Silicon Surface morphologies after femtosecond laser irradiation
Introduction
Over the past several decades, ultra-short-pulse laser irradiation of silicon sur-faces has been an active area of materials science research that has led to a number of unexpected observations and the forma-tion of new materials. The basic physics of this interaction is fully described in the article by Reis et al. in this issue of MRS Bulletin . Starting in the late 1970s, picosec-ond studies 1–8and later femtosecond pump-probe studies 9–21have been used to elucidate the specific mechanism of many processes, including electron–hole plasma formation,9,10,13,14melting,5,9,10,12abla-tion,19–21and ultrafast melting.16–18Ultra-fast melting—the disordering of a “cold”lattice within 100 fs of excitation due to covalent bond weakening upon excitation of more than 15% of the valence elec-trons 22–25—is a phenomenon unique to irradiation with high-intensity femtosec-ond laser pulses, because these pulses are shorter than the electron–phonon relax-ation time. Technologically, ultrashort laser irradiation offers an alternative method for annealing ion-implanted semiconductors.6,7
Early research on the surface morphol-ogy resulting from picosecond laser irra-diation near the melting threshold revealed the formation of ripples on the surface with a wavelength related to the wavelength of the laser.4,26These so-called laser-induced periodic surface structures (LIPSS)27–30are similar to rip-ple structures observed on a variety of ma-terials after irradiation with one or more pulses from a wide range of laser systems (including femtosecond, picosecond, and nanosecond pulses) and are well under-stood. In short, when the laser pulse is en-ergetic enough to fully melt the surface,the incident pulse interferes with light scattered from defects at the surface, set-ting up an inhomogeneous melt depth and the formation of capillary waves,which freeze in place.30Recently , a num-ber of groups have reported the formation of micro- and nano-sized structures result-ing from irradiation with femtosecond laser pulses.31–41
The majority of this research deals with the interaction of a single laser pulse with a flat surface. Consequently , the interaction
of single laser pulses with various materi-als is generally well understood. The in-teraction of multiple laser pulses with a surface that is altered by each pulse, in contrast, is currently not well understood.In this article, we will present two studies on surface morphology resulting from ir-radiation with multiple femtosecond laser pulses. In the next section, we discuss the evolution of micrometer-sized cones on a silicon surface irradiated with hundreds of femtosecond laser pulses in an atmos-phere of sulfur hexafluoride (SF 6) and other gases. After that, we discuss the for-mation of blisters or bubbles at the inter-face between a thermal silicon oxide and the silicon surface after irradiation with one or more femtosecond laser pulses.
Formation of Micrometer-Sized Cones on Silicon via Femtosecond Laser Irradiation
For the past ten years, the Mazur group has extensively studied the surface mor-phology and subsequent properties of sil-icon surfaces irradiated with femtosecond laser pulses in a variety of environments.In 1998, we published the initial discovery that a flat silicon surface is transformed into a forest of quasi-ordered micrometer-sized conical structures (Figure 1) upon irradiation with several hundred femto-second laser pulses in an atmosphere of sulfur hexafluoride (SF 6).42Shortly after-ward, we reported the dependence of cone height on laser fluence and pulse du-ration.43In the subsequent years, we stud-ied the ability of the microstructured surfaces to absorb nearly all incident light in the ultraviolet, visible, and near-in-frared (250–2500 nm) as a result of sulfur being trapped in the material during irra-diation 44–47and successfully employed the process to create silicon-based infrared photodetectors.48We also studied the mor-phology and properties that result from microstructuring silicon in a variety of other environments, including gaseous N 2, Cl 2, H 2, H 2S, Ar, and SiH 4, as well as vacuum, liquid water, and air.45,49–52When microstructured in air, the resulting sur-face photoluminesces.49
The surface morphology after micro-structuring depends strongly on the variables involved in femtosecond laser ir-radiation, including the number of inci-dent laser pulses, laser fluence (energy per unit area), wavelength, pulse duration,and the ambient gas (or liquid) species and pressure. Despite the large variation in surface morphologies obtained by varying experimental parameters, each re-sulting surface follows a similar pattern of evolution with increasing number of
S ilicon Surface
Morphologies after Femtosecond Laser Irradiation
Brian R. T ull, James E. Carey, Eric Mazur ,Joel P . McDonald, and Steven M. Y
alisove
https://www.wendangku.net/doc/682352238.html,/bulletin
