Optical Materials 66 (2017) 281e286

Contents lists available at ScienceDirect

Optical Materials
journal homepage: www.elsevier.com/locate/optmat

Tuning the nonlinear response of (6,5)-enriched single-wall carbon

nanotubes dispersions

stegui a, E.C.O. Silva b, A.L. Baggio b, R.N. Gontijo c, J.M. Hickmann d, C. Fantini c,
O.S. Are
M.A.R.C. Alencar e, E.J.S. Fonseca a, b, *
a
Centro de Tecnologia-CTEC, Universidade Federal de Alagoas, Maceio
, AL 57072-970, Brazil
b
Instituto de Física, Universidade Federal de Alagoas, 57061-970, Maceio
, AL, Brazil
c
Departamento de Física, Universidade Federal de Minas Gerais, Belo Horizonte, MG 30123-970, Brazil
d
Instituto de Física, Universidade Federal do Rio Grande do Sul, 91501-970 Porto Alegre, RS, Brazil
e
Departamento de Física, Universidade Federal de Sergipe, Sa~o Cristo ~o, SE 49100-000, Brazil

va

a r t i c l e i n f o a b s t r a c t

Article history: Ultrafast nonlinear optical properties of (6,5)-enriched single-wall carbon nanotubes (SWCNTs) disper-
Received 14 September 2016 sions are investigated using the thermally managed Z-scan technique. As the (6,5) SWCNTs presented a
Received in revised form strong resonance in the range of 895e1048 nm, the nonlinear refractive index ðn2 Þ and the absorption
31 January 2017
coefficients ðbÞ measurements were performed tuning the laser exactly around absorption peak of the
Accepted 6 February 2017
(6,5) SWCNTs. It is observed that the nonlinear response is very sensitive to the wavelength and the
spectral behavior of n2 is strongly correlated to the tubes one-photon absorption band, presenting also a
peak when the laser photon energy is near the tube resonance energy. This result suggests that a suitable
Keywords:
Single wall carbon nanotubes
selection of nanotubes types may provide optimized nonlinear optical responses in distinct regions of the
Z-scan electromagnetic spectrum. Analysis of the figures of merit indicated that this material is promising for
Nonlinear optical ultrafast nonlinear optical applications under near infrared excitation.
Non linear refraction index © 2017 Elsevier B.V. All rights reserved.

1. Introduction
After the development of efficient methods of separation of
single-wall carbon nanotube (SWCNT) species, new fundamental
problems and applications exploring the direct influence of the
SWCNTs nature in one specific physical property can now be
investigated [1,2]. Among a variety of applications, SWCNTs have
emerged as a unique nanostructure which may be responsible for
the new generation of nonlinear optical devices. This renders
SWCNTs as one of the most suitable candidate for photonic and
optoelectronic nanodevices [3].
Several papers have been published exploring the study of
SWCNT due to its large third-order nonlinearity and ultrafast
electronic response as a consequence of delocalized p-electrons
cloud along the tube axis [4e8]. In this line, a comprehensive
experimental study on specific SWCNT specie is desirable to
delineate the nonlinear optical properties of this particular
nanometric system. Especially, aiming its use for particular appli-
cations where the electronic and optical properties can be
optimized.
SWCNTs are direct-band gap materials, with a gap that depends
on the diameter and chirality [9]. Many of the applications that
might make use of the electronic properties of these materials
require individual (n,m) nanotubes with a given chirality for opti-
mum performance. In fact, an efficient resonant absorption occurs
only in SWCNTs whose diameter corresponds to an electronic
transition in resonance with the photon energy. For the sake of
developing of photonic devices based on SWCNT, it is extremely
important to explore its nonlinear optical properties for a specific
(n,m) specie.
In the present paper we explore the nonlinear properties for a
(6,5)-enriched SWCNT sample, showing that the nonlinear
response is very sensitive to the wavelength when the resonance of
tube corresponds to the resonance at the one-photon energy
excitation.

