Professional Documents
Culture Documents
Textbook Advances in Quantum Chemistry Ratner Volume John R Sabin and Erkki J Brandas Eds Ebook All Chapter PDF
Textbook Advances in Quantum Chemistry Ratner Volume John R Sabin and Erkki J Brandas Eds Ebook All Chapter PDF
https://textbookfull.com/product/advances-in-the-theory-of-
quantum-systems-in-chemistry-and-physics-1st-edition-erkki-j-
brandas-auth/
https://textbookfull.com/product/quantum-systems-in-chemistry-
and-physics-progress-in-methods-and-applications-1st-edition-
erkki-j-brandas-auth/
https://textbookfull.com/product/advances-in-organometallic-
chemistry-volume-65-pedro-j-perez-eds/
https://textbookfull.com/product/advances-in-organometallic-
chemistry-volume-66-pedro-j-perez-eds/
Advances in Organometallic Chemistry Volume 67 Pedro J.
Pérez (Eds.)
https://textbookfull.com/product/advances-in-organometallic-
chemistry-volume-67-pedro-j-perez-eds/
https://textbookfull.com/product/advances-in-heterocyclic-
chemistry-volume-113-1st-edition-alan-r-katritzky/
https://textbookfull.com/product/quantum-mechanics-for-chemistry-
seogjoo-j-jang/
https://textbookfull.com/product/ideas-of-quantum-chemistry-
volume-1-from-quantum-physics-to-chemistry-3rd-edition-lucjan-
piela/
https://textbookfull.com/product/advances-in-organometallic-
chemistry-1st-edition-pedro-j-perez/
EDITORIAL BOARD
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek
permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the Copyright
Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by
the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices,
or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety and
the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of
products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-812888-6
ISSN: 0065-3276
Adam P. Ashwell
Northwestern University, Evanston, IL, United States
Asbjørn Bols
Department of Chemistry, University of Copenhagen, Copenhagen, Denmark
Eric A. Buchanan
University of Colorado, Boulder, CO, United States
R. Benny Gerber
The Hebrew University of Jerusalem, Jerusalem, Israel; University of California, Irvine, CA,
United States
Thorsten Hansen
Department of Chemistry, University of Copenhagen, Copenhagen, Denmark
Zdeněk Havlas
University of Colorado, Boulder, CO, United States; Institute of Organic Chemistry and
Biochemistry, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Barak Hirshberg
The Hebrew University of Jerusalem, Jerusalem, Israel
Joshua Jortner
School of Chemistry, Tel Aviv University, Tel Aviv, Israel
Isidore Last
School of Chemistry, Tel Aviv University, Tel Aviv, Israel
Jan Linderberg
Aarhus University, Aarhus C, Denmark; Henry Eyring Center for Theoretical Chemistry,
University of Utah, Salt Lake City, UT; Quantum Theory Project, University of Florida,
Gainesville, FL, United States
Josef Michl
University of Colorado, Boulder, CO, United States; Institute of Organic Chemistry and
Biochemistry, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Kurt V. Mikkelsen
Department of Chemistry, University of Copenhagen, Copenhagen, Denmark
Jens Oddershede
University of Southern Denmark, Odense, Denmark; University of Florida, Gainesville, FL,
United States
John F. Ogilvie
University of Southern Denmark, Odense, Denmark; Simon Fraser University, Burnaby,
BC, Canada; Universidad de Costa Rica, San Jose, Costa Rica
ix
x Contributors
Stine T. Olsen
Department of Chemistry, University of Copenhagen, Copenhagen, Denmark
John W. Perram
School of Mathematics and Statistics, University of New South Wales, Kensington, NSW,
Australia
Mark A. Ratner
Northwestern University, Evanston, IL, United States
John R. Sabin
University of Southern Denmark, Odense, Denmark; University of Florida, Gainesville, FL,
United States
Michael Sabio
Stevens Institute of Technology, Center for Healthcare Innovation, Hoboken; 3D-2Drug,
Fair Lawn, NJ, United States
Stephan P.A. Sauer
University of Copenhagen, Copenhagen, Denmark
George C. Schatz
Northwestern University, Evanston, IL, United States
Sid Topiol
Stevens Institute of Technology, Center for Healthcare Innovation, Hoboken; 3D-2Drug,
Fair Lawn, NJ, United States
PREFACE
It is with the greatest pleasure that we present this volume in tribute to Mark
Ratner on the occasion of his 75th birthday for his seminal contributions to
theoretical chemistry during the last half century.
Mark Ratner began his career as an undergraduate in chemistry at
Harvard University, where he earned the AB degree in 1964. He then
did a PhD in chemistry at Northwestern University in 1969, working with
Prof. G. Ludwig Hofacker, followed by postdoctoral work with Prof. Jan
Linderberg in Århus, Denmark. Mark took his first job at New York
University, and in 1975 moved to Northwestern University where he
remains. Presently, Mark Ratner is Lawrence B. Dumas Distinguished
University Professor Emeritus Departments of Chemistry, Materials Science,
and Engineering at Northwestern University. While at Northwestern, he has
held several positions such as Chairman of the Chemistry Department and
Associate Dean of the College of Arts and Sciences.
Mark Ratner is a very personable fellow, which makes him an ideal
colleague to work with. He is friendly and helpful, and, perhaps most impor-
tant, an original and deep thinker in many aspects of theoretical chemistry;
he leads a scientific life dominated by a quest for understanding of chemical
structure and function at the nanoscale and the theory that will describe and
predict such in many nanoscale subfields. These characteristics have led to
a world-class research program under Mark Ratner’s direction. Mark Ratner
is not only a producer of world-class research, but he is prolific. Presently
he has produced over 800 technical papers, 5 edited volumes, and 4 books.
xi
xii Preface
It should be noted that of the books, two of them are coauthored by his son,
Daniel Ratner, and the other two are written together with George Schatz.
In addition, he has given over 100 named lectures around the world from
Denmark, to Israel, to Japan, and on.
