You are on page 1of 53

Advances in Physical Organic

Chemistry Volume 48 First Edition


Williams
Visit to download the full and correct content document:
https://textbookfull.com/product/advances-in-physical-organic-chemistry-volume-48-fir
st-edition-williams/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Advances in Physical Organic Chemistry 47 1st Edition


Ian H. Williams And Nicholas H. Williams (Eds.)

https://textbookfull.com/product/advances-in-physical-organic-
chemistry-47-1st-edition-ian-h-williams-and-nicholas-h-williams-
eds/

Advances in Physical Organic Chemistry 50 1st Edition


Ian H. Williams And Nicholas H. Williams (Eds.)

https://textbookfull.com/product/advances-in-physical-organic-
chemistry-50-1st-edition-ian-h-williams-and-nicholas-h-williams-
eds/

Advances in Clinical Chemistry Volume 66 First Edition


Makowski

https://textbookfull.com/product/advances-in-clinical-chemistry-
volume-66-first-edition-makowski/

Advances in Clinical Chemistry Volume 68 First Edition


Makowski

https://textbookfull.com/product/advances-in-clinical-chemistry-
volume-68-first-edition-makowski/
Chemistry³ Introducing inorganic organic and physical
chemistry 3rd Edition Andrew Burrows

https://textbookfull.com/product/chemistry%c2%b3-introducing-
inorganic-organic-and-physical-chemistry-3rd-edition-andrew-
burrows/

Worked Solutions in Organic Chemistry First Edition


Coxon

https://textbookfull.com/product/worked-solutions-in-organic-
chemistry-first-edition-coxon/

Introductory Organic Chemistry and Hydrocarbons-A


Physical Chemistry Approach 1st Edition Caio Lima Firme
(Author)

https://textbookfull.com/product/introductory-organic-chemistry-
and-hydrocarbons-a-physical-chemistry-approach-1st-edition-caio-
lima-firme-author/

Advances in Heterocyclic Chemistry Volume 118 1st


Edition Ramsden

https://textbookfull.com/product/advances-in-heterocyclic-
chemistry-volume-118-1st-edition-ramsden/

Advances in Clinical Chemistry, Volume 73 Gregory S.


Makowski

https://textbookfull.com/product/advances-in-clinical-chemistry-
volume-73-gregory-s-makowski/
ADVISORY BOARD

B. Feringa University of Groningen, The Netherlands


S. Fukuzumi Osaka University, Japan
E. Juaristi CINVESTAV-IPN, Mexico
J. Klinman University of California, Berkeley
C. Perrin University of California, San Diego
Z. Rappoport The Hebrew University of Jerusalem, Israel
H. Schwarz Technical University, Berlin, Germany
C. Wentrup University of Queensland, Australia
VOLUME FORTY EIGHT

ADVANCES IN
PHYSICAL ORGANIC
CHEMISTRY

Edited by

IAN H. WILLIAMS
Department of Chemistry,
University of Bath,
Bath, United Kingdom

NICHOLAS H. WILLIAMS
Department of Chemistry,
University of Sheffield,
Sheffield, United Kingdom

Amsterdam • Boston • Heidelberg • London


New York • Oxford • Paris • San Diego
San Francisco • Singapore • Sydney • Tokyo
Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier
The Boulevard, Langford Lane, Kidlington, Oxford, OX51GB, UK
32, Jamestown Road, London NW1 7BY, UK
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
225 Wyman Street, Waltham, MA 02451, USA
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA

First edition 2014

Copyright Ó 2014 Elsevier Ltd. All rights reserved.

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: 978-0-12-800256-8
ISSN: 0065-3160

For information on all Academic Press publications


visit our website at http://store.elsevier.com/

Printed and bound in UK


CONTRIBUTORS

Annette D. Allen
Department of Chemistry, University of Toronto, Toronto, ON, Canada
Weifang Hao
Department of Chemistry and Biochemistry, Utah State University, Logan, UT, USA
Zhao Li
Department of Chemistry and Biochemistry, Utah State University, Logan, UT, USA
David J. Nelson
West CHEM/Department of Pure & Applied Chemistry, University of Strathclyde,
Glasgow, UK
Vernon D. Parker
Department of Chemistry and Biochemistry, Utah State University, Logan, UT, USA
Jonathan M. Percy
West CHEM/Department of Pure & Applied Chemistry, University of Strathclyde,
Glasgow, UK
Michael A. Robb
Chemistry Department, Imperial College, London, UK
Thomas T. Tidwell
Department of Chemistry, University of Toronto, Toronto, ON, Canada

vii j
PREFACE

This volume of Advances in Physical Organic Chemistry provides our readers


with opportunities to think afresh about a wide range of topics across the
subject.
Vernon Parker has contributed chapters in volumes 19 and 20, and
following on from his long experience in electrode reactions and kinetics,
he has developed new kinetic methods for the study of fundamental organic
reaction mechanisms. In this volume he is joined by Zhao Li and Weifang
Hao to describe these experimental methods and the data treatment to probe
reactions where pseudo first-order behavior is conventionally assumed. By
exploiting the accuracy and number of data points that can be gathered
routinely with modern spectroscopic instrumentation, the approach to
steady state of a reaction that involves a kinetically significant intermediate
can be highlighted. Analyzing how the extent of reaction varies with time
provides a method to test whether a reaction involves a single step, or is
more mechanistically complex. These data and their analysis provide a
challenge to several fundamental reactions in organic chemistry, and the
authors suggest alternative mechanisms complemented by computational
studies.
David Nelson and Jonathan Percy have taken on the substantial task of
reviewing the literature on metathesis reactions as applied to organic chem-
istry from the perspective of the physical organic chemist. They have dis-
tilled out the structural factors that affect the processes involved in a
quantitative way, identifying the opportunities for the mechanistic chemist
in this field. They consider both the effect of catalyst and substrate structure
on reactivity, and illuminate the blend of stability and reactivity that under-
pins this widely used component of the synthetic organic chemist’s toolbox.
Mike Robb is a leading exponent of the computational chemistry of
electronically excited states and their intersections with ground states of
molecules undergoing chemical reaction, as occur during the course of pho-
tochemical reactions. His chapter, In this molecule there must be a conical inter-
section, distils the underlying theory and practical experience of high-level
numerical computation on several instructive examples in order to demon-
strate that considerable chemical insight about the nature of conical intersec-
tions can be obtained from a simple valence-bond analysis. The emphasis is
to present a way of thinking that can be used not only to rationalize the

, ix j
x Preface

results of computations of photochemical mechanisms but also to make


predictions.
As a former Editor of this series, Tom Tidwell is no stranger to APOC
but he makes a welcome return, this time with long-time collaborator
Annette Allen, with their chapter on Structure and mechanism in ketene chem-
istry. While it should be no surprise that more than half of the citations in this
up-to-date review of substituent effects in stability and reactivity of ketenes,
and the preparation and reactions of ketenes, are from work published since
2010, it is perhaps less expected that such a significant proportion comes
from physical chemistry journals. This illustrates the interdisciplinary nature
of physical organic chemistry: to quote Allen and Tidwell, “Ketenes were
discovered just when the discipline of physical organic chemistry was devel-
oping, and there has been a synergistic and highly beneficial interaction
between these reactive intermediates and the development of the means
for their study.”
New experimental techniques, novel reactions, modern computational
methods, fresh light on interesting molecules: physical organic chemistry
continues to surprise and stimulate. Read, enjoy, reflect, and be inspired!

IAN H. WILLIAMS
NICHOLAS H. WILLIAMS
CHAPTER ONE

Is the Single-Transition-State
Model Appropriate for the
Fundamental Reactions of
Organic Chemistry?
Experimental Methods and Data
Treatment, Pertinent Reactions,
and Complementary
Computational Studies
Vernon D. Parker1, Zhao Li, Weifang Hao
Department of Chemistry and Biochemistry, Utah State University, Logan, UT, USA
1
Corresponding author: E-mail: vernon.parker@usu.edu

Contents
1. Introduction 2
2. Kinetic Methods to Differentiate between Single-Step and Complex Mechanisms 4
2.1 The Basis for the Differentiation of Simple and Complex Mechanisms 4
2.2 Experimental Methods and Data Processing 7
2.3 The Use of Isosbestic Points to Differentiate between Single-Step 22
and Complex Reaction Mechanisms
3. SN2 Reactions in the Gas Phase and in Solution 28
3.1 SN2 Reactions in the Gas Phase 29
3.2 SN2 Reactions in Solution 34
4. Proton Transfer Reactions of Simple and Aryl Nitroalkanes in Solution 40
and in the Gas Phase
4.1 Proton Transfer Reactions in Water 43
4.2 Computations of the Proton Transfer Reactions of Nitroalkanes 47
5. Hydride Transfer Reactions of NADH/NADþ Model and Related Systems 49
5.1 NADH/NADþ Models 49
6. Computation Studies of Electrophilic Aromatic Substitution 62
6.1 Nitration with Nitronium Ions 62
6.2 Electrophilic Bromination of Benzene in the Gas Phase and in Carbon 68
Tetrachloride Solvent
7. Conclusions 73
Acknowledgments 75
References 75

© 2014 Elsevier Ltd.


j
Advances in Physical Organic Chemistry, Volume 48
ISSN: 0065-3160 All rights reserved. 1
http://dx.doi.org/10.1016/B978-0-12-800256-8.00001-1
2 Vernon D. Parker et al.

Abstract
The experimental recording of spectrophotometric data suitable for detailed kinetic
analysis and reaction mechanism determination is covered early in the chapter with
examples of experiments and subsequent data treatment on a variety of different
chemical systems. Several new methods to distinguish between single-step and com-
plex reaction mechanisms are described in detail. The fundamental organic reactions
covered in more detail were limited to recent work on four general reaction types.
These include (1) proton transfer reactions of nitroalkanes and related compounds,
(2) SN2 reactions of CH3-X where X is the leaving group, (3) electrophilic aromatic sub-
stitution, and (4) hydride transfer reactions of NADH/NADþ model systems. In all cases
possible, discussion of the experimental studies was complemented by computa-
tional studies on the same reaction systems. The experimental kinetic studies all
involved short-lived intermediates and could only distinguish the complex mecha-
nisms involved from the mechanism with a single transition state. Computational
studies confirmed the complex nature of the mechanisms and provided structures
for the likely intermediates in the experimental reactions. The energies of the nonco-
valently bonded intermediates were highly dependent on the computational
method and therefore could not confirm the apparent Gibbs free energies of the
transition states.

