High spectral resolution multiplex CARS spectroscopy using chirped pulses
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Chemical Physics Letters 387 (2004) 436–441
www.elsevier.com/locate/cplett
High spectral resolution multiplex CARS spectroscopy
using chirped pulses
K.P. Knutsen, J.C. Johnson, A.E. Miller, P.B. Petersen, R.J. Saykally *
Department of Chemistry, University of California, D31 Hildebrand, Berkeley, CA 94720-1460, USA
Received 6 February 2004
Published online: 10 March 2004
Abstract
A simple technique for achieving high spectral resolution coherent anti-Stokes Raman scattering (CARS) spectra with a fem-
tosecond laser system is presented. A linearly chirped and stretched (10 ps) pump pulse generates CARS signal only when
overlapped in time with the Stokes pulse (90 fs), creating a ‘temporal slit’ that defines the spectral resolution of the technique.
Multiplex CARS spectra for liquid methanol and liquid isooctane are presented, demonstrating a spectral resolution of better than
5 cm1 . This new chirped (c-CARS) technique should prove useful for chemically-selective imaging applications, as it significantly
reduces the non-resonant background contribution.
Ó 2004 Elsevier B.V. All rights reserved.
1. Introduction the advantage of utilizing femtosecond lasers due to
their low average powers. However, a femtosecond pulse
Coherent anti-Stokes Raman scattering (CARS) is a width implies low spectral resolution (e.g., 200 cm1
third-order process wherein the interaction of three relative to sample line widths of ca. 15 cm1 ), which
beams, xp and xp0 (the usually degenerate ‘pump’ beams), reduces the resonant to non-resonant CARS signal, and
and xs (the ‘Stokes’ beam) generates the anti-Stokes sig- hence, the chemical selectivity.
nal at the frequency xas ¼ 2xp xs . By tuning the energy Conventional experimental designs employing
difference xp xs over a molecular resonance, typically a narrow bandwidth (picosecond) pulsed lasers achieve
vibrational transition, one can achieve a large enhance- sample-limited spectral resolution, but have the disad-
ment in the signal, relative to the accompanying non- vantage of requiring frequency scanning to generate a
resonant background. The CARS signal is dependent spectrum, increasing acquisition time. Recent compel-
upon both the type and environment of chemical bonds ling demonstrations of multiplex CARS have all em-
and is therefore a powerful tool for studying chemical ployed broadband (femtosecond) light sources to
properties of a system. simultaneously excite several vibrational modes in the
Recently there has been much interest in the devel- sample, followed by a probe pulse that is of narrow
opment of CARS microscopy for chemically-selective bandwidth. This can be accomplished using a synchro-
imaging of complex (e.g., biological) systems using en- nized femtosecond/picosecond laser combination,
dogenous chromophores, and both near-field [1] and wherein the picosecond laser defines the spectral reso-
far-field [2,3] imaging studies of cells have been reported. lution [2,4], but a disadvantage of this technique is the
Current limitations of these techniques include low sig- necessity to synchronize the two separate laser systems,
nal levels and heating of the samples, both suggesting which increases the complexity of the experiment. An-
other approach, demonstrated in sum frequency gener-
ation spectroscopy, is to disperse the bandwidth of a
*
Corresponding author. Fax: +1-510-642-8369/8566. femtosecond pulse (e.g., with prisms or gratings), and
E-mail address: saykally@uclink4.berkeley.edu (R.J. Saykally). then introduce a slit in the dispersed beam to decrease its
0009-2614/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2004.02.049K.P. Knutsen et al. / Chemical Physics Letters 387 (2004) 436–441 437
bandwidth, effectively converting the femtosecond pulse regenerative amplifier (90 fs, 1 kHz, Spectra Physics,
into a picosecond pulse [5]. Spitfire) before entering an optical parametric amplifier
Here we describe a simple variation on the latter (OPA) (90 fs, 60 lJ at 2100 nm, Light Conversion,
design, wherein the spectral resolution of our instrument TOPAS) which generates the near-infrared idler pulse
is defined by a ‘temporal slit’ rather than a mechanical for the Stokes beam (xs ). A small amount (5%) of the
slit. Through the use of chirped laser pulses, the spectral s-polarized 800 nm beam from the amplifier is picked off
resolution is defined by the temporal overlap of the 90 fs by a beam-splitter before entering the OPA. This pulse
pulse with a temporally chirped pulse stretched to sev- makes four passes between two gratings (800 nm blaze,
eral picoseconds. The advantages of this technique are 830 grooves/mm, Edmund Industrial Optics), stretching
that the spectral resolution can be easily adjusted via the the pulse in time due to the induced linear negative
grating separation to match the natural line widths of temporal chirp, resulting in the blue frequencies of the
the sample transitions, thereby increasing the ratio of pulse leading the red frequencies in time. The near-in-
resonant to non-resonant CARS signal. Since all of the frared (2100 nm) s-polarized idler beam from the OPA is
pulses originate from the same laser, this technique also frequency-doubled by a BBO crystal to a p-polarized
obviates the need for synchronizing two separate laser 1050 nm (xs ) beam, and temporally overlapped with a
systems. fraction of the chirped 800 nm pulse [the degenerate
The use of chirped pulses in nonlinear optical spec- pump (xp ) and probe (xp0 ) pulse] using a delay stage.
