1. Introduction
Increasing the contrast ratio between a laser pulse and any residual amplified radiation has important implications for a wide range of applications. Many ultra-high intensity laser matter interaction experiments require the target to be at solid density at the arrival of the laser pulse. This is particularly true for ultrathin foil targets used in laser acceleration schemes. Any pre-pulse preceding the main pulse may interfere with this requirement due to pre-plasma formation. Prepulses originating from regenerative amplifiers or from amplified spontaneous emission (ASE) are a common feature of high intensity laser experiments. In laser laser-solid target interactions the laser intensities may be well above
, extremely high contrast ratios for the investigation of laser-matter interactions are demanded in order to prevent target ablation prior to the arrival of the main pulse. Several methods have been put to use for this purpose: second harmonic generation after pulse compression [
1Y. Huang, C. Zhang, Y. Xu, D. Li, Y. Leng, R. Li, and Z. Xu, “Ultrashort pulse temporal contrast enhancement based on noncollinear optical-parametric amplification,” Opt. Lett. 36(6), 781–783 (2011). [CrossRef] [PubMed]
], polarization wave generation (XPW) [
2A. Jullien, O. Albert, F. Burgy, G. Hamoniaux, J. P. Rousseau, J. P. Chambaret, F. Augé-Rochereau, G. Chériaux, J. Etchepare, N. Minkovski, and S. M. Saltiel, “10-10 temporal contrast for femtosecond ultraintense lasers by cross-polarized wave generation,” Opt. Lett. 30(8), 920–922 (2005). [CrossRef] [PubMed]
], plasma mirrors [
3S. Backus, H. C. Kapteyn, M. M. Murnane, D. M. Gold, H. Nathel, and W. White, “Prepulse suppression for high-energy ultrashort pulses using self-induced plasma shuttering from a fluid target,” Opt. Lett. 18(2), 134–136 (1993). [CrossRef] [PubMed]
], and relativistic plasma shutters [
4S. A. Reed, T. Matsuoka, S. Bulanov, M. Tampo, V. Chvykov, G. Kalintchenko, P. Rousseau, V. Yanovsky, R. Kodama, D. W. Litzenberg, K. Krushelnick, and A. Maksimchuk, “Relativistic plasma shutter for ultraintense laser pulses,” Appl. Phys. Lett. 94(20), 201117 (2009). [CrossRef] [PubMed]
]. These methods have several drawbacks: low transmission, beam dispersion, and low durability which, for example, makes plasma mirrors unsuitable for high repetition systems. Moreover, reaching extremely high intensities requires tight focusing, for which the common approach is to use a large radius laser beam with conventional optical components such as mirrors. For focusing intense ultra short pulses, the intensity on the optics must stay below the damage threshold of these components. Thus this approach limits the geometrical characteristics of the focused beam such as the focal spot size and focal length.
In this work we demonstrate the use of a plasma lens that is almost transparent to a high-intensity laser pulse, but disperses the low-intensity pre-pulses and does not have the focusing limitations of conventional optics for high power lasers. Compared with conventional optics, plasma lenses have a substantially higher damage threshold [
5C. B. Schaffer, A. Brodeur, and E. Mazur, “Laser-induced breakdown and damage in bulk transparent materials induced by tightly focused femtosecond laser pulses,” Meas. Sci. Technol. 12(11), 1784–1794 (2001). [CrossRef]
]. We have recently demonstrated focusing of a laser pulse by a plasma channel microlens with a hollow plasma density profile [
6Y. Katzir, S. Eisenmann, Y. Ferber, A. Zigler, and R. F. Hubbard, “A plasma microlens for ultrashort high power lasers,” Appl. Phys. Lett. 95(3), 031101 (2009). [CrossRef]
]. A plasma channel lens is a short cylindrical column of plasma with a density minimum on its axis. The experimental results were consistent with analytical and simulation estimates of the focal length of a plasma channel lens [
7R. F. Hubbard, B. Hafizi, A. Ting, D. Kaganovich, P. Sprangle, and A. Zigler, “High intensity focusing of laser pulses using a short plasma channel lens,” Phys. Plasmas 9(4), 1431–1442 (2002). [CrossRef]
]. Moreover, their focusing properties can be tailored to the incident laser intensity by exploiting the ionization properties of the background gas, as demonstrated here.
Section 2 discusses the theoretical basis for the plasma lens and presents simulations of the experiment. Section 3 discusses the experimental results. Section 4 presents our conclusions.
