printing plates, an exposure of the printing plates has been found in the preferred interval of 0.1 GW/cm² and 3GW/cm², more preferably between 0.5GW/cm² and 2.5GW/cm², most preferably between 0.5GW/cm² and 2GW/cm². In the time domain, i.e. in terms of pulse duration, this zone is found to be in the interval of pulse durations of approximately 1 ps to 20ps for a laser with a quasi continuous power of 3W and a wavelength of 355nm. It is understood that these values depend on the material which is exposed. In the above example, the material was a thin film polymer on an aluminium substrate. For pulses substantially smaller than 1 ps the multiphoton ionization process is the dominating process.

It is interesting to note that the ionization process of the material can be controlled quite accurately in the overlap zone. This is very important for printing applications, because it allows controlling the exposure vs. ablation effects as illustrated above.

Hence, for a given type of printing plate, the above threshold values may be determined by exposing the printing plate with laser pulses of different powers per unit area and according to a predetermined dot pattern. By subsequent examination of the dot pattern on the exposed and, if applicable, developed plates, the lower threshold value can be accurately determined as the lowest power per unit area at which a satisfactory printing result is achieved. Similarly, the higher threshold value, at which ablation of the substrate sets in, can accurately be determined by a test with different power values per unit area.

One should further notice that the required amount of energy from the laser source is a factor 30 or more smaller for the multiphoton ionisation process than what would be expected for an exposure by the Joule process, even when loss factors due to e.g. thermal diffusion, spontaneous emission, etc. are included. Experiments have shown that printing plates could be exposed by laser pulses with 1 GW/cm² and a pulse duration of 10ps, corresponding to an energy per unit area of 10mJ/cm², even though the printing plates were specified to require 300mJ/cm², 400mJ/cm² and even 800mJ/cm², respectively.

As illustrated in fig. 5, for a laser with a given quasi continuous-wave power, the power per unit area is increased when the pulse duration is decreased, thereby yielding a higher pulse peak power. This decrease of the pulse peak duration has a further positive effect on the printing quality, as will be illustrated with reference to fig. 6. Fig. 6 illustrates the effect of the pulse duration on the print quality. When the laser beam is scanned over the printing surface, the impact light moves across the printing surface of the photosensitive material while pixel data is transferred. In particular, the laser beam moves a certain distance during the exposure time required for exposing a single pixel, thereby resulting in an elliptic shape of the pixels. By using an ultrashort pulse this ellipticity of the pixel is no longer visible to the eye, because the pixel will become almost perfectly round. Fig. 6 schematically illustrates a pixel 612 generated with an exposure time in the microsecond range, a pixel 613 generated with an exposure time in the nanosecond range, and a pixel 614 generated with an exposure time in the picosecond range. Hence, when the laser beam is scanned in the direction indicated by arrow 611 , the pixel shape becomes increasingly elongated, the longer the exposure time.

This phenomenon of elliptic spots sets a limit for how good a resolution on the print can be obtained in the scan direction 611. This is due to the fact that pixels start to melt into each other. This means that a resolution of e.g. 204 lines per inch in the scan direction often is considered as very good. With the multiphoton process described herein, this limitation is overcome, as ultrashort pulses are sufficient for an exposure. Consequently, the same resolution is present in all directions.

Fig. 7 schematically shows a block diagram of an internal drum image setter utilising ultra short light pulses. The image setter comprises a laser source 720 that generates a pulsed laser beam 721. The laser beam 721 is directed through an acousto-optic modulator (AOM) 722 or other modulating device which modulates the laser beam 721 based on the image data to be printed. It is an advantage of the short pulse durations that the requirements for the length of the rise/fall times of the modulator are less strict than for known internal drum systems with dwell times of 5-10ns. Consequently, conventional AOMs may be used without the risk of damaging the AOM due to small focus spots even at high power levels. An advantageous control method and apparatus for the modulator will be described in greater detail below. The resulting modulated beam 723 is directed into a cylindrical drum 724 along the centre axis of the drum 724. A prism 725 is mounted inside the drum 724 such that the prism 725 deflects the incoming beam 723 towards the inner surface of the drum. A printing plate 735 to be exposed is mounted on the inner surface of the drum with the photosensitive layer facing inwardly towards the prism 725. Hence, the deflected laser beam 736 hits the printing plate at a predetermined pixel position 728. The prism 725 is mounted on a shaft 727 such that the prism is rotated around the centre axis by a motor 726, causing the deflected laser beam 736 to scan the surface of a printing plate in a main scanning direction, i.e. the illuminated pixel position 728 moves along a circumferential scan line.

