Open Access
Add to BibSonomyAbstract
We demonstrate that the interaction of intense femtosecond light on a plain solid substrate can be substantially altered by a few micron layer coating of bacterial cells, live or dead. Using E. Coli cells, we show that at an intensity of 1016W cm−2, the bremsstraahlung hard x-ray emission (up to 300 keV), is increased by more than two orders of magnitude as compared to a plain glass slab. Particle-in-cell simulations carried out by modeling the bacterial cells as ellipsoidal particles show that the hot electron generation is indeed enhanced by the presence of microstructures. This new methodology should pave way for using microbiological systems of varied shapes to control intense laser produced plasmas for EUV/x-ray generation.
© 2012 OSA
1. Introduction
Intense, femtosecond laser created plasmas [1] are unique tabletop point-sources of pulsed energetic electrons [2], ions [3] and x-rays with energies extending well into the MeV region [4]. These emissions, originating from preferentially heated plasma electrons [5] are of great importance to diverse scientific, technological and medical applications such as radiography [6], cancer therapy [3] and lithography [7]. Nanoscale structuring of the target surface by particle deposition [8] or sub-lambda gratings [9] enhances these emissions via local field amplification caused by scattering and/or surface plasmon coupling. Such artificial target preparation, however, suffers from limitations in the range of structures achievable, apart from the restrictions imposed by the manufacturing process [10]. Besides, laser intensities required for hot electron generation using such targets [11] are currently large enough to limit the source repetition rate to <100 Hz. Hot electrons are generated by a number of processes: a) resonance absorption (RA), b) vacuum heating, c) J × B heating [5] etc.,. These methods of interaction undergo changes in different manner depending on the target and the laser intensity. For example RA occurs when the strong laser field drives plasma electron waves, which damp out to produce hot electrons [12]. RA is excited by p-polarized light that is incident at an angle (typically 30–45 degrees) to the normal of the target. Experimental and simulation evidence indicates that RA occurs efficiently above 1015 W cm−2 and about 30% of the laser light can be coupled to the hot electron generation [5]. The hot electron energies are characterized by well established scaling laws, such as the Forslund scaling law [13], which yields, Thot (keV) = 14 (Tc I λ2)1/3, where Tc is the bulk plasma temperature in keV, I is the laser intensity in units of 1016 W cm−2 and λ is the laser wavelength in μm. The local light intensity that produces plasmas can be very different from the incident light intensity if the target is structured with micro- or nanoparticles and scatter light to bring out modifications in the local field.
A method that enhances the absorption of the laser energy and therefore, the ability to increase hot electron/x-ray emission, is of paramount importance. This has been addressed in two ways. The first relies on optimizing laser parameters such as intensity, pulse width, polarization and temporal shape. A second method aims to engineer the target for efficient absorption and subsequent channeling of the absorbed radiation to increase a specific process, x-ray emission, for example. This latter approach has just begun to be realized [8, 9]. For instance, a 15μm liquid droplet is shown to produce two orders of magnitude amplification in the local fields and 60 fold enhancement in the hard x-ray yield [4]. Similarly, a velvet target [14] with complex nanostructuring yields a 50-fold enhancement in soft x-ray emission. In both cases the complexity of target preparation and experimental procedures involved in maintaining the target in vacuum are major limitations. Scattering of light in nano- or micro-structures of the target can change the nature of interaction and efficiency of hot plasma generation. It is this aspect that we exploit in the present experiments by using micron sized bacterial cells, namely E. coli. We demonstrate a novel structured target that liberates these sources from the above constraints. A target consisting of a few micron-thick layer of a ubiquitous microbe, Escherichia coli (E. coli), catapults the brightness of hard x-ray bremsstrahlung emission (up to 300 keV) by more than 100 fold at an incident laser intensity of 1016 W/cm2. We model the bacteria as ellipsoidal particles placed on a solid substrate and particle-in-cell (PIC) [15] simulations demonstrate the effect of the size and shape of the target in bringing out the enhancement. This combination of laser plasmas and biological targets can lead to turn-key, multi-kilohertz and environmentally safe sources of hard x-rays. It would also trigger the exploration of an unlimited diversity of biological structures in nature as targets for hard x-ray generation.
