Efficient generation of Xe K-shell x raysby high-contrast

interaction with submicrometer clustersYukio Hayashi,1,* Alexander S. Pirozhkov,1 Masaki Kando,1 Yuji Fukuda,1 Anatoly Faenov,2 Keigo Kawase,1

Tatiana Pikuz,2 Tatsufumi Nakamura,1 Hiromitsu Kiriyama,1 Hajime Okada,1 and Sergei V. Bulanov1,3

1Quantum Beam Science Directorate, Japan Atomic Energy Agency, Kizugawa, Kyoto 619-0215, Japan2Joint Institute for High Temperature of the Russian Academy of Sciences, Moscow 127412, Russia

3A. M. Prokhorov Institute of General Physics, Russian Academy of Sciences, Moscow 119991, Russia*Corresponding author: hayashi.yukio@jaea.go.jp

Received January 26, 2011; revised March 14, 2011; accepted March 24, 2011;posted March 30, 2011 (Doc. ID 141755); published April 27, 2011

The interaction between a 25TW laser and Xe clusters at a peak intensity of 1 × 1019 W=cm2 has been investigated. XeK-shell x rays, whose energies are approximately 30keV, were clearly observed with a hard x-ray CCD at 3:4MPa.Moreover, we studied the yield of the Xe K-shell x rays by changing the pulse duration of the laser at a constant laserenergy and found that the pulse duration of 40 fs is better than that of 300 fs for generating Xe K-shell x rays. © 2011Optical Society of AmericaOCIS codes: 320.7110, 340.7480, 350.5400.

Laser-plasma x-ray sources are becoming promisingtools for various fields because of their compactness,high brightness, short pulse duration, spatial coherence,and so forth. By focusing a short-pulse laser onto a target,photons at energies ranging from hundreds of eV to keVare relatively easily obtained from a micrometer-sizedsource.One of the applications of laser-plasma x rays is x-ray

imaging [1–4]. Using laser-plasma x rays, Gordon et al.performed time-gated imaging [1] and Tillman et al. de-monstrated biological imaging [2] about 15 years ago. Re-cently, Chen et al. have shown a phase contract imagingof a spider [3], and Fukuda et al. have carried out nano-structural imaging with a lithium fluoride detector [4].For imaging of thick samples or medical applications,several tens of keV photons are required. Such x rayswere observed using a metal target. In metal targets, deb-ris from the irradiated target contaminates focusingoptics, and this issue makes the laser-plasma x raysgenerated from metal targets difficult for practicalapplications.As an alternative, cluster targets are being studied

because they have gaslike and high-density-like solidcharacteristics. X rays radiating from K-, L- and M-shellvacancies in ions have been measured from Ar [5] and Kr[6] clusters, and L- and M-shell radiations have been mea-sured from Xe clusters [7,8]. Because the energies of XeK-shell x rays (∼30 keV) are higher than those of the light-er rare gases, Xe can be suitable for x-ray imaging of athick sample.In this Letter, we report the generation of Xe K-shell x

rays from laser–plasma interaction. We considered thatintense Xe K-shell x rays can be produced, if Xe clustersare not destroyed by the prepulse and interact with theintense main pulse. Therefore, we decided to utilize ahigh-contrast Ti:sapphire chirped-pulse amplification(CPA) laser system, i.e., the J-KAREN laser [9], for x-ray generation. The use of double CPA with an opticalparametric chirped-pulse preamplifier and Ti:sapphireamplifier leads to extremely high temporal contrast, over

1 × 10−11. This laser system was installed in the KansaiPhoton Science Institute of the Japan Atomic EnergyAgency. The experiment was carried out with the useof a specially designed nozzle and a photon-countingx-ray CCD.

Detailed experimental conditions and the characteris-tics of K-shell x rays will be presented. The laser, with∼1 J pulse energy, was focused on the clusters, andthe focal spot diameter (1=e2) was 18μm × 36 μm. Themaximum value of the peak laser irradiance in ourexperiment was 1 × 1019 W=cm2 (average irradiance:4 × 1018 W=cm2), which corresponded to the laser pulseduration of 40 fs.

Since the irradiance of the prepulse was less than1 × 108 W=cm2, the Xe clusters survived until the main la-ser pulse arrived. The cluster nozzle was specially de-signed with fluid simulation to produce large-diameterclusters [10]. Our experiment was carried out at reservoirpressures of 3.4 and 4:1MPa, and that of 4:1MPa corre-sponds to the mean cluster diameter of approximately0:08 μm. An x-ray CCD was utilized to measure the XeK-shell x rays and bremsstrahlung x rays generated bythe laser-cluster interaction. The CCD was placed 35°from the laser propagation direction in a horizontal plane(xy plane) and 45° from a vertical axis (z direction). Thesolid angle of the x-ray CCD was 0:96 × 10−4 sr.