The prism 725 is further mounted movably along the centre axis, thereby allowing a successive scanning of parallel scan lines. Hence, after a full rotation of the prism 725, the prism is moved along the centre axis of the drum such that during a subsequent rotation of the prism, the deflected laser beam 736 scans a different scan line parallel to the previous scan line.

The image data 729 to be printed is fed into an image signal processing circuit 730 which generates control signals 732 and 733 for controlling the acousto-optic modulator 722 and the motor 726, respectively. The image signal processing circuit 730 receives a master clock signal 731 from the laser source 720. Each master clock signal pulse (e.g. a TTL signal or the like) from the laser 720 corresponds to one laser pulse. Based on the image data 729 and triggered by the master clock signal 731 , the image signal processing circuit 730 determines whether the current laser pulse 721 should be transferred to the printing plate 735 via the AOM (e.g. if the image data is TTL High) or be discarded (e.g. if the image data is TTL Low), and controls the AOM accordingly. The image signal processing circuit 730 further controls the rotational speed of the prism 725 in synchronisation with the repetition rate of the master clock signal 731. This means that the laser repetition rate and the rotational speed of the prism is synchronised in order to obtain a given pixel resolution. The circuit 731 keeps further track of the direction of the rotation of the prism 725 and its position along the drum axis, as indicated by feedback signal 734. Hence, under the control of the circuit 730, the image setter transfers the image data to corresponding positions inside the drum.

It is an advantage of the above systems that it employs a single laser source. As mentioned above, it is desirable to control the power in the multiphoton ionization printing process, as small fluctuations in the power can be seen immediately by the eye. Therefore it is advantageous to use a scanning system for the image setter where all pixels are illuminated by the same light source, thereby providing a more uniform exposure.

Preferably, the laser source should have a predetermined constant or only slowly varying power. The pulse duration of the laser is preferably chosen as short as possible. Examples of suitable laser sources include solid-state diode-pumped lasers, in particular passively mode locked lasers, the Spectra Physics Vangaurd, Lumera UPL 20, 1064 (FCS 355), the Coherent Paladin 355 or like.

The typical specifications of these lasers is that they have a quasi continuous-wave power (qcw) of 2-4W_qcw. The radiated beam is typically pulsed with a repetition rate of 80-160MHz, depending on the length of the inside cavity crystal, and corresponding to a pulse peak power of 2-4kW. Furthermore, these types of lasers can be frequency tripled from 1064nm to 355nm. The beam quality facto M² is close to one, i.e. the laser beam can be focused to a very small spot.

In one embodiment an internal drum system is chosen, which has an internal drum circumference of 1.6m. In such an embodiment, a laser repetition rate of 80MHz and a rotational speed of the scanning prism/mirror of 30.000 rpm allow an exposure of each pixel with a single laser pulse. Similarly, a 160MHz repetition rate corresponds to a rotational speed of 60.000rpm. The corresponding printing speed is 160 million pixels per second.

In general, the repetition rate of the laser source becomes in correlation with the rotational speed of the image/prism determines how many laser pulses hit each pixel. It is preferred that this number is as low as possible in order to increase the number of exposed pixels per unit time.

It is noted that, in alternative embodiments wherein the laser is not passively mode locked, the master clock signal can be changed to come from the spinning prism, instead. Consequently, in this embodiment, the position of the prism determines the repetition rate of the laser source.

It is further understood that the multiphoton ionisation printing process described herein may also be used in connection with other types of image setters, such as a polygon system, and external drum systems or the like.

Fig. 8 schematically shows an arrangement for generating a modulated laser beam. The arrangement comprises a laser source 820 that generates a pulsed laser beam 821. The laser beam 821 is directed through an acousto- optic modulator (AOM) 822 that modulates the laser beam 821 based on a modulation signal 829. For example, the modulation signal may represent image data of a bit pattern to be printed, as described herein.

Hence, the output beam 823 is a pulsed modulated beam. In the printing process described herein the resulting pulsed modulated beam 823 is scanned over a photosensitive medium causing the medium to be exposed according to the modulation signal 829.