1. INTRODUCTION
In 1997 Professor George S. Hammond published a short article,
Physical Organic Chemistry after 50 years: It has changed, but is it still there?1
In this article, Hammond summarized the remarkable developments in
the subdiscipline since it began with the work of L.P. Hammett.2 No
mention was made of the development of pseudo-first-order kinetic
methods that was the most important kinetic technique used by the vast
majority of physical organic chemists. Aside from the development of
very rapid kinetic techniques, essentially no improvements were made in
the methods for which so many rate constants were measured for linear
free energy relationships. This is evident from the classic book first pub-
lished in 1961, Kinetics and Mechanism, by J.W. Moore and R.G. Pearson
3rd edition published in 1981.3 This deficiency continued for the
remainder of the twentieth century. The digital revolution occurred dur-
ing the time period of the development of physical organic chemistry. The
latter had a great influence on the work of the physical chemists studying
rapid kinetics but did not filter down to the physical organic chemists
measuring rate constants of slower reactions. The objective of this chapter
is to fill this gap by introducing work on the fundamental reactions of
Single-Transition-State Model: Methods and Studies 3

organic chemistry using new methods which take advantage of modern


digital acquisition of absorbance–time data and extensive new data analysis
of pseudo-first-order kinetics.
The development of new kinetic methods for the study of fundamental
organic chemistry reaction mechanisms followed more than 30 years of
research on electrode reactions and electrode kinetics by the Parker
group.4–25 During the latter period it was observed that the electrode-
generated radical cations and radical anions invariably undergo complex re-
action mechanisms. At the same time, the fundamental reactions of organic
chemistry such as proton transfer, hydride transfer, E2 elimination, and SN2
reactions, continued to be considered single-step processes. It was univer-
sally accepted that bimolecular reactions pass through Eigen encounter com-
plexes26 but these were correctly left out of the mechanisms proposed since
these complexes have extremely short life-times in solution and are held
together very loosely.
In general, experimental kinetic studies of reactions taking place on
the millisecond time scale or longer time scales can readily be carried
out to give semi-quantitative data using the methods developed over
the past 15 years.4–23Single-step and complex reaction mechanisms are
readily differentiated by these procedures. At the present time, experi-
mental kinetic studies cannot provide detailed mechanisms with the iden-
tification of all steps and the structures of the intermediates that may be
present.
The huge effort put into the continued development of quantum chem-
istry by theoreticians over the past 50 years has resulted in readily available
programs, which can be used to compute energies that compare very closely
to those measured by experiment. These programs are user-friendly enough
to allow non-theoreticians to carry out competent computational chemistry
investigations. It was more recently25,26 recognized that new experimental
kinetic studies coupled with computations on the same systems provide a
powerful combination of methods to define the details of fundamental
organic reaction mechanisms.
In this chapter a number of complex mechanisms that are still
regarded as single-step processes by the majority of the organic chemistry
community are discussed using experimental and computational data.
Many of the studies involving computations and experimental data
from the author’s laboratory have not yet been published at the time
of this writing.
4 Vernon D. Parker et al.

2. KINETIC METHODS TO DIFFERENTIATE BETWEEN


SINGLE-STEP AND COMPLEX MECHANISMS
Before considering the experimental methods to differentiate single-
step and complex mechanisms, it is instructive to consider how the
concentration of the product changes as the reactant is consumed. For
the single-step mechanism, for each molecule of reactant consumed a
molecule of product is formed. This 1/1 relationship results in the plot
on the left in Figure 1.1. On the other hand, for the complex mechanism
as shown on the right below, the number of product molecules formed
always lags behind the number of reactant molecules consumed. The first
molecular act in the reaction is the formation of an intermediate (I), which
consumes a reactant molecule without yielding a product molecule. This
phenomenon is the fundamental difference between a single-step and a
complex mechanism.

2.1 The Basis for the Differentiation of Simple and Complex


Mechanisms
The basic relationship for the experimental determination of first-order rate
constants is the extent of reaction (E.R.)–time profile where E.R. is defined
either as ([R]o  [R])/[R]o (decay of reactant concentration) or as [P]/[R]o
(evolution of product concentration). With these definitions in mind, we see
that the scale on both x and y axes in the two plots in Figure 1.1 are actually
apparent E.R. if [A]o is unity. In spectrophotometric kinetic analysis con-
centrations are not measured and it is necessary to assume that absorbance

k1 kf kp
A + B Product A + B I Product
kb

0.6 one-step mechanism 0.6 two-step reversible


consecutive mechanism
k1' = k1[B]0 = 0.1 s-1 -1
0.4 0.4 kf' = kf[B]0 = 0.2 s
[P] / mM

[P]/mM

[A]0 = 1.0 mM -1
kb = kp = 1 s
0.2 0.2 [A]0 = 1.0 mM
steady state
0.0 0.0
non-steady state
-0.2 -0.2
0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6
([A]0 - [A]) / mM ([A]0 - [A]) / mM

Figure 1.1 Theoretical [P] versus ([A]o  [A]) profiles for the reactions following (left)
simple one-step and (right) two-step reversible consecutive mechanisms, calculated
under pseudo-first-order conditions, namely [B]o >> [A]o.
Single-Transition-State Model: Methods and Studies 5

Scheme 1.1 The two-step reversible consecutive mechanisms.

is directly proportional to the concentration of the species monitored. Thus,


we start with the premise that absorbance is due to only reactant or only
product at the wavelengths where reactant decay or product evolution,
respectively, is monitored.
Under conditions where absorbance is due to only the primary reactant
or product, it is of interest to examine how the complex mechanism affects
the E.R.–time profile. A convenient parameter to use for this purpose is the
time ratio t0.5/t0.05 where t0.5 and t0.05 are the times for E.R. to equal 0.5 and
0.05, respectively.6 For a simple single-step mechanism t0.5/t0.05 is a con-
stant, 13.51, independent of the magnitude of the microscopic rate constant
(k1) for the reaction. For the complex mechanism shown in Scheme 1.1,
there are three rate constants (kf, kb, and kp). Since in this chapter we are
dealing with pseudo-first-order kinetics, rather than expressing the forward
rate as kf, we adopt the term kf[B]o where [B]o is the initial concentration of
the reactant in excess (B), which is assumed to remain constant in the kinetic
experiment or calculation. The intermediate in the reaction is called the
“reactant complex” since it is most often a noncovalently bound complex
between reactants A and B.
The data in Table 1.1 are the values of the time ratio as a function of the
magnitudes of the relative rate constants. The value of kp is held constant and
in the calculation of the time ratios shown is equal to unity. Table 1.1 hold
for all values of kp when kb and kf[B]o are expressed as multiples of kp. Since
the limiting value of the time ratio is 13.51 for the single-step mechanism,
any value differing from 13.51 provides a measure of how great the devia-
tions due to the complex mechanism are under the influence of the specific
values of the relative rate constants.
There are significant deviations from 13.51 in all columns in Table 1.1
with the exception of the last column (kp/kb ¼ 0.001), which indicates
that reactions with this set of kinetic parameters would be at steady state.
At the other extreme, in the first data column (kp/kb ¼ 1000), there are sig-
nificant deviations for all kf[B]o. However, limiting values are also
approached in this case, where kb no longer affects the kinetics.
The calculations of the data in Table 1.1 were carried out using the in-
tegrated rate equations6a in Eqns (1.1)–(1.3). In these equations, [R], [I], and
6 Vernon D. Parker et al.

Table 1.1 Time ratios t0.5/t0.05 for the pseudo-first-order reversible consecutive
mechanism with kp held constant at 1/s1 over a range of 106 in the rate
constant ratio (kp/kb) and a range of 105 in kf[B]o
kp/kb
kf[B]o 1000 100 10 1 0.1 0.01 0.001

512 13.05 13.06 13.06 13.06 13.07 13.19 13.46


256 12.62 12.62 12.62 12.63 12.69 13.04 13.47
128 11.85 11.85 11.85 11.87 12.06 12.94 13.49
64 10.61 10.61 10.62 10.68 11.20 12.96 13.50
32 9.00 9.01 9.02 9.16 10.31 13.09 13.50
16 7.41 7.41 7.43 7.69 9.74 13.23 13.51
8 6.12 6.12 6.16 6.56 9.72 13.35 13.51
4 5.28 5.28 5.34 5.89 10.25 13.43 13.51
2 4.85 4.86 4.93 5.68 11.12 13.47 13.51
1 4.72 4.73 4.83 5.89 12.00 13.49 13.51
0.5 4.85 4.87 5.01 6.56 12.65 13.50 13.51
0.25 5.28 5.31 5.52 7.69 13.05 13.51 13.51
0.125 6.12 6.15 6.47 9.16 13.27 13.51 13.51
0.0512 7.85 7.90 8.31 11.07 13.41 13.51 13.51
0.0256 9.49 9.54 9.95 12.14 13.46 13.51 13.51
0.0128 11.15 11.18 11.49 12.82 13.49 13.51 13.51
0.00512 12.38 12.39 12.56 13.21 13.50 13.51 13.51

[P] are as previously defined and (l1 þ l2) is equal to (kf þ kb þ kp), l1l2 is
equal to kfkp, and (kf  l1) (kf  l2) ¼ kfkb.
n
½R ¼ ½Ro ðl2  l1 Þ1 ðkb þ kp  l1 Þexpð  l1 tÞ
o (1.1)
 ðkb þ kp  l2 Þexpð  l2 tÞ
n o
½I ¼ ½Ro kf ðl2  l1 Þ1 expð  l1 tÞ  expð  l2 tÞ (1.2)
n h io
½P ¼ ½Ro 1 þ kf kp ðl1  l2 Þ1 expð  l1 tÞ=l1  expð  l2 tÞ=l2
(1.3)
Another advantage of conditions where only reactant or product absorbs
at the wavelength monitored is that all three rate constants in Scheme 1.1
can be evaluated by fitting experimental E.R.–time profiles to the set of
rate constants most consistent with the experimental data.6a
Single-Transition-State Model: Methods and Studies 7

When an intermediate absorbs at the wavelength monitored, it is no


longer possible to carry out the fitting procedure described in the previous
paragraph. If the extinction coefficient of the intermediate were available, a
data fit could be attempted but the inclusion of another parameter would
seriously affect the reliability of the fit. For reactions following the mecha-
nism in Scheme 1.1, it is highly likely that the “reactant complex” will
adsorb in the same wavelength region as the reactant and a comparable
“product complex” will absorb in the same wavelength region as the
product.

