Tha Devalopmant of the Arrhenius Equation
Keith J. Laidler
Ottawa-Carleton Institute for Research and Graduate Studies in Chemistry, University of Ottawa,
Ottawa, Ontario, Canada K 1N 6B4
Today the influence of temperature on the rates of chemical
reactions is almost always interpreted in terms of what is now
known as the Arrhenius equation. According te this, a rate
constant £ is the product of a pre- exponential (“frequency”)
factor A and an exponential term
# = A—e ” ^” (I)
where R ic the gas constant andñ is the activation energy. The
apparent activation energy Supp is now de firied (1 j in terms
of this equation as
(2)
Often A and ñ can be treated as temperature independent.
For the analysis of more precice rate-temperature data, par-
ticularly those covering a wide temperature range, it is
usual to allow A to be proportional to T raised to a power
m, so that the equation becomes
ñ = A’T“ e —”^” (3)
where now A’ is temperature independent. Recently Gardiner
(£) hac shown that eqn. 13) applies satisfactorily, and better
than eqn. (1), te a number of reactions; he has also discussed
reasons for the failure of eqn. (1). In modern compilations (3)
the procedure often employed is to use eqn. (1) for data of
lower precision or where the temperature range is limited, and
to analyze more precise data in terms of eqn. (3).
Although the temperature dependence of rates is now al-
most universally interpreted in terms of these equations, for
a period of over 60 years (ca. 1850 to ca. 1910), in spite of
considerable experimental effort, there was much uncertainty
and controversy. Mellor’s famous textbook of kinetics (4)
published in 1904 quotes Ostwald as saying that temperature
dependence “is one of the darkest chapters in chemical me-
chanics.” In van’t Hoff’s pioneering textbook of physical
chemistry (5) published in t898 the evidence was reviewed
with respect to a number of empirical relationships between
/z and T, and the author concluded his survey by saying that
“It is so far impossible to choose between the above equa-
tions.”
Some of these early struggles with the problem have re-
cently been reviewed by King (6, 7), who in a very interesting
way has covered both scientific and historical aspects. Besides
being of considerable historical interest, a study of the early
work on temperature dependence provides us with consider-
able insight as to how scientific progress is made not merely
from an analysis of data but from theoretical considerations
of broad applicability.
The Empirical Approach: A Brief Histor ical Survey
W ilhelmy (8) appears to have been the first to propose,
in 1850, an equation relating the rate constant of a reaction
to the temperature. He used the Celsius temperature 0, and
his equation can be written as1
494 Journal of Chemical Education
# = AF” (I + C 6) (4)
In terms ef the absolute temperature T, the equation can be
written as
(5)
In 1862 Betthelot (9) presented the equation
in \ = A’ + IJ’f or k = A eh " (6)
A considerable number of workers supported this equation,
which was taken seriously at least until 1908 (J 0).
In 1881 Warder (i i , J2) proposed the relationship
l« + k 1(b — 1’) = c (7)
where a, 6, and c are constants, and later Mellor (4) pointed
out that if one multiplies out this fr›rmula, expands the
factor in T and accepts only the first term, the result is
k = a’ + b’T" (8)
Equation (7) received some support from results obtained by
Urech (28—JS) and by Reicher (J6, 17), but it did not prove
at all popular.
In 1883 Schwab (i 8), working in van’t Hoff's
laboratories in Amsterdam, proposed the relationship
In /z - A’ — + DT ‹› # — A ‘ ^ —””"'” (9)
—e
and demonstrated that it applied to the conversion of dibro-
mosuccinic acid into bromomaleic acid and to the reaction
between hydroxide ions and monochloroacetate ions. In the
following year van’t Hoff's famous textbook on chemical
dynamics (7 9) appeared, and it contained a discussion of
the temperature dependence of equilibrium constants in
terms of an equation that we can now express in the form
bU°
(10)
bT AT2
where K is the concentration equilibrium constant and k U°
the standard internal energy change. Van’t Holt pointed out
that since K is the ratio of rate constants i and /z — i r re-
action in forward and reverse directions, these rate constants
are also expected to obey an equation of the same form
E
bT
where E is an energy change that relates to the particular
re- action.