pulses. After approximately 500 pulses, a quasi-ordered array of conical structures is formed. The resulting cones are ap-proximately 1–15 μm high and spaced by 1–5 μm (with the exception of surfaces ir-radiated in water, where the cones are an order of magnitude smaller and denser 53).For linearly polarized light, the cones have an elliptical base, with the long axis of the ellipse perpendicular to the polarization axis of the incident light. In the remainder of this section, we describe the general stages of cone formation that take place for silicon irradiated in SF 6using specific conditions and then generalize this forma-tion mechanism for other experimental conditions. These specific (“standard”)conditions are an n -doped Si(111) wafer,260 μm thick, with resistivity ρ?8–12 Ωm, irradiated by a 1-kHz train of 500–600 laser pulses (duration, 100 fs; cen-tral wavelength, 800 nm; spot size, 150 μm FWHM yielding a fluence of 8 kJ/m 2; lin-early polarized perpendicular to the op-tics table), in 67 kPa of SF 6. More detailed experimental procedures can be found in our other papers.43,46
Figure 1 shows a series of scanning elec-tron microscope (SEM) images that illus-trate how the final morphology evolves
20 m
μa b c d
e
f
g
h
i j k l
m n o p
q
r s t
Figure 1. Scanning electron micrographs of a silicon surface after the following number of femtosecond laser pulses: (a)1, (b)2, (c)3, (d)4,(e)5, (f)6, (g)7, (h)8, (i)9, (j)10, (k)12, (l)15, (m)20, (n)30, (o)50, (p)70, (q)100, (r)200, (s)400, and (t)600. Each SEM image is taken at a 45° angle to the surface with the same magnification. Adapted from Reference 54.
with an increasing number of incident laser pulses. Cross sections of the irradi-ated samples show that the tips of the cones are at or below the original surface, indicating that the cones are formed by net ablation rather than by deposition and growth.43This finding is consistent with our laser fluence (8 kJ/m2) exceeding both the melting and ablation thresholds for silicon (1.5 kJ/m2and 3 kJ/m2, respec-tively21). As each pulse is absorbed by the sample, the energy density is greatest at the surface and then decreases deeper into the sample. This density profile causes the topmost layer to be ablated away and the material beneath it to melt and then resolidify to form the resulting surface. A simple, uniform ablation process, how-ever, cannot explain the surface morphol-ogy, as the valleys between the cones are an order of magnitude smaller than the laser spot size.
The formation of the surface morphol-ogy can be separated into two parts, an early stage (1–10 pulses) and a late stage (>10 pulses).54The first pulse causes small defects that are randomly distributed over the surface (Figure 1a). Their circular shape suggests that they result from a burst bubble that is frozen in place upon resolidification of the melt. These bubbles can be attributed to a local increase in va-porization of the silicon melt due to de-fects or impurities at the surface. Apart from these random circular features, the ablation and melting after the first laser pulse appear to be uniform.
After the second pulse, a distinct ripple pattern appears (Figure 1b). The wave-length of the ripple is close to the central wavelength of the incident laser, and the long axis of the ripple is perpendicular to the laser polarization, in agreement with the ripple formation observed in LIPPS. At high fluence, the interference between the incident beam and light scattered by minor surface defects results in inhomo-geneous energy deposition. Ablation and melt formation occur at non-uniform depths, creating capillary waves with the wavelength of the laser. Rapid resolidi-fication subsequently freezes the ripple structure in place.
Figure 2 quantifies the evolution of the periodic patterns at the surface. The graphs to the right of the SEM images are Fourier transforms of the intensity of the SEM image in the horizontal and vertical direc-tions. A peak in the Fourier transform (indicated by an arrow) represents a peri-odicity in the surface at that frequency (corresponding to a periodic distance in the image). In the Fourier spectrums of Figures 2b–2d, the spectrum for the previ-ous figure is shown in gray for comparison.After two pulses (Figure 2a), the Fourier
transforms reveal a distinct periodicity of
the ripple pattern in the horizontal direc-
tion, but not in the vertical direction. After
five pulses (Figure 2b), the ripple pattern
disappears, along with the peak in the
horizontal Fourier spectrum, but a larger
periodicity of 2 μm develops in the vertical
direction. Visually, the ripple pattern is
replaced by small beads spaced ap-
proximately 2 μm apart. After 10 pulses
(Figure 2c), the periodicity in the vertical
direction shifts to larger distances (smaller
frequencies), to a wavelength of 3.5 μm.
At this point, the periodicity of the final
surface morphology is established and the
early stage of formation is over. As these
pictures show, the periodicity established
in the early stage begins with a LIPSS-like
ripple formation with a wavelength of
the laser (Figure 1b), and then changes to
a quasi-periodic array of beads with a
larger wavelength (Figures 1c–1j). The na-
ture of the formation of beads and their
persistence throughout the ablation
process will be discussed after the late-
stage formation.