2. Theoretical basis and simulation
The refractive index of a plasma is given by the expression and is mostly determined by the electron density. Here, is the plasma frequency, is the laser angular frequency, is the plasma electron density is the electron mass and is the magnitude of the electron charge. The transverse variation of the electron density determines the focusing properties of the plasma by introducing a transverse variation in . In the absence of such a variation, the laser pulse will expand over a characteristic distance termed the Rayleigh length, where is the laser spot size, and is the wavelength. As demonstrated in numerous experiments, a plasma channel with a density minimum on axis can guide a laser pulse over many Rayleigh lengths.
In addition, a short plasma channel can in principle act as a microlens with a well- defined focal length. For an axially symmetric parabolic density variation, i.e.,
where
is the plasma frequency at the center of the channel
and
is the characteristic channel radius, the focal length
of a plasma channel lens can be calculated analytically and is given by Eq. (10) in [
7R. F. Hubbard, B. Hafizi, A. Ting, D. Kaganovich, P. Sprangle, and A. Zigler, “High intensity focusing of laser pulses using a short plasma channel lens,” Phys. Plasmas 9(4), 1431–1442 (2002). [CrossRef]
]. The focal length scales with the Rayleigh length of the laser pulse at when it enters the plasma channel lens and is a function of the on-axis plasma density, the axial thickness of the lens plasma
and the characteristic channel radius. Equation (11) of the same reference provides an analytical estimate of the focal spot size
which scales with the initial laser spot size
These analytical estimates neglect relativistic self-focusing in the lens plasma and assume that
The plasma lens electron density is a function of propagating laser intensity due to ionization; hence the lens can be made to affect low-intensity (pre-pulse) and high-intensity components of a laser pulse differently. In our experiment, a plasma lens was created using a low power nanosecond laser to ablate the surface of a Boron-Nitride [
8Y. Hirayama and M. Obara, “Ablation characteristics of cubic-boron nitride ceramic with femtosecond and picosecond laser pulses,” J. Appl. Phys. 90(12), 6447–6450 (2001). [CrossRef]
] (BN) composite disk with a hole drilled in the center, the resulting spatial plasma profile has a hollow profile with low density plasma nearer the center Such a microlens has the advantage that proper timing between the creation of the lens and main beam, allows for control over the focal length. During the interaction of a sufficiently intense femtosecond laser the Boron and Nitrogen ions previously generated by the ablating laser [
6Y. Katzir, S. Eisenmann, Y. Ferber, A. Zigler, and R. F. Hubbard, “A plasma microlens for ultrashort high power lasers,” Appl. Phys. Lett. 95(3), 031101 (2009). [CrossRef]
], will go through further ionization. Considering these ions electron configurations and ionization potentials, the low and high intensity components of the pulse propagate through different plasma density profiles and accordingly, have different focusing properties. For intensities below the multiphoton ionization threshold the original plasma density is unaffected, and the focusing properties of the plasma lens are preserved. Thus, with proper choice of plasma lens parameters, the relatively low-intensity pre-pulse is strongly focused by the lens and then rapidly disperses, causing significant intensity reduction on the target placed at distance of several Rayleigh lengths from the plasma lens. Upon arrival of the high-intensity main pulse, the hollow profile of the plasma lens is almost instantaneously flattened by the leading edge of the main pulse. For laser intensities below
the ionization process will stop once all the electrons, except the last 1s
2 electrons, will be removed. Thus, the plasma lens becomes a layer with a relatively uniform transverse profile (infinite focal length) that is almost transparent to the main laser pulse. The concept is illustrated in
Fig. 1.
Fig. 1 Schematic representation of differences in propagation between the prepulse and main pulse. The prepulse is strongly focused by the plasma lens and then disperses while leaving the plasma distribution unaffected. As the main pulse arrives it further ionizes the plasma and neutral gas, thus destroying the plasma lens and propagating mostly unaffected.
We simulated the propagation of the prepulse and main laser pulse through a plasma lens using the HELCAP code [
9P. Sprangle, J. R. Peñano, and B. Hafizi, “Propagation of intense short laser pulses in the atmosphere,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 66(4), 046418 (2002). [CrossRef] [PubMed]
,
10J. R. Peñano, P. Sprangle, B. Hafizi, A. Ting, D. F. Gordon, and C. A. Kapetanakos, “Propagation of ultra-short, intense laser pulses in air,” Phys. Plasmas 11(5), 2865–2874 (2004). [CrossRef]
] for parameters similar to our experimental set up. The plasma lens was modeled as a parabolic transverse density variation as shown in
Fig. 2(b) (thin solid curve) with a longitudinal extent of
a plasma density of
at the center of the lens and
The pre-pulse and main pulse were initially collimated and taken to have Gaussian transverse and longitudinal profiles with spot size radius of
The pulse durations of the main pulse and pre-pulse were
In our simulations we assume that the origin of the prepulse is the leakage from regenerative amplifier, the duration of this prepulse is about the same as the main pulse but it arrives about 10nsec before. In order to reduce the computation time of the simulation the two pulses were separated temporally by
The peak intensity of the main pulse was
while the pre-pulse had a factor of
lower intensity. The laser wavelength was
and the total laser energy was
Multiphoton ionization was assumed to be the dominant ionization mechanism for the boron nitride gas. In addition to the hollow plasma, it was assumed that the ablation process produces a uniform gas fill.