The laser source 820 may be any suitable pulsed laser source, e.g. a passively mode-locked laser, an actively mode-locked laser, a Q-switched laser, or the like. In a preferred embodiment, the laser pulses have a duration in the picoseconds range, e.g. between 0.1 and several hundred picoseconds. In particular, in the printing application described herein, it is preferred to operate with ultra-short laser pulses, e.g. in the range of 0.1 to 100ps, preferably 1-20ps, more preferably 1-12 ps. Examples of suitable laser sources include solid-state diode-pumped lasers, in particular passively mode locked lasers, the Spectra Physics Vangaurd, Lumera UPL 20, 1064 (FCS 355), the Coherent Paladin 355 or like. It is understood that the invention may also be worked with femtosecond lasers and attosecond lasers. It is expected that these types of lasers will become available in the future at reasonable prices and complexity.

The AOM 822 comprises a modulating material 840, such as a glass, onto which the laser beam 821 is focussed with a predetermined spot size as indicated by the dotted line 844. The AOM further comprises a piezo-electric transducer 841 attached to the material 840, such that acoustic waves packets 843 generated by the transducer propagate through the material 840 along a predetermined direction 842 and intersect the laser beam 821 inside the material 840. The acoustic wave packets 843 modify the refractive index of the material 840, thereby causing a diffraction of the laser beam 821 in response to the waveform of the acoustic wave generated by the transducer 841. Hence, the modulator outputs the output beam 823 or a diffracted beam 845, depending on the presence/absence of a wave packet in the region 844. For example, the beam 845 may be discarded and the beam 823 may be utilised as modulated output beam.

According to one embodiment, the acoustic-waves have an RF carrier frequency, e.g. 350 MHz, which is modulated by a constant modulation frequency resulting in a sequence of wave pulses. The modulation frequency corresponds to the repetition frequency of the pulsed laser beam, such that the duty cycle of the RF signal 832 is 50%. Typical repetition frequencies are 50-200 MHz, e.g. 80-160MHz. The modulation of the laser beam 821 is achieved by selectively imposing a phase shift on the RF signal 832 relative to the laser pulses and in response to the modulation signal 829, i.e. a phase shift of the wave pulses relative to the laser pulses. Consequently, the driver circuit 830 controlling the transducer 841 receives a master clock signal 831 indicative of the pulses of the pulsed laser beam 821. In the arrangement of fig. 8, the master clock signal 831 is generated by the laser source, e.g. a passive mode-locked laser. In other embodiments, the master clock signal may be generated by a different signal source, e.g. a control circuit controlling an active mode locked laser.

Hence according to this embodiment, the AOM is controlled by means of a phase shift induced in a carrier signal having a constant duty cycle, thereby avoiding the limitations related to finite rise and fall times of the transducer signal. This allows a modulation of high-power laser beams with ultra-short pulses without damaging the modulator.

It is a further advantage that the AOM material and the transducer receive a constant average power of the RF signal, irrespective of the content of the modulation signal. Hence, the temperature of the material and transducer remains substantially constant, irrespective of the modulation signal.

In an alternative embodiment, in particular an embodiment using an actively mode-locked laser, the phase relative phase shift between the acoustic wave and the laser pulses can be induced by phase shifting the laser pulses.

Fig. 9 illustrates the RF signal relative to the laser pulses. Curve 931 illustrates the master clock signal that corresponds to the pulsed laser beam. The master clock signal comprises an ON signal 949 at regular intervals corresponding to the repetition rate of the laser pulses. Curve 921 corresponds to the actual waveform of the laser beam that enters the AOM. The laser beam comprises a sequence of short laser pulses 959 at a rate corresponding to by the master clock signal 931. Curve 929 represents the modulation signal. In the example of fig. 9, the modulation signal is ON during time interval 951 and OFF during time intervals 950 and 952. As described above, the driver circuit receives the master clock signal 931 and the modulation signal 929 and generates an RF signal 932 comprising wave pulses 953, 954, and 955 of an RF frequency carrier wave, e.g. with a 350 MHz carrier frequency. The wave pulses are triggered by the master clock such that the RF signal 932 has a constant duty cycle, in this example a duty cycle of 50%.