Van’t Hoff acknowledged having received inspiration for
these ideas from a paper by Pfaundler (20), whe as early as
’ To avoid confusion, the early equations have all been converted to
a consistent notation, and natural logar ithms have been used
throughout. Gonstants such as A. A’, B, c, a, r, o, e, b, a' , and 6' are
temperature independent.
1867 had given a qualitative discuesion of equilibrium and large to be explicable on the basis of how temperature
rates in terms of molecular motions and Maxwell’s law of affects the molecular translational energies, or of the
distribution of molecular speeds. Pfaundler saw clearly that temperature dependence of the viscosity of the medium. He
when chemical equilibrium is established, forward and reverse concluded that the explanation must be that an equilibrium
reactions are occurring at equal rates. Maxwell’s law had is established between normal and active reactant
led to the conclusion that the fraction of molecules having molecules, and that this equilibrium shifts in the manner
energy greater than a critical value £ is equal to predicted by van’t HofPs equation (eqn. (10)). He did not
exp(—E/RT) and can increase markedly with temperature. concern himself with a pos- sible temperature dependence
Pfaundler suggested that only those molecules possessing of this energy difference; in fact he tacitly assumed no
more than a critical energy E could undergo chemical temperature dependence and therefore favored the simple
change. Pfaundler’s work is now largely forgotten, but he eqn. (1).
was responsible for important con- cepts that are usually An equation in which the pre-exponential factor also has
attributed to van’t Hoff and Arrhe- nius. a temperature dependence was first proposed in 1893 by
Van’t Hoff did not assume AU° and £ to be temperature Kooij (29), whose equation may be expressed as
independent; he also considered the possibility that Aft°
and
£ might involve a temperature independent term and in ad-
dition a negative term proportional to U. He thus This equation was supported by Trautz (80) in 1909. As
considered two possible forms for the temperature pre- viously noted, this equation (equivalent to eqn. (3)) is
dependence of h, eqn. perhaps the most satisfactory of the equations; data that do
(9) and also the simpler relationship not fit the simple eqn. (1) usually fit eqn. (13) very
B precisely.
—
OF (t2) In 1865—67 Harcourt and Esson (8J—8J) made
pioneering contributions to chemical kinetics, being the
We have seen that eqn. (9) wae supported by the work of first to relate reactant and product concentrations to the
Schwab (18); it was also supported by work of Spohr (2J), time by means of equations that they obtained by
van’t Hoff and Reiches (2P), Buchb0eh (28), and integrating the differential equations for rates of reaction.
Wegscheider (24). Some 30 years later they col- laborated again on a very
In the meantime Hood was making careful kinetic long paper (34) presented as a Bakerian Lecture to the
studies of a number of reactions in 9o1ution, and in a paper Royal Society. In this publication they gave very precise
published in 1885 (25) he interpreted his results in terms of results for the reaction between hy- drogen peroxide and
Berthelot’s formula, eqn. (6). This equation remained hydrogen iodide, obtained from 0°C to 50°C at intervals of
popular for another 20 or more years; a number of papers about 5°. They presented a very detailed analysis of the
supported it, including those of Tammann (P6), Remsen temperature dependence and concluded that it was best
and Reid (27), and Perman and Greaves (i0). represented by equations of the form
A very important contribution to the problem was made
in In # = A’ -r C In T 'or # = AT* (14)
1889 by Arrhenius (28), whose approach was rather different This type of relationship did not receive much
encourage- from that of the other workers. Starting with van’t Holt’s ment from others, but it was supported by some work
of Veley equation (eqn. (11)) Arrhenius pointed out that the magni- (35) on the affinity constants of bases. One
particularly in- tudes of the temperature effects on rates are usually much too teresting feature of Harcourt and Esson’s work
is that they
Table 1. Summary of Temqerature•Dependenoe Equatiozss
Differential Expression Eqn. Number Refer-
Form Integrated Form for k in Text Supported by
d In k _ B -I- GT +
—B + 15 van’I Hoff, 1898
dF " W T Bodenstein, 1899
d In £ _ 8 + CT 30
13 Kooij, 1893
dT " T' Trautz, 1909 IB.