During late-stage formation, from
pulse 10 to several hundred pulses (Fig-
ures 1j–1t), material is preferentially ab-
lated on the sides of the beads, creating
the resulting conical microstructures with
the beads at their tips. The beads act to
concentrate the light into the valleys be-
tween them. Light that hits the sides of the
beads has a high angle of incidence, and
because reflectivity increases for high an-
gles of incidence, this light is reflected into
the valleys, raising the incident fluence
and increasing the ablation rate.55As the
conical structures become steeper, the effect
is intensified. A final Fourier analysis of the
conical microstructures after 500 pulses is
shown in Figure 2d. The periodicity
moves to slightly longer wavelengths, and
the average spacing of the conical mi-
crostructures is 3.5–4 μm.
Toward the end of the evolution (Fig-
ures 1q–1t), the tips of the cones start to
become wider than regions immediately
below the tip, giving the appearance of a
sphere perched on top of a cone. This ob-
servation may be consistent with some re-
deposition of vapor material at the tip
while the surface is molten via vapor–
liquid–solid growth.56Similar observations
have been made on conical structures
grown on silicon with nanosecond laser
pulses, which also exhibit spheres at the
tip.57–59For nanosecond pulses, the tips of
the conical structures protrude well above
the original surface, suggesting that
growth is a more dominant formation
mechanism than ablation for nanosecond
pulses. A systematic comparison between
cones formed with femtosecond and nano-
second laser pulses appears in Reference 47.
The transition from the ripple structure
to beads is not fully understood, but sev-
eral factors may contribute to the develop-
ment of beads. First, the ripple structure is
essentially a half cylinder on the surface,
and subsequent melting may cause this
structure to bead up. A liquid cylinder that
is longer than its radius is unstable and
collapses into equal-sized equally spaced
drops, a phenomenon known as cylindri-
cal collapse.60Second, during ultrashort
laser irradiation of silicon, the velocity of
the resolidification front is extremely
high.12As a result, a high concentration of
defects can be trapped in the solidified
material, including vacancies, interstitials,
and elements from the background gas
(Rutherford backscattering reveals that
the resolidified surface contains approxi-
mately 0.5–1% of the ambient species
present during the irradiation46,54). If the
melt depth is non-uniform across the sur-
face, the concentration of defects may also
be non-uniform. A recent study61predicts
that the trapping of defects can be inho-
mogeneous during femtosecond laser ab-
lation; interstitials collect in extended
regions, where the surface height is a max-
imum and vacancies collect at minima.
Regardless of the details of the defect dis-
tribution, the inhomogeneous nature of
the surface creates a non-uniform melting
temperature profile preferentially protect-
ing those regions with a higher melting
temperature. It is possible that these re-
gions lead to the formation of beads. The
shift in bead periodicity from 2 μm to
3.5 μm at the end of the early stage can be
attributed to larger beads being protected
by this mechanism while smaller beads
are ablated away.
The specific surface features that de-
velop depend to some degree on the con-
ditions, but the formation mechanism
follows the overall trend described in the
previous few paragraphs for a wide range
of experimental parameters. In nearly
every case, the surface morphology devel-
ops a ripple structure on the order of the
wavelength of the incident light after only
a few pulses. Different conditions, how-
ever, lead to differences in the periodicity
and size of the beads that develop, the
number of pulses required to create bead-
like structures, and the final shape of the
cones. These factors depend on the surface
tension of the liquid silicon that forms a
protective bead, which in turn depends on
melt depth, cooling rate, gas species, and
pressure as well as the inclusion of ele-
ments from the background gas.
For example, when the incident laser
pulses are frequency-doubled so they
have a wavelength of 400 nm, the result-ing LIPSS ripples have a wavelength of 400 nm, and sharp conical structures de-velop in a manner similar to the formation of structures with 800-nm light, except that they are smaller and their density is dou-bled.54In vacuum, ripples develop after a few pulses, but it takes up to 50 pulses for the ripples to coarsen into beads. The re-sulting conical structures are broader and more blunt, with a slightly wider spac-ing than cones formed in SF 6.54
These
Figure 2. Scanning electron micrographs of a silicon surface after (a)2, (b)5, (c) 10, and (d) 500 femtosecond laser pulses. The micrographs were taken normal to the surface. The graphs are Fourier transforms of the intensity of each SEM image in the horizontal (center) and vertical (far right) directions. A peak in the Fourier transform (indicated by an arrow) represents a periodicity in the surface at that frequency
(corresponding to a periodic distance in the image). In the Fourier spectra of Figures 2b–2d, the spectrum for the previous figure is shown in gray for comparison. Adapted from Reference 54.
