Page 146 - Academic Press Encyclopedia of Physical Science and Technology 3rd Chemical Engineering

P1: LLL Final
Encyclopedia of Physical Science and Technology EN003H-565 June 13, 2001 20:37

208 Coherent Control of Chemical Reactions

STIRAP Transfer of population by means of Stimulated A number of strategies have been developed over the
Raman Adiabatic Passage, using a pump and Stokes past few decades to overcome the limitations of bulk
laser. Population in a three-level system is completely kinetics. One method, known as mode-selective chemistry,
transferred without populating the intermediate state if exploits the idea that a molecule may have an eigenstate
the Stokes laser precedes the pump laser in a “counter- that strongly overlaps a desired reaction coordinate. De-
intuitive” order. positing energy into that degree of freedom may selec-
Wave packet A localized wave function, consisting of a tively enhance the reaction of interest. In the following
non-stationary superposition of eigenfunctions of the section we give a number of examples using localized
time-independent Schr¨odinger equation. nuclear or electronic motion to enhance a particular pro-
cess. Although this approach does not depend intrinsically
on the coherent properties of light and therefore lacks
COHERENT CONTROL refers to a process in which the one of the characteristic features of coherent control, it
coherent properties of an electromagnetic field are used to has played an historic role in the development of control
alter the motion of a microscopic object such as an elec- techniques.
tron, atom, or molecule. The controlled process may be Aninherentlimitationofmode-selectivemethodsisthat
categorized according to the degree of freedom that is Nature does not always provide a local mode that coin-
manipulated. For example, a laser beam with carefully cides with the channel of interest. One way to circumvent
tailored properties might be used to control the motion of the natural reactive propensities of a molecule is to ex-
electrons within an atom or molecule, thereby populating ploit the coherence properties of the quantum mechanical
specific eigenstates, or to create electronic wave packets wave function that describes the motion of the particle.
with interesting spatial and temporal properties. Another These properties may be imparted to a reacting molecule
possibilityistheuseofcoherentlighttocontrolthestretch- by building them first into a light source and then trans-
ing and bending modes of a molecule, thereby altering its ferring them to the molecular wave function by means of
chemical reactivity. These are both examples of the control a suitable excitation process.
of internal degrees of freedom. Alternatively, a coherent Two qualitatively different (though fundamentally re-
light source might be used to orient a molecule in space so lated) strategies for harnessing the coherence of light were
that a particular bond is pointing in a chosen direction. An- developed in the mid-1980s. The first, proposed and devel-
other possibility is to use a focused laser beam to control oped by Paul Brumer and Moshe Shapiro, is a molecular
the translational motion of an atomic or molecular beam, analogue of Young’s two-slit experiment, in which two
perhaps focusing the particles to a small volume or steer- coherent excitation paths promote the system to a com-
ing them in a new direction. In a condensed phase, a laser mon final state. Variation of the relative phase of the two
might be used to alter the direction of an electric or ion cur- paths produces a modulation of the excitation cross sec-
rent. These are illustrations of control of external degrees tion. This method does not rely on the temporal properties
of freedom. In all of these examples, the coherence of a of the light source and may in principle use a continuous
light wave is transferred to a material target so as to alter laser. We refer to this approach as coherent phase control
the dynamical properties of the target in a controlled way. and describe it in detail in Section III.
The second approach, proposed by David Tannor and
Stuart Rice, and further developed by them and others in-
I. OVERVIEW cluding Ronnie Kosloff and Herschel Rabitz, uses very
short pulses of light to prepare a wave packet that evolves
In this article we approach the topic of coherent control in time after the end of the pulse. After a suitable delay,
from the perspective of a chemist who wishes to maximize an interrogating pulse projects out the product of interest.
the yield of a particular product of a chemical reaction. The Wave packet control may be thought of as a generalization
traditional approach to this problem is to utilize the prin- of mode-selective chemistry in which a short (and there-
ciples of thermodynamics and kinetics to shift the equilib- fore broadband) pulse of light produces a localized non-
rium and increase the speed of a reaction, perhaps using a stationary state that evolves in a predetermined fashion.
catalyst to increase the yield. Powerful as these methods Wave packet methods have been used with considerable
are, however, they have inherent limitations. They are not success to control electronic, vibrational, and rotational
useful, for example, if one wishes to produce molecules motion of a variety of simple systems. One of the very
in a single quantum state or aligned along some spatial powerful properties of this approach is that it is possible
axis. Even for bulk samples averaged over many quan- to use automated learning algorithms to tailor the laser
tum states, conventional methods may be ineffective in pulses to create wave packets with desired properties. De-
maximizing the yield of a minor side product. tails of wave packet control are given in Section IV.