* Corresponding author.Centro de Tecnologia-CTEC, Universidade Federal de

2. Material and methods

, AL 57072-970, Brazil.
Alagoas, Maceio
E-mail address: eduardo@fis.ufal.br (E.J.S. Fonseca). Nanotube dispersion was prepared by adding 100 mg of

http://dx.doi.org/10.1016/j.optmat.2017.02.013
0925-3467/© 2017 Elsevier B.V. All rights reserved.
282 O.S. Ar
estegui et al. / Optical Materials 66 (2017) 281e286

CoMoCAT SG65i, a (6,5)-enriched SWCNT powder, to 100 mL of a
70 mM Sodium Dodecyl Sulfate (SDS) solution. This dispersion was
tip sonicated at 40 W for a total of 3 h with a 5 min interval every
30 min. After sonication, the solution was centrifuged at 10000 rpm
for 1 h followed by an ultracentrifugation at 40000 rpm for another
1 h to remove bundles and impurities. The resulting dispersion was
then used for separation by column chromatography [10,11].
Chromatographic column were made using 10 mL plastic sy-
ringes filled with gel Sephacryl S-200 HR and a small amount of a
70 mM SDS solution. After that, 1 mL of the SWCNTs dispersion was
added to the column for the separation process. As the affinity of
metallic SWCNT with the gel is negligible, its interaction with gel
beads is very weak, thus they flow through the column without
being stuck on the beads [10,11]. Semiconducting SWCNT on the
other hand has a chiral dependent interaction with the gel, being
attached to the gel beads. Therefore, metallic and low affinity
semiconducting SWCNT run through the column, leaving only the
SWCNT that strongly interact with the medium. The (6,5) SWCNT
has a strongly interaction with the gel, and as it is present in a large
quantity in the sample. Thus, large quantity of this chirality is stuck
on the gel beads when the nanotube dispersion runs through the
column. Finally, 175 mM solutions of SDS is added to the column to
remove the SWCNT stuck on the gel [11]. It is important to mention
that using the column chromatography method is possible to
produces samples with higher nanotube concentration as
compared with other separation methods such as density gradient
ultracentrifugation (DGU). Fig. 1(a) shows the optical absorption
spectrum of the (6,5)-enriched sample. The spectrum is dominated
by the peaks at 985 and 570 nm associated, respectively, with the
E11 and E22 optical transitions of the (6,5) nanotube. Other nano-
tube species present in the sample are responsible for the small
absorption bands observed in the spectrum.
To completely characterize the chirality distribution of the (6,5)-
enriched SWCNT dispersion, Raman spectra of the radial breathing
modes (RBM) were measured in a wide range of excitation energies
(2.54e1.90 eV). Fig. 1(b) presents some of the Raman spectra ob-
tained. From the resonance Raman profiles, and the cross section
corrected RBM intensity profiles, it is possible to obtain the relative
concentration of the SWCNT species present in the sample,
following the procedure described in Ref. [12]. The (n,m) species