troscopy has been explored previously to study the dy- The two beams are then focused to a 10 lm spot by an
namics of molecules in solution [6,7], and when used in objective (Olympus, 10x, 0.3 N.A.) and collinearly excite
conjunction with a Michelson interferometer, chirped the sample with 3 lJ (xp ) and 1 lJ (xs ). An identical
pulses provide a simpler alternative to pulse-shaping [8] collection objective collimates the CARS signal into a
for obtaining high spectral resolution four-wave mixing fiber optic that directs the signal to a spectrograph/CCD
spectra and to control the coherence of the excited states (Roper Scientific, spectral resolution 0.1 nm), typically
for gas phase [9,10] and condensed phase samples [11]. integrating 300–400 laser shots.
While similar in concept, the stepping of the delay be-
tween the xp0 and xs pulses and the added analysis of 2.2. Autocorrelation of chirped pulses
the resultant signal makes those designs more compli-
cated than that which we describe in this Letter. In order to characterize the extent of the chirp ef-
In addition, Cheng et al. [2] described the effect that a fected by the grating pair, a homebuilt Michelson
chirped pulse has on the observed spectra in their mul- autocorrelator was employed to measure the pulse width
tiplex CARS setup, noting that the overall spectral of the unchirped and chirped xp pulses as a function of
bandwidth of their CARS signal decreased for increased grating separation (Table 1). The temporally chirped
chirp rate of the Stokes pulse, attributing this to the pulse width is directly proportional to the grating sep-
decreased overlap of the spectral components of the aration [12], described by:
chirped pulse. However, we demonstrate below that 2bðk=dÞdk
more interesting effects occur when the probe pulse is dk s ¼ 2
; ð1Þ
chirped, namely increased spectral resolution and en- cd½1 ðk=d sin cÞ
hanced resonant signal to background. where dk s is the variation of group delay, b is the slant
distance between the gratings, d is the grating constant,
dk is the bandwidth of the pulse, c is the speed of light,
2. Experimental and c is the angle of incidence (0° in our design).
CARS signal is generated only when the pulses
2.1. Optical design for c-CARS overlap in time on the sample, as defined by the shorter,
unchirped 90 fs xs pulse. Therefore, an effective ‘tem-
Femtosecond pulses for the pump and Stokes beams poral slit’ is introduced into the optical path since the
are generated by a home-built Ti:sapphire oscillator (40 90 fs pulse temporally overlaps with only a small portion
fs, 300 mW, 88 MHz). Seed pulses are amplified by a of the much longer xp pulse. Due to the temporal chirp,
Table 1
Autocorrelation results and calculated pulse widths for chirped and unchirped xp pulses
Grating separation (cm) Measured pulse width (ps) Calculated pulse width (ps) Spectral resolution (cm1 )
0 0.09 0.01 – 300
7.5 9.4 0.2 9.1 3
13.5 22.4 0.4 22.0 1
34.1 53 1 55.6 0.5438 K.P. Knutsen et al. / Chemical Physics Letters 387 (2004) 436–441
2000
2x idler, ωs 1800
-1
Chirped 800 nm, ω p
1600
2940 cm
-1 2825 cm
CARS signal (a.u.)