Perhaps Mark Ratner’s best known work is in the field of molecular
electronics. In 1974, Mark Ratner and his at that time student, Ari Aviram,
produced a work on molecular rectifiers [Bull. Am. Phys. Soc. 19, 341
(1974); Chem. Phys. Lett. 29, 277 (1974)] which is probably his best known
work. This was the first time it had been shown, as inconceivable as it was by
most experts in the field, that a single molecule could act as a rectifier. From
their work, a whole field of molecular electronics has blossomed. However,
Mark Ratner’s work has not been confined to this particular subfield, as he
has significantly contributed in many other related areas. For instance, he
has contributed substantially to work on molecular electron transfer and
electron transport in organic polymers as well as to various aspects of
molecular dynamics and relaxation.
Mark Ratner’s achievements have won him many honors. Most presti-
gious, he is a member of the National Academy of Science (2002) and a
foreign member of the Royal Danish Academy of Science (2004). In
addition he has received honorary Doctor of Science degrees from the
Hebrew University of Jerusalem (2004) and the University of Copenhagen
(2010). In 2010, Mark Ratner was also celebrated and honored with a fest-
schrift published in the Journal of Physical Chemistry C (volume 114, #48).
It has certainly been a great honor for those of us whom have had
interaction, scientific and social, with Mark Ratner, to collect this tribute
to him. He has been a vibrant member of the Quantum Chemistry Club of
those individuals who have practiced the science for nearly a half century,
and we wish him the very best for the future.
The papers presented here are written by Mark Ratner’s friends and
coworkers to demonstrate their respect for him and his work.
Best wishes, Mark!
By: JOHN R. SABIN
€
ERKKI J. BRANDAS
CHAPTER ONE
Contents
1. Introduction 2
2. Theoretical Framework, Advantages, and Limitations 5
2.1 Vibrational Self-Consistent Field 5
2.2 Time-Dependent Self-Consistent Field 6
2.3 Mixed Quantum–Classical Methods 8
2.4 Classical Separable Potentials 10
3. Recent Developments and Applications 12
3.1 Ab Initio Classical Separable Potentials 12
3.2 Mixed Quantum–Classical Methods 18
4. Summary and Future Directions 20
Acknowledgments 22
References 22
Abstract
Methods that can accurately describe the quantum dynamics of large molecular sys-
tems have many potential applications. Since numerical solution of the time-dependent
Schro€dinger equation is only possible for systems with very few atoms, approximate
methods are essential. This paper describes the development of such methods for this
challenging time-dependent many-body quantum mechanical problem. Specifically,
we focus on the development of mean-field theories, to which Mark Ratner has contrib-
uted greatly over the years, such as the time-dependent self-consistent field method,
mixed quantum–classical methods, and the classical separable potentials method.
The advantages and limitations of the different variants of mean-field theories are
highlighted. Recent developments, aimed at applying mean-field methods for large sys-
tems, and their applications are presented. Issues where further methodological
advancement is desirable are discussed. Examining the tools available so far, and the
recent progress, we conclude there are promising perspectives for future development
of mean-field theories for quantum dynamics with applications to realistic systems in
important chemical and physical processes.
1. INTRODUCTION
Classical molecular dynamics, in which the motion of the atoms is
described using Newton’s equations of motion, is one of the most widely
applied methods in chemistry. This includes methods in which the interac-
tions between the atoms are described using empirical force-fields1 as well as
ab initio molecular dynamics (AIMD),2–5 which directly employs potential
energy surfaces (PES) obtained from quantum chemical methods to the clas-
sical nuclear equations of motion. The scope of applications is overwhelm-
ing and ranges from the study of biological molecules,6,7 properties of water
and ice,8–10 materials science,11 chemical reaction dynamics,12 and many
more. Describing the motion of the nuclei using classical mechanics is often
an excellent approximation at typical chemical conditions, e.g., room tem-
perature. However, in many cases, nuclear quantum effects (NQE) such as
zero-point energy (ZPE), tunneling, and quantum interference may play a
significant role, especially for light atoms. Important example is liquid water
and ice, and a detailed review on the importance of NQE for water and ice
was published recently by Ceriotti et al.13 Other examples include molecules
embedded in superfluid helium droplets,14,15 the kinetic isotope effect in
enzyme catalysis,16–18 and more. As a result, for these processes and others,
a quantum description of the nuclei is essential.
Ideally, one would like to numerically solve the time-dependent
Schr€ odinger equation (TDSE) for the nuclei using a suitable Born–
Oppenheimer (BO) PES. However, due to the exponential scaling of the
computational effort with the number of degrees of freedom,19 this is an
extremely difficult task. State-of-the-art calculations in full dimensionality
are currently limited to systems with approximately 4–6 atoms,20 such as
the recent study of electron photodetachment from F(H2O).21 Since most
chemical and physical processes of interest occur in larger systems, approx-
imate methods for quantum nuclear dynamics are necessary.
Several different approximate methods are available which address, con-
ceptually, NQE. This includes semiclassical methods such as the semiclassi-
cal initial value representation method by Miller22 and the Gaussian
wavepacket method by Heller23,24 as well as methods based on the
Mean-Field Methods for Time-Dependent Quantum Dynamics of Many-Atom Systems 3
agreement with numerical solution of the nuclear TDSE for several pro-
cesses such as vibrational predissociation of I2He45 as well as photodissoci-
ation of Xe-HI46 and Ar-HCl.47 While the computational complexity of the
TDSCF method is significantly reduced in comparison to numerical solu-
tion of the TDSE, the evaluation of the mean potentials involves multi-
dimensional integrals over all modes of vibration except one, which are
the computational bottleneck of TDSCF. As a result, TDSCF was only
applied to relatively small systems.
An approximation which, when justified, can significantly decrease the
computational complexity of TDSCF is the treatment of a small subset of the
system using TDSCF, while treating the other degrees of freedom in the
classical limit of TDSCF. This family of methods is often referred to as mixed
quantum–classical methods.29 More precisely, the quantum degrees of free-
dom evolve in time according to a TDSCF-like equation; however, the mean
potential for the quantum subsystem is obtained from classical trajectories.