Fig. 1. (a) Absorption Spectrum of the (6,5) enriched SWCNT sample. (b) Raman spectra of the (6,5)-enriched SWCNT sample showing the radial breathing modes recorded at
different excitation laser energies. The main chiralities observed are indicated on the figure. (c) Relative distribution of SWCNT species in the enriched sample.
 O.S. Ar
estegui et al. / Optical Materials 66 (2017) 281e286
distribution in the sample is presented in Fig. 1(c). It is observed
that the (6,5) SWCNT corresponds to 60% of the chiralities present
in the sample. The (7,5) SWCNT, whose absorption at 1025 nm
appears in Fig. 1(a) as a shoulder of the E11 absorption of the (6,5)
SWCNTs, is the second most abundant corresponding 20% of
SWCNTs in the sample. Other three chiralities (6,4), (7,6) and (8,3)
are also present in smaller amounts. Raman results were combined
with optical absorption spectra in order to obtain a trustworthy
concentration of the SWCNT species present in the sample. the
absolute concentration of the (6,5) species can be obtained from the
absorption spectra using the reported values for oscillator strength
for this nanotube [13,14] and the equation
Sn;m
Cc ¼ B
f $l
where Cc is the concentration of carbon atoms, Sn,m is the integrated
area of the optical absorption peak associated with the (n,m)
SWCNT, l is the optical path of the cuvette and f is the oscillator
strength of the (6,5) SWCNT. B is a constant whose value is
3.48  105 mol cm/L,eV [13,14]. The value obtained for the ab-
solute concentration of the (6,5) SWCNT is around (9.0 ± 0.5) mg/
mL.
3. Experimental
The nonlinear optical characterization was performed using the
same procedure of Ref. [15]. The experimental setup is shown in
Fig. 2(a). The excitation source used was a mode-locked Ti:Sapphire
laser, linearly polarized, delivering pulses of 120 fs (FWHM) at
80 MHz repetition rate tuned between the absorption band of
samples with SWCNTs (895e1048 nm). The laser beam was
Fig. 2. Sketch of steps used to separate thermal from electronic effects. Experimental setup (a), temporal window (b) and (c) 3D graph of a temporal evolution of the Z-scan curve
with prefocal and postfocal positions.
283
modulated by a chopper (14 Hz frequency and 0.09 duty cycle) and
focused on to the sample by a convergent lens of 6,5 cm focal
length. The samples consisted of 1 mm width quartz cell filled with
colloids of enriched (6,5) SWCNT. The cell was moved around the
lens focal plane (z ¼ 0). The light transmittance was then measured
by a closed (without) aperture photodetector as a function of the
sample position for nonlinear refraction (absorption) measure-
ments. The nonlinear absorption was performed using the inset
showed in Fig. 2(a). The detected signal was temporally analyzed by
digital oscilloscope and then processed by a computer.
Fig. 2(b) shows some typical temporal windows, of the prefocal
and postfocal in some -z and þ z positions, respectively. The red
curve is an extrapolation for t ¼ 0 [see equation (2)] and Fig. 2(c)
shows a typical 3D graph of a complete temporal evolution of the Z-
(1)
scan curve generated by the signal change in each sample position.
4. Theory
Among different Z-scan techniques-based [16e18], the one
presented by Gnoli, Razzari and Righini [18] has been used as an
efficient technique enable to separate the instantaneous (elec-
tronic) nonlinear optical response from the cumulative thermal
effect. In order to understand this methodology, let us assume that
a high repetition rate laser modulated by a chopper is employed as
the excitation source for a Z-scan measurement and define t ¼ 0 as
the time instant that the chopper begins to unblock the laser beam.
In this case, the thermo-optical contribution to transmittance
measured by the z-scan technique is negligible for time instants
near to zero but increases with time, due to the cumulative heating
pulse by pulse, as can be described by
284 O.S. Ar
estegui et al. / Optical Materials 66 (2017) 281e286

Iðx; tÞ
Tðx; tÞ ¼
Iðx; 0Þ
2 3
1 4h 2qx 5;
¼ 1 þ qTan i (2)
2 t c ð xÞ 2
ð2q þ 1Þ2 þ x 2qt þ 2q þ 1 þ x

In this equation, q is the phase-shift induced thermally, x ¼ z=z0

is the normalized distance, z0 corresponds to the Rayleigh range of
the laser beam, q is the order of the multiphoton process and tc ðxÞ is
the characteristic thermal lens time.
The electronic contribution, on the other hand, is always present
each time a laser pulse interacts with the nonlinear medium.
However, its contribution is usually smaller than the thermal
response, being purely detectable, theoretically, at t ¼ 0. Although
this measurement cannot be made directly using the proposed
experimental setup, Gnoli et al. [18] figured out that the Z-scan
curve free of thermal effects can be obtained extrapolating the time
evolution of the transmittance curves for t ¼ 0. Hence, the value of
n2 can be obtained fitting the normalized transmittance curve at t ¼
0 employing the standard equation of Z-scan method [16].

