Overheated femtosecond plasma in highly porous ...

Overheated femtosecond plasma in highly porous silicon

M.S.Dzhidzhoev*, S.A.Gavrilov**, V.M.Gordienko*, P.M.Mikheev*, A.B.Savel'ev*, A.A.Shashkov*, T.M.Vlasov*, and R.V.Volkov*

* - International Laser Centre & Physics Faculty, MV. Lomonosov Moscow State University, Vorobyevy goiy, Moscow, 1 19899, Russia, phone: +7-095-9395318, e-mail:[email protected] ** - Moscow Institute ofElectronic Techniques

ABSTRACT We present results on plasma formation in porous silicon (cluster like solid with mean cluster size of 3 rim, mean density 0. 1-0.2 of crystalline silicon) by femtosecond laser pulses at intensity above 10 TW/cm2. We deduced hot electron temperature as high as 8 keV and fast ions of at least 2 MeV energy. Keywords: femtosecond laser plasma, hard x-rays, hot ions, porous silicon

1. DTRODUCTION A typical value for the mean electron temperature obtained in femtosecond laser plasma interaction with bulk solid targets at intensities below iO' W/cm2 does not exceed 1 keV while relatively large part of the electrons gain much higher temperature of 2-5 keV due to "resonant absorption", "vacuum heating", etc1 . Novel possibilities arise while using cluster jet as a target 2,3 Here one can achieve higher electron temperatures, efficient production of hard x-rays, and other. The key thsadvantage of cluster jet is relatively low mean density and presence of backing pressure due to unclustered atoms or molecules, that creates underdense plasma and alters the interaction regime of superstrong laser pulse with overdense plasma. From that point cluster like solids are good choice for this target type keeping all the advantages of cluster jet, being consisted of small cluster of 3-5 mn in diameter, and having mean density of 0.5-0. 1 of the bulk without backing pressure of free atoms or molecules. It should be noticed here that the phenomena taking place under interaction of superstrong laser pulse with cluster like solids badly depend on porosity P (ratio ofbulk to porous densities)4'5. In this paper we show that using nanostructured highly porous silicon as a target one can overheat the plasma under "moderate" intensity of 1O6 W/cm2, thus obtaining hard x-ray production above 5 keV and high flux of hot 200-2000 keV ions.

2. EXPERIMENTAL SETUP AND METHODS In performing experiments we used femtosecond laser system6 which provides intensity in excess of 1016W/cm2 with intensity contrast ratio better than iO. Figure 1 represents basic set of plasma diagnostics devices, including diffraction grating and photomultiplier tube (PMT) to detect SH on reflection from plasma. Hard X-rays were detected after 100 im Be output window (transparency 0. 1 for quanta energy E - 2.3 keV) by means of NaI(Tl) crystal along with another PMT. A set of filters was used to distinguish between different spectral bands from 2.5 up to 40 keV. Ion current was measured by simple time-of-flight MCP detector with temporal resolution of 20 ns.

We used double-channel registration scheme to estimate from our hard X-ray data electron temperature. For each - X-ray quantum energy) the output signal could be written as: channel with transparency function H1(E) (E =

1,2 = f(Te)JH12(E).e

•dE

SPIE Vol. 3735 • 0277-786X199/$1O.OO 249

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/18/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

(assuming that quantum energy is much higher

than photorecombination threshold Eth-2.5

Time-of-flight MCP detector

keV for deeply ionised Si). To get the value for Te directly from single channel data one needs

200 fs

PIN diode

to know function f(Te) = Ne Z T'2 (Ne electron concentration and Z - mean ion

616nm

charge). At the same time for the ratio r of two

channels with different filters one can easily 2 obtain the following:

r= JH1(E).edE J H,(E) edE

target

Hence measuring in one shot signals from two channels one can get electron temperature.

Soft X-rays

Diffraction grating

PIN diode

3. EXPERIMENTAL RESULTS AND DISCUSSION

Fig.2 shows the result of the procedure Fig. 1. Experimental scheme described in the previous section for two samples (c-Si and por-Si). It should be noticed here that this Figure collects results for different filter pairs. The first filter was always the same - 100 im Be, while the second one was one of the following: 180 m Al (E>9keV), 13 pm Ta (E>l 1 key), i mm Al (E>17 keV), 26 im Ta (E>20 keV).

For all the intensities the value far above 1 keV

(typical value for mean electron temperature in

10000

Th,eV

+

S

8000

femtosecond plasma) was deduced for the electron

Si Por - Si

temperature. This should be treated as the consequence of hot electron production in plasma.