2.2 Experimental Methods and Data Processing


The kinetic responses of a wide variety of reactions will be covered without
mechanistic discussions. Instead, the results of the application of three
different pseudo-first-order procedures on the data will be discussed. The
data presented for each reaction result from a single set of stopped-flow ex-
periments carried out over the course of one day. In all cases, the data pre-
sented are a fraction of the body of data available in our laboratory for each
of these reactions but at the same time are representative of the latter. The
purpose of this type of presentation is to allow readers to assess the data
without being influenced by the author’s views on the mechanistic implica-
tions of the data. The three methods are listed below.
1. Half-Life Dependence of Conventional Pseudo-First-Order Rate Constants.
The procedure involves converting absorbance–time profiles (Abs–t)
to ln (1  E.R.)–time profiles which is the traditional way to obtain
a first-order rate constant. The ln (1  E.R.)-time profiles are then
subjected to linear least squares analysis on five different segments of
the data over time ranges corresponding to 0–0.5 half-lives (HL, when
E.R. ¼ 0.5), 0–1 HL, 0–2 HL, 0–3 HL, and 0–4 HL. The response of
the single-step mechanism to this procedure is that the five apparent
pseudo-first-order rate constants are equal within experimental error.
This is a very simple, but effective method to differentiate single-step
and complex reaction kinetics.
2. Procedure for the Estimation of Instantaneous Rate Constant–Time Profiles.
This method is highly dependent upon gathering enough data to ensure
that experimental error and instrumental noise do not affect the results
significantly. This is readily accomplished using stopped-flow
spectrophotometry to gather the data. The method used in the
author’s laboratory is to gather three independent sets of absorbance–
time data (2000 points each) from 20 stopped-flow shots for either
8 Vernon D. Parker et al.

reactant decay or product evolution over about the first HL of the


reaction. The raw data are first averaged and converted to ln
(1  E.R)–time profiles as before. The linear least squares analysis is
then carried out over the following point segments: 1–3, 1–4, 1–5, 1–
6, ., 1–2000. This gives 1998 apparent instantaneous rate constant
(kIRC) at the middle points of the segments. The three independent
kIRC-time profiles are finally averaged and the result plotted and
compared to the three plots of the data before the final averaging. In
the event that two of the three plots are very close to each other
while the third plot deviates substantially, a fourth independent set of
data are gathered and if the kIRC-time profile is closer to that of the
first two, the deviating set of data is not included in the average.
3. Sequential Pseudo-First-Order Linear Correlation. In our laboratory this
method involves linear correlation of 24 segments of data, each begin-
ning with point 1 in the following sequence; 1–11, 1–21, 1–31, 1–41,
1–51, 1–101, 1–201, 1–301, 1–401, ., 1–1801, 1–1901 from the ln
(1  E.R.)-time profiles. First-order rate constants are calculated for
the 24 segments, and kapp is obtained for each segment and is regarded
to be at the time of the middle point. This procedure can be modified
to best fit the needs of a particular study. The plots obtained from this
procedure resemble those obtained by the instantaneous rate
procedure. In order to obtain statistical data, rather than averaging the
20 absorbance–time profiles, the analysis can be carried out on each of
the profiles and standard deviations (SD) calculated for kapp of each
segment.
In order to illustrate the various methods, eight experimental systems
were chosen. For each example only one set of 20 stopped-flow shots
were used. As mentioned above three sets of data are generally used in
the study of a reaction mechanism. The methods can be illustrated more
easily without getting into a discussion about possible deviations, which
becomes important when discussing mechanisms.
Example 1. The SN2 Reaction of Phenoxide Ion with Methyl Iodide in Aceto-
nitrile at 298 K (Scheme 1.2). The data in Table 1.2 illustrate the application
of the reaction. The value of kapp shows a moderate decrease (about 10%) in
going from 0.5 to 4 HL. Obviously, these changes are much too small to
distinguish between the simple and complex mechanisms.
The application of method (2), instantaneous rate constant analysis, for
the reaction of phenoxide ion with methyl iodide in acetonitrile is illustrated
in Figure 1.2. Here, we see that the plot on the left of kIRC vs time falls
Single-Transition-State Model: Methods and Studies 9

Scheme 1.2 The SN2 reaction of phenoxide ion with methyl iodide.

Table 1.2 Changes in the slopes and the intercepts with the extent of reaction
during conventional pseudo-first-order analysis of the reaction of phenoxide ion
with methyl iodide in acetonitrile at 298 K
kapp/s1 Intercept/s1 Number of HL

0.0125 0.0116 0.5


0.0125 0.0123 1.0
0.0121 0.0223 2.0
0.0117 0.0445 3.0
0.0111 0.0832 4.0

steeply from an initial value of 0.07 to the plateau value of 0.013 s1 in about
2.5 s. This transformation in the plot clearly shows the initial stages of the
complex mechanism proceeding toward the steady state.
The application of method (3) (the 24-point sequential analysis) on the
same reaction is shown in Table 1.3. The value of kapp begins at 0.0444 s1
and the decreases to 0.0157 s1 in segment 6 before holding steady at the
plateau value for the remaining of the analysis. The SD over the 20 measure-
ments are relatively high (equal to about 23% of kapp) for segment 1 and then
decrease to 3.8% at the plateau. The cause of this phenomenon is that the

0.060 [Mel] = 51.4 mM in AN at 298 k


0.04 treated with 1.4 mM PO
[Mel] = 51.4 mM in AN at 298 k
0.045 treated with 1.4 mM PO [PO-] = 0.5mM, 305nm

kIR C /s-1 [PO-] = 0.5mM, 305nm


0.03
0.030 kapp /s-1

0.02
0.015

0.01
0 5 10 15 20 25 0 5 10 15 20 25
time/s time/s

Figure 1.2 Instantaneous rate constant (kIRC)-time plot (left) and a plot of the kapp-
segment data from Table 1.2 for the reaction between phenoxide ion and methyl
iodide in acetonitrile at 298 K (1.0 HL ¼ 48.6 s).
10 Vernon D. Parker et al.

Table 1.3 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of methyl iodide with phenoxide ion
in acetonitrile at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

0.14 0.0444 0.01016 1 10.41 0.0130 0.00005 13


0.27 0.0324 0.00433 2 11.71 0.0130 0.00005 14
0.40 0.0263 0.00191 3 13.01 0.0130 0.00005 15
0.53 0.0230 0.00117 4 14.31 0.0130 0.00005 16
0.66 0.0207 0.00081 5 15.61 0.0130 0.00005 17
1.31 0.0157 0.00034 6 16.91 0.0130 0.00005 18
2.61 0.0136 0.00015 7 18.21 0.0130 0.00005 19
3.91 0.0132 0.00007 8 19.51 0.0130 0.00005 20
5.21 0.0130 0.00006 9 20.81 0.0130 0.00006 21
6.51 0.0130 0.00005 10 22.11 0.0130 0.00006 22
7.81 0.0130 0.00005 11 23.41 0.0129 0.00006 23
9.11 0.0130 0.00005 12 24.71 0.0129 0.00006 24

number of data points that kapp values are derived from in segment 1 is only
11 and the number data points in the analysis steadily increases to 1999 in
going to segment 24.
The data in Table 1.3 are plotted on the right in Figure 1.2. In
comparing the plots in Figure 1.2 we see that they are very nearly superim-
posable. This in no way suggests that the two plots are redundant, but rather
provides a confirmation of results obtained from two very different methods.
Example 2. Proton Transfer Reactions of the Simple Nitroalkanes. The well-
known phenomenon referred to as the Nitroalkane anomaly was named by
Kresge27 to describe the inverse relationship between equilibrium and ki-
netic acidities of nitromethane, nitroethane, and 2-nitropropane in aqueous
solution. Kresge, and all others who have studied these reactions since
treated the data as arising from a simple single-step process. We have
recently19 reinvestigated the reactions and observed significant deviations
from pseudo-first-order behavior. Here, we take 2-nitropropane (2-NP)
as an example and show how detailed pseudo-first-order kinetics distinguish
between single-step and multistep mechanisms for the reactions of the sim-
ple nitroalkanes with hydroxide ion in water.
The HL dependence of kapp for the proton transfer reaction between
2-NP and hydroxide ion in water (Scheme 1.3) obtained by the application
of the pseudo-first-order linear least squares procedure is illustrated by the
data in Table 1.4. Once again, the change in kapp as a function of number
of HL analyzed is modest (about 10% of kapp) but the trend of decreasing
Single-Transition-State Model: Methods and Studies 11

Scheme 1.3 The proton transfer reaction of 2-nitropropane (2-NP) with hydroxide ion
in water.

Table 1.4 Changes in the slopes and the intercepts with the extent of reaction
during conventional pseudo-first-order analysis of the reaction of 2-nitropropane
(0.2 mM) with hydroxide ion (100 mM) in water at 298 K
kapp/s1 Intercept/s1 Number of HL

0.0182 0.0017 0.5


0.0171 0.0104 1.0
0.0168 0.0178 2.0
0.0165 0.0270 3.0
0.0163 0.0404 4.0

rate constants with increases in number of HL is significant and reproducible.


The observed intercepts are much more sensitive to the deviations from
first-order kinetics than observed for kapp. The deviation of both of these
quantities from that expected for first-order kinetics is a clear indication of
the operation of a complex mechanism.
The 24-point sequential analysis of Abs–t data over 1 HL (Table 1.5,
Figure 1.3(c)) for the reactions of 2-NP does not only verify the time-
dependence of kapp but also provides SD of the latter derived from repet-
itive stopped-flow shots. As observed before, the SD are greatest for the
initial kapp values but decrease steadily over the first six segments and
then goes to a nearly constant value equal to 1% or less of the correspond-
ing kapp.
The kIRC–time profile for product formation during the reaction of
2-NP (0.2 mM) with hydroxide ion (100 mM) in water (Figure 1.3(a))
sharply decreases from about 0.03 s1 at short times to a more slowly
decreasing profile near 0.015 s1 at 1 HL. The expected result for the
one-step mechanism is a straight line with zero slope, while that for a com-
plex mechanism increases from zero at short times when only product ab-
sorbs at the experimental wavelength. On the other hand, an initial
decreasing kIRC–time profile at a wavelength that the product but not the
reactant absorbs indicates that decay of a reaction intermediate is responsible
for the initial sharp decrease in absorbance.
12 Vernon D. Parker et al.

