“Single-beam pumped” coherent anti-Stokes Raman scattering on carbon nanotubes thin films excited through surface plasmons

doi:10.1016/j.physe.2007.07.022

Physica E: Low-dimensional Systems and Nanostructures

Volume 40, Issue 7, May 2008, Pages 2380-2385

https://doi.org/10.1016/j.physe.2007.07.022 Get rights and content

Abstract

In this paper, we propose an interpretation for an abnormal anti-Stokes Raman emission observed on nanometric thin films of different materials and in particular carbon nanotubes. We demonstrate that under a tight-focusing of the excitation light, a coherent anti-stokes Raman scattering (CARS) emission is produced, resulting from a wave mixing process between the incident laser light (ω_l) and Stokes Raman light (ω_s) generated by a surface enhanced Raman scattering (SERS) mechanism. Although the Stokes/anti-Stokes intensity ratio has been explained differently, we present here the results which corroborate the CARS emission. They can be summarized as follows: (i) a square relationship between the CARS signal intensity and the film thickness; (ii) a square relationship between the CARS signal intensity and the exciting laser intensity; (iii) a dependence of the CARS intensity on the numerical aperture (NA) of the microscope objective used for the detection of the anti-Stokes emission. Such effects are not specific to carbon nanotubes and have been observed with other materials accommodated in similar conditions on rough metallic surfaces acting as SERS supports.

Introduction

Giant optical fields locally generated in nanoscale metal structures through the local excitation of surface plasmons (SPs) underlie a variety of non-linear optical processes. Metallic particles of 5–100 nm size as well as metallic films with a roughness-type structure in the range of 10–100 nm are convenient supports to confine the electromagnetic (EM) energy in subwavelength-sized regions. Using silver and gold, reproducible enhancements of the Raman signal of the order of 10²–10⁴ on various polymeric and semiconducting materials have been demonstrated.

Raman scattering has been widely used to characterize single-walled carbon nanotubes (SWNTs). The main results will not be recalled here [1], [2]. We will focus our attention to the unusual anti-Stokes Raman effect characterized by an abnormal intensity, sometimes increasing with the vibrational wave number, large differences in line profiles in Stokes and anti-Stokes sides and some discrepancies between the Stokes and anti-Stokes frequencies [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15]. In normal thermal equilibrium, the whole structure of the Stokes and anti-Stokes Raman spectra is described by the Maxwell–Boltzmann law $\frac{I_{anti ‐ Stokes}}{I_{Stokes}} = {(\frac{ω_{0} - Ω}{ω_{0} + Ω})}^{4} \exp {(\frac{hc Ω}{kT})}^{- 1},$ where ω₀ is the exciting laser frequency, Ω is the wave number of Raman line (cm⁻¹), c is the light velocity, h is the Planck constant and k is the Boltzmann constant.

High values of the anti-Stokes/Stokes intensity ratio (I_as/I_s) and different line shapes of the G band are the most characteristic features which have been reported in the anti-Stokes Raman spectra of SWNTs. Briefly speaking, the asymmetry of the Stokes and anti-Stokes spectra was generally interpreted as resulting from a double-resonant Raman scattering effect produced by the excitation of different (n, m) nanotubes with the incident and Stokes scattered photons [4], [5], [6], while the strong deviation of the anti-Stokes/Stokes intensity ratio (I_as/I_s) has been explained as resulting from a non-thermal overpopulation of the higher vibration states by the large photon density of the excitation light combined with an enhancement of the Raman cross-section [15].

Recently, this abnormal anti-Stokes emission was accurately identified as having the characteristics of coherent anti-Stokes Raman scattering (CARS) [16], [17], [18], [19], [20]. This effect is not uniquely related to SWNTs in specific experimental conditions when the energy of excitation coincides with the energy of electronic transitions, but can be observed in very different materials, due to the fact that at resonance, the third-order dielectric susceptibility differs from zero, χ⁽³⁾≠0, which is basically a requirement necessary in the generation of non-linear optical processes [16], [17].