Fitting data from the Fig.2 we obtained scaling for c-Si sample. This appears to be in Th good agreement with different theories of hot electron

S

6000

production on sharp boundary (so-called vacuum heating) or on the smooth one (resonant absorption). For the lauer case the scaling should be slightly lower

+

4000 +

-

+

Th

'

O.3 •

For

the porous sample the

experimentally obtained scaling is even higher , and this is a direct signature of plasma Th

+

2000

overheating in this sample.

I, \V/cm2* 1016

0

0.4

0.6

0.8

I .0

I .2

Fig. 2 Hot Electron temperature Th dependency on laser intensity Ifor c-Si (crosses) andpor-Si (full circles).

Fig. 3 shows the dependencies of ion current on

time for two laser intensities and porous and crystalline Si samples. The vertical dashed line marks

laser pulse position on the time axis. The ion front velocity deduced from its temporal delay with respect to laser pulse t for Si V1 3* cm/c (20 keV fast

ion energy) is in a good agreement with the simple

250


J arbUn.

estimations for hydrodynamic expansion velocity of plasma with Te 800 eV and mean

ion charge Z

12. Comparing results for

porous and ciystalline Si samples at the same intensity one could see significant decrease in temporal delay t. That gives for intensity 1016 W/cm2 at least one order of magnitude increase

in ion front velocity for porous Si, which corresponds to the fast ions energy above 2 MeV. The appearance of these hot ions could be attributed for cluster Coulomb explosion and local field enhancement. It should be noticed here, that the same figure of at least 400 keV was obtained in 7,8 for fast ions energy resulted from cluster explosion.

4.CONCLUSIONS

Hence, our investigation of fast ions production and hard x-ray generation from plasma created in highly porous Si confirmed that this material could be successfully used in

femtosecond laser plasma interaction. The phenomena taking place under this interaction

0

1

ts4

2

differ in some chief points from interaction with bulk materials and are similar to those

occurs in the case of cluster jet plasma interaction.

Besides silicon itself, large amount of hydrogen atoms and OH groups are absorbed on the surface of silicon cluster. This could strongly affect the interaction process or either cause new physical phenomena. For instance, hydrogen atoms could be substituted by deuterium ones. Ion kinetic energy emerging from cluster explosion should transform to thermal energy through ion-ion collisions, thus giving rise to direct D-D fusion: D+D - 3He(O,82 Mefr) + n (2,45 MeV). The cross-section for this reaction reaches 4*1028 cm2 for D ion mean energy ED30 keV . Hence, the number of D+D reactions and consequently, the number of neutrons in one laser shot, would be i0 (1) ions concentration 4021cm3, plasma volume 10b0 cm3, plasma cooling time 10 ps). Fig. 3 The dependencies of ion current Jon time t.

5. ACKNOWLEDGMENTS Authors greatly acknowledge support from Russian Foundation for Basic Research (grant #96-02-l9146a) and State Scientific Programs "Fundamental Metrology" and "Russian Universities".

6. REFERENCES 1. Gibbon P., Forster R., Plasma Physics Control. Fusion, 38, 769 (1996). 2. C.Wulker, W.Theobald, D.R.Gnass, et. a!., Appl.Phys.Lett., 68, 1338 (1996). 3. A.McPherson, K.Boyer, C.K.Rhodes, J.Phys.B: At.Mol.Opt.Phys., 27, L637 (1994). 4. R.V.Volkov, V.M.Gordienko, M.S.Dzhidzhoev, et al., Quantum electronics, 28, 1 (1998) 5. V.M.Gordienko, M.S.Dzhidzhoev, M.A.Joukov,, et.al., Superstrong Fields in Plasmas, Ed. : M.Lontano, G.Mourou, F.Pegoraro, E.Sindom, AlP Conf.Proc. v.426, p.241, AlP, New-York. 6. R.V.Volkov, V.M.Gordienko, M.S.Dzhidzhoev, et a!., Quantum electronics, 27, 1081 (1997). 7. T.Ditmire, J.W.G.Tisch, E.Spnngate, eta!., Nature, 386, 54 (1997). 8. T.Ditmire, R.A.Snuth, J.W.G.Tisch, et.al., Phys.Rev.Lett., 78, 3121 (1997). 9. K.Brakner, S.Jorna, Laser driven fusion, KMS Fusion, mc, 1973.

251