Table 1.5 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of 2-nitropropane (0.2 mM) with
hydroxide ion (100 mM) in water at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

0.14 0.0250 0.0045 1 9.02 0.0177 0.0002 13


0.25 0.0223 0.0013 2 10.15 0.0175 0.0002 14
0.36 0.0218 0.0012 3 11.27 0.0174 0.0002 15
0.47 0.0216 0.0008 4 12.40 0.0172 0.0002 16
0.59 0.0214 0.0006 5 13.52 0.0171 0.0002 17
1.15 0.0209 0.0002 6 14.65 0.0170 0.0002 18
2.27 0.0201 0.0002 7 15.77 0.0169 0.0002 19
3.40 0.0195 0.0002 8 16.90 0.0168 0.0002 20
4.52 0.0189 0.0002 9 18.02 0.0168 0.0002 21
5.65 0.0185 0.0002 10 19.15 0.0167 0.0002 22
6.77 0.0182 0.0002 11 20.27 0.0166 0.0002 23
7.90 0.0179 0.0002 12 21.40 0.0166 0.0002 24

(a) (b)
0.0375 [PNM] = 0.0518 mM
[2 -NP] = 0.2 mM in water at 298 k 0.6
[NaOH] = 4.0 mM
[NaOH] = 0.10 M in H2 O at 293k
0.0300 240 nm 0.5 294 nm
kIRC /s-1 -1
kIRC /s
0.0225 0.4

0.0150 0.3
0 5 10 15 20 25 0.00 0.05 0.10 0.15 0.20 0.25
time/s time/s
(c) (d)
0.0375 [2 -NP] = 0.2 mM in water at 298 k [PNM] = 0.0518 mM
0.6
[NaOH] = 4.0 mM
[NaOH] = 0.10 M in H2 O at 293k
0.0300 0.5
294 nm
240 nm
kapp /s-1 kapp /s-1

0.0225 0.4

0.0150 0.3
0 5 10 15 20 25 0.0 0.2 0.4 0.6 0.8 1.0
time/s time/s

Figure 1.3 Instantaneous rate constant (kIRC)-time plots (upper row, a and b) and plots
(bottom row, c and d) of the kapp-segment data from Tables 1.4 and 1.6 for the reactions
of (left column, a and c) 2-nitropropane (0.2 mM) with hydroxide ion (100 mM)
(1.0 HL ¼ 40.6 s) and of (right column, b and d) phenylnitromethane (0.052 mM) with
hydroxide ion (4.0 mM) in water at 293 K (1.0 HL ¼ 1.593 s).
Single-Transition-State Model: Methods and Studies 13

Scheme 1.4 The proton transfer reaction of phenylnitromethane (PNM) with hydroxide
ion in water.

Example 3. Proton Transfer Reactions of Phenylnitroalkanes with Hydroxide


ion in Water. The proton transfer reactions of phenylnitromethane (PNM) in
water and in water–dimethylsulfoxide (DMSO) mixture have been studied
extensively by Bernasconi28,29 and the results have played a prominent role
in discussions of the principle of nonperfect synchronization.30–33 A recent
paper on proton transfer reactions of PNM’s was concerned with the impli-
cations on the nitroalkane anomaly.34
The nonconventional pseudo-first-order study of the reaction of
PNM with hydroxide ion in water (Scheme 1.4) is illustrated by the data
in Table 1.6 in which kapp decreases steadily from 0.440 over 1 HL to
0.356 s1 at 4 HL while the intercept increased over the same range from
0.001 to 0.136. The latter serves to point out that the intercept in this anal-
ysis is also a parameter which helps distinguish between single-step and
complex reaction mechanisms.
The 24-point sequential analysis provides more detail as illustrated in
Table 1.7 and Figure 1.3(d). The value of kapp in segment 1 (relating to
the 11 data points collected in 0–12 ms) was observed to be equal to
0.403 and then rise smoothly to a maximum value of 0.449 at segment 6
and then decrease steadily to 0.434 s1 at segment 24. The latter decrease
in the kapp with time is barely visible in plot of the data in Figure 1.3(d).
The value of SD changes from about 20% of kapp in segment 6 to <1% in

Table 1.6 Changes in the slopes and the intercepts with the extent of reaction
during conventional pseudo-first-order analysis of the reaction of
phenylnitromethane (0.0518 mM) with hydroxide ion (4.0 mM) in water at 293 K
kapp/s1 Intercept/s1 Number of HL

0.440 0.001 0.50


0.433 0.001 1.00
0.420 0.012 2.00
0.398 0.043 3.00
0.356 0.136 4.00
14 Vernon D. Parker et al.

Table 1.7 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of phenylnitromethane (0.0518 mM)
with hydroxide ion (4.0 mM) in water at 293 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

0.006 0.403 0.138 1 0.401 0.442 0.002 13


0.011 0.426 0.043 2 0.451 0.441 0.002 14
0.016 0.434 0.030 3 0.501 0.441 0.001 15
0.021 0.442 0.020 4 0.551 0.440 0.001 16
0.026 0.442 0.015 5 0.601 0.439 0.001 17
0.051 0.449 0.009 6 0.651 0.438 0.001 18
0.101 0.447 0.004 7 0.701 0.438 0.001 19
0.151 0.445 0.002 8 0.751 0.437 0.001 20
0.201 0.444 0.002 9 0.801 0.436 0.001 21
0.251 0.444 0.002 10 0.851 0.435 0.001 22
0.301 0.443 0.002 11 0.901 0.435 0.001 23
0.351 0.443 0.002 12 0.951 0.434 0.001 24

segment 24. This reaction apparently has added complexity in comparison to


previous examples.
The kIRC–time profile (Figure 1.3(b)) for product formation for the
reaction of PNM with hydroxide ion under these conditions presents
more detailed information about the initial stage (0–10 ms) of the reaction,
which shows an initial value close to 0.6 s1, then decreases sharply to about
0.34 s1 at about 10 ms before rising more slowly to 0.43 s1 at about 60 ms
and then remains approximately constant for the remainder of the reaction
period analyzed.
As mentioned above, kIRC-time plots and the plots of the kapp-segment
data should be identical. All of the 24 kapp data points in Figure 1.3(c) and (d)
can fall onto the corresponding kIRC-time plots (Figure 1.3(a) and (b),
respectively). By comparison of the two plots generated by methods (2)
and (3), the plot of 24 kapp data points is good at displaying the tendency
of changing in reaction kinetics, while the kIRC-time plot is more capable
to show the kinetic changes in detail, especially the very initial stage of re-
action, earlier than the first kapp-segment point was recorded. This character
is very important for the kinetic analyses of fast reactions.
Example 4. The Reaction of 1-Benzyl-1,4-dihydronicotinimide (BNAH)
with N-methylacridinium Ion (MAþ) in Acetonitrile. Detailed studies of the re-
actions of BNAH and analogs with substituted isoquinolinium cations were
published by Bunting and co-workers35–37 in the early 1980s and later
reviewed.38
Single-Transition-State Model: Methods and Studies 15

H H H H
CONH2 CONH2
AN
+
+
CIO4
N N N N
CIO4
H 2C Ph CH3 H 2C Ph CH3
BNAH MA+CIO4-
BNA CIO4 MAH

Scheme 1.5 The hydride transfer reaction of 1-benzyl-1,4-dihydronicotinimide (BNAH)


with N-methylacridinium ion (MAþ) in acetonitrile.

Table 1.8 Changes in the slopes and the intercepts with the
extent of reaction during conventional pseudo-first-order
analysis of the reaction of BNAH (7.2 mM) with MAþ (0.3 mM)
in acetonitrile at 298 K
kapp/s1 Intercept/s1 Number of HL

0.564 0.013 0.50


0.536 0.020 1.00
0.522 0.029 2.00
0.513 0.039 3.00
0.502 0.056 4.00

Pseudo-first-order analysis of the BNAH–MAþ reaction (Scheme 1.5)


illustrated in Table 1.8 reveals that kapp varies smoothly from 0.564 for
0.5 HL to 0.502 for 4 HL, approximately a change of 10% over that range.
The 24-point sequential 24-point pseudo-first-order analysis of the
BNAH–MAþ reaction is illustrated in Table 1.9. In this case reactant
(MAþ) decay was monitored at 430 nm and kapp was equal to 1.553 for
segment 1, 0.584 for segment 10 and finally 0.541 for segment 24. The
SD values were large in the early segments and were of the order of 1%
of the corresponding kapp from segment 11 onward.
The kIRC–time profile (Figure 1.4) has the form expected for reactant
decay in a complex mechanism decaying rapidly from an initial value of
close to 2.0 s1 before approaching a very slowly descending value at times
greater than about 1.5 s.
Example 5. The Reaction of N-Methyl-9,10-dihydroacride (MAH) with Tro-
pylium Ion (Trþ) in Acetonitrile. We are not aware of previous studies of the
reaction between MAH and Trþ (Scheme 1.6). This system has the exper-
imental advantage that Trþ does not absorb above 300 nm, leaving a large
wavelength range over which to monitor product (MAþ) formation in
acetonitrile.
The conventional pseudo-first-order analysis (Table 1.10) in this case
follows the general pattern that is emerging, a kapp value of 0.245 s1
16 Vernon D. Parker et al.

Table 1.9 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of BNAH (7.2 mM) with MAþ (0.3 mM)
in acetonitrile at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

0.008 1.553 0.574 1 0.294 0.561 0.005 13


0.012 1.393 0.225 2 0.331 0.557 0.004 14
0.016 1.236 0.122 3 0.367 0.555 0.004 15
0.019 1.106 0.118 4 0.403 0.553 0.004 16
0.023 1.002 0.100 5 0.439 0.551 0.004 17
0.041 0.795 0.022 6 0.475 0.549 0.004 18
0.077 0.679 0.015 7 0.512 0.548 0.004 19
0.113 0.626 0.016 8 0.548 0.546 0.005 20
0.150 0.599 0.014 9 0.584 0.545 0.005 21
0.186 0.584 0.011 10 0.620 0.543 0.006 22
0.222 0.573 0.008 11 0.656 0.542 0.006 23
0.258 0.567 0.006 12 0.693 0.541 0.006 24

2.0
[BNAH]= 7.2mMin AN at 298K
+ -
[MA CIO4 ] = 0.3mM, 430nm
1.5
kIR C /s-1

1.0

0.5
0.0 0.2 0.4 0.6 0.8
time/s
Figure 1.4 Instantaneous rate constant (kIRC)-time plot for the reaction of BNAH
(7.2 mM) with MAþ (0.3 mM) in acetonitrile at 298 K.

H H H H

_ AN
+ + BF +
+
N N _
BF
CH3 Tr + BF4- CH
TrH
MAH MA+CIO4-

Scheme 1.6 The hydride transfer reaction of N-methyl-9,10-dihydroacride (MAH) with


tropylium ion (Trþ) in acetonitrile.
Single-Transition-State Model: Methods and Studies 17

Table 1.10 Changes in the slopes and the intercepts with


the extent of reaction during conventional pseudo-first-
order analysis of the reaction of MAH (0.4 mM) with Trþ
(0.05 mM) in acetonitrile at 298 K
kapp/s1 Intercept/s1 Number of HL

0.245 0.0064 0.50


0.225 0.0199 1.00
0.206 0.0499 2.00
0.190 0.0973 3.00

(1 HL) and decreasing smoothly to 0.190 s1 (4 HL). The corresponding in-
tercepts increase from 0.0064 (0.5 HL) to 0.0973 (4 HL), a factor of about 15
indicating a great deal of curvature in the correlation.
The 24-point sequential analysis (Table 1.11) for product evolution
(MAþ) in this case, once again a sharply decreasing kapp from 0.699 at
segment 1 to a more slowly decreasing 0.266 s1 at segment 8 and finally
at segment 24, kapp equal to 0.222 s1.
The kIRC–time profile (Figure 1.5, left) allows for a more detailed illus-
tration of the pattern indicated above. It should be pointed out, once again,
that this general pattern is obviously inconsistent with a single-step mecha-
nism and that there is an intermediate that absorbs at the wavelength of the
experiment.