two examples illustrate how the forma-tion mechanism remains similar for differ-ent conditions. The first case shows that the wavelength of the laser determines the final density of the structures. The second example shows that the background gas has an effect on the number of pulses re-quired to form structures and their overall final shape. The ambient gas changes the surface tension of the molten silicon and determines the elements that become trapped in the molten silicon. For SF6, H2S, and Cl2, the final cones are sharp, as in Figure 1, but for N2, H2, and air, they are blunt, like the cones formed in vac-uum.45,52,54The difference may indicate that certain elements (such as S and Cl) provide more protection against ablation than other elements (N and H). Interface Effects from Femtosecond Laser Irradiation of SiO2-Coated Silicon Femtosecond-laser-induced damage on single-crystalline silicon has been exten-sively studied with a large variety of char-acterization tools, including atomic force microscopy, micro-Raman spectroscopy, and laser scanning microscopy.31,33,39,40,62–68 These studies typically measure damage thresholds and identify morphological damage features that depend on variables such as fluence (energy per unit area), temporal pulse width, polarization, wave-length, and angle of incidence. Few of these papers, however, address the role of the 2-nm-thick native oxide layer. We recently published69results which demon-strate the significant role that the 2-nm na-tive oxide plays on the morphology and damage threshold as compared with atomically clean Si. These results and the model that was developed to explain them motivated us to perform experi-ments on Si with thicker, thermally grown oxide films.
We studied thicker SiO2samples by growing thermal oxides of different thick-nesses.70Indeed, the morphology that is seen in the native oxide disappears for the thinnest thermally grown oxide (20 nm). Instead, an entirely new morphology—either blisters or craters—appears. Figure 3 presents atomic force mi-croscopy (AFM) images illustrating a
range of morphologies that are produced by varying the thermal oxide thickness, the laser fluence, and the number of laser pulses. For a given oxide thickness, blis-ters form at low fluence, 1–2 times the damage threshold of Si(100), while craters are formed as the fluence is increased be-yond a critical value. Unique to ultrafast-laser material interaction is the inherent reproducibility and control over blister and crater dimensions. Using multiple
pulses, blister height can be sequentially
increased (or pumped up) from 100 nm to
900 nm (Figures 3d–3g). Alternatively,
blister dimensions can be controlled by
changing the laser fluence of a single laser
pulse (Figures 3i–3l).
Blister formation in thin films result-
ing from laser irradiation is not a new
phenomenon, but the conditions and
mechanisms at play are quite different for
irradiation with femtosecond laser pulses
than for longer pulse durations.71,72Relax-
ation of residual stress plays a role but is
far from sufficient to explain the observed
blister dimensions.73–75
The mechanism responsible for blister
production by femtosecond laser pulses is Figure3. Atomic force microscopy images of femtosecond-laser-induced damage
features produced on thermally oxidized Si(100). (a)–(h)Oxide thickness ?1200nm.
(i)–(m)Oxide thickness ?147nm. (a)Blister; laser fluence ?7.6 kJ/m2. (b)Crater; laser
fluence ?13.2 kJ/m2. (c)AFM section analysis of features in (a)and (b). (d)–(g)Blister
features produced in 1200-nm-thick thermal oxide on Si(100) with laser fluence ?4.3 kJ/m2 produced using the number of laser pulses as follows: (d)2, (e)4, (f)6, and (g)8. (h)AFM
section analysis of features in (d)–(g). (i)–(l)Damage features produced on thermally oxidized Si(100) with a 147-nm-thick oxide. (i)Blister; laser fluence ?3.0 kJ/m2. (j)Blister; laser
fluence ?3.3 kJ/m2. (k)Blister; laser fluence ?4.0 kJ/m2. (l)Crater feature produced in
147-nm-thick thermal oxide on Si(100) with peak fluence ?19.2 kJ/m2. (m)AFM section
analysis of features in (i)–(l), demonstrating the heights of the blister features and depth of the damage crater.