d In /r _ B + OF
In /r = #‘ - - -I- DT 9 Schwab, 1883
âT " W 7
van't Hoff, 1884
Spohr. 1888
van't Hoff and Reiches. 1889
Buchbock. 1897
Wegscheider, 1899
d ln k G£ + OF
In â - A’ + C. In T + DY 16 is
28
C=Ae—” 12 van’t Half, 1884 29
dT H
Arrhenius, 1889
Kooij, 1893 34
d In k G 35
k=XW 14 Harcourt and Esson, 1895
ar ‘T 36
Veley, 1908
Harcourt and Esson, J912 fi'5
d In k _ 3T
Berthelot, 1862 38
dI"
Hood, 1885
Spring, 1887 40
Veley. 1889
Hecht and Conrad, 1889
Pendelbury and Seward, 1009
Tammann, 1897
Remsen and Reid, 1899
Bugarszk y, 1904
Perman and Greaves, 1908
Volume 61 Number 6 June J984 495
 made no assumpt,ir›n as to the absolute zero; they represented
the absolute temperature as + #, where # is the centigrade
(Celsius) temperature, and found to be 272.6°, in very good
agreement with the modern value of 27'1.15°. They had thus
established, quite independently of' thermodynamics, a kinet.ic
absolute zero of temperature, at which chemical reaction
would cease.
Most other workers, however, favored different tempera-
ture-dependence equations, and in 1912 Harcourt and Esson
returned to the fray with another lengthy paper (’36 ), in
which they analyzed many ‹›f the results obtained by others.
They claimed that eqn. (14) interprets the temperature
dependence better than the rival equations, and for some
reactions this claim is reasonable. For several reactions,
hr›wever , agreement can be obtained only if C is allowed to
be significantly tem- perature dependent—it sometimes
decreases with increasing T , but more often increases. To
explain deviations from their equation, they suggested that
“the conditions for a stable communication of heat energy
to chemical energy are not se- cured.” Harcourt and Esson’s
two papers make no reference at all to Arr henius’ ideas,
and make only a disparaging refer- ence to the arguments of
van’t Hoff.
In his 1898 book, van'I Hoff IN) pointed out that most of the
previously presented equations were special cases of the
equations
In \ - A’ — + L’ Tn T + DT or k A'' -e ' ^ -“”'*”'
(15)
They can be arrived at by dropping one or two of the constants
fi , C, or N . ’1’his is illustrated, and the equations and refer-
ences summarized, in Table 1. We will refer to the general
equation (eqn. 15) as a three-parameter equation, since three
parameters P, C, and D are concerned with the temperature
dependence and remain when the equation is put into its
differential form. Dropping one of the three constants leads
to three two-parameter equati‹,ns, eqns. (1.3) and (9) and a
third equation
which seems to have received no consideration or supp‹›rt.
Dropping two of the constants leads tr› the th ree one-param-
eter equations, eqns. (12), (14), and (6).
In the following year (1899) Bodenstein (42) published the
results of very careful measurements, over a wide temperature
range (283°—508° C), on the gas phase reaction between hy
drogen and iodine, on the reverse decompositi‹›n of hydrogen
iodide, and on the equilibrium H; + I_ 2HI. He analyzed
his results in terms of the three -parameter eqn. (15) and ob-
tained very gocd agreement. However, from the modern point
of view his analysis would hardly be considered satisfactory.
For example, for the combination of hydrogen and iodine he
fitted his temperature data to the formula
(note the extravagant use of significant figures). This means
that the preexponential factor varies with temperature ac-
cording to T ' , which now seems to be an unreasonably
strong dependence. Since his pre-exponential factor decreases
strongly with increasing temperature, Bodenstein’s parameter
BR corresponds to an activation energy (181.7 kJ mol ' ) that
is significantly higher than the presently accepted E zpg value
(3b , ‹') of I 71.4 kJ mol '
In 1908—1909 Trautz and Vo}kmann (4.? , 44) also
analyzed data of others on the basis of various formulae,
including the three-parameter eqn. (15). Their analysis also
leads tu ab- normally large negative temperature dependencies
of the pre-exponential t'actors, and therefore to apparent
activation energies that are t‹›o high according to modern
interpreta- tions.