Table 1.11 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of MAH (0.2 mM) with Trþ (50 mM) in
acetonitrile at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

0.014 0.699 0.027 1 1.001 0.242 0.001 13


0.026 0.563 0.014 2 1.126 0.239 0.001 14
0.039 0.488 0.009 3 1.251 0.237 0.001 15
0.051 0.442 0.008 4 1.376 0.235 0.001 16
0.064 0.407 0.007 5 1.501 0.233 0.001 17
0.126 0.326 0.005 6 1.626 0.231 0.001 18
0.251 0.281 0.003 7 1.751 0.229 0.001 19
0.376 0.266 0.002 8 1.876 0.228 0.001 20
0.501 0.258 0.002 9 2.001 0.226 0.001 21
0.626 0.252 0.002 10 2.126 0.225 0.001 22
0.751 0.248 0.002 11 2.251 0.223 0.001 23
0.876 0.245 0.001 12 2.376 0.222 0.001 24
18 Vernon D. Parker et al.

0.0015
[MAH] = 40 mM
0.8 [BQCN+] = 0.5 mM
[MAH] = 0.4 mM in AN at 298K 0.0012 in AN at 298 K
0.6 [Tr+ ] = 0.05 M, 430nm kIRC /s-1 294 nm

kIRC /s-1
0.0009
0.4

0.2 0.0006

0.0 0.5 1.0 1.5 2.0 2.5 0 100 200 300 400 500
time/s time/s

Figure 1.5 Instantaneous rate constant (kIRC)-time plots for the reactions of (left) MAH
(0.4 mM) with Trþ (50 mM) (1.0 HL ¼ 2.87 s), and of (right) MAH (40 mM) with BQCNþ
(0.5 mM) in acetonitrile at 298 K (1.0 HL ¼ 602 s).

Example 6. The Reaction of 1-Benzyl-3-cyanoquinolinium Ion (BQCNþ)


with N-Methyl-9,10-dihydroacridine (MAH) in Acetonitrile. The reaction be-
tween MAH and BQCNþ in acetonitrile (Scheme 1.7) has been considered
to be an authentic example of a direct hydride transfer reaction mainly due
to work of Kreevoy and coworkers.39 We presented an alternative three-
step mechanism in 200311 that was challenged by Perrin and Zhao in
2008.40 The mechanism has been found to be even more complex than
we first11 believed and was shown to involve an oxygen-catalyzed chain re-
action even when carried out in a glove-box under a nitrogen atmosphere.22
The data presented here is for the reaction between MAH (40 mM) with
BQCNþ (0.5 mM) in acetonitrile half-saturated with air (Table 1.12). The
latter was achieved by mixing the MAH solution from which air was
removed with a BQCNþ solution in acetonitrile saturated with air in the
mixing chamber of the stopped-flow spectrometer. The 24-point sequential
analysis in this case is distinctively different from any previously presented
one. In Table 1.13 we see that kapp first increases to a maximum value
(0.000,866 s1) before decreasing slowly to 0.000,831 s1 at segment 8
and then increases smoothly for the remainder of the time period analyzed
with a value of 0.001,356 s1 at segment 24.

H H H H
_ H
BF4 CN CN
+
+
AN +
+
N N _
N BF4 N
CH3 H 2C Ph CH3 H2C Ph
+
MAH BQCN MA +CIO4- BQCNH

Scheme 1.7 The hydride transfer reaction of 1-benzyl-3-cyanoquinolinium ion


(BQCNþ) with N-methyl-9,10-dihydroacridine (MAH) in acetonitrile.
Single-Transition-State Model: Methods and Studies 19

Table 1.12 Changes in the slopes and the intercepts with


the extent of reaction during conventional pseudo-first-
order analysis of the reaction of MAH (40 mM) with BQCNþ
(0.5 mM) in acetonitrile half-saturated with air at 298 K
kapp/s1 Intercept/s1 Number of HL

0.00095 0.029 0.50


0.00116 0.066 1.00
0.00148 0.152 2.00
0.00171 0.241 3.00
0.00186 0.312 4.00

Table 1.13 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of MAH (40 mM) with BQCNþ
(0.5 mM) in acetonitrile half-saturated with air at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

2.5 0.000830 0.00003 1 180.2 0.000914 0.00001 13


4.7 0.000866 0.00001 2 202.7 0.000948 0.00001 14
7.0 0.000865 0.00001 3 225.2 0.000983 0.00001 15
9.2 0.000859 0.00001 4 247.7 0.001020 0.00001 16
11.5 0.000854 0.00001 5 270.2 0.001058 0.00001 17
22.7 0.000842 0.00001 6 292.7 0.001097 0.00002 18
45.2 0.000834 0.00001 7 315.2 0.001138 0.00002 19
67.7 0.000831 0.00001 8 337.7 0.001181 0.00002 20
90.2 0.000832 0.00000 9 360.2 0.001224 0.00002 21
112.7 0.000838 0.00000 10 382.7 0.001268 0.00002 22
135.2 0.000856 0.00000 11 405.2 0.001312 0.00003 23
157.7 0.000883 0.00001 12 427.7 0.001356 0.00003 24

The kIRC–time profile (Figure 1.5, right) shows an important detail


hinted at in Table 1.13. An initial sharp increase in kIRC from the initial
to a maximum value nearly 50% greater than the former is shown. From
the maximum, kIRC first decreased slightly and then began a steady increase
for the remainder of the analysis period.
Example 7. The Reaction of Bromoacetonitrile with Phenoxide Ion in Aceto-
nitrile. Bromoacetonitrile (BrAN) is not a common alkyl halide but ClAN
has been used in recent publications concerned with the mechanism of
the SN2 reaction.41,42 The advantage of studying SN2 reactions of BrAN
is that it is much more reactive than a typical alkyl halide.
The data for the pseudo-first-order analysis of the reaction of phenoxide
ion with BrAN (Scheme 1.8) illustrated in Table 1.14 conform to the
20 Vernon D. Parker et al.

Scheme 1.8 The SN2 reaction of bromoacetonitrile with phenoxide ion in acetonitrile.

Table 1.14 Changes in the slopes and the intercepts


with the extent of reaction during conventional
pseudo-first-order analysis of the reaction of BrAN
(5.0 mM) with PO (0.50 mM) in acetonitrile at 298 K
kapp/s1 Intercept/s1 Number of HL

0.0350 0.0125 0.50


0.0346 0.0134 1.00
0.0336 0.0244 2.00
0.0318 0.0565 3.00
0.0289 0.1353 4.00

Table 1.15 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of BrAN (5.0 mM) with PO (0.50 mM)
in acetonitrile at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

0.059 0.1038 0.0188 1 4.306 0.0357 0.0004 13


0.113 0.0841 0.0139 2 4.844 0.0356 0.0004 14
0.167 0.0687 0.0089 3 5.381 0.0355 0.0003 15
0.220 0.0598 0.0059 4 5.919 0.0355 0.0003 16
0.274 0.0542 0.0039 5 6.456 0.0354 0.0003 17
0.543 0.0426 0.0012 6 6.994 0.0354 0.0003 18
1.080 0.0382 0.0005 7 7.532 0.0353 0.0003 19
1.618 0.0369 0.0005 8 8.069 0.0352 0.0003 20
2.156 0.0364 0.0005 9 8.607 0.0352 0.0003 21
2.693 0.0361 0.0004 10 9.144 0.0351 0.0003 22
3.231 0.0359 0.0004 11 9.682 0.0351 0.0003 23
3.768 0.0358 0.0004 12 10.220 0.0350 0.0003 24

pattern observed in earlier tables with kapp equal to 0.0350 s1 (0.5 HL) and
decreasing smoothly to 0.0289 s1 (4 HL).
In segment 1 of the 24-point sequential analysis (Table 1.15), kapp for
reactant (PO) decay was equal to 0.1038 s1 with an SD of 0.0188.
Both quantities decreased sharply over the next few segments to
0.0382 s1 and 0.0005, respectively, in segment 7. The trend continued
but with much smaller changes for the remainder of the reaction period.
Single-Transition-State Model: Methods and Studies 21

0.0012
0.100
[BrAN] = 6 mM in AN at 298 K
0.0008
pretreated with 1 mM PO
0.075 [PO] = 1mM, 310nm kIRC /s-1
pretreated with 0.5 mM BrAN
kIRC /s-1
0.0004 [Pyr] = 1.25 M in AN at 298 K
0.050 [BBr] = 40 mM, 345nm

0.0000
0.025 400
0 2 4 6 8 10 12 0 100 200 300 500
time/s time/s

Figure 1.6 Instantaneous rate constant (kIRC)–time plots for the reactions of (left) BrAN
(5.0 mM) with PO (0.50 mM) (1.0 HL ¼ 19.0 s), and of (right) pyridine (1.25 M) with
benzyl bromide (40 mM) in acetonitrile at 298 K (1.0 HL ¼ 697 s).

The kIRC–time profile for the reaction of phenoxide ion with bromoa-
cetonitrile (Figure 1.6, left) shows an initial steep rise to a maximum at short
times before decreasing toward a very slightly decreasing or nearly constant
value. The formation of intermediate dominants in the initial sharp rise in
kIRC is a possible explanation of this phenomenon.
Example 8. The Reaction of Pyridine with Benzyl Bromide in Aceto-
nitrile (Scheme 1.9).
The pseudo-first-order analysis data in Table 1.16 for the reaction of pyr-
idine with benzyl bromide do not fit the general pattern in correlation quan-
tities expected for product evolution in a complex mechanism, and this is
due to the fact that reactant decay was monitored. In this case, it would
appear from the kapp values that a steady state may be approached late in
the reaction.
The data from the 24-point sequential analysis in Table 1.17 is complex
with kapp ¼ 0.000,174 s1 in segment 1, going to a minimum
(0.000,117 s1) and then increasing steadily to 0.001,066 s1 at segment
24. We have not yet studied this interesting process in detail. It is of interest
to note that the SD values are greater than the corresponding kapp values sug-
gesting that they should be ignored until a more thorough study has been
carried out.
The kIRC–time profile for the SN2 reaction between benzyl bromide and
pyridine (Figure 1.6, right), after the initial small value at point 1, a

Scheme 1.9 The SN2 reaction of pyridine with benzyl bromide in acetonitrile.
22 Vernon D. Parker et al.