Accordingly, the classical degrees of freedom are not described by a wave
function, and instead evolve in time classically under the influence of a force
which is the quantum mechanical average of the full potential over the quan-
tum degrees of freedom. Using quantum–classical TDSCF and employing an
additional approximation, describing each mode using Gaussian wavepackets,
low temperature clusters such as Ar13 and (H2O)548 and inelastic scattering of
argon off (H2O)1149 were studied.
Another approach which was successful in incorporating quantum effects
in significantly larger systems is the CSP method.31 In this method, which is
based on the TDSCF method, the quantum mechanical mean potential
governing the time-dependent dynamics of each mode is replaced by an
average potential obtained from classical trajectories simulations, using prop-
erly sampled initial conditions. A key point here, as well as in mixed
quantum–classical dynamics, is that the initial conditions sampled for the
classical trajectory simulations represent the classical analog for the quantum
mechanical process studied. The main difference between CSP and mixed
quantum–classical methods is that in CSP while all degrees of freedom are
described quantum mechanically, there is no quantum force acting on the
nuclei in the classical trajectories used to build the effective potentials for
each mode. The average potentials obtained from classical molecular
dynamics simulations implicitly couple the different modes of vibration to
each other through the time-dependence of the potential. As a result, the
single-mode equations are not solved self-consistently in the CSP method.
While the lack of self-consistency may give rise to errors, it was shown that,
Mean-Field Methods for Time-Dependent Quantum Dynamics of Many-Atom Systems 5
for several test cases, CSP gives results in good agreement with TDSCF
results.31 Using CSP, much larger systems were studied such as the electron
photodetachment in C60 .50 In this paper, we will also present a recent
extension of CSP, denoted ab initio CSP (AICSP),51 which uses potentials
directly obtained from quantum chemical methods. Recent application to
test the validity of AICSP to vibrational spectroscopy of realistic systems will
be discussed.
The remainder of the paper is organized as follows: Section 2 presents the
theory of VSCF, TDSCF, mixed quantum–classical methods, and the CSP
method. Section 3 describes new methodological developments, which are
aimed at describing large systems using ab initio PES, and their recent appli-
cations. Section 4 of the paper summarizes and presents current challenges of
mean-field methods and possible future directions.
Next, it is assumed that the total nuclear wave function can be represen-
ted by a simple product of single-mode functions
Y
N
Χ ðnÞ ðQ1 , …, QN Þ ¼ χ j ðnÞ Qj (3)
j¼1
Since the effective potential for each mode depends on all other modes,
the VSCF equations are solved self-consistently. The main computational
difficulty in applying VSCF lies in the evaluation of the multidimensional
integrals involved in calculating the effective potentials. This computational
complexity is significantly reduced if the potential can be approximated, to a
reasonable extent, by a pairwise approximation:
X
N
diag XN X
N
coup
V ðQ1 , …, QN Þ ¼ Vj Qj + Vij Qi , Qj (6)
j¼1 i j>i
^ iℏ @
hδΧjH jΧi ¼ 0 (9)
@t
where δΧ is the variation in Χ with respect to the single-mode function χ k,
we get the following TDSCF single-mode equations
procedure60 of writing the single-mode wave function using the polar rep-
resentation (given in Eq. (12)), deriving the equations for the amplitude
Aj(Qj, t) and phase Sj(Qj, t) and taking the classical limit in which ℏ ! 0.
i
ϕj Qj , t ¼ Aj Qj , t eℏSj ðQj , tÞ (12)
Plugging Eq. (12) into Eq. (10) and separating the real and imaginary
parts, we get the equations for the amplitude and phase
2
@Sj Qj , t 1 @Sj Qj , t
+ + V TDSCF Qj , t
@t 2 @Qj j
ℏ2 @ 2 A j Qj , t
¼ (13)
2Aj Qj , t @Qj2
@Aj Qj , t @Aj Qj , t @Sj Qj , t
¼
@t @Qj @Qj
1 @ 2 Sj Qj , t
Aj Qj , t (14)
2 @Qj2
So far, Eqs. (13) and (14) are completely analogous to the TDSCF equa-
tions. Now, however, we can take the classical limit to get an equation for
the classical phase SCj (Qj, t)
!2
@SjC Qj , t 1 @Sj Qj , t
C
+ + V TDSCF Qj , t ¼ 0 (15)
@t 2 @Qj j
The equation for the classical amplitude is identical to Eq. (14), except
that the classical phase appears instead of the quantum mechanical phase in
the equation. Eq. (15) is simply the Hamilton–Jacobi equation for a particle
evolving in time classically under the influence of the potential
V TDSCF
j Qj , t . We note that V TDSCF
j Qj , t is, as before, the quantum
mechanical average of the full potential over all other degrees of freedom.
Thus, the classical modes are evolved under the influence of a force which
originates from the other quantum degrees of freedom. Accordingly, the
TDSCF equations for the quantum mechanical modes are not changed,
but the averaging of the potential over the classical modes is now done using
classical trajectories. We note that in the derivation given here, starting from
the TDSCF equations and taking the classical limit, complete separability is
10 Barak Hirshberg and R. Benny Gerber
@ψ ðr, tÞ ^
iℏ ¼ T r + V^ ðr, RðtÞÞ ψ ðr, tÞ (17)
@t
where T^ r is the kinetic energy operator for the quantum nuclei and V^ is the
potential energy operator.
In both derivations, the quantum mechanical equations are solved self-
consistently with the equations of motion for the classical subsystem. We
note that while mixed quantum–classical methods allow simulations of large
systems, the quantum mechanical subsystem must still be small.
In cases where the energy or action for the classical subsystem is high,
describing some modes classically and others quantum mechanically works
very well, as shown for the photodissociation of HI in Xe clusters.61,62
However, due to the fact that some modes are not described by a wave func-
tion, significant errors can arise, such as negative absorption cross sec-
tions.63,64 Other possible discrepancies can results from the fact that the
energy for the classical modes can drop below the ZPE of the mode.