496 Journal of Chemical Education
I t is a curious hist‹iri‹'al fact that in spite of the controversy
that had been raging f'or some 60 years, by the second decade
of the present century all f‹irmulae except eqns. (1) and (3)
had been q uiet ly dropped. For example, the inn portant
papers of Trautz (4.5) and Lewis (4ñ) on the significance of
the pre- exponent,ia1 factor use ‹in ly the simple Arr henius
equati‹ n (eqn. (I )). As far as the Harc‹›urt-Esson equation is
concerned, I am nt›t aware of any papers that qur›te it except
as an his- torical anomaly (see, for example, ref. (47)).
The most important reason f'or the acceptance of the
Arr henius equations is that all ot the other relationships are
theoretically sterile. The parameter B in eqn. (12) is related
to an energy E which can be related to the height of an energy
barrier for the reaction. The parameters C’ and D that appear
in the Harcour t-Esson and Berthelot formulations, on the
other hand, cannot easily be related to any meaningful
physical quantity. The undoub ted fact that, for cert.ain re-
actions, the Harcourt-Esson equation libs the data better than
other equations is not a compelling argument in view of the
fact that it does n ot lead to any understanding of the way in
which a chemical reaction r›ccurs.
Comparison of lhe Empirical Equations
At first sight it appears surprising that equations as widely
different as some of those listed in Table 1 can all give a rea
sonably good bit to the same experimental data. We may first
focus attention on the three one-parameter eqns. (12), (14),
and (6). A tit t.o these separate equations is confirmed if the
following plots are linear:
eqn. (12): ln k against 1/£
eqn. {] 4); lr k against ln /
eqn. ( 8): ln k against T
The reason that all ct these plots can give reasonably good
linear fits with the same data is that over the narrow tem-
perature ranges usually em ployed in kinetic studies, 1/'r, ln
T, and T are more ter less linearly related to each other. For
practical reast›ns the temperature range of a kinetic study is
often ‹›nly about 40° —50°, and it is sometimes even less.
Moreover, the temperature range is often somewhere around
room temperature and is t'requently much higher. For exam-
ple, if a study is carried tuft from 0°C to 50°C, the range is
275 K to 323 K, and the variati‹›n in absolute temperature is
therefore only about 20*r.
Figure 1 shows ln T and T plotted against 1/1', for the
temperature range from 0°C try 100°C. The plot ot In T against
I IT shows little curvature, and if the range were only 40° or
so, as in many experiments, the curvature would be hardly
detectable. If we espand ln T about the mid point of the range
('323 K) and accept only t he first term, the result is
In t I/K) = 4.78 + 3.1 0 'x J0 "(T/K) {18)
Over the 0--100° C range this equation represents the rela-
tionship with an err‹ r of less than 0.2'7.
The plot of 7’ against 1/T shows a little more curvature. In
reality such a plot is a hyperbola (see inset in Fig. 1), but in
the 0- 100° C range one is at the far reaches of the hyperbola
where there is little ctir vat ure. An expansion i›I 1 /T about
the mid point of the range leads to
j
-619x10—"-9f9x10—"(T/K) (19)
Over the 0—100° C range this equation is valid to within 3%.
It follows that if In k is plotted against any one of the three
functions 1/T, In T, and ’F and a straight line is obtained, the
plot wi11 be almost linear if either of the other two functions
is employed. This is illustrated in Figure 2 for the very reliable
data of Harcourt and Esson (34) on the reaction between
hydrogen peroxide and hydrogen iodide, the temperature
range being 0°C to 50° C. The open circles are for a plot of In

08-
8 6 8.8 3.O 3. 8 3 4 3.6
Figure 1. Plots of In 17/ II and T/ K against 1/(TI fi), lor \he temperature range
0”—100°C. The inset shows a plot of T/ K against 1/{ 7/ for the range 0 K to
10 K.
Ta6}e 2. RWatJonsfiip between tfie ArrfienTus Activat*an Energy
£( /3'Z*d Tn mid F) and the Temperature Parameters
Equation Activation energy E
15
13
16
12
14 RCT RDTv4'
fi against In T, corresponding to the t'ormula favored by
Harcourt and Esson (eqn. (14) ), while the t'ilIed circles are for
an Arrhenius plot of ln fi against 1/T. It is certainly true, as
Harcourt and Esson concluded, that the In # against In T plot
is more linear than the other, so that their preference for eqn.