Table 1.16 Changes in the slopes and the intercepts


with the extent of reaction during conventional
pseudo-first-order analysis of the reaction of pyridine
(1.25 M) with benzyl bromide (40.0 mM) in
acetonitrile at 298 K
kapp/s1 Intercept/s1 Number of HL

0.00079 0.024 0.50


0.00094 0.052 1.00
0.00104 0.086 2.00
0.00109 0.113 3.00
0.00112 0.134 4.00
0.00114 0.152 5.00

Table 1.17 Apparent rate constants and standard deviations (SD) obtained by the
24-point sequential analysis of the reaction of pyridine (1.25 M) with benzyl
bromide (40.0 mM) in acetonitrile at 298 K
Time/s kapp/s1 SD/s1 Segment Time/s kapp/s1 SD/s1 Segment

2.34 0.000174 0.00023 1 170.21 0.000769 0.00004 13


4.46 0.000126 0.00020 2 191.46 0.000809 0.00005 14
6.59 0.000117 0.00014 3 212.71 0.000840 0.00006 15
8.71 0.000128 0.00011 4 233.96 0.000869 0.00007 16
10.84 0.000147 0.00008 5 255.21 0.000896 0.00008 17
21.46 0.000243 0.00004 6 276.46 0.000922 0.00008 18
42.71 0.000381 0.00002 7 297.71 0.000947 0.00009 19
63.96 0.000479 0.00002 8 318.96 0.000971 0.00010 20
85.21 0.000557 0.00002 9 340.21 0.000994 0.00011 21
106.46 0.000620 0.00003 10 361.46 0.001018 0.00012 22
127.71 0.000674 0.00003 11 382.70 0.001042 0.00013 23
148.96 0.000721 0.00003 12 403.95 0.001066 0.00014 24

maximum is observed at point 4 (kIRC ¼ 0.000188) before the initial kapp


decay as indicated in Table 1.17. The same reservation indicated above
applies to the early kapp-segment number analysis.

2.3 The Use of Isosbestic Points to Differentiate between


Single-Step and Complex Reaction Mechanisms
Isosbestic points,43 commonly observed in UV–vis,44–67 IR,68–70 fluores-
cence,71,72 and electronic73–76 spectroscopies, are defined as the wavelength,
wave number, or frequency at which the total absorbance of a sample does not
change during a chemical reaction or a physical change of the sample.43
Single-Transition-State Model: Methods and Studies 23

A simple example of the occurrence of an isosbestic point is when two


absorbing substances, one of which can be converted into the other (a reac-
tant and a product), have the same molar extinction coefficient at a given
wavelength.
Discussions of isosbestic points began as early as 1937.44 The first com-
plete isosbestic point theory was offered by Cohen and Fischer in 1962.45
Their conclusion is that the appearance of an isosbestic point in a closed sys-
tem requires that the changes in the concentrations of the various compo-
nents be linearly related. Although criticized for lack of considerations of
temperature and solvent composition,46 Cohen and Fisher’s theory is gener-
ally accepted in practical kinetic studies with respect to the developments in
both theoretical47–49 and methodological50–52 aspects.
In order to avoid complications due to overlapping of absorbance,
kinetic measurements using spectrophotometry are most conveniently car-
ried out at wavelengths where only product or reactant absorbs. However,
analyses near isosbestic points can also be carried out qualitatively and quan-
titatively. For example, isosbestic points have been widely used as reference
points for the calibration of spectrophotometers77 and for qualitative analysis
of the composition of a binary mixture.78
According to the theory developed by Cohen and Fisher, the existence
of well-defined isosbestic points (true isosbestic points) has often been
considered to be proof of a single-step conversion53–60 or a parallel mecha-
nism.61 On the other hand, the absence of a true isosbestic point, but rather a
series of steadily shifting intersection points, has been attributed to compli-
cations in the reaction mechanism.62–66 The shift of isosbestic wavelength
has been utilized to estimate the conversion fraction of a precursor to the
product.79 Recently, more and more chemical physicists73–76 have paid
attention to the occurrence of isosbestic points, and compared these to
the fingerprints of electronic correlations.76
For a consecutive reaction, such as A / B / C, Cohen and Fischer45
suggested that true isosbestic points do not occur, which is the criterion often
used to differentiate the parallel and consecutive mechanisms.52 Bunnett80
presumed that spectral evidence for the formation of intermediate B might
be obtained at the isosbestic point between A and C absorbance, if the inter-
mediate B accumulates before converting to C.
In our study19 of the proton transfer reactions of three simple nitroal-
kanes (nitromethane, nitroethane and 2-nitropropane), absorbance–time
profiles due to the formation of reactive intermediates were observed at
the isosbestic points where changes in reactant and product absorbance
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Le mystère du
tigre
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: Le mystère du tigre


roman

Author: Maurice Magre

Release date: December 10, 2023 [eBook #72370]

Language: French

Original publication: Paris: Albin Michel, 1927

Credits: Laurent Vogel and the Online Distributed Proofreading Team


at https://www.pgdp.net (This file was produced from
images generously made available by the Bibliothèque
nationale de France (BnF/Gallica))

*** START OF THE PROJECT GUTENBERG EBOOK LE


MYSTÈRE DU TIGRE ***
MAURICE MAGRE

LE MYSTÈRE
DU TIGRE
ROMAN

ALBIN MICHEL, ÉDITEUR


PARIS — 22, RUE HUYGHENS, 22. — PARIS
DU MÊME AUTEUR

ROMANS

Priscilla d’Alexandrie. (Albin Michel, éditeur.)


La Luxure de Grenade. (Albin Michel, éditeur.)
L’Appel de la Bête. (Albin Michel, éditeur.)
Le Roman de Confucius. (Fasquelle, éditeur.)
Le Livre des Lotus entr’ouverts. (Fasquelle, éditeur.)
La Vie amoureuse de Messaline. (Flammarion, éditeur.)
Vie des Courtisanes. (La Nouvelle Revue Critique.)
Les Colombes poignardées. (J. Fort, éditeur.)
La Tendre Camarade. (J. Fort, éditeur.)
Il a été tiré de cet ouvrage :

40 exemplaires sur papier de Hollande


numérotés à la presse
de 1 à 40
75 exemplaires sur vergé pur fil Vincent Montgolfier
numérotés à la presse
de 1 à 75
L’édition originale a été tirée sur Alfa Impondérable
des Papeteries Sorel-Moussel.

Droits de traduction, de reproduction, de représentation théâtrale et d’adaptation


cinématographique réservés pour tous pays.

Copyright 1927 by Albin Michel.