4DF0 x
TðxÞy1 þ   ; (3)
x þ 9 x2 þ 1
2

pffiffiffi
where DF0 ¼ 2kn2 I0 Leff is the induced nonlinear phase-shift, I0 is
the laser intensity at the lens focal plane, Leff ¼ ð1  ea0 L Þ=L is the
effective length, L is the sample length and the modulus of the
beam wave vector is k. Whenever the medium presents a negative
nonlinear refraction, the value of n2 is negative and the profile of
the transmittance curve displays a maximum for a sample position
x ¼ 1:7 and a minimum when the sample is placed at the
x ¼ þ1:7. On the other hand, positive nonlinear refraction implies
that n2 > 0 and the maximum (minimum) occurs at occurs at
x ¼ þ1:7 (x ¼ 1:7).
It is worth mentioning that the transmittance values measured
during the time period between t ¼ 0 and the chopper rise time
must be disregarded in this methodology. This is a necessary pro-
cedure because within this period of time the laser beam is partially
blocked by the chopper and Eq. (2) is no longer valid to describe its
behavior. Nevertheless, the time evolution of the Z-scan traces can
be followed from the opening rise time onwards [19].
The medium nonlinear absorption can be investigated by
removing the detector aperture. In this configuration, the trans-
mitted signal as a function of the sample relative position can be
described by Ref. [16].
X
∞
½  q0 ðx; 0Þm
TðxÞ ¼ ; (4)
m¼0 ðm þ 1Þ3=2
Where q0 ðx; tÞ ¼ bI0 ðtÞLeff =ð1 þ x2 Þ, b is the nonlinear absorption
coefficient and I0 ðtÞ is the laser intensity at the focal plane.
5. Results and discussion
Fig. 1(a) shows the absorption spectra of the (6,5)-enriched
sample. The colloids present a strong absorption ranging of
895 nme1048 nm associated with E11 S excitonic transitions. The
interesting point here is that this resonance coincides with the
wavelength range of the laser used.
Fig. 3(a) shows, a typical result of the reconstructed Z-scan
curves at t ¼ 0s using pulsed laser with high repetition tuned at
wavelength 974 nm at 30 mW average power. The analyses of the
results were performed using Eq. (2) at t ¼ 0 [18]. Eq. (3) was used,
Fig. 3. Z-scan curves at t ¼ 0 for the (a) semiconducting SWCNTs colloids measured at
974 nm. Open squares correspond to the closed aperture experimental data and the
red lines are the fittings. Open circles corresponds to the open aperture measurements
results. (b) jn2 j dispersion curves for semiconducting SWCNTs colloids, black squares
are the experimental data and the dashed lines are guides to the eyes. (For interpre-
tation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
after the Z-scan curve reconstructed, to calculate the values of the
nonlinear refraction n2 of the electronic origin. The peak-valley
curve [Fig. 3(a)] is due to the intense negative nonlinear refrac-
tion of electronic origin ðn2 < 0Þ. The nonlinear absorption was not
able to be detected. Indeed, the nonlinear absorption coefficient of
these colloids must be smaller than our experimental limit, which
was equal to 1:6  1010 cm=W, in modulus.
Varying the laser wavelength from 895 to 1048 nm it is possible
to reconstruct Z-scan curves at t ¼ 0 for different excitation
wavelength. Fig. 3 (b) shows the dispersion curves of jn2 j, for
different wavelength at 30 mW of incident average power. A sig-
nificant increase on the nonlinear refractive index modulus as the
excitation wavelength approaches to the one-photon absorption
resonance is clearly observed. Indeed, the spectral behavior of jn2 j
curve presents a strong correlation with observed absorbance
spectrum of (6,5) SWCNT within this wavelength range.
As described in Ref. [20], several resonance peaks can be
observed on the third-order nonlinear response of carbon nano-
tubes due to multiphoton transitions between the Van Hove sin-
gularities. For third-harmonic generation, for instance, three
photon transitions play the most important role, however, from
their theoretical calculations, one photon and two photon transi-
tions may also contribute to this effect, but in distinct spectral
ð3Þ
ranges. Indeed, for the ciiii ðu; u; u; uÞ, one-photon [8] and two
photon transitions [15,21] would also give an important contribu-
tion to the generation of spectral peaks on the nonlinear responses
related to this tensor element. From our results, the direct
 O.S. Ar
estegui et al. / Optical Materials 66 (2017) 281e286 285