CARS is a four-wave mixing process in which the anti-Stokes light (ω_as) results from the parametric coupling of the incident laser light (ω_l), the Stokes light (ω_s; ω_s<ω_l) and a probe light (ω_p). if ω_l=ω_p the CARS experiments appear as a degenerate four-wave mixing process that reduces to an energy transfer between the two pump waves, ω_l and ω_s.

As a general rule, the two exciting wavelengths ω_l and ω_s, are supplied by two independent lasers or are generated by parametric frequency conversion. Both alternatives make the experimental setup rather complex and careful adjustments must be carried out to assure the implementation of phase matching geometries.

Recently, anti-Stokes Raman spectra having the characteristics of a CARS emission were observed on different materials using a simple experimental setup under conditions of tight focusing of a single laser excitation beam [16], [17], [18]. In this case, the second pump wave is achieved by: (i) the surface enhanced Raman scattering (SERS) mechanism that supplies ω_s pump light discretely distributed over all Raman transitions [16], [17] and (ii) the use of only one laser light with a wide emission band, which passed through a dispersion system—diffraction grating—simultaneously supplying the two pump waves, ω_l and ω_s [18]. In the former case, the SERS mechanism operating over a wide spectral range furnishes ω_s for all Raman lines while in the latter case, ω_s is determined by the spectral width of the exciting light, so that the CARS effect is limited to low-wavenumber vibrational modes situated in the range of 50–400 cm⁻¹ [18].

Using the micro-SERS technique and the anti-Stokes/Stokes intensity ratio (I_as/I_s) as controlling parameter, we demonstrate the existence of two types of anomalies in the anti-Stokes spectra of SWNTs that are featured by a bigger and a smaller value of (I_as/I_s) ratio as compared to the Maxwell–Boltzmann formula. In the former case, a wave mixing process operates between the incident laser light (ω_l) and Stokes Raman light (ω_s) generated by a SERS mechanism. In the latter case, the smaller value of (I_as/I_s) observed only for the Raman bands associated with RBM of isolated nanotubes, indicates a cooling process appearing in the semiconducting nanotubes when they are resonantly excited. This point will not be discussed here.

Section snippets

Experimental

SERS spectra excited with different laser power at 676.4 and 1064 nm were recorded in air, in backscattering geometry, using a Jobin Yvon T64000 Raman-spectrophotometer and a Bruker FT Raman S/100 spectrophotometer, respectively. Both spectrometers were fitted with a microprobe allowing the laser light to be focused to a dot on the sample with micrometer accuracy. A microscope objective of 0.95 and 0.55 numerical aperture (NA) was used. The Stokes and anti-Stokes pair spectra were recorded under

Results and discussion

In the plane-wave approximation and for a non-absorbing medium, the CARS intensity depends non-linearly on the incident pump intensity I_l $I_{CARS} \propto N_{A} ω_{as}^{2} d^{2} | χ^{(3)} |^{2} I_{l}^{2} I_{s} \sin c^{2} (| Δ k | \frac{d}{2}),$ where χ⁽³⁾ is the third-order non-linear dielectric susceptibility, d is the sample slab thickness, sin c(x) means sin(x)/x and NA is the numerical aperture of the collecting lens. The coherent nature of the CARS process is reflected by fulfilling the phase-matching condition |Δk|×d≪π, where Δk=k_as−(2k_p−k_s) and k_as, k_s

Conclusion

Using the anti-Stokes/Stokes Raman intensity ratio (I_as/I_s) as analyzing parameter, the comparison with its value predicted of the Maxwell–Bolzmann formula reveals a higher value for the anti-Stokes Raman lines of SWNTs. We demonstrate that under a tight focusing of the excitation light, a CARS emission resulting from a wave mixing process between the incident laser light (ω_l) and Stokes Raman light (ω_s) generated by an SERS mechanism is produced. The CARS emission is argued by: (i) a square

Acknowledgments

Work performed in the frame of the Scientific Cooperation between the Laboratory of Crystalline Physics of the Institute of Materials in Nantes and the Laboratory of Optics and Spectroscopy of the National Institute of Materials Physics in Bucharest.