2. Experiment
A regular strain of E. coli bacteria DH5alpha was grown overnight in a suspension culture in minimal media. We used both live or chemically fixed and ultraviolet (UV) radiation-attenuated cells. Though the effect is similar, most of the experiments were carried out with fixed and attenuated cells to avoid generation of any virulent strains due to radiation exposure. Fixation is carried out by the use of a mixture of 4% formaldehyde and 2.5 % gluteraldehyde solutions. To coat the bacteria on an optically polished BK-7 glass substrate, it was at first painted with 1 mg/ml poly-L-lysine solution and then air-dried for a few minutes. This created a charged hydrophilic surface on the substrate and helped to form a uniform coating of the bacterial cells. The cell-suspension was then applied on the poly-L-lysine coated glass, allowed to dry in a laminar flow hood and irradiated in a suitable chamber with an appropriate UV dose (250 mJ of 280–300 nm UV) to fully attenuate all the bacterial cells. The coated target slabs were then left to dry in a desiccator. Thus fixed, the E. coli cells remained structurally unaffected under a 10−5 Torr vacuum for over an hour. A profilometer scan of the the bacterial coatings showed that the coating was fairly uniform with thickness variations between 1–2 μm, which is equivalent to 2–3 layers of E. coli cells.
A conceptual picture of the novel hard x-ray source is given in Fig. 1. A femtosecond laser beam (40 fs, 800 nm) was focused with an f/3 off-axis parabolic mirror at a 45° incident angle to a 17 μm spot (intensity of the order of 1016 W/cm2) on a solid glass plate coated with a few-microns thick layer of E. coli cells. The target was rastered to expose a fresh portion to each laser shot. The x-ray measurements were carried out with a 2” thick NaI (Tl) scintillation detector coupled with conventional electronics. The detector system was gated with a 20 μs gate pulse generated in synchronous to the laser trigger. The detector was placed at a distance of about 70 cm from the target across a 5 mm glass window. Pile-up free detection was ensured by shielding the detector in a 10 mm thick lead housing with an appropriate-sized aperture, such that the count rate was less that 1 count in 10 pulses. Under otherwise identical conditions, the x-ray energy and yield were measured for over 10,000 laser shots on both the bacteria-coated portion and the uncoated portion to make a direct comparison.
Fig. 1 Schematic of the experimental apparatus. A 40 fs pulse of 800 nm light is focused on the target to intensities of about 1016 W cm−2 using a off-axis parabolic mirror. The target is glass substrate coated with a few micron layers of E. coli bacteria. Inset shows an electron microscope (EM) image of E. coli.
3. Results and discussion
E. coli cells are ellipsoidal in shape (about 1.8 μm along the major axis and ≈ 0.7μm along the minor axis) [16]. Each cell also expresses numerous flagella, 25 nm in diameter and 10–20 μm in length from their cell exterior [16]. However, the fixation process sheds the flagella and therefore only the ellipsoidal structure of the cells was exploited in the current experiments. E. coli has no known sub-cellular structures. Hence, we assume that the majority of the 0.7 femtolitre volume is filled with optically uniform cytoplasm. For the purpose of this work, these cells can also be viewed as microparticles with well-defined sizes that are filled with atoms with a low atomic number Z.
To evaluate the effect of E. coli on hard x-ray emission, the laser light is first focused on the plain solid glass, used as a reference. X-rays are detected with NaI (TI) detector placed at about 70 cm from the target and sampled through an appropriately sized lead aperture such that a pile-up free measurement is made. The target is then moved so that the light is focused on the portion of the glass with the bacterial coating, under otherwise identical (like laser intensity, focal waist etc.) conditions and the x-ray measurements are repeated. If the x-ray flux is very large, the aperture in the 10mm thick lead housing is reduced to ensure that the count rate is less than one per second and is pile-up free. Figure 2(a) shows the hard x-ray spectrum for the glass target with the bacterial coating (blue circles) at an intensity of 5× 1016 W cm−2. Under same conditions, focusing light on the plain target yielded much smaller flux of x-ray and are hardly visible (pink triangles) in Fig. 2(a). The total yield of hard x-rays integrated over 50–300 keV, is about 120 times larger in the former case.