PREMIÈRE PARTIE

LA FUMERIE DE SINGAPOUR

Il y a dans le vieux quartier de Singapour une rue avec deux


pentes qui forme la bosse d’un chameau. Au sommet de cette
bosse, parmi les maisons lépreuses écrasées les unes contre les
autres, s’ouvre une porte grossièrement sculptée dont la partie haute
représente un mufle de félin et que l’on appelle à cause de cet
emblème, la porte du Tigre.
Une des deux pentes de la rue descend vers un bassin
abandonné du port où l’on relègue les sampans hors d’usage et les
jonques à demi-mortes. Et à l’endroit où la rue bossuée aboutit au
quai étroit, une pierre aiguë surgit du sol, nommée par la population
chinoise et malaise, la dent du requin.
En vérité, ce ne pouvait être que dans cette rue où tout était à
l’image de l’animal, que moi, le fils de commerçants en bêtes
empaillées, devenu le dompteur intrépide de bêtes vivantes, je
devais voir s’allonger sur mon âme la première ombre de ma
destinée étonnante.
— C’est à la porte du Tigre ! me dit Ali le Macassar, qui connaît
aussi parfaitement les hommes de Singapour que les forêts de
l’Archipel et qui prétend que les uns sont aussi sauvages que les
autres, lorsque je lui demandai de m’indiquer la fumerie d’opium la
plus couleur locale de la ville. Dans le quartier pourri qui enveloppe
d’une couronne de lèpre l’eau du vieux bassin en décomposition, il
n’y avait, d’après Ali le Macassar, qu’un point unique, une seule
porte à franchir, la porte du Tigre.
— La fumerie vaut par l’homme qui la tient, ajouta-t-il. Là, il y a
un homme.
L’homme était un misérable Chinois obséquieux pareil à tous
ceux que je connaissais. Il faillit se rompre en deux pour saluer en
voyant des Européens franchir la porte du Tigre.
Oui, moi, je franchis cette porte, je montai un escalier gluant, je
me mêlai à la plus abjecte racaille de Singapour pour plaire à un sot,
à mon cousin de Goa qui faisait son premier voyage d’affaires dans
les îles et voulait, disait-il, s’instruire en toutes choses, comme si un
sot de naissance peut jamais s’instruire.
Certes, quand j’eus pénétré dans cette salle basse où l’odeur de
l’opium se mêlait à une odeur nauséabonde de sueur humaine, il
était encore temps et j’aurais dû obéir à mon instinct — j’aurais dû
tomber à coups de cravache sur les Malais et les Chinois étendus ;
j’aurais dû les jeter sur la bosse du chameau, j’aurais dû menacer
d’une correction semblable mon cousin le sot. Le risque eût été nul.
Personne n’eût osé se mesurer avec moi. Chacun se serait enfui
dès qu’il m’aurait reconnu.
Or, on m’avait reconnu. Une voix, à mon entrée, prononça :
— C’est Rafaël Graaf, le fameux dompteur.
Et ce fut la nuance d’admiration que je perçus dans ces syllabes
qui atténua ma colère et mon dégoût pour les êtres déchus que
j’étais venu voir. Les chuchotements se turent, je surpris sur les têtes
des fumeurs étendus, quelques légères inclinaisons, quelques
mouvements de paupière marquant la surprise ou le respect et j’allai
docilement me coucher sur une natte à côté d’une petite lampe que
me désigna le propriétaire du lieu. Car c’est la vanité qui dirige
presque toutes nos actions. Puis ces événements devaient se
dérouler, ces personnages devaient apparaître.
Il arrive, lorsqu’on lit un livre, qu’on trouve le sujet résumé dans
quelques lignes au début de l’ouvrage, avec l’indication du mystère
qui occupera l’esprit pendant toute la lecture. De même le hasard
place très souvent au commencement de la vie une scène
synthétique où sont réunis les personnages qui doivent vous
influencer par la suite et où se pose l’énigme qui vous fera vivre et
mourir. Le sot n’était qu’un instrument, la porte du Tigre n’était que le
seuil du chemin, car il fallait que le but fût atteint.
— Est-ce que vous savez que ce sont des moines Bouddhistes
qui ont porté les premiers l’opium en Chine ?
— Je l’ignorais.
— Un traité de morale dont la traduction remonte à la dynastie
des Tang l’affirme. Ce même traité attribue au Bouddha lui-même
l’invention de la pipe et la méthode pour préparer le suc du pavot.
J’éclatai ostensiblement de rire en entendant ces paroles
stupides murmurées non loin de moi et comme celui qui avait parlé
ne semblait pas s’apercevoir de ma gaîté, je soufflai encore avec
bruit et mis sur mon visage une expression de hautain mépris.
Cet homme n’avait jeté sur moi qu’un seul regard clair et profond
où il n’y avait ni curiosité ni respect et il s’était remis à rouler avec un
soin minutieux une boulette brune comme si ma présence non loin
de lui n’avait aucune importance.
La vague clarté de la lampe auprès de laquelle il se trouvait me
permettait de voir ses traits. Il n’était ni Chinois, ni Malais. Peut-être
Hindou. Il s’exprimait en anglais avec un léger accent et un
chantonnement dans la voix. Je trouvai à la réflexion qu’il avait le
type mongol et j’eus envie de lui chercher quelque mauvaise
querelle, d’allonger le pied et de l’en frapper, ou de lancer mon
chapeau sur sa lampe afin de la culbuter.
Mais, à ce moment, mon attention fut distraite. J’eus la sensation
qu’il y avait un visage de femme européenne qui se dressait parfois
au fond de la salle. Je crus entrevoir de grands yeux clairs remplis
d’une allégresse de curiosité et la ligne délicate d’un cou ambré. Une
femme européenne dans ce bouge, était-ce possible ?
L’homme continuait à parler sans s’occuper de moi et je
l’entendis qui disait :
— Les hommes sont d’autant plus malheureux qu’ils éprouvent
plus de haine, d’autant plus heureux qu’ils aiment davantage.
Et, répondant à une parole du personnage qui était en face de lui
et que je n’avais pas entendue, il ajouta :
— Oui, développer en soi l’amour ! Mais c’est difficile. L’opium qui
est l’esprit du règne végétal peut nous y aider. Il y a d’autres plantes
et d’autres secrets et les hommes les ignorent. De même, qu’il y a
plusieurs qualités de pensées, il y a des sucs d’herbes et des
racines avec des propriétés différentes.
Au Mexique, sur la moisissure des pierres, croît la plante peyotl
qui donne la clairvoyance de l’avenir. Dans les forêts du Siam, et là
seulement, on peut trouver une graminée rougeâtre qui procure un
état de transe et aide au dédoublement de l’âme et du corps. Par
l’opium, absorbé avec mesure, l’homme est mis sur la voie où il
découvre sa parenté avec l’espèce animale. Et il y a aussi les
crissements de certains insectes, les chants de certains oiseaux,
comme le rohi-rohi, dans lesquels, si nous savions écouter, nous
pourrions trouver des enseignements, des moyens de nous
développer.
Mon cousin ne fumait pas pour la première fois. Je le vis à
l’habileté avec laquelle il roulait régulièrement en cônes les boulettes
d’opium et à la satisfaction qu’il laissait éclater sur son visage en
lançant au plafond de grandes bouffées de fumée.
Il me tendit une pipe. J’eus un haussement d’épaules pour
exprimer que l’opium ne pourrait exercer aucune action sur mon
robuste tempérament. Mais alors il sourit avec malice et je pensai
qu’il supposait intérieurement que je craignais un effet quelconque
de la drogue sur la netteté de mes idées. Je me hâtai de fumer la
pipe qu’il me tendait. Mes aspirations furent maladroites et le sourire
de mon cousin resta malicieux.
Or, rien n’est irritant comme le sourire d’un sot.
Je voulus montrer qu’un homme de ma trempe n’est pas modifié
par une absorption quelconque et j’invitai mon cousin à me préparer
quelques pipes successives que j’aspirai d’une seule bouffée et dont
je n’éprouvai ni plaisir ni déplaisir.
— J’aime mieux la chasse à l’éléphant dans les forêts de Bornéo,
dis-je.
Je revenais d’un voyage de chasses à Célèbes et à Bornéo et
j’étais passé maître dans l’art d’approcher l’éléphant et de le tirer à
quelques pas.
— Plus l’animal est intelligent et plus il est agréable de le tuer,
ajoutai-je.
Ce fut seulement parce que ma bouche était sèche que je ne
crachai pas dans la direction du Mongol, dont j’avais senti le regard
clair posé sur moi. Je me contentai de me gratter avec force et
d’enrouler ma veste d’alpaga autour de moi pour bien montrer que je
redoutais la vermine qui devait grouiller sur le corps de mes voisins.
Mon cousin ne s’intéressait vraiment qu’aux diverses variétés
d’écailles dont son père faisait commerce à Goa. Je lui énumérai,
malgré cela, un grand nombre de mes exploits cynégétiques, étant
soudain saisi d’une envie de récits, d’un désir d’être écouté avec
admiration en retraçant des aventures dangereuses.
Le temps passa. Je parlais exprès assez haut pour troubler la
tranquillité des autres fumeurs. Quelques-uns se levèrent et sortirent
sans cependant oser laisser voir leur mécontentement. La femme
européenne que j’avais cru apercevoir dans l’obscurité, apparut de
nouveau, ayant sur son visage la même expression de gaîté et de
curiosité. Je faillis plusieurs fois l’interpeller en la priant de venir
s’étendre à côté de moi pour me montrer comment elle était faite.
Mais les idées se pressaient avec abondance dans mon cerveau et
je continuai à parler pour mon cousin qui ne m’écoutait pas.
La notion de l’heure disparut en moi et toute la nuit coula comme
un instant, sous le plafond bas, avec l’odeur épaisse de l’opium,
l’odeur des hommes, et ce je ne sais quoi de poivré, de pourri et de
printanier qui, par la fenêtre entr’ouverte, venait du port.
De ce personnage dont les traits calmes m’étaient
insupportables, je n’entendis plus qu’une phrase et qui me parut
sans importance :
La vieille loi de Manou dit :
Celui qui a tué un chat, un geai bleu, une mangouste ou un
lézard doit se retirer au milieu de la forêt et se consacrer à la vie des
bêtes jusqu’à ce qu’il soit purifié.
Je ne savais pas ce que c’était que la vieille loi de Manou et
d’ailleurs il importait peu.
Mon âme était paisible, il y nageait seulement, comme une
barque sur un lac, la nécessité d’offenser ce fumeur à figure de
mongol.
Or, comme l’air commençait à blanchir par l’approche du matin,
un lézard, un de ces lézards familiers qui hantent les habitations des
hommes, glissa parmi les formes étendues, lentement et sans
frayeur. Il me frôla, puis s’éloigna et je le vis qui tournait autour du
haïssable fumeur.
Mais alors mes oreilles furent choquées par un imperceptible
sifflement. Ce sifflement partait des lèvres de l’homme, et le lézard,
en l’entendant, sans être ébloui par la clarté de la lampe, se
rapprocha de lui et je vis même une main effilée, une main aux
doigts trop longs, dont la forme m’était singulièrement répugnante,
caresser avec une sorte d’amour, la tête du lézard.
La bête charmée fit encore deux ou trois tours, revint se faire
caresser, repartit.
Comme un ressort mon pied se détendit. Il y eut un léger
craquement. La queue du lézard écrasé fit encore deux ou trois
sauts et j’éprouvai la plénitude que donne une action nécessaire que
l’on vient d’accomplir.
Je dus fermer les paupières durant quelques secondes. Quand je
les rouvris, il y avait non loin de moi une lampe entre deux nattes
vides. Le corps du lézard n’était plus au bout de mon talon.
Quelqu’un avait emporté le petit cadavre.
Je me mis à ricaner :
— Cet imbécile l’a peut-être pris pour l’enterrer.
Je secouai mon cousin. Il sortit derrière moi en chancelant. J’eus
la sensation d’un rire clair comme un égrènement de perles qui
résonnait dans l’ombre et je crus encore en franchissant la porte,
voir, sous un sarong malais, le buste d’une femme qui se soulevait.
Mais il était trop tard pour m’en occuper. Je désirais surtout respirer
l’air pur.
Dehors, la fraîcheur était exquise. Une grande voile déchirée
claquait au bas de la bosse du chameau. On entendait au loin, dans
les ruelles, les cris des premiers marchands d’agar-agar. Je m’étirai.
J’aurais voulu me battre avec quelqu’un. Je cinglai l’air avec ma
cravache. Un homme doit toujours avoir une cravache avec lui.
L’opium ne m’avait décidément fait aucun effet. Comme j’étais fort !
Quelle joie j’avais à vivre !
LE COBRA ET LE CRAPAUD

