Регистрация / Вход
Прислать материал

Laser writing with sub-nanosecond burst of femtosecond pulses.

Name
Andrey
Surname
Okhrimchuk
Scientific organization
D. Mendeleyev University of Chemical Technology of Russia
Academic degree
PhD
Position
Leading Research Scientist
Scientific discipline
Physics & Astronomy
Topic
Laser writing with sub-nanosecond burst of femtosecond pulses.
Abstract
A method of efficient laser micromachining in bulk of fused silica and sapphire with a burst of femtosecond pulses separated by 10 – 100 ps is proposed and demonstrated. We used Fabry-Perot cavity to split the NIR 180 fs pulse into a burst with the same duration of each pulse in the burst. The new method exploits strongly localized absorption by transient electronic excitations prepared by the first pulse in the burst. In comparison with a single pulse an increased refractive index change and birefringence is obtained by a single burst
Keywords
laser direct writing, refractive index change, birefringence, burst of femtosecond pulses.
Summary

Laser writing with sub-nanosecond burst of femtosecond pulses

A. Okhrimchuk*1, S.Fedotov1, I. Glebov1, V. Sigaev1 and P. Kazansky1,2

1D. Mendeleyev University of Chemical Technology of Russia, Miusskaya square 9, Moscow 125047, Russia

2Optoelectronics Research Centre, University of Southampton, Southampton SO17 1BJ, UK.

 

1 Introduction

Femtosecond laser direct writing and micromachining attract considerable interest for a variety of applications ranging from integrated optics and microfluidics to printable flat optics and multi-dimensional optical data storage. Micromachining with temporally shaped pulses or femtosecond pulse trains that exploit material dependent relaxation processes has been applied for creating sub-wavelength features via control of ionization processes on a sub-picosecond time scale [1], and writing waveguides with a circular cross-section via control of a heat deposition with sub-microsecond bursts of femtosecond pulses [2]. There is lack of experiment sub-nanosecond burst. Nevertheless an enhancement of strength of the nanograting form birefringence was observed under writing with train of pairs of pulses, when interval between pulses was less then nearly 50 ps presumably due to increased absorption of self-trapped excitons (STE) [3]. Results of this work allowed us to suggest that one or few bursts of pulses, separated by few decades of picosecond would efficiently produce material modification, and this efficiency would go down with pulse separation increase according to STE decay. In our experiments we have generated such bursts, and found that it is a really efficient tool for refractive index modification and forming of a birefringence structure. However we observed an unexpected dependency of the refractive index change upon pulse separation interval, which is inconsistent with simple decay of STE.

2 Writing with sub-nanosecond burst

IR laser femtosecond modification creating permanent refractive index change is initiated by multiphoton ionization, which is followed by linear absorption by the electron plasma, multiphoton absorption of self-trapped excitons (STE) and avalanche ionization [4]. In the experimental setup we used Fabry-Perot cavity to split the 180 fs pulse at 1030 nm into a burst of pulses (Fig.1). Pulse separation in the burst Δt was equaled to 10 -70 ps defined by the cavity length. The first pulse with the highest energy was followed by train of pulses with exponentially decreasing pulse energy. Effective burst duration could be achieved about 10 times greater compared to the duration of tailored pulse produced by the third-order chirp without stretching of each pulse in the burst [1]. Such temporal form of the burst allowed localizing modification in the beam waist due to reduction the energy of the first pulse down to the inscription threshold energy, and at the same time strong modification is produced due to absorption of the subsequent pulses by electron-hole plasma or STE [3-5].

 

Fig.1 Schematics trace of a burst of femtosecond pulses, generated by Fabry-Perot cavity with the 75 % reflection coefficient of each mirror.

 

Absorption of the subsequent pulses is more efficient, than the first pulse absorption, because the lower number of photons is required for an absorption act for subsequent pulses by transient electronic excitation. On the other hand this absorption is localized in the region prepared by the first pulse, thus strong localized modification with refractive index change is produced.

3 Charaterization of modifications and discussions

We found, that burst of pulses produce modification of refractive index with higher phase delay than one pulse by factor of 2 - 5, and the phase delay monotonically increases with increase of pulse separation interval Δt (Fig.2) in the range of 0-20 ps and then the dependence is saturated. In fused silica the single pulse modification with pulse energy just above inscription threshold causes the positive refractive index change, and pulses with higher energies produce negative changes of index, while bursts of pulses always produce modification with negative refractive index change for burst energies above the inscription threshold energy as in fused silica, so in sapphire doped with Ti3+.

     

 

(a)

(b)

Fig. 2. Phase delay due to the single burst modification of refractive index in fused silica (a) and in sapphire doped with Ti3+ (b) in dependency on the pulse separation interval. Zero interval corresponds to a single pulse. Positive phase delay corresponds to negative refractive index change in our notations. Labels on the graph denotes burst energy. Focusing lens - NA=0.65.