a combination of compressive stress relax-ation, softening of the thermal oxide via electron heating and conduction, and mo-mentum transfer from the underlying substrate to the oxide layer.75The interac-tion of an above-damage-threshold fem-tosecond laser pulse with the Si(100) substrate produces a dense electron–hole plasma in silicon and energetic electrons that expand in all directions from the exci-tation region.76These energetic electrons reach thermal equilibrium with the lattice as they transfer their kinetic energy to the lattice in a few picoseconds, heating the substrate to about 5000 K.20Due to its proximity to the highly heated substrate as well as heating by electrons scattered from the substrate, the thermal oxide film is left in a softened state. This softening al-lows the ablated substrate material to push the oxide film upward, forming a blister. This model is consistent with the picture of near-threshold ablation pre-sented in the article by Reis et al. in this issue. It is also consistent with time-resolved pump-probe imaging experi-ments of near-threshold ablation of metals and semiconductors, where interference phenomena suggest that a moving liquid layer is expelled.19,21,77
By overlapping blisters produced by femtosecond laser pulses in a 1200-nm-thick thermal oxide layer on Si(100), one can produce channels that can be used in microfluidics applications. The top sur-face of the channel is the delaminated thermal oxide film and the freshly oxi-dized 40-nm film of Si; the bottom surface is the underlying Si(100) substrate, which has a thermal oxide that forms as the molten Si solidifies in the presence of air.75,78A single pass of the sample through the focused laser beam at a speed of 10 mm/s and a laser fluence of 3.5 kJ/m2 with a focused laser beam diameter of 55±5 μm on the sample produces channels 24 ±1 μm wide and 355 ±45 nm in height. To make larger channels, we laterally overlapped single-pass channels, produc-ing channels with widths exceeding 300μm. The height of the channels is a func-tion of channel width (see Figure 4). We also produced linear channels with lengths exceeding 10 mm, as well as other channel geometries, including intersec-tions, corners, and curves. SEM images of the end of a channel are presented in Fig-ures 4i–4j, showing the morphology of the substrate or bottom surface of the chan-nels. Similar channels have been pro-duced by selectively delaminating
diamond-like carbon films via litho-graphic techniques.74
Compared with other femtosecond-laser-based micromachining techniques for producing microfluidic channels, the
technique discussed here has the advan-
tage of producing little debris at relatively
fast writing speeds.79,80In general, direct-
writing techniques are simpler than litho-
graphic techniques, because the channels Figure4. Nano-and microfluidic channels produced with the femtosecond laser direct-write technique with laser fluence of 3.5 kJ/m2at a single-pass scan rate of 1cm/s. (a)Optical
microscope image of channels produced with (from left to right) a single pass to 5 over-
lapped (overlap ?15mm) passes. (b)Simple grid device produced with the femtosecond laser direct-write technique. (c)–(g)AFM images of channels produced with single and mul-tiple passes. (c)Single pass. (d)Six passes. (e)AFM cross section analysis of channels;
(c) and (d) are indicated. (f)AFM of corner channels. (g)AFM of channel intersection.
(h)Plot of channel height as a function of channel width, with results of linear fit shown.
(i)SEM image (tilt ?59°) of end of channel written to the edge of sample. The interior of the
channel was exposed by intentionally fracturing the delaminated glass at the edge of the
sample. (j)SEM image from inset in (h), showing the roughness on the substrate surface
and bottom surface of the delaminated thermal oxide film.
can be created with a single processing step, allowing adjustments to the fluidic network design to be implemented quickly without the need to produce a new mask. The channels produced via femtosecond-laser-induced delamination of thermal oxide films from Si(100) sub-strates exhibit a noncircular cross section, which is quite different from those pro-duced by other techniques.79,81However, the delaminated glass is very thin and may require a layer of poly(dimethylsilox-ane) or other material to make these frag-ile systems more robust.
Summary
The interaction of intense, ultrafast laser pulses with materials produces a variety of damage morphologies that depend on both the laser irradiation and the material of interest. Many other interesting mor-phological phenomena have been ob-served and characterized, including laser-induced periodic surface structures or ripples,34,82,83gratings produced by in-terfering two laser beams on a surface,84,85 and so-called nanobumps and nanojets in thin gold films on quartz substrates.86A number of recent papers report on the ef-fect of liquid35,87and gaseous88,89environ-ments on the resulting morphology during machining. Controlling damage morphology is essential for improving mi-cromachining capabilities. The resulting damage morphologies can also prove use-ful in their own right, as in the case of the two studies presented in this article. The intersection of materials research and ul-trafast optical science is producing many valuable fundamental scientific results, and the trend is expected to evolve as new and exciting discoveries are made. References
1.P.L. Liu, R. Yen, N. Bloembergen, and R.T. Hodgson, Appl. Phys. Lett.34(1979) p.864.
2.J.M. Liu, R. Yen, E.P. Donovan, N. Bloember-gen, and R.T. Hodgson, App. Phys. Lett.38 (1981) p.617.