Fig. 2 Bremsstrahlung x-ray spectrum up to 300 keV, measured for different types of targets at a laser intensity of 5 × 1016 W cm−2. The topmost curve (blue circles) represents the yield from the E. coli coated target and it is evident that this is significantly larger than the emissions from the uncoated (pink triangles) and homogenate coated targets (magenta squares). The integrated yield from the E. coli coated target is about 120 times larger than that from the uncoated glass. The homogenate shows only 23 times larger yield. False color EM images of the intact E. coli cells and the homogenate are shown in (b) and (c) respectively.
To prove that the shape of the microbacterial coating is vital to the x-ray generation efficiency, we carried out experiments under otherwise identical conditions with homogenized bacteria. The bacterial solution is ultrasonicated to break the ellipsoidal cell structures and a coating of the broken bacterial suspension is made on the plain glass solid slab. Glass slab coated with the homogenized bacteria thus have the similar chemical composition but are devoid of the ellipsoidal microstructures. X-ray measurements, as described above are repeated to measure changes in relative yield both with the plain glass target and glass coated with normal intact bacteria. Most of the bacterial cells were disrupted by the homogenization process, as can be seen in the EM image given in Fig. 2(c). The homogenized bacterial layer of similar thickness produced x-rays with much diminished efficiency (magenta squares) as seen in Fig. 2(a). Marginal increase in x-ray yield with the homogenized bacteria is attributed to the presence of partially broken structures of bacteria and also a small percentage of unhomogenized normal bacteria.
X-rays are produced by bremstraahlung of the hot electrons generated in interaction of the intense ultra short light and a measure of the electron temperature can be obtained by a Maxwell-Boltzmann fit of the energy spectrum. Figure 3 shows the x-ray spectrum measured over the entire energy range. Solid line shows the Maxwell-Boltzmann fit to the x-ray emission from the bacteria-coated target. The hot electron temperature from the fit is about (57 ± 2) keV for the bacteria-coated target. Similar measurement with uncoated glass target show that the hot electron temperature with plain glass is about 2.5 times smaller under similar conditions of intense laser pulse interaction.
Fig. 3 X-ray spectrum from E. coli coated target measured over the entire energy range and Maxwell- Boltzmann fit to data shown as a solid line. The exponential fit has an electron temperature of 57±2 keV
In order to reproduce the effect of bacterial coating on a solid slab we use two-dimensional PIC simulations. Bacteria are modelled as elliposidal plasma particles of appropriate size placed on a solid slab. Computations are carried out on a 1000 × 1000 grid with a uniform grid size of Δ = λ/40 to simulate the experimental observations. A solid slab of 20λ × 5λ with ellipsoidal particles of a similar size as E. coli (0.7 μm × 1.8 μm) are illuminated by normally incident light of wavelength λ = 800 nm and an intensity of 1016 W/cm2 to simulate conditions close to the experiment. Assuming the slowly-varying envelope approximation [17, 18], a continuous Gaussian beam with a transverse field component
with a focal width w0 = 20λ, Rayleigh range and radius of curvature is numerically excited at one end y = yb and propagated across the computational box. The laser field is switched off after an interaction time of approximately 35 fs. A uniform initial electron density 2nc is assumed for the slab and the ellipsoids, where nc is the critical electron density (∼ 1×1021 cm−3 at 800 nm).In order to simulate experimental conditions, simulations are performed with an increasing number of ellipsoidal structures in front of the solid slab in the focal spot. Figure 4 shows how an increase in the number of ellipsoids in the simulations increases the hot electron yield as well as the temperature. For the solid slab without ellipsoids, the electron temperature is about (5±1) keV, with an insignificant higher temperature component (solid black curve in Fig. 4). On the other hand, with eleven ellipses, nearly the entire focal area would be filled with the bacterial cells and therefore, the hotter electron temperature (red solid curve) is about 70 ±2 keV, which is reasonably close to the experimental measurements (57 keV with bacterial coating). In addition, the hotter electron temperature in simulations is also about 2.5 times that of the uncoated target, close to the experimental measurements. Further, while the temperature remains similar with larger number of ellipsoids, the hot electron yield increases very effectively. The inset in Fig. 4 shows the hot electron yield above 50 keV for different number of ellipses on the slab. For eleven ellipses the hotter electron yield is about 50 times that of the solid slab. In the experiments, however, the number of ellipsoidal bacterial cells was more and thus the enhancement obtained was about 120 fold. Since the hot electron component increases with the number of ellipsoids, addition of monolayers of ellipsoids is perhaps required in the simulations to quantitatively reproduce the experimentally measured enhancement. However, even with the simplistic model assumed, the simulations reasonably reproduce the enhancement in the hotter electron yield as measured by the x-ray spectrum.