When the modulation signal 929 is OFF, i.e. during intervals 950 and 952, the signal bursts 953 and 955 are in phase with the master clock pulses. When the modulation signal 929 is ON, i.e. during interval 951 , the driver circuit induces a 180 deg. phase shift Δ into the RF signal 932 relative to the master clock signal 931. Consequently, while the laser pulses 959 are in phase with the RF pulses 953 and 955, the laser pulses 959 are out-of phase with the pulses 954, thereby causing a diffraction of the laser pulses in the AOM material depending on the modulation signal 929. Curve 923 illustrates the waveform of the resulting laser beam with pulses 948 in accordance with the modulation signal 929.

It is understood that the laser pulses and the wave pulses 953 have a further relative phase in the region within the material where the laser pulses intersect the wave pulses. This additional phase is constant and depends on the geometry of the material and the position of the transducer.

In the following, an alternative embodiment is described, in which the AOM is controlled by means of a pulse width modulated signal.

According to this embodiment, a modulation device as described in connection with fig. 8 is used. As described above, the acoustic-waves generated in the modulation medium 840 have an RF carrier frequency, e.g. 350 MHz, which is modulated by a constant modulation frequency resulting in a sequence of wave pulses. The modulation frequency corresponds to the repetition frequency of the pulsed laser beam. Typical repetition frequencies are 50-200 MHz, e.g. 80-160MHz. However, according to this embodiment, the modulation of the laser beam 821 is achieved by selectively varying the pulse width of the wave pulses 843 between narrow pulses and broad pulses such that the laser pulses of the laser beam 821 only intersect/overlap with the broad wave pulses but not with the narrow pulses due to their relative phase shift, as will be described in greater detail with reference to fig. 10.

Fig. 10 illustrates a pulse width modulation of the control signal for controlling a modulator device. In particular, fig. 10 illustrates the RF signal relative to the laser pulses. Curve 931 illustrates the master clock signal that corresponds to the pulsed laser beam. As described in connection with fig. 9, the master clock signal comprises an ON signal 949 at regular intervals corresponding to the repetition rate of the laser pulses. Curve 921 corresponds to the actual waveform of the laser beam that enters the AOM. The laser beam comprises a sequence of short laser pulses 959 at a rate corresponding to by the master clock signal 931. Curve 929 represents the modulation signal. In the example of fig. 10, the modulation signal is ON during time interval 951 and OFF during time intervals 950 and 952.

According to this embodiment, the driver circuit receives the master clock signal 931 and the modulation signal 929 and generates an RF signal 1032 comprising wave pulses 1053, 1054, and 1055 of an RF frequency carrier wave, e.g. with a 350 MHz carrier frequency.

When the modulation signal 929 is OFF, i.e. during intervals 950 and 952, the signal bursts 1053 and 1055 have a width W and a relative phase with the master clock pulses 949 such that the laser pulses 959 and the signal bursts 1053 and 1055 overlap in time within a predetermined region of the modulation medium. As above, the laser pulses 959 and the wave pulses 1053, 1054, and 1055 may have a further relative phase in the region within the material where the laser pulses intersect the wave pulses, which is disregarded in fig. 10 for simplicity. This additional phase is constant and depends on the geometry of the material and the position of the transducer. When the modulation signal 929 is ON, i.e. during interval 951 , the driver circuit causes the wave pulses 1054 to have a shorter width w but the same relative phase as the pulses 1053, such that the laser pulses 959 do not interfere with the short pulses 1054. Consequently, while the laser pulses 959 interfere with the RF pulses 1053 and 1055, the laser pulses 959 do not coincide with the pulses 1054 due to their constant phase difference. As a result, the laser pulses 959 are diffracted in the AOM material depending on the modulation signal 929. Curve 923 illustrates the waveform of the resulting laser beam with pulses 948 in accordance with the modulation signal 929.

It is understood that, in the above embodiments, the modulator may be operated such that the laser pulses which interact with the wave pulses in the medium are utilised rather than the pulses that do not interfere. This would correspond to the reversal of the modulation signal 929 in figs. 9 and 10.

The acousto-optic modulator described herein can advantageously be used in a printing apparatus as described herein, since this printing apparatus utilises ultra-short laser products of high power.