X
Ntraj
V CSP
k ðQk , t Þ ¼ Wα V Q1α ðtÞ,…, Qk , …, QN
α
ðt Þ (18)
α¼1
are determined from Eq. (18). In order to obtain stationary frequencies and
wavefunctions for each mode, a reasonable approach seems to be to average
the time-dependent effective potentials over a time period τ, to obtain effec-
tive time-independent potentials, given by Eq. (20). τ should obviously be
long compared with the time-scale of the relevant mode
Zτ
e CSP 1
V k ð Qk Þ ¼ V CSP
k ðQk , t Þ: (20)
τ
0
e k ðQk Þ
ℏ2 @ 2 ϕ
e CSP e e
+V k ðQk Þϕk ðQk Þ ¼ Ek ϕk ðQk Þ: (21)
2 @Qk 2
3.1.4 Applications
The recent application of AICSP51 focused on CH, NH and OH stretching
modes in amino acids such as glycine, alanine, and proline as well as the
guanine–cytosine (G–C) pair of nucleobases. We note that while these sys-
tems are already nontrivial in size (for example, the G–C complex has
81 vibrational modes), admittedly, the AICSP method was developed with
the intent of applying it to much larger systems, such as peptides. However,
if the method performs well for amino acids, the building blocks of peptides
16 Barak Hirshberg and R. Benny Gerber
and proteins, this will encourage applications also for other biological mol-
ecules. The anharmonic frequencies obtained using AICSP were in very
good agreement with experimentally measured values.51 Using MP2 poten-
tials, the deviation from experimental values for nonhydrogen-bonded OH
and NH stretching frequencies was smaller than 1%. For the hydrogen-
bonded OH stretches, the agreement was good, with deviations from exper-
imental values being approximately between 2% and 4%. In this overview,
we focus on features seen in the time-dependent effective potentials as well
as on the properties of the effective potential averaged over time, from
which the stationary vibrational frequencies are obtained. The differences
between the effective single-mode potentials, the harmonic approximation,
and the diagonal anharmonic potential, given by Eq. (22), are also discussed.
We present specific results for conformer I of alanine, shown in Fig. 1, as a
representative example
diag
Vk ðQk Þ ¼ V ð0, …, Qk , …, 0Þ (22)
While the first application of AICSP was to stationary states, as was men-
tioned earlier, analyzing the time-dependent potentials still provides dynam-
ical information on classical vibrational energy flow in the system. Fig. 1
shows the different time-dependent single-mode effective potentials for sev-
eral modes of conformer I of alanine. Several interesting features can be seen:
First, fluctuations in the time-dependent effective potentials are seen initially
for all modes. However, for approximately 600 fs these fluctuations are
rather mild. Then, for the symmetric NH2 stretching mode and the OH
stretching mode, the fluctuations seem to increase and the potential becomes
softer than at t ¼ 0. This suggests that for short timescales, due to the initial
conditions chosen, energy flow between the modes is indeed minor. How-
ever, after a certain period of time, energy starts to flow significantly
between some of the modes, which results in larger fluctuations of the effec-
tive potentials. These increased fluctuations are not seen for the asymmetric
NH2 stretching mode, which fluctuates at a similar amplitude up to a sim-
ulation time of 1 ps. This may suggest that the OH stretching mode and the
symmetric NH2 stretch are more significantly coupled to other modes in the
system. The significant energy flow between the modes may also provide
information on vibrational linewidths, and we intend to pursue such analysis
in the near future.
Fig. 2 shows different time-averaged effective potentials obtained using
Eq. (20) from the time-dependent single-mode potentials presented in
Fig. 1. The shape of time-averaged effective potentials provides information
on the nature of the different modes. For example, it shows that the OH
stretch is significantly more anharmonic than the NH2 stretches. This agrees
with experimental measurements70 since a significant redshift of approxi-
mately 170 cm1 between the harmonic frequencies calculated at the
MP2 level of theory and the experimental value is seen for the OH stretching
mode, while a much lower shift of approximately 30–55 cm1 is seen for the
NH2 stretching modes. As was mentioned previously, the stationary fre-
quencies obtained using the time-averaged potentials are in very good agree-
ment with experimental measurements.
A comparison of the time-averaged effective potentials with the har-
monic potential and the diagonal anharmonic potential also provides addi-
tional information on the different modes. The comparison for the different
modes of conformer I of alanine is given in Fig. 3. The main feature seen is
that the time-averaged potentials are often very similar to the diagonal
18 Barak Hirshberg and R. Benny Gerber
Fig. 2 Time-averaged effective single-mode potentials obtained using Eq. (20) for dif-
ferent vibrational modes of conformer I of alanine. Blue diamonds represent the sym-
metric NH2 stretch, orange circles the asymmetric NH2 stretch, and yellow squares
present the OH stretching mode. Results were obtained using MP2/aug-cc-pVDZ level
of theory.
potential. This is not surprising since, especially for fundamentals, the diag-
onal accounts for a significant portion of the anharmonic correction, which
is often called the intrinsic anharmonicity. The time averaging over the clas-
sical trajectories accounts for the different interactions between the modes.
This effect is subtle, but very important in order to obtain spectroscopically
accurate results for modes with significant anharmonicity, such as OH
stretches. A comparison to the harmonic potentials shows that indeed,
the asymmetric NH2 stretch can be well approximated by the harmonic
approximation even for energies well above the fundamental transition. This
agrees with the small deviation of the harmonic approximation from the
experiment for this mode which is about 1% at the MP2 level of theory.
On the other hand, this analysis also emphasizes the significant anharmonic
character of the OH stretching mode for which anharmonic treatment is
essential.
between the nuclei using ab initio potentials directly. Such potentials are
required in order to accurately describe complex systems, for which empir-
ical potentials are not generally available. Several developments in this spirit,
presented in this paper are (1) The AICSP method, which combines the
CSP method for quantum nuclear dynamics with ab initio PES, and (2)
the QWAIMD method, which is a mixed quantum–classical method where
the trajectories for the classical subsystem are performed using potentials
obtained directly from quantum chemical methods. AICSP was applied
recently to the calculation of stationary anharmonic vibrational frequencies
of amino acids and the G–C pair of nucleobases,51 for which ab initio poten-
tials are essential in order to obtain quantitative agreement with experimen-
tal measurements. QWAIMD was applied, for example, to study hydrogen
tunneling in a model for an enzyme active site.78
These recent developments open a way to several exciting future
directions and to a wide range of applications, such as ultrafast time-
domain spectroscopy measurements, the calculation of vibrational lin-
eshapes, electron photodetachment in large systems, and transition state
dynamics. For example, while several applications, mainly by McCoy
and coworkers,57–59 were performed using mixed quantum–semiclassical
TDSCF to study transition state dynamics of ClHCl in Ar clusters, the
application of mean-field methods for transition state dynamics has so
far been limited mainly to systems for which an accurate empirical poten-
tial in the vicinity of the transition state was available. The direct use of
ab initio potentials can significantly increase the range of applicability of
mean-field methods to study transition state dynamics in different systems.