Fig. 4 Electron spectrum derived from the 2D-PIC simulations with varied number of ellipsoid particle ion the solid as indicated in the legend. Slab refers (black square) refers to electron spectrum with solid slab and eleven refers (red circles) to calculation with eleven ellipsoid particles on the solid slab. The solid line show a two temperature exponential fit to the simulated data. See the text for details. Inset in (b) shows the hot electron yield above 50 keV for calculations with different number of ellipses. The solid slab yield is normalized to one to obtain a relative enhancement.
A deeper analysis is warranted to find the physics insights that cause the enhanced hard x-ray efficiency with the bacteria. Scattering of light affecting the local fields could be playing a role in the interaction of light with bacteria. However, since the particles are closer to the wavelength of the light, Mie scattering may not increase the local intensity by a large fraction. Anharmonicity in the interaction potential brought out by the ellipsoidal particles could play a role in hot electron generation. More work in this regime is necessary to probe the complex target used here for plasma generation and we hope that the two orders of magnitude raise in x-ray generation reported here would strongly influence such an endeavor.
4. Conclusion
In conclusion, we have demonstrated that biological microparticles can strongly influence the plasma generation in the interaction of intense ultra short pulses. Relative efficiency in hard x-ray generation increase by 100 times due to the presence of a few microns layer of E.coli. Shape of the microbial structures are shown to be crucial. We believe that a new paradigm in x-ray generation from intense laser-matter interaction has been initiated through this work where biology is used to address a physics problem. A number of new questions of both technological and fundamental interest arise from these experiments – for example, the role of the atomic constituents of the cell. If the bacterial cells are doped with high Z-atoms, it is very likely that the x-ray emission can be further boosted. The smoothness of the cell wall or the presence of nanostructure on the cell surface could also affect the plasma evolution. Thus, biological cells with well-ordained microstructures can be exploited as effective targets for hotter plasma generation and a resultant bright point-source of radiation with energy up to the hard x-ray regime (a few hundred keV). Natural cells provide a myriad of microstructures that can be well exploited for this purpose. Apart from the coccus (spherical), spirillum (spiral) or filamentous bacteria, even more complex shapes due to stalks and appendages in species like caulobacter, Myxococcus or Streptomyces can be exploited [19].
Acknowledgments
We thank Jasmine Seth for the help in initial set of experiments, Lalit Borde and Seema Shirolikar of the TEM facility for electron microscopic images. We also thank C. Danani and TBM group of IPR for executing simulations in their workstations. MK thanks DST, Government of India for a Swarnajayanti Fellowship and GRK acknowledges a DAE-SRC-ORI grant.