-1 -1
Spectrometer/ 1400 (38 cm FWHM) (25 cm FWHM)
objective CCD
BBO
Grating 1 1200
sample
objective 1000
800
CARS
600
Fiber optic Idler from OPA
400
(a) Unchirped 800 nm from amplifier Grating 2
200
0
3200 3000 2800 2600
-1
(a) Raman shift (cm )
10 ps
100 fs
ωp ω as
ωs
E ωp
vasym
(b) t
vsym
Fig. 1. (a) Optical design for chirped CARS. The 800 nm (xp ) pulse
from the amplifier passes 4 times between two gratings, stretching the (b)
vgs
pulse in time due to an induced linear negative temporal chirp that is
directly proportional to the grating separation. The transform-limited Fig. 2. (a) CARS spectrum of the asymmetric and symmetric C–H
near-infrared (2100 nm) idler pulse from the OPA is doubled by a BBO stretches of liquid methanol at 2940 and 2825 cm1 , respectively.
crystal to 1050 nm (xs ) and is temporally overlapped with the chirped (b) Multiplex CARS energy level diagram for liquid methanol. The
pulse using a delay stage. (b) A ‘temporal slit’ where only a fraction of chirped pump (xp ) pulse is gated by the 90 fs Stokes pulse (S ) and thus
the xp bandwidth generates CARS signals. its effective spectral bandwidth is determined by the extent of the chirp.
The spectrally broad Stokes pulse (xs ) simultaneously excites both
vibrational levels, and the spectrally narrow the chirped probe pulse
(xp ) resolves the two virtual state transitions, generating the observed
this overlap also corresponds to only a small fraction of chirped CARS spectrum.
the xp frequency bandwidth, as shown in Fig. 1b. Eq.
(2) describes the effective spectral resolution of the
CARS output (xas ¼ 2xp xs ) in the limiting case that only minimal spectral distortion resulting from inter-
the chirped (xp ) pulse is much longer than the xs pulse: ference with the non-resonant electronic CARS back-
ground. The energy level diagram in Fig. 2b shows how
Dts
Dxp ¼ Dxspec ; ð2Þ the observed multiplex chirped CARS spectrum for li-
Dtp
quid methanol is generated. Both the symmetric and
where Dts is the pulse width of the unchirped xs pulse, anti-symmetric C–H stretches are excited by the xp and
Dtp is the pulse width of the chirped (xp ) pulse, Dxp is xs pulses, followed by the chirped probe pulse (xp ),
the spectral bandwidth of the chirped (xp ) pulse and which is effectively spectrally-narrowed by the temporal
Dxspec is the resulting effective spectral resolution of the slit, then resolves the two excited state transitions, gen-
CARS signal. The spectral resolution determined for erating the observed chirped CARS spectrum. The
each grating separation is given in Table 1. measured line width (FWHM) of the symmetric C–H
stretch at 2825 cm1 is 25 cm1 , while that for the
asymmetric stretch is 38 cm1 . The spontaneous Raman
3. Results and discussion spectrum of methanol also exhibits a broader asym-
metric stretch line width of 30–40 cm1 [13]. The line
3.1. c-CARS of liquid methanol and liquid isooctane widths measured here remain the same for all grating
separations used, implying that homogeneous broaden-
As shown in Fig. 2, the resonance enhancement ing mechanisms limit the spectral resolution.
achieved with high spectral resolution chirped CARS The narrowest observed spectral resolution was
dominates the CARS spectrum of liquid methanol, with achieved for liquid isooctane (Fig. 3). Numerous ali-K.P. Knutsen et al. / Chemical Physics Letters 387 (2004) 436–441 439
5
3.5x10
the maximum CARS signal for liquid methanol remains
-1
centered at the same wavelength. The CARS center
5
3.0x10 2963 cm
wavelength does not change as xs is adjusted since it resu-
CARS signal (a.u.)
-1 -1
2906 cm 2901 cm
5
2.5x10 lts from the summation of xp with the fixed C–H vibra-
5 tional frequencies, which are not affected by the xs
2.0x10
wavelength (see Fig. 4b). However, as the maximum
5 2920 2910 2900 2890
1.5x10 -1 spectral intensity of the xs pulse is shifted across the vibra-
2930 cm
5
1.0x10 2875 cm
-1 tional energy levels, the relative intensities and shapes of
4 the two C–H peaks change due to different excited popula-
5.0x10
tions of the two vibrational energy levels and to varying
0.0
contributions from the non-resonant background, caus-
3100 3000 2900 2800 2700 ing the distortion of the resonant CARS peaks [4].
