Multiphoton Intrapulse Interference 3: Probing Microscopic Chemical Environments
Johanna M. Dela Cruz, Igor Pastirk, Vadim V. Lozovoy, Katherine A. Walowicz, and
Marcos Dantus*
Department of Chemistry, Michigan State UniVersity, East Lansing, Michigan 48824
ReceiVed: July 23, 2003; In Final Form: September 3, 2003
Phase modulation, through its influence on the probability of two-photon excitation at specific frequencies,
is used to probe a molecule’s microscopic chemical environment. The spectral phase required is designed
according to the principles of multiphoton intrapulse interference. We show experimental results where phase
modulation of 20 fs pulses from a titanium sapphire laser oscillator controls the intensity of two-photon
induced fluorescence from the pH-sensitive dye 8-hydroxypyrene-1,3,6-trisulfonic acid (HPTS) in solution.
We show that the dependence of fluorescence yield on the phase can be utilized as a means of distinguishing
molecules in different environments. Consequently, this method can be used to achieve selective multiphoton
microscopy. To illustrate this capability, we show images where phase modulation was used to selectively
excite HPTS in acidic or basic microscopic regions.
1. Introduction
This work represents a continuing effort from our group
focused on the design of tailored femtosecond pulses to achieve
control of nonlinear optical excitation in large molecules based
on the concept of multiphoton intrapulse interference (MII).
1-3
Our goal is to elucidate well-defined and reproducible pulse
shapes that can be used to enhance or suppress particular
nonlinear optical transitions in large molecules such as laser
dyes and proteins in solution. In this paper, we demonstrate
the use of MII to probe the local and microscopic environment
of molecules by selective two-photon laser induced fluorescence
(LIF).
There has been much interest in the use of tailored ultrafast
laser pulses to control the excitation of large molecules in
condensed phase.
4-7
Coherent control of large molecules in the
condensed phase has been considered very challenging because
of the large numbers of intra- and intermolecular degrees of
freedom involved.
8
Further complications arise from the ex-
tremely fast homogeneous (<100 fs) and inhomogeneous (<10
fs) electronic dephasing in these systems.
9
Many of these studies
have been based on the concept of optimal pulse shaping,
whereby a learning algorithm optimizes the spectral phase of
the laser field.
10-13
These methods have the advantage of
convergence to an “optimal” pulse shape. However, because
of the sensitivity of high-intensity multiphoton excitation to a
large number of experimental parameters (for example, pulse
intensity, saturation, transverse-mode quality, pulse bandwidth,
and spectral phase which is affected by transmission through
lenses and windows and reflection from dielectric mirrors), it
is difficult to learn principles from these experiments that can
be applied consistently to different samples in different labo-
ratories.
The influence of phase on two-photon excitation was explored
by Broers et al., who proposed that certain phase functions could
focus the second harmonic spectrum in the frequency domain
like a Fresnel lens focuses light in the spatial domain. They
demonstrated this principle by controlling the two-photon
excitation of Rb atoms.
14
Meshulach et al. used the same
principle to demonstrate coherent control on the two-photon
excitation of Cs atoms.
15
Their work focused on the narrow
two-photon absorption resonance; they showed that the coherent
two-photon enhancement was absent in the broadband excitation
of large molecules in solution.
15
Work from our group has
demonstrated that the changes introduced by pulse shaping on
two- and three-photon transitions arise from destructive interfer-
ence and not on focusing in the frequency domain.
2
The
theoretical framework developed by our group went beyond the
narrow resonance condition and allowed us to develop MII for
controlling multiphoton excitation in large molecules in solu-
tion.
1,2
The fundamental advantage of our work was the early
realization that very short pulses, with concomitant large
bandwidths, would be needed to control the nonlinear excitation
of complex systems. We realized that in the condensed phase
inhomogeneous broadening would prevent the coherent ma-
nipulation of vibronic levels, as can be done in the gas phase.
Therefore, we developed a method to control the delivery of
energy to the molecule that was less dependent on intramolecular
dynamics. MII has now proven to be a robust method that is
amenable to a number of applications.
3
The first application of phase tailored pulses for sensing pH
involved intense linearly chirped laser pulses that saturated the
one-photon resonant transition of a pH-sensitive dye.
16
The ratio
between the LIF signal obtained for positively and negatively
chirped pulses was found to vary from 1.9 to 1.2 for different
pH values. However, this ratio did not vary monotonically, being
the same for pH 7 and pH 9.
The method presented here is completely different from the
early experiments of Buist et al.
16
Our method involves very
weak pulses inducing two-photon transitions. The samples are
thus transparent at the carrier wavelength of the pulse, the
transitions are not saturated, and the results are independent of
probe molecule concentration. The mechanism we used for pH
probing is understood and theoretical simulations are in close
agreement with the data.
The probability for two-photon excitation, S
(2)
, using phase-
modulated pulses at frequency ω ) Δ + ω
0
, where ω
0
is the * To whom correspondence should be addressed. E-mail: dantus@msu.edu.
53 J. Phys. Chem. A 2004, 108, 53-58
10.1021/jp036150o CCC: $27.50 © 2004 American Chemical Society
Published on Web 11/25/2003