3.J.M. Liu, R. Yen, H. Kurz, and N. Bloember-gen, Appl. Phys. Lett.39(1981) p.755.
4.D.Y. Sheng, R.M. Walser, M.F. Becker, and J.G. Ambrose, Appl. Phys. Lett.39(1981) p.99.
5.K.L. Merkle, H. Baumgart, R.H. Uebbing, and
F. Phillipp, Appl. Phys. Lett.40(1982) p.729.
6.Y.I. Nissim, J.Sapriel, and J.L. Oudar, Appl. Phys. Lett.42(1983) p.504.
7.Y. K anemitsu, I. Nakada, and H. K uroda, Appl. Phys. Lett.47(1985) p.939.
8.W.K. Wang and F. Spaepen, J.Appl. Phys.58 (1985) p.4477.
9.J.M. Liu, H. Kurz, and N. Bloembergen, Appl. Phys. Lett.41(1982) p.643.
10.D. von der Linde and N. Fabricius, Appl. Phys. Lett.41(1982) p.991.
11.L.A. Lompre, J.M. Liu, H. K urz, and N. Bloembergen, Appl. Phys. Lett.43(1983) p.168.12.P.H. Bucksbaum and J.Bokor, Phys. Rev.
Lett.53(1984) p.182.
13.L.A. Lompre, J.M. Liu, H. K urz, and
N. Bloembergen, Appl. Phys. Lett.44(1984) p.3.
14.H.M. Vandriel, L.A. Lompre, and N.
Bloembergen, Appl. Phys. Lett.44(1984) p.285.
15.I.W. Boyd, S.C. Moss, T.F. Boggess, and A.L.
Smirl, Appl. Phys. Lett.46(1985) p.366.
16.C.V. Shank, R. Yen, and C. Hirlimann, Phys.
Rev. Lett.51(1983) p.900.
17.C.V. Shank, R. Yen, and C. Hirlimann, Phys.
Rev. Lett.50(1983) p.454.
18.H.W.K. Tom, G.D. Aumiller, and C.H.
Britocruz, Phys. Rev. Lett.60(1988) p.1438.
19.K. Sokolowski-Tinten, J.Bialkowski, A.
Cavalleri, D. von der Linde, A. Oparin,
J.Meyer-ter-V ehn, and S.I. Anisimov, Phys. Rev.
Lett.81(1998) p.224.
20.A. Cavalleri, K. Sokolowski-Tinten,
J.Bialkowski, M. Schreiner, and D. von der
Linde, J.Appl. Phys.85(1999) p.3301.
21.D. von der Linde and K. Sokolowski-
Tinten, Appl. Surf. Sci.154(2000) p.1.
22.P. Stampfli and K.H. Bennemann, Progr.
Surf. Sci.35(1990) p.161.
23.P. Stampfli and K.H. Bennemann, Phys. Rev.
B42(1990) p.7163.
24.P. Stampfli and K.H. Bennemann, Phys. Rev.
B46(1992) p.10686.
25.P. Stampfli and K.H. Bennemann, Phys. Rev.
B49(1994) p.7299.
26.P.M. Fauchet and A.E. Siegman, Appl. Phys.
Lett.40(1982) p.824.
27.J.E. Sipe, J.F. Young, J.S. Preston, and H.M.
Vandriel, Phys. Rev. B27(1983) p.1141.
28.J.F. Young, J.S. Preston, H.M. Vandriel, and
J.E. Sipe, Phys. Rev. B27(1983) p.1155.
29.J.F. Young, J.E. Sipe, and H.M. Vandriel,
Phys. Rev. B30(1984) p.2001.
30.H.M. Vandriel, J.E. Sipe, and J.F. Young,
J.Lumin.30(1985) p.446.
31.J.Bonse, S. Baudach, J.Kruger, W. Kautek,
and M. Lenzner, Appl. Phys. A74(2002) p.19.
32.M. Hirano, K. Kawamura, and H. Hosono,
Appl. Surf. Sci.197(2002) p.688.
33.A.P. Singh, A. K apoor, K.N. Tripathi, and
G.R. Kumar, Opt. Laser T echnol.34(2002) p.37.
34.F. Costache, S. K outeva-Arguirova, and
J.Reif, in Gettering and Defect Engineering in
Semiconductor T echnol ogy, edited by H. Richter
and M. Kittler (Trans-Tech, Zurich, 2004) p.635.
35.G. Daminelli, J.K ruger, and W. K autek,
Thin Solid Films467(2004) p.334.
36.C.W. Hee, B.K.A. Ngoi, L.E.N. Lim, K.