Table 1
Optical properties of the (6,5)-enriched SWCNT.

l (nm) Pulse's linewidth (nm) Pulse's time duration (fs) a0 (cm1) jn2 j (1013 cm2/W) jWj

895 10.67 110.44 0.113 1.54 12.07

906.5 10.73 112.53 0.112 1.66 12.94
915.5 10.75 114.66 0.130 1.63 10.85
928 10.99 115.28 0.187 1.79 8.12
937 11.33 113.90 0.272 2.07 6.42
947.5 12.00 110.00 0.466 2.34 4.19
958 12.28 109.86 0.740 3.14 3.51
966 12.37 110.98 0.986 3.71 3.08
974 12.42 112.33 1.215 3.95 2.64
986 12.46 114.79 1.380 4.54 2.64
999 13.05 112.48 1.140 4.35 3.02
1009 13.52 110.74 0.824 4.06 3.87
1019 13.59 112.33 0.563 3.19 4.40
1029 13.19 117.97 0.380 2.77 5.60
1036.5 13.04 121.14 0.274 2.63 7.32
1048 13.20 122.35 0.151 1.99 9.98

comparison of absorbance and jn2 j spectra indicates that one-
photon transitions, associated with E11 S excitonic transitions, are
giving the main contribution to the nonlinear refraction displayed
by this SWCNTs specie.
Aiming ultrafast photonic applications, this colloid all-optical
switching's figures of merits, W ¼ Dnmax =la0 and T ¼ 2a2 l=n2
were evaluated. These conditions are known as Stegeman figures
[22]. Here, Dnmax is the maximum refractive index change
achievable, limited by saturation [23]. Carbon nanotube can be
considered useful for all-optical switching devices when satisfy the
conditions jWj > 0:27 and jTj < 1. Table 1 shows the figures of merit
jWj for the (6,5)-enriched sample obtained at different excitation
wavelengths. The figures of merit T could not be determined due to
the fact that the nonlinear absorption coefficients of the studied
SWCNTs were much smaller than our system resolution
1:6  1010 cm=W [15,24]. As can be observed, this SWCNTs kind
satisfies the figures of merit jWj condition at all investigated
wavelength range.
It is worthwhile to compare the all-optical switching perfor-
mance of the colloidal system investigated in the present study
with other promising nonlinear optical materials. In Table 2, it is
summarized the results for the figure of merit jWj reported previ-
ously in literature of several media. As can be observed, in com-
parison with other materials, the colloid containing carbon
nanotubes investigated in the present work presents values of jWj
as high as other promising optical materials, such as organic
compounds [22,25], colloidal metallic nanoparticles [26,27], glasses
[22,25,28e30], and colloidal carbon nanotubes [15]. Therefore,
these results indicate that these materials have a good potential for
the development of switching applications in femtosecond regime.
Recently, other figures of merit analyzing multiphoton absorp-
tion have been proposed and employed in the evaluation of the
photonics materials application's potential [31e33]. Nevertheless,
as the investigated SWCNTs colloids did not present a detectable
nonlinear absorption, it was not possible to calculate or estimate
these figures of merits in this work, and a direct comparison with
ref. [31e33] could not be performed.
6. Conclusion
In conclusion, we measured the negative nonlinear refractive
index ðn2 < 0Þ of electronic origin of colloids containing mainly the
semiconducting single (6,5) SWCNT, using the thermally managed
Z-scan technique. The jn2 j dispersion curves along the main ab-
sorption bands Vis-NIR showed a similar behavior with the band
resonance point by point, which indicates that one-photon ab-
sorption related to the E11 S excitonic transitions give the main
contribution to the observed nonlinear effect. The evaluated values
for the all-optical switching figure of merit W results indicate that
these semiconducting SWCNTs are promising materials for ultrafast
nonlinear optical. Moreover, our results suggest that using a suit-
able choice of the nanotubes species, it is possible to produce
optimized media for nonlinear optical applications at distinct re-
gions of the electromagnetic spectrum.
Acknowledgments
The authors are thankful to CAPES, PROCAD, Procad-NF, CNPq/
MCT, Pronex/FAPEAL, FAPEMIG and FINEP.