Toujours j’ai passionnément aimé faire souffrir les bêtes. Petit,


j’arrachais les ailes des mouches et je les faisais défiler sur le sable
de la vérandah où je jouais. A dix ans, je m’étais fabriqué un arc
avec des flèches aiguës en bois de sandal dont je criblais les bœufs
et les chiens qui s’enfuyaient à ma vue, comme à la vue d’un
monstre redoutable.
En ce temps-là, l’île de Singapour n’était pas encore entièrement
défrichée comme aujourd’hui, et la forêt y luttait avec les cottages
hâtivement construits, les carrés de terre labourée. C’était sur les
confins des plantations qu’avec quelques gamins de mon âge j’allais
assouvir ma soif de mort animale. Très vite j’avais été habile à tirer
de l’arc. Mais c’est quand mon père me fit cadeau d’un fusil Devisme
que commencèrent mes véritables exploits.
Je venais d’obtenir un prix d’instruction religieuse et le pasteur
qui venait parfois dîner chez nous avait déclaré que si j’étais ignare
en toutes choses, j’avais la connaissance innée de Dieu, ce qui est
l’essentiel. Car j’eus, dès mon jeune âge, un mépris profond pour les
livres et ceux qui les lisent, mépris que j’ai gardé en avançant dans
la vie.
L’expérience m’a enseigné qu’il n’y a d’hommes intelligents et
utiles que ceux qui sont rebelles à l’instruction et tournent toutes
leurs facultés vers l’action.
Je me flatte d’avoir jeté aux ordures, à part une Bible que je n’ai
d’ailleurs jamais ouverte, les quelques ouvrages anglais et portugais
qui traînaient dans notre maison. Ne jamais rien lire ! Quelle force
puissante pour le caractère ! J’empêchais mes employés d’aller
chercher « le Courrier de Malacca » quand il arrivait le dimanche, et
pour ma part, je me faisais raconter, oralement, les événements
historiques, notamment, la révolte des Cipayes de 1857, pour ne pas
risquer d’être influencé moi-même par la stupidité de ceux qui
écrivent.
Le fusil Devisme fit merveille. Je tirais les oiseaux au vol et je
cassais à cent pas la tête des serpents. Je reçus les enseignements
des meilleurs chasseurs de Singapour qui ne connaissaient rien à la
chasse, je m’en aperçus plus tard — et à quinze ans je me mettais à
l’affût avec eux pour ma première chasse au tigre.
Je peux dire que je suis un des hommes les plus courageux de
tous ceux qu’il m’a été donné de connaître. On reconnaît un homme
courageux à la capacité qu’il a d’avouer les peurs qu’il a éprouvées.
J’ai eu peur, certes, mais je l’ai dit, je l’ai dit hautement, sinon aux
autres, ce qui m’aurait nui, du moins à moi-même, ce qui est
l’important. Par cette connaissance de ma propre peur, je suis
devenu courageux et j’ai accompli les exploits qui m’ont rendu
célèbre de Bornéo jusqu’aux côtes de Coromandel et même plus
loin.
C’était le moment où les tigres commençaient à diminuer dans
l’île de Singapour. Le Résident organisait perpétuellement des
chasses et comme il était l’ami de mon père, j’y étais convié et j’en
devins même l’acteur principal. Je me souviens que lorsque le navire
de guerre français l’Amazone fit escale dans le port, il fut convenu,
durant un dîner, que chaque officier tirerait son tigre et que ce serait
moi, malgré ma jeunesse, qui réglerais toutes ces chasses.
Tout cela n’a aucune importance et je ne le dis que pour mémoire
et afin de faire connaître mon extraordinaire précocité de tueur
d’animaux sauvages. Je me hâte d’ajouter que les officiers français
quittèrent Singapour sans avoir pu tirer un coup de fusil et que ce ne
fut qu’un an plus tard qu’il me fut donné de tuer mon premier tigre.
Car ces créatures ont un si prodigieux pouvoir de se dissimuler, que
même dans les lieux où elles abondent, comme Malacca et Java, on
peut les chasser très longtemps, sans jamais les rencontrer, quitte à
se trouver, un soir, face à face avec elles, au moment où l’on s’y
attend le moins. Mais je dirai, ailleurs, les mœurs de ces êtres
mystérieux et féroces et quel enseignement je tirai de cette
connaissance.
Les premiers tigres que je vis étaient empaillés, dans les
magasins de mon père. Il y en avait de toutes tailles et de toutes
provenances. Il y avait les tigres noirs de l’Himalaya que l’on appelle
noirs, bien qu’ils soient plus jaunes que les autres, parce qu’ils sont
censés être les incarnations d’une sorte de déesse hindoue qui, elle,
est noire et que l’on appelle, je crois, Kali.
Il y avait ceux du Bengale qui ont sur la queue quinze anneaux
noirs, exactement, sur fond blanchâtre, et ceux de la Mongolie qui
ont douze anneaux noirs exactement, sur fond jaunâtre.
Il y avait ceux du Siam qui ont la gueule allongée, ceux de
Malacca qui sont gigantesques et ceux qui viennent de Zanzibar et
qui sont ridiculement petits parce que ce ne sont pas des tigres,
mais de simples panthères déguisées en tigres.
Il y avait aussi toutes sortes d’animaux sauvages, des crocodiles,
des serpents, des lions de Perse, des hyènes, parfois un fourmilier,
et toutes les variétés d’oiseaux de proie de l’Asie. Ils occupaient une
immense galerie vitrée adossée à notre habitation et qui donnait sur
les jardins.
Je regardais souvent leurs silhouettes quand je jouais et je me
souviens qu’une force intérieure m’obligeait à me glisser dans la
galerie pour arracher une plume de ci de là, piquer un mufle avec un
bâton pointu, tirer une oreille, injurier l’ennemi impuissant.
Le propriétaire d’une ménagerie qui devait de l’argent à mon
père, mourut insolvable, et celui-ci hérita de ses animaux et de son
matériel.
Pendant que dura la procédure, il dépensa pour la nourriture des
fauves et d’un jeune éléphant savant beaucoup plus d’argent que la
valeur de sa créance. Le désir de retrouver les sommes avancées lui
donna l’idée d’adjoindre à son commerce de peaux et de bêtes
empaillées un commerce de bêtes vivantes. Le commencement de
sa grande fortune date de là. Il fit installer dans ses immenses
jardins qui s’étendaient en bordure du quartier chinois une série de
cages devant lesquelles, deux fois par an, quand arrivaient les
bateaux de Macao et de Shanghaï, défilaient les grands marchands
chinois fournisseurs des ménageries de la Chine. Car ce peuple qui
semble au premier abord purement commerçant et borné dans ses
conceptions a une curiosité extraordinaire pour toutes les espèces
animales et je crois que les collections zoologiques les plus
curieuses de l’univers se trouvent chez certains riches Mandarins de
Canton et de Pékin. Je note en passant que les plus grands succès
que j’ai obtenus dans mes exhibitions d’animaux sauvages furent
dans ces dernières villes et cela a contribué grandement à me
prouver l’intelligence des Chinois que j’avais d’abord méconnus.
Il fallut bientôt transformer complètement les jardins. Outre les
cages, il y eut des volières, des cahutes, des gourbis, des fosses,
des hangars, des étables, des écuries, des habitations sur pilotis
dans un étang artificiel, des bassins entourés de treillis pour les
sauriens. On construisit une sorte de ville, avec ses rues et ses
remparts, ses perchoirs et ses colombiers où les habitations étaient
aménagées selon le caractère et les mœurs des différents habitants,
mammifères, pachydermes, solipèdes, plantigrades, bimanes,
ruminants, herbivores et carnassiers.
C’est peu après ces transformations que se placent les
événements terribles qui contribuèrent à augmenter ma haine pour
ces bêtes dont la vie et la mort m’enrichissaient.
Ma mère était une sainte. Toutes les mères sont des saintes en
principe, mais je crois que ma mère l’était plus que les autres. Elle
était Portugaise aussi et s’était fait enlever très jeune par un
capitaine au long cours. Ce capitaine, un certain Pinto qui lui donnait
les marques du plus vif amour l’installa à Singapour dans une
ravissante villa du quartier anglais et s’en alla faire certaines
livraisons de cargaisons à Batavia et à Madras. Il ne revint jamais et
jamais ma mère n’en entendit parler. Au bout d’une année,
désespérée et sans argent, elle se demandait ce qu’elle allait
devenir, quand elle rencontra mon père et l’épousa. Elle connut avec
lui un bonheur parfait, mais elle ne put jamais se défendre d’une
admiration naïve pour ce Pinto si mystérieusement disparu. Elle me
raconta souvent des histoires pleines de fantaisie qu’elle tenait de lui
et à son insu elle me communiqua son admiration.
Quand je fus plus grand et que je fus en état de comprendre les
choses, mon admiration se changea en colère pour le séducteur qui
avait osé prendre une jeune fille à Lisbonne et la déposer à
Singapour, sans plus se soucier d’elle. J’aurais voulu le rencontrer et
lui dire son fait. Mais ma mère, dans la pureté de son âme
angélique, ne lui conservait aucune rancune.
La sainteté de ma mère s’exprimait physiquement par une facilité
extrême à rougir. Elle avait conservé un teint de peau extrêmement
clair qui se colorait en rose si on lui adressait la parole un peu
brusquement.
Cette facilité à rougir ne contribua pas peu à augmenter le grand
amour filial que je portais à ma mère. J’ai toujours considéré cette
particularité sanguine comme le signe extérieur d’une noble
élévation de sentiments, ce qui distingue l’élite vraie. Ce signe est,
du reste, bien gênant pour celui qui le porte. Je l’ai reçu de ma mère,
et malgré la trempe puissante de mon âme, malgré les soleils
asiatiques qui ont brûlé ma peau, il suffit souvent d’une parole
inattendue pour faire monter le sang à mon visage.
Ma mère, dans sa sainteté, souffrait de ne pas assez participer
aux charges du métier de son mari. Elle voulut jouer un rôle dans
l’éducation des animaux, et c’est ce qui la perdit, car on est perdu
par sa vertu avec autant de sûreté que par sa folie.
Un Malais nous ayant apporté un crapaud de dimensions
inusitées, qui venait de l’île Komodo, celle où l’on trouve toutes les
espèces monstrueuses, ma mère, dans sa bonté, se mit en tête de
l’apprivoiser.
Comme l’on fait d’ordinaire, elle commença par l’affamer et elle
l’enferma dans un vase étroit et long. J’avais toujours entendu parler
d’une sorte de projection haineuse qui part des yeux des crapauds
dans certains cas, mais je n’y avais pas cru. Ma mère en fut la
victime. Elle alla voir, au bout de trois jours, ce que devenait le
crapaud au fond de son vase étroit. Ni mon père, ni moi, n’étions là.
C’est une jeune servante malaise qui raconta ce qui était arrivé.
A peine ma mère s’était-elle penchée sur le vase que son visage
si pur refléta une expression d’horreur indicible. Tout son corps se
mit à trembler. Elle regardait fixement le crapaud comme si elle ne
pouvait détacher ses yeux de lui. La jeune Malaise accourut et fut
obligée de la tirer de toutes ses forces par derrière pour l’arracher à
sa contemplation. Elle mourut quelques minutes après sans avoir pu
prononcer une parole.
Il est à noter que des singes gibbons, qui remplissaient une cage
voisine, se mirent à jacasser de façon effroyable et à regarder
avidement dans l’espace comme s’il y avait un spectacle invisible.
Il est à noter aussi que le crapaud mourut en même temps que
ma mère.
Notre désespoir fut immense. Ni mon père, ni moi, nous ne
crûmes d’abord que le crapaud pouvait être pour quelque chose
dans cette mort inexplicable. Mais M. Muhcin, vieux marchand
bouddhiste, d’une honnêteté légendaire à Singapour, et d’une
sagesse reconnue, qui fréquentait notre maison, nous affirma que
les crapauds, quand ils ont atteint un point de fureur extrême
peuvent transmettre la mort par le regard, surtout quand il s’agit
d’une créature délicate et sans défense comme ma mère.
Pas tous les crapauds, ajouta-t-il. Car il y a des hiérarchies chez
les animaux comme chez les hommes. Il y a ceux qui commandent,
ceux qui obéissent, ceux qui ont pénétré certains secrets de la
nature et ceux qui les ignorent.
Et il se lança dans une théorie que je trouvai absurde à ce
moment-là et qui concluait presque à glorifier le crapaud meurtrier.
Je restai seulement convaincu qu’il y a dans la nature des choses
occultes qui dépassent le cerveau de l’homme et auxquelles il vaut
mieux ne pas penser.
Ma mère était catholique et mon père était protestant en sorte
que nous recevions également le pasteur, les jésuites français de la

You might also like