V enkatakrishnan, and W.L. Liang, Opt. Laser
T echnol.37(2005) p.93.
37.W. K autek, P. Rudolph, G. Daminelli, and
J.Kruger, Appl. Phys. A81(2005) p.65.
38.R. Le Harzic, H. Schuck, D. Sauer, T. Anhut,
I. Riemann, and K. Konig, Opt. Express13(2005)
p.6651.
39.T. Matsumura, A. K azama, and T. Yagi,
Appl. Phys. A81(2005) p.1393.
40.D.V. Tran, H.Y. Zheng, Y.C. Lam, V.M.
Murukeshan, J.C. Chai, and D.E. Hardt, Opt.
Lasers Eng.43(2005) p.977.
41.S.V. Zabotnov, I.A. Ostapenko, L.A.
Golovan, V.Y. Timoshenko, P.K. Kashkarov, and
G.D. Shandybina, Quantum El ectron.35(2005)
p.943.
42.T.-H. Her, R.J. Finlay, C. Wu, S. Deliwala,
and E. Mazur, Appl. Phys. Lett.73(1998) p.1673.
43.T.-H. Her, R.J. Finlay, C. Wu, and E. Mazur,
Appl. Phys. A70(2000) p.383.
44.C. Wu, C.H. Crouch, L. Zhao, J.E. Carey, R.J.
Younkin, J.A. Levinson, E. Mazur, R.M. Farrel,
P. Gothoskar, and A. Karger, Appl. Phys. Lett.78
(2001) p.1850.
45.R.J. Younkin, J.E. Carey, E. Mazur, J.A.
Levinson, and C.M. Friend, J.Appl. Phys.93
(2003) p.2626.
46.C.H. Crouch, J.E. Carey, M. Shen, E. Mazur,
and F.Y. Genin, Appl. Phys. A79(2004) p.1635.
47.C.H. Crouch, J.E. Carey, J.M. Warrender,
M.J. Aziz, E. Mazur, and F.Y. Genin, Appl. Phys.
Lett.84(2004) p.1850.
48.J.E. Carey, C.H. Crouch, M. Shen, and
E. Mazur, Opt. Lett.30(2005) p.1773.
49.C. Wu, C.H. Crouch, L. Zhao, and E. Mazur,
Appl. Phys. Lett.81(2002) p.1999.
50.M.A. Sheehy, “Femtosecond Laser Mi-
crostructuring of Silicon: Dopants and Defects,”
PhD dissertation, Harvard University (2004).
51.M. Shen, C.H. Crouch, J.E. Carey, and E.
Mazur, Appl. Phys. Lett.85(2004) p.5694.
52.M.A. Sheehy, L. Winston, J.E. Carey, C.M.
Friend, and E. Mazur, Chem. Mater.17(2005)
p.3582.
53.M. Shen, C.H. Crouch, J.E. Carey, R.J.
Younkin, E. Mazur, M.A. Sheehy, and C.M.
Friend, Appl. Phys. Lett.82(2003) p.1715.
54.J.E. Carey, “Femtosecond Laser Microstruc-
turing of Silicon for Novel Optoelectronic De-
vices,” PhD dissertation, Harvard University
(2004).
55.C. Wu, “Femtosecond Laser Gas–Solid In-
teractions,” PhD dissertation, Harvard Univer-
sity (2000).
56.R.S. Wagner and W.C. Ellis, Appl. Phys. Lett.
4(1964) p.89.
57.F. Sanchez, J.L. Morenza, R. Aguiar, J.C.
Delgado, and M. Varela, Appl. Phys. Lett.69
(1996) p.620.
58.F. Sanchez, J.L. Morenza, R. Aguiar, J.C.
Delgado, and M. Varela, Appl. Phys. A66(1998)
p.83.
59.A.J. Pedraza, J.D. Fowlkes, and D.H.
Lowndes, Appl. Phys. Lett.74(1999) p.2322.
60.T.M. Rassias, Ed., The Problem of Plateau: A
Tribute to Jesse Doug l as and Tibor Rado,
1st ed. (World Scientific, River Edge, NJ, 1992).
61.V.I. Emel’yanov and D.V. Babak, Appl. Phys.
A74(2002) p.797.
62.S. Amoruso, G. Ausanio, A.C. Barone,
E. Bruzzese, L. Gragnaniello, M. Vitiello, and
X. Wang, J.Phys. B 38(2005) p.L329.
63.J.Bonse, K.W. Brzezinka, and A.J. Meixner,
Appl. Surf. Sci.221(2004) p.215.