References and links
1. D. Giulietti and L. A. Gizzi, “X-ray emission from laser-produced plasmas,” Riv. del Nouv. Cim. 21, 1–93 (1998). [CrossRef]
2. J. Faure, Y. Glinec, A. Pukhov, S. Kiselev, S. Gordienko, E. Lefebvre, J. P. Rousseau, F. Burgy, and V. Malka, “A laser plasma accelerator producing monoenergetic electron beams,” Nature 431, 541–544 (2004). [CrossRef] [PubMed]
3. C. -M. Ma, I. Veltchev, E. Fourkal, J. S. Li, W. Luo, J. Fan, T. Lin, and A. Pollack, “Development of a laser-driven proton accelerator for cancer therapy,” Laser Phys. , 16, 639–646 (2006). [CrossRef]
4. M. Anand, S. Kahaly, G. Ravindra Kumar, M. Krishnamurthy, A. S. Sandhu, and P. Gibbon, “Enhanced hard x-ray emission from microdroplet preplasma,” Appl. Phys. Lett. 88, 18111 (2006).
5. P. Gibbon, Short pulse laser interactions with matter (Imperial College Press, London, 2005).
6. Y. Glinec, J. Faure, L. Le Dain, S. Darbon, T. Hosokai, J. J. Santos, E. Lefebvre, J. P. Rousseau, F. Burgy, B. Mercier, and V. Malka, “High resolution γ-ray radiography produced by a laser-plasma driven electron source,” Phys. Rev. Lett. 94, 025003–025007 (2004). [CrossRef]
7. M. Al-Rabban, M. Richardson, M. Scott, F. Gilleron, M. Poirier, and T. J. Blenski, EUV sources for lithography, SPIE Press, 299–337 (2006).
8. P. P. Rajeev, P. Taneja, P. Ayyub, A. S. Sandhu, and G. Ravindra Kumar, “Metal nanoplasmas as bright sources of hard x-ray pulses,” Phys. Rev. Lett. 90, 15003 (2003). [CrossRef]
9. S. Kahaly, S. K. Yadav, W. M. Yang, S. Sengupta, Z. M. Sheng, A. Das, P. K. Kaw, and G. Ravindra Kumar, “Near-complete absorption of Intense, ultrashort laser light by sub-λ gratings,” Phys. Rev. Lett. 101, 145001 (2008). [CrossRef] [PubMed]
10. K. E. Drexler, Nanosystems (Wiley, John & Sons, Inc., 1992).
11. C. L. Rettig, W. M. Roquemore, and J. R. Gord, “Efficiency and scaling of an ultrashort-pulse high-repetition-rate laser-driven X-ray source,” App. Phys. B , 93, 365–372 (2008). [CrossRef]
12. A. S. Sandhu, G. R. Kumar, S. Sengupta, A. Das, and P. K. Kaw, “Laser-pulse-induced second-harmonic and hard x-ray emission: role of plasma-wave breaking,” Phys. Rev. Lett. 95, 025005 (2005). [CrossRef] [PubMed]
13. D. W. Forslund, J. M. Kindel, and K. Lee, “Theory of hot-electron spectra at high laser intensity,” Phys. Rev. Lett. 39, 284–288 (1977). [CrossRef]
14. G. Kulcsár, D. AlMawlawi, F. W. Budnik, P. R. Herman, M. Moskovits, L. Zhao, and R. S. Marjoribanks, Intense picosecond x-ray pulses from laser plasmas by use of nanostructured “velvet” targetsPhys. Rev. Lett. 84, 5149–5152 (2000). [CrossRef] [PubMed]
15. M. Kundu and D. Bauer, “Nonlinear resonance absorption in the laser-cluster interaction,” Phys. Rev. Lett. 96, 123401 (2006). [CrossRef] [PubMed]
16. http://cpicanada.org/CCDB/cgi-bin/STAT_NEW.cgi.
17. M. A. Porras, “Diffraction effect in few-cycle optical pulses,” Phys. Rev. E 65, 026606 (2002). [CrossRef]
18. Z. Yan, Y. K. Ho, P. X. Wang, J. F. Hua, Z. Chen, and L. Wu, “Accurate description of ultra-short tightly focused Gaussian laser pulses and vacuum laser acceleration,” App. Phys. B 81, 813–819 (2005). [CrossRef]
19. M. T. Madigan, J. M. Martinko, and D. A. Stahl, Brock biology of microorganisms (Pearson Prentice Hall, New Jersey, 2005).