Figure 1 indicates the average spectrum of x-ray emis-sion from the laser-cluster interaction at the reservoirpressure of 3:4MPa. The peak laser irradiance was1 × 1019 W=cm2. To inhibit the interaction of multipho-tons, whose energies are less than 10 keV, with the singleCCD pixel, an Al 260 μm filter was placed in front of the x-ray CCD. The x-ray energy spectrum has a sharp peak inthe vicinity of 30 keV with a FWHM of 0:7 keV, as shownin Fig. 1. This value is larger than the energy resolution ofthe x-ray CCDð¼ 0:32 keVÞ. There are many papers thatindicate that the multiple ionization of Xe from Xe30þ toXe40þ has been reached by laser-cluster interaction [11].Ionization by ultrahot electrons from plasma could di-rectly remove electrons from the K-shell of such ions,

1614 OPTICS LETTERS / Vol. 36, No. 9 / May 1, 2011

0146-9592/11/091614-03$15.00/0 © 2011 Optical Society of America

and in such a case, the superposition of the radiation ofthese ions increases the FWHM of the spectra in thevicinity of the Kα line. Here, the energy resolution wasestimated as [12]

ΔE ¼ ω ×ffiffiffiffiffiffiffiffiffiffiffiffiffi8 log 2

p×

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiN2 þ F × E=ω

qkeV; ð1Þ

whereΔE is the energy resolution of the x-ray CCD, N isthe reading noise of the x-ray CCDð¼ 20Þ, F is the Fanofactor ð¼ 0:12Þ, E is the photon energy in keV, and ω isthe average energy per electron-hole pair in keV(¼ 3:62 × 10−3 keV). Assuming that the Xe K-shell x raysare produced isotropically, the Xe K-shell yield is esti-mated to be 3:1 × 107 photons/shot. Thus, the laser-to-Xe-K-shell x-ray conversion efficiency corresponds to1:5 × 10−7 for a 40 fs pulse width. When a Maxwellianelectron distribution with a temperature, Th, interactswith a target, the x-ray energy spectrum also follows aMaxwellian distribution with the same temperature Th[13]. Therefore we try to fit the x-ray energy spectrumby the Maxwellian gðEÞ,

gðEÞ ¼ AffiffiffiffiE

pexpðE=TeffÞ; ð2Þ

where A is constant and Teff is the effective temperatureof the x rays.As we can see from Fig. 1, the broad part of the x-ray

energy spectrum is explained by the two effective tem-peratures, 1.7 and 6:3 keV. Probably the electron energyspectrum follows the Maxwellian distributions with simi-lar temperatures.The dependence of the x-ray energy spectrum on the

laser pulse duration was investigated at 0:88 J on the tar-get at a reservoir pressure of 4:1MPa. The results ob-tained with the x-ray CCD with a 260 μm Al filter isshown in Fig. 2. The broad part of the spectrum in eachline is explained by the sum of two Maxwellian curves.The effective temperatures are 1.3 and 5:4 keV for the la-ser pulse duration of 40 fs, and 1.6 and 4:8keV for that of300 fs. The figure indicates that the Xe K-shell x-ray yieldat the laser pulse duration of 40 fs is clearly higher than

that at 300 fs. From this figure, the energy conversion ef-ficiency from the laser to the Xe K-shell x rays corre-sponds to 1:3 × 10−7 for 40 fs and 2:4 × 10−8 for 300 fs.This result differs from the results of many papers thatstate that the laser pulse duration of 200–300 fs is theoptimum condition for x-ray generation at a constant la-ser energy [6,14]. This may be understood by the fact thatthe energy of the K-shell x ray (∼30 keV) obtained in thisexperiment is higher than those reported previously;thus, higher energy electrons accelerated by the laserpulse are needed to ionize Xe inner-shell electrons. In or-der to confirm the assumption, we have performed simu-lations using the particles-in-cell code, as shown in Fig. 3.

Fig. 1. (Color online) Hundred-shot average x-ray energyspectrum with Xe cluster target at 3:4MPa. The peak laser in-tensity was 1 × 1019 W=cm2. The red line represents the fittingcurve of the broad part of the x ray; its equation is inserted inthe figure.

Fig. 2. (Color online) Dependence of 20-shot average x-ray en-ergy spectrum on pulse width at 4:1MPa: (a) 40 fs and (b) 300 fs.The fitting curves of the broad part of the x ray (red lines) andtheir equations are inserted in the figure. The focal spotdiameter (1=e2) was 18 × 36 μm2.

Fig. 3. (Color online) Energy spectrum of the electrons fromPIC simulations: 40 fs, blue line with closed squares; and 300 fs,red line with closed circles.