Fig. 2 (a) Simulated intensity contours as a function of time and transverse coordinate at (b) Transverse profile of electron density at (lighter solid curve) just before the arrival of the laser pulses, (dashed curve) at the leading edge of the main pulse, and (short-dashed curve) at the peak intensity of the main pulse. The transverse intensity profile of the main pulse is denoted by the thicker solid curve.
The laser pulses were propagated a distance of
corresponding to the distance after the lens at which the pulses were imaged experimentally.
Figure 2(a) shows the electron density contours at
as a function of time and transverse coordinate. It is seen that the plasma channel is present for
i.e., before the arrival of the main laser pulse, but is completely filled in by the ionization process early in the main pulse.
Figure 2(b) shows the transverse profile of electron density at
(lighter solid curve) just before the arrival of the laser pulses,
(dashed curve) at the leading edge of the main pulse, and
(short-dashed curve) at the peak intensity of the main pulse. The transverse intensity profile of the main pulse is denoted by the thicker solid curve.
Figure 3 displays the intensity profiles of the two laser pulses at
(panel a) before the plasma lens,
(panel b) near the focus of the pre-pulse, and at
(panel b) after the plasma lens.
Fig. 3 Simulated intensity contours as a function of time and transverse coordinate at (a) , (b) , and (c) for a laser pulse and pre-pulse passing through a plasma lens. (d) On-axis temporal profile of intensity at (dashed curve) and showing the reduction in pre-pulse amplitude.
The pre-pulse is observed to undergo focusing by the plasma lens with a focal length of
and is defocused and reduced in intensity when the pulses arrive at the imaging plane
. The main pulse is distorted by ionization on its leading edge but remains relatively intact.
Figure 3(d) displays the on-axis intensity profiles corresponding to panels (a) and (c) and shows the pre-pulse intensity reduced by a factor of
at the imaging plane.
3. Experiment
In our experiment, a BN disk with a hole drilled in its center is irradiated by a single Nd:YAG laser pulse (
,
,
FWHM) to create the initial plasma lens. The disk has a
outer diameter and a
thickness. A capillary hole with a
diameter is drilled through the center of the disk (
Fig. 4).
Fig. 4 Experimental setup. Two photodiodes were used to measure the contrast ratio: photodiode 1 was set before the plasma lens, measuring a leak from an input mirror; photodiode 2 was placed after the plasma lens measuring a reflection off a beam splitter. Each photodiode output was divided into two channels with two different scales, depicting the prepulse and main pulse intensities, thus determining their ratio.
The Nd:YAG laser pulse was focused such that it will ablate a diameter of
around the capillary entrance in order to ensure uniform fluence at the perimeter of the drilled hole. All measurements were done in a vacuum chamber using backing pressure of
Torr. To characterize the plasma lens temporal and spatial characteristics, we measured the plasma radial density profile along the capillary by imaging the capillary exit along a slit at the entrance of a spectrometer, such that only a thin strip, one diameter long, would pass at a
magnification [
11T. Palchan, D. Kaganovich, P. Sasorov, P. Sprangle, C. Ting, and A. Zigler, “Electron density in low density capillary plasma channel,” Appl. Phys. Lett. 90(6), 061501 (2007). [CrossRef]
]. The spectrum was captured by a fast gated (
) ICCD camera (Andor Tech., DH520). 1D hydro code “Hyades” simulation for this scenario showed typical plasmas temperature of
which determines that the Stark line broadening is predominant over all other broadening effects. By measuring the broadening of doubly ionized Boron emission lines
[
12R. H. Huddlestone and S. L. Leonard, Plasma Diagnostic Techniques (Academic Press, 1965).
,
13H. R. Griem, Plasma Spectroscopy (McGraw-Hill, 1964).
] we were able to determine the spatial plasma density distribution at various times.
Evaluation of plasma lens focusing characteristics, after passage of the ablating pulse, was conducted using a Ti:Sapphire laser pulse ( central wavelength). Plasma measurements were made with and without the main beam in order to explore the plasma density distribution differences in the ablated plasma. The main beam was focused on the capillary entrance and then imaged to determine its shape and radius. In order to observe the effect of the plasma microlens on the pre-pulse, we deliberately readjusted the contrast ratio between the pre-pulse, which originated in the regenerative amplifier before the main pulse, to so it would be within the dynamic range by a fast photodiode.