-1
Raman shift (cm )
3.3. Effect of delay on c-CARS spectra
Fig. 3. CARS spectrum of liquid isooctane. Inset: two CH3 symmetric
stretch transitions, separated by 5 cm1 , demonstrating that a spectral
resolution of ca. 3 cm1 is achieved. When the delay between the chirped xp pulse and the
xs pulses is varied, the temporal slit selects different xp
frequencies, causing the overall CARS spectrum to shift
phatic C–H vibrations were observed. Note specifically since the center CARS wavelength is determined by the
that the CH3 symmetric stretch appears as two clearly sum of the vibrational frequencies and the pump fre-
resolved and reproducible peaks, with a splitting of only quency, as mentioned above (see Figs. 1b and 5). All
5 cm1 (Fig. 3 inset). These two peaks remain resolvable delay times are reported relative to the delay producing
for all grating separations, and follow the expected po- the largest resonant CARS signal (t ¼ 0). The wave-
sitions when the Stokes frequency and relative delay length shift is directly proportional to the relative delay,
between the pulses are varied. as determined by the product of the xp bandwidth and
the ratio of delay time to xp pulse width. For example,
3.2. Effect of varying xs wavelengths for a relative delay of 3.4 ps, there would be a 109 cm1
shift in the center CARS wavelength [300 cm1 * (3.4 ps/
Resonance enhancement for the observed C–H vibra- 9.4 ps) ¼ 109 cm1 ]. This is observed in Fig. 5a by the
tions is confirmed by generating the chirped CARS near overlap of the two resonantly enhanced methanol
spectra with different xs wavelengths. As shown in C–H vibrations for 3.4 ps relative delay, as predicted
Fig. 4a, for xs wavelengths ranging from 1075 to 1025 nm, by their Raman spacing (115 cm1 ). This consistency
40000
35000 1075 nm
CARS signal(a.u.)
1062.5
30000
1050 nm
25000 1037.5 nm
1025 nm
20000
15000
10000
5000
0
640 650 660 670
(a) wavelength (nm)
E
(b)
Fig. 4. (a) Observed liquid methanol CARS spectra for xs between 1025 nm and 1075 nm. CARS spectra are centered at the same wavelength;
however, the relative intensities of the two C–H peaks change due to different excited populations as the maximum intensity of the xs pulse shifts
across the vibrational energy levels. (b) Energy level diagrams for decreasing wavelength of xs (from left to right).440 K.P. Knutsen et al. / Chemical Physics Letters 387 (2004) 436–441
-1.7 ps delay (x2)
t=0 (m ax intensity)
40000 -1 +1.7 ps delay (x2)
2940 cm +3.4 ps delay (x10)
CARS signal (a.u.)
35000
30000 -1
2825 cm
25000
20000
15000
10000
5000
0
640 645 650 655 660
(a)
wavelength (nm )
E
(b)
Fig. 5. (a) Scaled spectra of liquid methanol for adjusting the delay between xp to xs . For increased delays, the 2825 cm1 peak increases in relative
intensity compared to the 2940 cm1 peak, and the center CARS wavelength shifts to the red. A delay of 3.4 ps results in an overlap of the 2825 cm1
and 2940 cm1 peaks. (b) Energy level diagrams for adjusting delay.
further establishes the mechanism of the observed the chirp, which through its overlap with the unchirped
chirped CARS spectrum with our technique. 90 fs Stokes pulse, defines the spectral resolution of the
Recent theoretical studies by Naumov and Zheltikov system. Ultimately, more theoretical work is necessary
proposed two different experimental techniques em- to rigorously describe the spectral resolution achieved
ploying chirped pulses to obtain CARS signals [14,15]. with the experimental design we describe here.
The first of these, recently demonstrated by Lang et al.