14) is justified or
The slope of the plot /z against In Z' leads to a vaIue of
20.4 1'or Harcourt and Esson ’s parameter C; i.e., their tem-
perature dependence is expressed by
The Arrhenius parameter N in eqn. (12) has an average value
of’ 6.05 X 10* K, corresp‹›nding to an activation energy of 50.3
kJ mol I . Table 2 shows the relationships between the
Arr henius activation energy defined by eqn. (2), and the pa-
rameters in the various equations. In terms of the Harcourt
and Essun parameter C, the activation energy is RR"F and will
therefore range from 46.3 kJ mol ' at 0°C to 54.8 kJ mol i at
50° G. This, however, as shown by the plot in Figure 2, does not
correspond to a very large curvature, and might easily have
been overlooked if the data had been less precise.
Most kinetic data can be interpreted almost equally well
by any rat“ the three one-parameter equations. This being so,
the introduction of' an additional parameter into any ct these
equations will probably give a completely reliable fit; thus, the
*! 32 33 34 3.5 3.6
-&20
- 6.C
-6?
5.0
- 880
I n ( T/ K)
Figure 2. Plots for the data of Harcourt and Esson (34) for the reaction
— 260
between hydrogen perox ide and hydrogen iodide. The units of the rate
3.8 constant are not stated but are probably dm’ mol— ’ min— . The natural
logarithm of the va/ue k" of the rate constant is plotted against In (7//
(open circles) and against 1/(7/K} (filled circles).
three two-parameter equations, eqns. (13), (9), and (16),
will be equally reliable. The three-parameter eqn. (15) will
hardly ever be significantly better than any of the two-
parameter equations; indeed, as we have noted with
reference to Bo- denstein ’s analysis, the three-parameter
equation may be
disadvantageous in that there may be a variety of assignments
of parameters that will lead to equally good agreement with
experiment.
lt is to be emphasized that finding out which equation
gives the best fit It› experimental data is not neeessarily the
most satisfactory way of' mak ing scientific progress. A
significant point is that certain plots are inherently m ore li
kely to be linear than certain other plots. The more often
one takes logarithms, the more linear do plots become. This
is well il-
lustrated by an equation of the form xy — constant. A simple
pltit of J' against z gives a hyperbola, whereas a plot of ln
against ln i gives a straigh t line; plots of In y against z , or of
y against ln z , give lines uf m uch less curvature than the hy-
per bola.
Harcourt arid Esson’s plots ct' I n £ against In T are inher-
ently more likely to be linear than pltits of In i against F or
1/T, liecause the variation in In T ‹›ver any range is consider-
ably less than that ot' ?’ or of 1/T. Thus in the 0 to 50°C range,
T and I /7’ vary IU' 17%, whereas ln T only varies by 3‘fo.
Harcourt and Esson ’s work is nevertheless of considerable
interest and signiI'icance. Perhaps their most important
contribution was to demonstrate the kinetic absolute zero.
Although their equation was not the most fruitful one, it did
represent the temperature dependence very well for certain
reactions, including the one they studied in detail. The
Arr henius eq nations (eq ns. (1) and (3)) give zero rate at the
absolute zero because as T • 0, —E!R T - and theref'ore exp(—
ñ/ft T) approaches zero. In the same way ln T • — as
F 0; therefore according to the Harcourt-Esson formula,
as "/’ . In K — and therel'ore /i approaches zero. Thus,
although their formula was not a fruitful one, in that it did not
lead to a useful theoretical concept of a chemical reaction, it
did lead to their important discovery ‹af the kinetic absolute
zero.