We first characterized the operation of the plasma lens at low power using a pulse train from an Ti:Sapphire oscillator. Each pulse had a duration of and an energy of The beam was imaged into the fast gated () ICCD camera. The Ti:Sapphire pulse energy was simultaneously measured after the lens by a fast photodiode. The measurement of the lens transmission revealed that the conditions are optimal for focusing after the ablating laser reaches the capillary surface. At this delay time, the plasma density profile has a low minimum value on axis
We then propagated the high-intensity main laser pulse (with a pre-pulse) and measured the plasma distribution after passage of the main pulse. This measurement revealed that the transverse variation of the plasma density, and hence the focusing properties of the plasma lens was eliminated, presumably due to further ionization by the main pulse (
Fig. 5).
Fig. 5 Measured plasma density before arrival of the main pulse (red) shows the hollow profile corresponding to different optical path lengths as a function of distance from capillary center causing focusing at low intensities. Plasma density after arrival of the main pulse (blue) showing a smoothing the hollow profile.
In order to compare the contrast ratio before and after passage through the plasma lens the photodiodes output were divided into two different input sockets of an oscilloscope. Each input was set to measure a different scale corresponding to sought intensity. Thus out of a single pulse 4 measurements were made: pre pulse and main pulse before and after the plasma lens. According to these measurements the contrast ratios were calculated. Measurement of the pre-pulse contrast ratio after passing through the plasma lens showed an order of magnitude increase compared to before the plasma lens (
Fig. 6).
Fig. 6 Prepulse suppression. Blue line shows contrast ratio in the absence of the plasma lens. Green line shows contrast ratio with plasma lens. The contrast ratio increases an order of magnitude when propagating through the lens.
Imaging the beam after propagation through the plasma lens, showed that the prepulse is focused by it and expands after the focal point to an extent that is undetectable by the imaging camera; whereas the main pulse is passes mostly unaffected. An image of the pulse shows a dramatic increase in diameter of the low intensity components, while at the same time, slight focusing of the high intensity component (
Fig. 7).
Fig. 7 (a)Image of high intensity laser pulse without plasma lens (PL). (b) high intensity pulse with PL and ND filter of OD = 1.5 added. c)low intensity without PL. d) low intensity with PL.
Energy transmittance was measured to be The disk shows an erosion of per shot, so that a millimeter thick disk can endure around a hundred shots before needing replacement.
4. Conclusions
We have experimentally demonstrated the use of a plasma lens to significantly reduce pre-pulse of a high-intensity ultra-short laser pulse while transmitting > 90% of the laser pulse energy. These results can be interpreted through the following mechanism: the relatively low intensity pre-pulse is focused by the plasma microlens and later defocused. In this stage the focal length is less than
The pre-pulses intensity is too low to affect the electron density, therefore leaving the plasma microlens intact. As the main pulse arrives, the intensity of its leading edge is high enough for further ionization, thus smoothing the plasma distribution and effectively reducing or eliminating the plasma lens focusing effect. As shown in
Fig. 5, this effect is restrained, i.e. a defocusing lens is not created by a peak in plasma density at the center of the capillary. The restraining effect can be explained by the structure of Boron electron configurations: the first three (2p and 2s
2) electrons can be removed by the main pulse laser, where as the remaining 1s
2 electron requires an intensity exceeding that of the laser (same explanation goes for the Nitrogen). This creates a different divergence for the pre-pulse and the main pulse, allowing for focusing and divergence of the pre-pulse before the desired target and eliminating any residual radiation interaction before arrival of the main pulse. These effects are illustrated in the simulation results shown in
Figs. 2 and
3. Our experimental results support this interpretation. The plasma density measured before the arrival of the main pulse shows the presence of a plasma channel, while after the passage of the main pulse, the plasma density is relatively uniform. Correspondingly, the prepulse is observed to be substantially suppressed, which the main pulse propagates through relatively unaffected. An alternate explanation can be attributed to fast plasma shuttering . The plasma ablated from the walls moves toward the capillary center and generates a parabolic electron density profile on axis of the capillary, causing strong focusing of the prepulse. By the time that the main pulse arrives (after 12.5nsec) the electron density non-uniformity has reduced considerably allowing the main pulse to pass almost undisturbed.
Either of these mechanisms can be utilized for pre-pulse suppression for high intensity laser-matter interactions and can be viewed as an ultrafast dynamic optical switch. This switch reduces the fluence of the pre-pulse reaching the target while leaving the main pulse mostly unaffected.
In principle, the plasma lens can be tuned by controlling the electron density, making it possible to move the position of the focus without physically moving any optical component. In addition, the high durability and slow deterioration of BN makes it suitable for high repetition laser systems.