This may also allow quantitative comparison of mean-field methods with
experimental measurements, which has not been done extensively in ear-
lier studies. At the same time, a major challenge that remains is finding
extensions of the method. Currently, methods such as AICSP are very
much limited to cases for which normal coordinates can be used and
the dynamics are dominated by a single minimum on the potential. Appli-
cation of these methods for cases in which more than one local minimum is
involved is problematic, since the separable ansatz will no longer be ade-
quate. It is an interesting challenge to develop methods that use other coor-
dinates which maintain the separability of the nuclear wavefunction for
longer timescales and more complicated dynamics. In these cases, tech-
niques used for stationary vibrational spectroscopy, such as the use of inter-
nal coordinates53 or local coordinates52 may prove very useful.
22 Barak Hirshberg and R. Benny Gerber
ACKNOWLEDGMENTS
We dedicate this paper to Mark Ratner. R.B.G. gratefully acknowledges very stimulating
collaborations with Mark, including in the field of this paper, and many illuminating
discussions and exchanges of ideas. B.H. is supported through an Adams Fellowship of the
Israel Academy of Sciences and Humanities. We thank Dr. Laura McCaslin for her
comments on the manuscript.
REFERENCES
[1] Rapaport, D. C. The Art of Molecular Dynamics Simulation, Cambridge University Press:
Cambridge, 2004.
[2] Marx, D.; Hutter, J. Ab Initio Molecular Dynamics: Theory and Implementation. In
Grotendorst, J., Ed.; Modern Methods and Algorithms of Quantum Chemistry, 1,
2000; pp 301–449.
[3] Car, R.; Parrinello, M. Unified Approach for Molecular Dynamics and Density-
Functional Theory. Phys. Rev. Lett. 1985, 55(22), 2471–2474.
[4] Hutter, J. Car-Parrinello Molecular Dynamics. Wiley Interdiscip. Rev. Comput. Mol. Sci.
2012, 2(4), 604–612.
[5] K€uhne, T. D. Second Generation Car-Parrinello Molecular Dynamics. Wiley Interdiscip.
Rev. Comput. Mol. Sci. 2014, 4(4), 391–406.
[6] Karplus, M.; McCammon, J. A. Molecular Dynamics Simulations of Biomolecules.
Nat. Struct. Biol. 2002, 9(9), 646–652.
[7] Ufimtsev, I. S.; Luehr, N.; Martinez, T. J. Charge Transfer and Polarization in Solvated
Proteins from Ab Initio Molecular Dynamics. J. Phys. Chem. Lett. 2011, 2(14),
1789–1793.
[8] Hassanali, A. A.; Cuny, J.; Verdolino, V.; Parrinello, M. Aqueous Solutions: State of the
Art in Ab Initio Molecular Dynamics. Philos. Trans. A. Math. Phys. Eng. Sci. 2014,
372(2011), 20120482.
[9] Lin, I.-C.; Seitsonen, A. P.; Tavernelli, I.; Rothlisberger, U. Structure and Dynamics of
Liquid Water from Ab Initio Molecular Dynamics—Comparison of BLYP, PBE, and
revPBE Density Functionals with and without van Der Waals Corrections. J. Chem.
Theory Comput. 2012, 8(10), 3902–3910.
[10] Matsumoto, M.; Saito, S.; Ohmine, I. Molecular Dynamics Simulation of the Ice
Nucleation and Growth Process Leading to Water Freezing. Nature 2002,
416(6879), 409–413.
[11] Hafner, J. Ab-Initio Simulations of Materials Using VASP: Density-Functional Theory
and beyond. J. Comput. Chem. 2008, 29(13), 2044–2078.
[12] Gerber, R. B.; Shemesh, D.; Varner, M. E. E.; Kalinowski, J.; Hirshberg, B. Ab Initio
and Semi-Empirical Molecular Dynamics Simulations of Chemical Reactions in Iso-
lated Molecules and in Clusters. Phys. Chem. Chem. Phys. 2014, 16, 9760–9775.
[13] Ceriotti, M.; Fang, W.; Kusalik, P. G.; McKenzie, R. H.; Michaelides, A.;
Morales, M. A.; Markland, T. E. Nuclear Quantum Effects in Water and Aqueous Sys-
tems: Experiment, Theory, and Current Challenges. Chem. Rev. 2016, 116(13),
7529–7550.
[14] Paesani, F.; Viel, A.; Gianturco, F. A.; Whaley, K. B. Transition from Molecular Com-
plex to Quantum Solvation in 4He_{N}OCS. Phys. Rev. Lett. 2003, 90(7), 73401.
[15] Grebenev, S.; Sartakov, B. G.; Toennies, J. P.; Vilesov, A. F. Spectroscopic Investiga-
tion of OCS (P-H2)n (n ¼ 1–16) Complexes inside Helium Droplets: Evidence for
Superfluid Behavior. J. Chem. Phys. 2010, 132(6), 064501.
[16] Gao, J.; Truhlar, D. G. Qunatum Mechanical Methods for Enzyme Kinetics. Annu.
Rev. Phys. Chem. 2002, 53(1), 467–505.
Mean-Field Methods for Time-Dependent Quantum Dynamics of Many-Atom Systems 23
[17] Vardi-Kilshtain, A.; Nitoker, N.; Major, D. T. Nuclear Quantum Effects and Kinetic
Isotope Effects in Enzyme Reactions. Arch. Biochem. Biophys. 2015, 582, 18–27.