Table 2
Values of jWj for different materials. References

Material Wavelength (nm) jWj References [1] J.-H. Han, G.L. Paulus, R. Maruyama, D.A. Heller, W.-J. Kim, P.W. Barone,
C.Y. Lee, J.H. Choi, M.-H. Ham, C. Song, Nat. Mater. 9 (2010) 833e839.
(6,5)-enriched SWCNT 895e1048 2.64e12.94 This work
[2] G.A. S

afar, R.N. Gontijo, C. Fantini, D.C. Martins, Y.M. Idemori, M.V. Pinheiro,
Metallic SWCNTs 755e825 0.6e6.2 [15] K. Krambrock, J. Phys. Chem. C 119 (2015) 4344e4350.
Semiconducting SWCNTs 755e825 13.2e190 [15] [3] F. Bonaccorso, Z. Sun, T. Hasan, A. Ferrari, Nat. Photon. 4 (2010) 611e622.
Semiconductor (GaAs) 1.06  103 <2.8 [22] [4] Y.C. Chen, N.R. Raravikar, L.S. Schadler, P.M. Ajayan, Y.P. Zhao, T.M. Lu,
Semiconductor (AlGaAs) 0.8e1.56  103 2.5e8 [22] G.C. Wang, X.C. Zhang, Appl. Phys. Lett. 81 (2002) 975e977.
Organics (PTS criystal) 1.06  103 13 [22] [5] N. Kamaraju, S. Kumar, B. Karthikeyan, B. Kakade, V.K. Pillai, A.K. Sood,
Organics (Poly-4BCMU) 1.3  103 0.3 [22] J. Nanosci. Nanotechnol. 9 (2009) 5550e5554.
Organics (DAN crystal) 0.63  103 26 [22] [6] N. Kamaraju, S. Kumar, A.K. Sood, S. Guha, S. Krishnamurthy, C.N.R. Rao, Appl.
Glass (SiO2) >1.06  103 >103 [22] Phys. Lett. 91 (2007).
Colloidal gold nanoparticles 793 >1.2 [26] [7] T. Hasan, Z.P. Sun, F.Q. Wang, F. Bonaccorso, P.H. Tan, A.G. Rozhin, A.C. Ferrari,
Colloidal silver nanoparticles 793 >0.06 [27] Adv. Mater. 21 (2009) 3874e3899.
[8] D.R.B. Valada ~o, D.G. Pires, M.A.R.C. Alencar, J.M. Hickmann, C. Fantini,
Tellurite glasses 810 0.20e0.53 [28,29]
M.A. Pimenta, E.J.S. Fonseca, Opt. Mater. Express 2 (2012) 749e756.
Antimony orthophosphate glasses 800 5e10 [30]
[9] M.S. Dresselhaus, G. Dresselhaus, R. Saito, A. Jorio, Phys. Rep. 409 (2005)
286 O.S. Ar
estegui et al. / Optical Materials 66 (2017) 281e286