[16], maps the transient CARS spectrum in one laser
shot by delaying a chirped probe pulse after the initial 4. Conclusions
pump and Stokes pulses to spatially disperse the tem-
poral evolution of a CARS signal. While this technique We have shown that the two-pulse chirped-CARS
can extract dynamical information, it cannot produce a technique allows for very simple control of experimental
high spectral resolution CARS spectrum. The second spectral resolution. The novelty of the technique is that
method incorporates two chirped pulses and one un- it renders bandwidth reduction unnecessary, instead
chirped femtosecond pulse to obtain high spectral res- exploiting only the chirp of the probe to achieve high
olution CARS spectra in one shot [14,15]. This proposed spectral resolution. As with other multiplex CARS
scheme is closer to the technique described here; how- techniques, by dispersing the CARS signal onto a CCD,
ever, their method of achieving high spectral resolution we can observe the entire (ca. 300 cm1 ) CARS spec-
comes primarily from the spatial overlap of large trum in a single laser shot, without the need for ad-
(3 cm beam waist) chirped beams with spatially justing the delay of the pulses, thus allowing for direct
smaller chirped (or unchirped) beams. The geometrical comparisons between peak intensities within the
requirements of such an experiment make it unfeasible broadband region of interest. Through the use of dif-
for CARS microscopy, thus our design is more suitable fraction gratings to induce the linear chirp, we can
for application to biologically relevant systems. readily adjust the experimental spectral resolution to
The second limitation on the spectral resolution de- match the line width of the vibrations in the sample,
scribed in their paper results from the temporal overlap increasing the ratio of resonant to non-resonant CARS
of only the first two (pump and Stokes) pulses, similar to signals through more efficient use of the laser power.
earlier studies that used chirped pulses to control the The simplicity of this experimental design enables high
bandwidth of second harmonic generation from a BBO spectral resolution CARS spectra to be easily obtained
crystal [17]. This bandwidth calculation entirely neglects with femtosecond laser systems, and should prove useful
contributions from the final probe pulse, including for imaging with chemically-selective contrast.K.P. Knutsen et al. / Chemical Physics Letters 387 (2004) 436–441 441
Acknowledgements [5] L.J. Richter, T.P. Petralli-Mallow, J.C. Stephenson, Opt. Lett. 23
(1998) 1594.
[6] E.T.J. Nibbering, D.A. Wiersma, K. Duppen, Phys. Rev. Lett. 68
This work was funded by the Experimental Physical (1992) 514.
Chemistry Program of the National Science Founda- [7] K. Duppen, F. deHaan, E.T.J. Nibbering, D.A. Wiersma, Phys.
tion. K.P.K. and A.E.M. are supported by graduate Rev. A 47 (1993) 5120.
fellowships through Lawrence Livermore National [8] N. Dudovich, D. Oron, Y. Silberberg, Nature 418 (2002) 512.
Laboratory, and P.B.P. is supported by the Danish [9] V.V. Lozovoy, B.I. Grimberg, E.J. Brown, I. Pastirk, M. Dantus,
J. Raman Spectrosc. 31 (2000) 41.
Research Agency. J.C.J. is a ChevronTexaco Research [10] R.A. Bartels, T.C. Weinacht, S.R. Leone, H.C. Kapteyn, M.M.
Fellow. Murnane, Phys. Rev. Lett. 88 (2002) 033001.
[11] E. Gershgoren, R.A. Bartels, J.T. Fourkas, R. Tobey, M.M.
Murnane, H.C. Kapteyn, Opt. Lett. 28 (2003) 361.
References [12] E.B. Treacy, IEEE J. Quantum Electron. QE-5 (1969) 454.
[13] L.K. Iwaki, D.D. Dlott, J. Phys. Chem. A 104 (2000) 9101.
[1] R.D. Schaller, J. Ziegelbauer, L.F. Lee, L.H. Haber, R.J. [14] A.M. Zheltikov, A.N. Naumov, Quantum Electron. 30 (2000)
Saykally, J. Phys. Chem. B 106 (2002) 8489. 606.
[2] J. Cheng, A. Volkmer, L.D. Book, X.S. Xie, J. Phys. Chem. B 106 [15] A.N. Naumov, A.M. Zheltikov, J. Raman Spectrosc. 32 (2001)
(2002) 8493. 960.
[3] M. Hashimoto, T. Araki, S. Kawata, Opt. Lett. 25 (2000) 1768. [16] T. Lang, M. Motzkus, J. Opt. Soc. Am. B 19 (2002) 340.
[4] M. M€ uller, J.M. Schins, J. Phys. Chem. B 106 (2002) 3715. [17] K. Osvay, I.N. Ross, Opt. Commun. 166 (1999) 113.You can also read