[18] Wang, L.; Isborn, C. M.; Markland, T. E. Chapter Fifteen—Simulating Nuclear
and Electronic Quantum Effects in Enzymes. In: Gregory A., Voth Ed.; Methods in
Enzymology, Academic Press, 2016, Vol. 577, pp 389–418, ISSN 0076-6879, ISBN
9780128053478, http://dx.doi.org/10.1016/bs.mie.2016.05.047.
[19] Kosloff, R. Time-Dependent Quantum-Mechanical Methods for Molecular Dynam-
ics. J. Phys. Chem. 1988, 92(8), 2087–2100.
[20] Zhang, D. H.; Guo, H. Recent Advances in Quantum Dynamics of Bimolecular Reac-
tions. Annu. Rev. Phys. Chem. 2016, 67(1), 135–158.
[21] Otto, R.; Ma, J.; Ray, A. W.; Daluz, J. S.; Li, J.; Guo, H.; Continetti, R. E. Imaging
Dynamics on the F + H2O ! HF + OH Potential Energy Surfaces from Wells to Bar-
riers. Science 2014, 343, 396–399.
[22] Miller, W. H. The Semiclassical Initial Value Representation: A Potentially Practical
Way for Adding Quantum Effects to Classical Molecular Dynamics Simulations.
J. Phys. Chem. A 2001, 105(13), 2942–2955.
[23] Heller, E. J. Time-Dependent Approach to Semiclassical Dynamics. J. Chem. Phys.
1975, 62(4), 1544.
[24] Heller, E. J. Frozen Gaussians:A Very Simple Semiclassical Approximation. J. Chem.
Phys. 1981, 75(6), 2923.
[25] Cao, J.; Voth, G. A. The Formulation of Quantum Statistical Mechanics Based on the
Feynman Path Centroid Density. I. Equilibrium Properties. J. Chem. Phys. 1994,
100(7), 5093–5105.
[26] Cao, J.; Voth, G. A. The Formulation of Quantum Statistical Mechanics Path
Centroid Density. II. Dynamical Properties. J. Chem. Phys. 1993, 1994(100),
5106–5117.
[27] Habershon, S.; Manolopoulos, D. E.; Markland, T. E.; Miller, T. F. Ring-Polymer
Molecular Dynamics: Quantum Effects in Chemical Dynamics from Classical Trajec-
tories in an Extended Phase Space. Annu. Rev. Phys. Chem. 2013, 64, 387–413.
[28] Gerber, R. B.; Buch, V.; Ranter, M. A. Time-Dependent Self-Consistent Field
Approximation for Intramolecular Energy Transfer. I. Formulation and Application
to Dissociation of van Der Waals Molecules. J. Chem. Phys. 1982, 1982(77), 3022.
[29] Messina, M.; Coalson, R. D. Time-Dependent Hartree Wave Packet Dynamical Tech-
niques for Computation of Electronically Excited State Optical Spectra of Many-Body
Quantum Systems. J. Chem. Phys. 1989, 90(8), 4015.
[30] Makri, N.; Miller, W. H. Time-Dependent Self-Consistent Field (TDSCF) Approxi-
mation for a Reaction Coordinate Coupled to a Harmonic Bath: Single and Multiple
Configuration Treatments. J. Chem. Phys. 1987, 87(1987), 5781.
[31] Jungwirth, P.; Gerber, R. B. Quantum Dynamics of Large Polyatomic Systems Using a
Classical Based Separable Potential Method. J. Chem. Phys. 1995, 102, 6046.
[32] Tully, J. C. Mixed Quantum-Classical Dynamics. Faraday Discuss. 1998, 110, 407–419.
[33] Gerber, R. B.; Ratner, M. A. A Semiclassical Self-Consistent Field (SC SCF) Approx-
imation for Eigenvalues of Coupled-Vibration Systems. Chem. Phys. Lett. 1979, 68(1),
195–198.
[34] Bowman, J. M. Self-Consistent Field Energies and Wavefunctions for Coupled Oscil-
lators. J. Chem. Phys. 1978, 68(2), 608.
[35] Chaban, G. M.; Jung, J. O.; Gerber, R. B. Ab Initio Calculation of Anharmonic Vibra-
tional States of Polyatomic Systems: Electronic Structure Combined with Vibrational
Self-Consistent Field. J. Chem. Phys. 1999, 111(5), 1823.
[36] Roy, T. K.; Gerber, R. B. Vibrational Self-Consistent Field Calculations for Spectros-
copy of Biological Molecules: New Algorithmic Developments and Applications. Phys.
Chem. Chem. Phys. 2013, 15(24), 9468–9492.
Another random document with
no related content on Scribd:
have been developed in membrane overlying one of the dentigerous
bones mentioned, without having become anchylosed to the bone.
When the tooth is fixed to the bone the attachment has generally
been effected by the ossification of the bone of the tooth, but in
some fishes a process of the bone projects into the cavity of the
tooth; in others the teeth are implanted in alveoli. In these, again,
frequently a process of bone rises from the bottom, on which the
tooth rests.
Many fishes, especially predatory fishes with long, lancet-shaped
teeth, have all or some of the teeth capable of being bent inwards
towards the mouth. Such “hinged” teeth resume at once the upright
position when pressure is removed from them. They are, however,
depressible in one direction only, thus offering no obstacle to the
ingress, but opposing the egress of prey. Mr. C. S. Tomes has shown
that the means by which this mechanism is worked are different in
different fishes; for whilst, in the Pediculati and Gadoids (Hake) the
elasticity resides solely in the tissue of the hinge (the tooth being as
resilient as ever after everything else is severed), in the Pike the
hinge is not in the least endowed with elasticity, but the bundles of
fibres proceeding from the interior of the dentine cap are exceedingly
elastic.
As regards texture the teeth of fishes show considerable
variation. The conical teeth of the Cyclostomes and the setiform
teeth of many Teleosteans consist of a horny albuminous substance.
The principal substance of the teeth of other fishes consists of
dentine, with numerous dividing and anastomosing tubercles,
sometimes covered by a stratum of unvascular dentine. An enamel-
like substance has been observed on the crown of the teeth of
Sargus and Balistes, and an ossification of the capsule of their
matrix covers the enamel with a thin coating of cement. The teeth
either possess a cavity in which the matrix is received, or, more
frequently, they are solid, in which case vascular canals of the
underlying bone are continued into the substance of the tooth. In the
teeth of some fishes numerous sets of canals and tubes are so
arranged that they do not anastomose with one another, each set
being surrounded by a layer of dentine and cement. These
apparently simple teeth are evidently composed of numerous small
teeth, and called compound teeth.