Example:

Exposure tests have been carried out with a CDI Vangaurd external drum system and a rotational speed of 1350 rpm. The laser source was a Vangaurd pulsed UV (355nm) laser with a power of 4Wqcw (3Wqcw measured on the drum). The pulse width was 10-12ps and the repetition rate was 80MHz, resulting in peak powers in the range of some kWs. The laser was modulated by an AOM by means of which the output power of the laser could be scaled from OWqcw to 3Wqcw on plate. This scaling was done by changing the diffraction efficiency of the AOM. The first order beam from the AOM was used for the exposure. All exposure jobs were done in a resolution of 2540 dpi (10μm spot). The exposure adjustment was done by a power sweep with a 0-100% raster bitmap. The results were inspected by a dot-meter. From these measurements the optimal power for the 1350rpm rotation speed was selected. The final print job was corrected according to the dot measurements by a correction curve. The same print job was bone on all plates. The print plates were developed manually.

The following powers per unit area were found to provide a good exposure:

It is noted that the apparatus described herein may advantageously be applied to the exposure of printing plates used in the graphical industry, e.g. in a computer-to-plate (CtP) or computer-to-film (CtF) process.

It is further noted that the invention has been described mainly in connection of exposing/ablating printing plates having a layered structure. However, the printing by exposure or ablation through multiphoton ionisation as described herein may be applied to other types of printing plates, including Flexo, polyester, films etc.

Furthermore, the described multiphoton ionisation process may be used for other applications where a selective processing/ablation of one or more layers of a multilayer structure is desired, e.g. the selective removal of layers of a multilayer structure.

It is noted that the control means described herein can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed microprocessor, computer, or the like. In the device claims enumerating several control means, processing means, and the like, several of these means can be embodied by one and the same item of hardware, e.g. a suitably programmed microprocessor or computer. The mere fact that certain measures are recited in mutually different dependent claims or described in different embodiments does not indicate that a combination of these measures cannot be used to advantage.

It should be emphasized that the term "comprises/comprising" when used in this specification is taken to specify the presence of stated features, integers, steps or components but does not preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

Claims

CLAIMS:

1. A modulator device for modulating a pulsed laser beam in response to a modulation signal, the pulsed laser beam comprising a sequence of laser pulses at a predetermined pulse rate, the modulator device comprising a modulation medium and a driver circuit for applying a control signal to the modulation medium; wherein the driver circuit is adapted to receive the modulation signal and a clock signal indicative of the pulse rate of the laser beam and to generate the control signal comprising signal pulses at a frequency determined from the clock signal; wherein the control signal is modulated by the modulation signal.

2. A modulator device according to claim 1 , wherein the modulator device is an acousto-optic modulator comprising a modulating medium and a wave generator adapted to generate a modulating wave in the modulating medium; and wherein the driver circuit is adapted to control the wave generator to generate the modulating wave as a carrier wave modulated by the control signal.

3. A modulator device according to claim 2, wherein the wave generator is a piezo-electric transducer.

4. A modulator device according to any one of claims 1 through 3, wherein the control signal has a phase shift relative to the clock signal; and wherein the driver circuit is adapted to modulate the control signal by controlling the relative phase shift in response to the modulation signal.

5. A modulator device according to claim 4, wherein the modulator device is an acousto-optic modulator comprising a modulating medium and a wave generator adapted to generate a modulating wave in the modulating medium; wherein the driver circuit is adapted to control the wave generator to generate the modulating wave as a carrier wave modulated by the control signal to comprise a sequence of wave pulses; and wherein the driver circuit is adapted to control the phase shift such that the laser pulses overlap with the wave pulses in the modulating medium dependant on the modulation signal.

6. A modulator device according to any one of claims 4 through 5, wherein the driver circuit is adapted to generate the control signal with a constant average duty cycle.

7. A modulator device according to any one of claims 1 through 3, wherein the pulsed control signal is pulse width modulated responsive to the modulation signal.

8. A modulator device according to claim 7, wherein the modulator device is an acousto-optic modulator comprising a modulating medium and a wave generator adapted to generate a modulating wave in the modulating medium; wherein the driver circuit is adapted to control the wave generator to generate the modulating wave as a carrier wave modulated by the control signal to comprise a sequence of wave pulses; and wherein the driver circuit is adapted to control the pulse width of the wave pulses such that, responsive to the modulation signal, the laser pulses overlap with the wave pulses in the modulating medium.

9. A method of modulating a pulsed laser beam in response to a modulation signal, the pulsed laser beam comprising a sequence of laser pulses at a predetermined pulse rate, the method comprising - receiving the modulation signal and a clock signal indicative of the pulse rate of the laser beam; - generating a control signal comprising signal pulses at a frequency determined from the clock signal; wherein the control signal is modulated by the modulation signal. - applying the control signal to a modulation medium.