References (25)

M.S. Dresselhaus et al.
Carbon
(2002)
M.S. Dresselhaus et al.
Phys. Rep.
(2005)
P.-H. Tan et al.
Phys. Rev. B
(1998)
S.D.M. Brown et al.
Phys. Rev. B
(2000)
K. Kneipp et al.
Phys. Rev. Lett.
(2000)
C. Thomsen et al.
Phys. Rev. Lett.
(2000)
A.K. Sood et al.
J. Appl. Phys.
(2001)
A.G. Souza-Filho et al.
Phys. Rev. B
(2001)
A. Jorio et al.
Phys. Rev. B
(2001)
S.-L. Zhang et al.
Phys. Rev. B
(2002)

L.G. Cancadao et al.

Phys. Rev. B

(2002)

V. Zolyomi et al.

Phys. Rev. B

(2002)

Cited by (14)

Coherent anti-Stokes Raman scattering: Spectroscopy and microscopy
2011, Vibrational Spectroscopy
Citation Excerpt :
Baltog et al. [176] provided an interpretation for an abnormal anti-Stokes Raman emission observed on nanometric thin films of different materials and in particular carbon nanotubes. Such abnormality is characterized by an abnormal intensity, sometimes increasing with the vibrational wave number, large differences in line profiles in Stokes and anti-Stokes sides and some discrepancies between the Stokes and the anti-Stokes frequencies [177]. Under a tight-focusing of the excitation light, a CARS emission is produced, resulting from a wave mixing process between the incident laser light and the Stokes Raman light generated by a surface enhanced Raman scattering (SERS) mechanism.
In this review the basis, recent developments and applications of coherent anti-Stokes Raman scattering (CARS) in the fields of spectroscopy and microscopy are dialed with. The nonlinear susceptibility of the investigated molecule induced by pump and Stokes laser beams employed in the CARS technique is discussed. The relation between the nonlinear susceptibility, the different CARS laser intensities and the phase matching condition between them is also presented. The structure of CARS spectrum is analyzed as a function of the physical characteristics of the different employed lasers. This includes laser half widths, interference effects, cross-coherence and saturation of the resultant CARS signal by stimulated Raman scatter process (SRS). The different broadening mechanisms for CARS spectral line such as pressure and Doppler broadening are demonstrated. The recent progress in CARS for the in situ reaction flame diagnosis due to its suitability for detection of vibrational–rotational excited gas molecules present in the electronic ground state is discussed. CARS diagnosis for liquid- and solid-phases including the progress in polymeric materials is considered. The applications of CARS microscopy are reviewed in the view of its recent advances to study chemical and biological systems.
Abnormal anti-Stokes Raman emission as a coherent anti-Stokes Raman scattering-like process in disordered media
2011, Journal of Physics B: Atomic, Molecular and Optical Physics
Plasmon-supported two-photon excited vibrational sensing and imaging
2017, Recent Developments In Plasmon-Supported Raman Spectroscopy: 45 Years of Enhanced Raman Signals
Abnormal anti-Stokes Raman scattering and coherent backscattering as manifestation of Anderson localization of light in nonlinear mesoscopic materials
2014, Optical Engineering
Nonlinear features of surface-enhanced Raman scattering revealed under non-resonant and resonant optical excitation
2014, Journal of Optics (United Kingdom)
Plasmonics for Enhanced Vibrational Signatures
2013, Challenges and Advances in Computational Chemistry and Physics

View all citing articles on Scopus

View full text

“Single-beam pumped” coherent anti-Stokes Raman scattering on carbon nanotubes thin films excited through surface plasmons

Abstract

Introduction

Section snippets

Experimental

Results and discussion

Conclusion

Acknowledgments

Carbon

Phys. Rep.

Phys. Rev. B

Phys. Rev. B

Phys. Rev. Lett.

Phys. Rev. Lett.

J. Appl. Phys.

Phys. Rev. B

Phys. Rev. B

Phys. Rev. B

Phys. Rev. B

Phys. Rev. B