     

      Phase delay increase on the scale of 0 - 20 ps is inconsistent with known electronic relaxation processes, in particular, with dynamics of formation and decay of STE in SiO2, and electron-hole plazma and self-shrunk excitons (SSE) in Al2O3 [6]. We suggest that the observed increase reflects intrinsic dynamics of excitation relaxation inside STE and SSE. Probably there are two channel for localized excitons transformation in presence of intense laser pulse: 1) back excitation to electron-hole plazma; 2) decay to stable structure defects causing refractive index change, and we suggest that after last transformation the probability of recurrent excitation to electro-hole pair dramatically drops down. It was shown that in fused silica under high density of excitation accompanying STE Frenkel defects are further transformed into stable defects [7]. Taking into account that only the second channel of STE and SSE transformation exists between pulses in the burst, one can conclude that, density of the stable defects increases with enhancement of pulse separation interval. In Al2O3 the phase delay dependence is less pronounced, than in SiO2. This is obviously connected with lower role of SSE in defect formation, as lifetime of free electron in this crystal is about 100 ps. In sapphire the subsequent pulses in the burst must be absorbed by electron-hole plazma too, and holes could be localized on Ti3+ ions with creation of stable Ti4+ ions.

      Increased efficiency of burst absorption is confirmed in measurements of one pulse/burst transmittance T in dependency on pulse /burst energy under the same focusing conditions in fused silica (Fig.3). Energy meter head was placed closely behind the thin glass plate so that 100% of laser radiation focused in the plate at the 100 μm depth by a lens with NA=0.65, went through the plate and then collected by the head. The glass plate was placed on a high precision translation stage, which was moving perpendicular to the focused laser beam, thus each laser shot produced an excitation in a new glass area. Transmittance curve of a single pulse has a recognizable shape [3], while dependence of absorbance for the burst is dramatically differed with a sharp increase at the energy close to modification threshold and higher absorption at higher energies.

      It was found, that in contrary to a single pulse, absorption of the burst with the same total energy is strong enough to produce a birefringent micro-structure in fused silica. It consists of two elongated birefringent spots with parallel slow axes separated by area without birefringence (Fig.4). The slow axis of birefringence and direction of elongation of inscribed pattern follow polarization of the femtosecond beam.

Fig. 3. Absorbance ln(T) of a pulse (black points) and a burst (red points) in fused silica in dependency on the pulse energy or a first pulse in the burst energy. Intervals between pulses equal to 10 ps (red points), 20 green points, 40 ps (cyan points) and 70 ps (blue points). Focusing lens - NA=0.65.

 

Fig. 4. Microscopic pictures of birefringent spots, produced by the burst of femtosecond pulses. Colors code direction of slow axis in accordance with the legend. Red arrows show direction of laser beam polarization. Burst energy is equaled to 110 nJ. Focusing lens - NA=0.65.

 

Fig. 5. Retardance under single burst modification in fused silica in dependency on the pulse separation interval.

     

Dependence of birefringence retardance upon the interval between pulses in the burst is not so definite as for phase delay, and its slope depends on the burst energy (Fig.5). There is small difference in retardance for different energies at the 10 ps interval, and diversity arises when the interval increases. There is no detectable birefringence for a single pulse.

 

Conclusion

In conclusion we have proposed and demonstrated a method of efficient micromachining with the burst of laser pulses separated by time interval 10 - 100 ps. Such burst produces enhanced refractive index change and a birefringent structure even with one laser shot due to increased pulse absorption. Nature of unexpacted dependence of refractive index change upon pulse separation interval in the burst could be connected with relaxation processes in localized excitons.

 

This work was supported by Advanced Research Foundation and Ministry of Education of Science of Russian Federation, grant #14.Z50.31.0009.

References

  1. J. Hernandez-Rueda, N. Gotte, J. Siegel, M. Soccio, B. Zielinski, C. Sarpe, M. Wollenhaupt, T. A. Ezquerra, T. Baumert, and J. Solis, “ACS Appl. Mater. Interfaces 7, 6613–6619 (2015).
  2. S. Rezaei, J. Li, and P. R. Herman, Opt. Lett. 40, 2064–7 (2015).
  3. S. Richter, F. Jia, M. Heinrich, S. Döring, U. Peschel, A. Tünnermann, and S. Nolte, Opt. Lett., 37, No. 4, 482–484 (2012).
  4. D. Grojo, M. Gertsvolf, S. Lei, T. Barillot, D. M. Rayner, and P. B. Corkum, Phys. Rev. B 81, 212301 (2010).
  5. F. Quere, S. Guizard, and Ph. Martin, Europhys. Lett. 56, 138–144 (2001).
  6. M. Kirm, G. Zimmerer, E. Feldbach, A. Lushchik, Ch. Lushchik, and F. Savikhin, Phys. Rev. B 60, 502–510 (1999).
  7. K. Tanimura, T. Tanaka, and N. Itoh, Phys. Rev. Lett. 51, 423–426 (1983).
  8. S. Guizard, P. Martin, Ph. Daguzan, G. Petite, P. Audebert, J. P. Geindre , A. Dos Santos and A. Antonnet, Europhys. Lett. 29, 401–406 (1995).