10. An apparatus for imaging a pixel on a photosensitive medium with a laser beam, the apparatus comprising a pulsed laser source for generating a pulsed laser beam; and

characterised in that the apparatus is adapted to direct a laser pulse towards the pixel position with a power per unit area large enough to induce a multiphoton ionisation process in the photosensitive medium.

11. An apparatus according to claim 10, wherein the power per unit area is larger than 0.1 GW/cm², preferably larger than 0.5 GW/cm².

12. An apparatus according to claim 11 , wherein the power per unit area is between 0.1 GW/cm² and 10 GW/cm², preferably between 0.5 GW/cm² and

5GW/CIT1², most preferably between 0.5 GW/cm² and 2GW/cm².

13. An apparatus according to any one of claims 10 through 12, wherein the apparatus id adapted to direct only one laser pulse to each pixel position.

14. An apparatus according to claim 13, wherein the apparatus comprises beam deflection means adapted to deflect the pulse laser beam in synchronism with the pulse rate of the laser source to cause each pixel position of the photosensitive medium to the exposed to only one laser pulse.

15. An apparatus according to any one of claims 10 through 14, wherein the apparatus is an internal drum plate making apparatus for the production of print plates.

16. An apparatus according to claim 15, comprising - a drum member defining a centre axis and having an internal surface for receiving a printing plate; - a pulsed laser source for injecting a pulsed laser beam into the drum member along said centre axis, the pulsed laser beam pulsing at a predetermined pulse rate ; - a deflector rotably mounted around the centre axis of the drum member and adapted to deflect the pulsed laser beam towards pixel positions along a circumferential scanning line on the internal surface of the drum member; - control means for controlling the deflector to rotate at a predetermined speed such that each pixel position along the circumferential scanning line is exposed by only one laser pulse.

17. An apparatus according to any one of claims 10 through 16, wherein the apparatus is adapted to focus the pulsed laser beam onto a spot with a diameter of 1 - 50 μm, preferably 5-20 μm, most preferably 10μm.

18. An apparatus according to claim 16, wherein the peak power of the laser pulse is between 0.1 kW and 10kW, preferably between 0.5KW and 5kW, more preferably between 0.5kW and 4kW.

19. An apparatus according to claim 18, wherein the peak power of the laser pulse is between 0.5 and 1.5 kW.

20. An apparatus according to any one of claims 10 through 19, wherein the laser source has a quasi continuous wave power of P_qcw= 1-5W, preferably between 2W and 4W.

21. An apparatus according to any one of claims 10 through 20, wherein the pulse duration of a single pulse is smaller than 20 ps, preferably smaller than 10ps, most preferably smaller than 5ps.

22. An apparatus according to any one of claims 10 through 21 , wherein the laser source is adapted to generate an ultra-violet laser beam, preferably in the range of 177nm-390nm, for example 355nm.

23. Use of the apparatus according to any one of claims 10 through 22 for the exposure of printing plates.

24. Use according to claim 23, wherein the wherein the laser source is adapted to generate an ultra-violet laser beam, preferably in the range of

177nm-390nm, for example 355nm.

25. Use according to claim 24, wherein the printing plate is a printing plate sensitive to visible light.

26. Use according to claim 23, wherein the printing plate is a printing plate sensitive to infrared light.

27. Use of a modulator device according to any one of claims 1 through 8 in an apparatus according to any one of claims 10 through 22.

28. A method of imaging a pixel on a photosensitive medium with a laser beam, the method comprising

29. A method according to claim 28, further comprising determining a lower threshold value of the power per unit area of the laser pulse, above which lower threshold value the laser pulse causes an exposure/ablation of the photosensitive medium; and controlling the power per unit area of the laser pulse to be higher than the determined lower threshold value.

30. A method according to claim 28 or 29, wherein the photosensitive medium has a layered structure comprising a substrate and a photosensitive layer; and wherein the method further comprises controlling the power per unit area of the laser pulse to be small enough to prevent a multiphoton ionisation process in the substrate layer.

31. A method according to claim 30, further comprising determining a higher threshold value of the power per unit area of the laser pulse, above which higher threshold value the laser pulse causes an ablation of the substrate; and controlling the power per unit area of the laser pulse to be lower than the determined threshold value.