64.A. Borowiec, M. Mackenzie, G.C. Weath-
erly, and H.K. Haugen, Appl. Phys. A76(2003)
p.201.
65.T.Y. Choi and C.P. Grigoropoulos, J.Heat
T ransfer–T rans. ASME126(2004) p.723.
66.F. Costache, S. K outeva-Arguirova, and
J.Reif, Appl. Phys. A79(2004) p.1429.
67.E. Coyne, J.P. Magee, P. Mannion, G.M.
O’Connor, and T.J. Glynn, App. Phys. A81
(2005) p.371.
68.B.K.A. Ngoi, K. V enkatakrishnan, E.N.L.
Lim, B. Tan, and L.H.K. Koh, Opt. Lasers Eng.35
(2001) p.361.
69.J.P. McDonald, A.A. McClelland, Y.N.
Picard, and S.M. Yalisove, Appl. Phys. Lett.86
264103 (2005).
70.J.P
. McDonald, V .R. Mistry , K.E. Ray
, and S.M. Yalisove, in “Thin Films—Stresses and Mechanical Properties XI,” edited by T.E.Buchheit, A.M. Minor, R. Spolenak, and K .Takashima (Mater. Res. Soc. Symp. Proc.875,Warrendale, PA, 2005) p.359.
71.J.R. Serrano and D.G. Cahill, J.Appl. Phys.92(2002) p.7606.
72.K. Xiao, Z.S. Guan, G.J. Wang, L. Jiang, D.B.Zhu, and Y .R. Wang, Appl. Phys. Lett.85(2004)p.1934.
73.J.W. Hutchinson and Z.Suo, in Adv. Appl.Mech. 29(1992) p.63.
74.D.B. Marshall and A.G. Evans, J.Appl. Phys.56(1984) p.2632.
75.J.P . McDonald, V .R. Mistry , K.E. Ray , and S.M. Yalisove, Appl. Phys. Lett. 88183113 (2006).76.K . Sokolowski-Tinten and D. von der Linde, Phys. Rev. B 61(2000) p.2643.
77.B. Rethfeld, K. Sokolowski-Tinten, D. von der Linde, and S.I. Anisimov , Appl. Phys. A 79(2004) p.767.
78.J.P . McDonald, V .R. Mistry , K.E. Ray , J.A.Nees, N.R. Moody , and S.M. Yalisove, Appl.Phys. Lett. 88153121 (2006).
79.K. Ke, E.F. Hasselbrink, and A.J. Hunt, Anal.Chem.77(2005) p.5083.
80.T.N. K im, K. Campbell, A. Groisman, D.Kleinfeld, and C.B. Schaffer, Appl. Phys. Lett.86201106 (2005).
81.D. Therriault, S.R. White, and J.A. Lewis,Nature Mater.2(2003) p.265.
82.A. Borowiec and H.K. Haugen, Appl. Phys.Lett.82(2003) p.4462.
83.T.H.R. Crawford, A. Borowiec, and H.K .Haugen, Appl. Phys. A 80(2005) p.1717.
84.B. Tan, N.R. Sivakumar, and K . V enka-takrishnan, J.Optics A 7(2005) p.169.
85.K. V enkatakrishnan, N.R. Sivakumar, and B. Tan, Appl. Phys. A 76(2003) p.143.86.F. Korte, J.Koch, and B.N. Chichkov , Appl.Phys. A 79(2004) p.879.
87.J.Ren, M. Kelly , and L. Hesselink, Opt. Lett.30(2005) p.1740.
88.W. Perrie, M. Gill, G. Robinson, P . Fox, and W. O’Neill, Appl. Surf. Sci.230(2004) p.50.
89.J.Sun and J.P . Longtin,J.Opt. Soc. Am. B 21(2004) p.1081.
?
■Trace and Ultra Trace Elemental Analysis Laboratory ■Bulk, Near Surface and Depth Profile Elemental Evaluation ■Conductive, Non Conductive and Semi Conductive Materials
Analysis from H to U (solids and powders)■Minimum sample requirements
(less than one gram required for 75 elements scan)■
Analytical Capabilities
?Trace Elemental Analysis by GD/MS ?Compositional Analysis by ICP-OES ?Metallography
? C and S by Combustion Technique ?N,O, H by Inert Gas Fusion
ISO 9002, NADCAP , GE—S400Lab facilities in US, Europe and Asia
Shiva Technologies Inc.
6707 Brooklawn Parkway
Syracuse, NY 13211Phone: 315-431-9900Fax: 315-431-9800sales@https://www.wendangku.net/doc/682352238.html,
https://www.wendangku.net/doc/682352238.html,