The teeth may be, and generally are, very different as regards
size or form in the different parts of the mouth; they may be also
different according to the age or sex of the fish (Raja). The teeth may
be few in number and isolated, or placed in a single, double, or triple
series, distant from one another or closely set; they may form narrow
or broad bands, or patches of various forms. As regards form, they
may be cylindrical or conical, pointed, straight, or curved, with or
without an angular bent near their base; some are compressed
laterally or from the front backwards; the latter may be triangular in
shape, or truncated at the top like incisors of mammals; they may
have one apex (cusp) only, or be bi- or tri-lobate (bi- or tri-cuspid); or
have the margins denticulated or serrated. Compressed teeth may
be confluent, and form a cutting edge in both jaws, which assume
the shape of a parrot’s beak (Fig. 53). In some the apex is hooked or
provided with barbs. Again, some teeth are broad, with flat or convex
surface, like molar teeth. With regard to size, the finest teeth are like
fine flexible bristles, ciliiform or setiform; or, if very short and
anchylosed to the bone, they appear only as inconspicuous
asperities of the bone. Very fine conical teeth arranged in a band are
termed villiform teeth; when they are coarser, or mixed with coarser
teeth, they are card-like (dents en rape or en cardes) (Fig. 54);
molar-like teeth of very small size are termed granular.
Fig. 53.—Jaws of Calliodon.
In all fishes the teeth are constantly shed or renewed during the
whole course of their life. In fishes which have compound teeth, as
the Dipnoi, Chimæroids, Scari,[14] Gymnodonts, as well as in those
which have apparently permanent teeth, as in the saw of Pristis, the
detrition of the surface is made up by a constant growth of the tooth
from its base. When the teeth are implanted in alveoli, they are
generally succeeded by others in the vertical direction, but in others
they succeed one another, side by side. In the majority of fishes the
new tooth is not developed (as in reptiles and mammals) in a
diverticulum of the sack of its predecessor, but like this from the free
surface of the buccal membrane. Generally there are more than one
tooth growing, which are in various stages of development, and
destined to replace the one in function. This is very conspicuous in
Sharks, in which the whole phalanx of their numerous teeth is ever
marching slowly forwards (or in some backwards), in rotatory
progress, over the alveolar border of the jaw, the teeth being
successively cast off after having reached the outer margin, and
fulfilled for a longer and shorter period their special function.
[The richest materials for our knowledge of the teeth of fishes are
contained in Owen’s “Odontography.” Lond. 1840. 8vo.]
Fig. 54.—Cardlike teeth of
Plectropoma dentex, with
canines.
The intestinal tract is divided into four portions: œsophagus,
stomach, small and large intestine; two or more of these divisions
may coalesce in fishes and become indistinguishable. But it is
characteristic of the class that the urinary apertures are constantly
situated behind the termination of the intestinal tract.
In Branchiostoma the whole intestinal tract is straight, and coated
with a ciliated mucous membrane. The wide pharynx passes into a
narrow œsophagus, this into a gastric cavity, the remainder being
again narrower and terminating in the anal aperture, which lies
somewhat to the left of the median line. The liver is represented by a
green coloured cœcal diverticulum of the stomachic dilatation. A
mesenterium is absent.
In the Cyclostomi the intestinal tract is likewise straight, and
without clearly defined divisions; however, in Petromyzon the
œsophagus shows numerous longitudinal folds, and the intestine
proper is provided with a single longitudinal fold. A mesentery, which
is present in the Myxinoids, is represented by a short median fold
only, by means of which the hindmost part of the intestine is fixed.
The Palæichthyes show differences in the structure of their
intestinal tract as considerable as are found among the Teleostei, but
they have that in common that the absorbent surface of their
intestine is enlarged by the development of a spiral valve, evidence
of the presence of which in extinct Palæichthyes is still preserved in
the fossilised fæces or coproliths, so abundant in some of the older
strata.
In Chondropterygians (Fig. 55) the stomach is divided into a
cardiac and pyloric portion, the former frequently terminating in a
blind sac, and the latter varying in length. The pyloric portion is bent
at its origin and end, and separated from the short duodenum (called
Bursa entiana in these fishes) by a valve; the ductus hepaticus and
pancreaticus enter the duodenum. This is succeeded by the straight
intestine provided with the spiral valve, the coils of which may be
either longitudinal and wound vertically about the axis of the
intestine, as in Carcharias, Galeocerdo, Thalassorhinus, and
Zygœna, or they may be transverse to that axis, as in the other
genera. The number of gyrations in the latter case varies: there may
be as many as forty. The short rectum passes into a cloaca, which
contains also the orifices of the urogenital ducts. Only the
commencement and end of the intestinal tract are fixed by
mesenterial folds.
In the Holocephali and Dipnoi, the intestinal tract is short,
straight, and wide, without stomachic dilatation, a pyloric valve, close
to which the ductus choledochus enters, indicating the boundary of
the intestine proper (Fig. 57, p). The spiral valve is perfect, and
makes from three (Chimæra) to nine (Ceratodus) gyrations. A cloaca
is present, as in Chondropterygians. A mesentery fixing the dorsal
side of the intestine is absent.
Fig. 55.—Siphonal stomach and spiral valve of
Basking-Shark (Selache). (After Home and Owen.)
a, Œsophagus; b, Cardiac portion of stomach; c, pyloric
portion; d, pouch intermediate between stomach
and duodenum, with circular valves at both ends; e,
Duodenum; f, Valve of intestine; g, Ductus
hepaticus; h, Spleen.
The other Ganoids resemble again more the Chondropterygians
in the structure of their intestinal tract. The stomach has always a
distinct pyloric portion, and has a still more complicated structure in
Acipenser. The duodenal portion receives the contents of
Appendices pyloricæ, which are confluent into a gland-like mass in
Acipenser, but separate in Polyodon, and numerous and short in
Lepidosteus, whilst Polypterus possesses one such appendage only.