May 1, 2011 / Vol. 36, No. 9 / OPTICS LETTERS 1615

This simulation result clearly explains our result thatXe K-shell yield depends on pulse duration. The genera-tion of Xe K-shell x rays requires the removal of the XeK-shell electron by fast electrons, which have energieshigher than the threshold energy of K-shell ionization(∼35 keV). Figure 3 indicates that the number of elec-trons above the threshold energy generated by the 40 fspulse duration is larger than that generated by the 300 fspulse duration because of the higher laser intensity.Therefore, it is natural that the 40 fs pulse duration isbetter than the 300 fs pulse duration for Xe K-shellx-ray generation in our case.In conclusion, we have measured the x-ray energy

spectrum at the peak intensity of 1 × 1019 W=cm2 withXe clusters generated at the reservoir pressure of3:4MPa. In addition to the broad x-ray energy spectrum,3:1 × 107 , Xe K-shell photons were clearly observed. Wealso found that the pulse duration of 40 fs is better thanthat of 300 fs from the point of view of Xe K-shell x-raygeneration. The energies of the Xe K-shell x rays areclose to the energy of an Sn K absorption edge (atomicnumber Z ¼ 50). Therefore, we consider that this Xe K-shell x-ray source is suitable not only for x-ray imaging ofa thick sample but also for x-ray imaging of a sample in-cluding high-Z ð45 < Z < 50Þ and low-Z ðZ ≪ 50Þ mate-rials. According to our result, 3:1 × 1010 Xe K-shellphotons can be generated by an accumulation of 100 sof laser shots at 10Hz under the same experimental con-ditions. We believe that this intensity is sufficient for x-ray imaging.

References

1. C. L. Gordon III, G. Y. Yin, B. E. Lemoff, P. M. Bell, andC. P. J. Barty, Opt. Lett. 20, 1056 (1995).

2. C. Tillman, A. Persson, C. G. Wahlstrom, S. Svanberg, andK. Herrlin, Appl. Phys. B 61, 333 (1995).

3. L. M. Chen, M. Kando, J. Ma, H. Kotaki, Y. Fukuda,Y. Hayashi, I. Daito, T. Homma, K. Ogura, M. Mori,A. S. Pirozhkov, J. Koga, H. Daido, S. V. Bulanov, T. Kimura,T. Tajima, and Y. Kato, Appl. Phys. Lett. 90, 211501(2007).

4. Y. Fukuda, A. Ya. Faenov, T. Pikuz, M. Kando, H. Kotaki,I. Daito, J. Ma, L. M. Chen, T. Homma, K. Kawase,T. Kameshima, T. Kawachi, H. Daido, T. Kimura, T. Tajima,Y. Kato, and S. V. Bulanov, Appl. Phys. Lett. 92, 121110(2008).

5. M. E. Sherrill, J. Abdallah Jr., G. Csanak, E. S. Dodd,Y. Fukuda, Y. Akahane, M. Aoyama, N. Inoue, H. Ueda,K. Yamakawa, A. Ya. Faenov, A. I. Magunov, T. A. Pikuz,and I. Yu. Skobelev, Phys. Rev. E 73, 066404 (2006).

6. R. C. Issac, G. Vieux, B. Ersfeld, E. Brunetti, S. P. Jamison,J. Gallacher, D. Clark, and D. A. Jaroszynski, Phys. Plasmas11, 3491 (2004).

7. H. Honda, E. Miura, K. Katsura, E. Takahashi, andK. Kondo, Phys. Rev. A 61, 023201 (2000).

8. A. McPherson, B. D. Thompson, A. B. Borisov, K. Boyer,and C. K. Rhodes, Nature 370, 631 (1994).

9. H. Kiriyama, M. Mori, Y. Nakai, T. Shimomura, H. Sasao,M. Tanoue, S. Kanazawa, D. Wakai, F. Sasao, H. Okada,I. Daito, M. Suzuki, S. Kondo, K. Kondo, A. Sugiyama,P. R. Bolton, A. Yokoyama, H. Daido, S. Kawanishi,T. Kimura, and T. Tajima, Opt. Lett. 35, 1497 (2010).

10. A. S. Boldarev, V. A. Gasilov, A. Ya. Faenov, Y. Fukuda, andK. Yamakawa, Rev. Sci. Instrum. 77, 083112 (2006).

11. T. C. Berkelbach, J. Colgan, J. Abdallah Jr., A. Ya. Faenov,T. A. Pikuz, Y. Fukuda, and K. Yamakawa, Phys. Rev. E 79,016407 (2009).

12. M. Nishiuchi, K. Koyama, H. Awaki, T. Tsuru, M. Sakano,K. Hamaguchi, H. Murakami, H. Tsunemi, K. Hayashida,S. Kitamoto, E. Miura, T. Dotani, M. Ozaki, M. Bautz,J. Doty, S. Kissel, R. Foster, and G. Ricker, Nucl. Instrum.Methods Phys. Res. A. 436, 79 (1999).

13. G. H. McCall J. Phys. D 15, 823 (1982).14. F. Dorchies, F. Blasco, C. Bonte, T. Caillaud, C. Fourment,

and O. Peyrusse, Phys. Rev. Lett. 100, 205002 (2008).

1616 OPTICS LETTERS / Vol. 36, No. 9 / May 1, 2011