47e99. [23] H.S. Nalwa, S. Miyata, Nonlinear Optics of Organic Molecules and Polymers,
[10] H. Liu, D. Nishide, T. Tanaka, H. Kataura, Nat. Commun. 2 (2011) 309. CRC Press, Boca Raton, Fla, 1997.
[11] K. Tvrdy, R.M. Jain, R. Han, A.J. Hilmer, T.P. McNicholas, M.S. Strano, ACS Nano [24] B. Valeur, Molecular fluorescence, in: Principles and Applications, first ed.,
7 (2013) 1779e1789. 2001.
[12] A. Jorio, A. Santos, H. Ribeiro, C. Fantini, M. Souza, J. Vieira, C. Furtado, J. Jiang, [25] M. Samoc, A. Samoc, B. Luther-Davies, Z. Bao, L. Yu, B. Hsieh, U. Scherf, J. Opt,
R. Saito, L. Balzano, Phys. Rev. B 72 (2005) 075207. Soc. Am. B 15 (1998) 817e825.
[13] F. Scho €ppler, C. Mann, T.C. Hain, F.M. Neubauer, G. Privitera, F. Bonaccorso, [26] H.P. Castro, H. Wender, M.A. Alencar, S.R. Teixeira, J. Dupont, J.M. Hickmann,
D. Chu, A.C. Ferrari, T. Hertel, J. Phys. Chem. C 115 (2011) 14682e14686. J. Appl. Phys. 114 (2013) 183104.
[14] J.K. Streit, S.M. Bachilo, S. Ghosh, C.-W. Lin, R.B. Weisman, Nano Lett. 14 (2014) [27] N.F. Corre^a, C.E. Santos, D.R. Valada
~o, L.F. de Oliveira, J. Dupont, M.A. Alencar,
1530e1536. J.M. Hickmann, Opt. Mater. Express 6 (2016) 244e250.
[15] A.C. Branda ~o-Silva, R.M.A. Lima, C. Fantini, A. Jesus-Silva, M.A.R.C. Alencar, [28] R.F. Souza, M.A. Alencar, J.M. Hickmann, R. Kobayashi, L.R. Kassab, Appl. Phys.
J.M. Hickmann, R.M. Jain, M.S. Strano, E.J.S. Fonseca, Carbon 77 (2014) Lett. 89 (2006) 171917-1e171923-3.
939e946. [29] R.F. Souza, M.A. Alencar, J.M. Hickmann, R. Kobayashi, L.R. Kassab, Appl. Phys.
[16] M. Sheik-Bahae, A.A. Said, T.H. Wei, D.J. Hagan, E.W. Van Stryland, Quantum Lett. 90 (2007) 089901.
electronics, IEEE J. 26 (1990) 760e769. [30] E. Falcao, C.B. de Araújo, C. Bosco, G. Maciel, L. Acioli, M. Nalin, Y. Messaddeq,
[17] M. Falconieri, G. Salvetti, Appl. Phys. B 69 (1999) 133e136. J. Appl. Phys. (2005) 5.
[18] A. Gnoli, L. Razzari, M. Righini, Opt. Express 13 (2005) 7976e7981. [31] P. Zhao, M. Reichert, D.J. Hagan, E.W. Van Stryland, Opt. Express 24 (2016)
[19] M. Falconieri, Anglais 1 (1999). 24907e24920.
[20] A.M. Nemilentsau, G.Y. Slepyan, A.A. Khrutchinskii, S.A. Maksimenko, Carbon [32] E. García-Ramírez, S. Almaguer-Valenzuela, O. S
anchez-Dena, O. Baldovino-
44 (2006) 2246e2253. Pantaleo
n, J. Reyes-Esqueda, Opt. Express 24 (2016) A154eA167.
[21] P.T. Araujo, S.K. Doorn, S. Kilina, S. Tretiak, E. Einarsson, S. Maruyama, [33] J.M. Hales, S. Barlow, H. Kim, S. Mukhopadhyay, J.-L. Bre
das, J.W. Perry,
H. Chacham, M.A. Pimenta, A. Jorio, Phys. Rev. Lett. 98 (2007) 067401. S.R. Marder, Chem. Mater. 26 (2013) 549e560.
[22] G.I. Stegeman, 1993, pp. 75e89.