A spiral valve is developed in the Sturgeons and Polypterus, but in
Amia, in which the intestine performs several convolutions, the four
gyrations of the valve are situated far back towards the end of the
intestine. In Lepidosteus the valve is rudimentary, and indicated only
by three raised lines crossing the terminal portion of the intestine. In
all these Ganoids the rectum has a separate opening, without
cloaca.
The structure of the intestinal tract of Teleosteous fishes is
subject to so numerous modifications that we should go beyond the
limits of the present work if we would attempt to enter into details.
Great differences in this respect may be found even in groups of the
same natural families. Frequently the intestinal tract remains of
nearly the same width throughout its course, and only the entrance
of the various ducts serves as a guide for the distinction of its
divisions. An intestine of such uniform width may be straight and
short, as in Scombresocidæ, Symbranchidæ, or it may be more or
less convoluted and long, as in many Cyprinidæ, Doradina, etc. On
the whole, carnivorous fishes have a much shorter and simpler
intestinal tract than herbivorous.
In the majority of Teleosteans, however, œsophagus, stomach,
duodenum, small intestine and rectum, can be more or less
distinctly, even externally distinguished.
There are two predominant forms of the stomach, intermediate
forms being, however, numerous. In the first, the siphonal, it
presents the form of a bent tube or canal, one-half of the horse-shoe
being the cardiac, the other the pyloric portion. In the second, the
cæcal, the cardiac division is prolonged into a long descending blind
sac, the cardiac and pyloric openings of the stomach lying close
together (Clupea, Scomber, Thynnus, etc.)
Fig. 56.—Siphonal Stomach and Pyloric
Appendages of a Female Salmon, 3⅓ feet
long. a a a, Pyloric appendages; ch, ductus
choledochus; oe, œsophagus; st, lower end of
stomach; p, pyloric region; i, ascending; and í’,
descending portion of intestine.
The duodenum receives always the hepatic and pancreatic
secretions, and, besides, those of the appendices pyloricæ, which, in
varying numbers (from 1 to 200), are of very common occurrence in
Teleosteans (Fig. 56). They vary also in length and width, and whilst
the narrowest serve only as secretory organs, the widest are
frequently found filled with the same contents as the intestine. When
few in number, each opens by a separate duct into the duodenum;
when their number is greater two or more coalesce into a common
duct; in the latter case the appendages cease to be free, and are
connected with one another by a more or less firm tissue.
Cœcal appendages at the end of the intestinal tract are of
exceedingly rare occurrence in fishes (Box). There is no cloaca in
Teleosteans.
In the majority of Teleosteous fishes the vent is situated on the
boundary between trunk and tail, behind the ventral fins. In a few it
lies farther backwards, not far from the caudal fin; more frequently it
is advanced forwards, under the middle of the abdomen or to the
scapular arch. In two fishes, Aphredoderus and Amblyopsis, it lies
before the pectoral fins.
A peritoneum envelops all the divisions of the intestinal tract
within the abdominal cavity. A broad, well-developed omentum has
hitherto been found in Gobiesox cephalus only.
Liver.—The existence of a liver in Branchiostoma as a long
diverticulum of the intestine has been mentioned above. In the
Myxinoids the liver is divided into two glandular bodies, an anterior
rounded smaller one, and a posterior larger one of an elongate
shape. The gall-bladder lies between both, and receives a cystic
duct from each of them. In the other fishes the proportionally large
liver is a single large gland, from which only now and then small
portions are found to be detached. It is either simple, or with a right
and left lobe, or with a third lobe in the middle; each lobe may have
incisions or subdivisions, which, however, are very inconstant. The
liver of fishes is distinguished by the great quantity of fluid fat (oil)
which it contains. The gall-bladder is but rarely absent, and attached
to the right lobe, or towards the centre; however, in some fishes it is
detached from the liver and connected with it by the cystic duct only.
The bile may be conveyed by one or more hepatic ducts into a
common duct which is continued towards the gall-bladder as ductus
cysticus, and towards the duodenum as ductus choledochus; or
some of the hepatic ducts enter directly the gall-bladder, or directly
the duodenum, without communicating with the common duct.
Individual variations in this respect are of common occurrence.
A pancreas has been found hitherto in all Chondropterygians,
Acipenser, and many Teleosteans. In the first it is a glandular mass
of considerable size behind the stomach, close to the spleen; its duct
leads into the duodenum. In the Sturgeons the pancreas is attached
to the duodenum, and opens close to the ductus choledochus. In
Silurus glanis it is very large, and the ductus choledochus passes
through its substance; it is smaller in Belone and Pleuronectes, and
situated in the mesentery; its duct accompanies the terminal portion
of the ductus choledochus. In the Salmon, which possesses a large
lobed pancreas, the duct is so intimately connected with the ductus
choledochus that both appear externally as a single duct only.
The spleen, which is substantially a lymphatic gland, may be
mentioned here, as it is constantly situated in the immediate vicinity
of the stomach, generally near its cardiac portion. With the exception
of Branchiostoma, it is found in all fishes, and appears as a rounded
or oblong organ of dark-red colour. In the Sharks frequently one or
more smaller pieces are detached from the principal body. In the
Dipnoi a thin layer of a very soft substance of brownish-black colour
below the mucous membrane of the stomach and upper part of the
intestine has been regarded as the homologue of the spleen (Fig.
57, m). In most Teleostei the spleen is undivided, and appended by
its vessels and a fold of the peritoneum to the pyloric bend of the
stomach or the beginning of the intestine.
Fig. 57.—Upper part of Intestine of Ceratodus.
The anterior wall of the intestine is opened, the liver
(c) and gall-bladder (e) being drawn forward. A slit is
made at n, through which part of the next
compartment of the spirally wound intestine may be
seen.
é, Mouth of ductus choledochus; f, stomach; i,
adipose agglomeration; l, first compartment of
intestinal spire; m, spleen; oe, lower part of
œsophagus, opened; p, double pyloric fold; q q,
glandular patches.
CHAPTER IX.
ORGANS OF RESPIRATION.