Volume 61 Number 6 June 1984 497
 Further Progress
After it had been recognized, in the second decade of the
present century, that eqns. (1) and (3) were the most satis-
factory, progress in kinetics was much more rapid. During the
last century chemists had been largely content with factual
information about the products and speeds of a chemical re-
action; the question of' the mechanisms of reactions was
hardly considered. Until the end of the nineteenth century
physical chemistry was still in its infancy, having made its
most im- portant advances in the treatment of solutions of
electrolytes. Ostwald and Arrhenius were the pioneers in this
field, and Harcourt played a similar pioneering role in
kinetics; in his lectures he was constantly emphasizing the
importance of finding cut how and why c hemical change
occurs (4B, 49). lt is unfortunate that, because he supported a
sterile treatment of temperature dependence, he was himself
unable to bring this laudable aim to a successful conclusion.
From the beginning ot the second decade of this century
one can discern three distinct lines of development in the
theory of kinetics, all based on the Arrhenius equation and all
coming together in 193fi with the formulation of transition-
state theory ISO, 62 ).
I) In t9t I Xch ns tamm and ScheFler (52), in a paper that nuw
seems years ahead of its time, attempted a thermodynamic
formulation of reaction rates. 'This work was further
developed by Brandsma and Scheffer (53—55).
2) In 1914 Marce lin (56, 57) treated a chemical reaction in terms
of' motion over what we would now call a “potential-energy
surface.” The dynamics of such motion was further treated by
Eyring and Polanyi (58 ) and by ke lzer and Wigner (59 j.
I n 1916 Trautz (SP), and in 1918 Lewis (67 ) developed a treat-
meit of the pre-exponential factor A in terms of the kinetic
theo ry ot gases. This treatment was supported and developed
by many work ers, particularly by H inshelwood (62) and Mo-
elwyn -Hughes (fiJ) .
A number of papers attempted to bridge the gap between
these rather different approaches. La Mer (64), for
example, combined the kinetic theories of collisions with
the thermo- dynamic formulations, and represented the
preexponential factor as the pr‹iduct of a collision
frequency and a term es p(kS t/Q), where AS* is the
entropy of activation.
It was in 1935 that the most general and comprehensive
treatment of reaction rates was formulated, independently
by Eyring (50) and by Evans and Polanyi (57 ). This theory,
now generally called tsarist tion-st etc th cor y , has been ex-
tensively applied to a wide variety of physical and chemical
processes (65, 6fi). Pacey (6‘7) has recently discussed the re-
lationships, which are sometimes not entirely straightforward
between the Arrhenius activation energy and the height of an
energy barrier.
Subsequent treatments of rates, such as those concerned
with the dynamics ot' motion over a potential-energy
surface, have been of' great value, but they are much more
complicated than transition-state theory and are less
general in their ap- plication. The results of calculations on
the basis of these treatments are to a good appr‹›ximation
consistent with transition-state thet›ry and with the
Arrhenius equation.
Acknowledgmenl
The author's best thanks are due to M. Christine King,
for very valuable discussions and suggestions regarding the
subject matter of this paper. After this paper was prepared
S. R. I•ogan’s very interesting article (68 ) on a related topic
appeared; the author is grateful to Logan for valuable sug-
gestions, some of' which have been used in revising the
present paper.
Literature Cited
t1) "Man ua1 of Sy m bols and ’l’ermin ology fvr Ph ysicocln emical Qu a ntities and Units;
Appendix V, Symbolism and Terminology in Chemica) 1inetics,” (rmommendations
prepared by ' J. I a'dle r ar'd •!*° published *n *•e Appl Chem., 53. 753
t2) Gardiner , W. C., Hcc. Chem. Yes., 1It, 826 (I 977).
(3) I‘"or example, (a) Bauloh, D. I.., Drysda)e, D. D.. add Home, D. G., “f4igh
Temperature Reaction Rate Data," J.Tniv. of Leeds, Vols. I- 5, l96g-70j (b)
Kondtatiex, y. g., "@@ Consorts of Gas Phase Reactions,” Office of Standard
Reference Data, National Bureau of S tanclarfis, Washington, t)C, 1972; tej K era,
J. A., and Moss, S. J., 'ccc Handbook of B'mo1ecu1ar and Termolecular Gas
Reactions," CRC php, Boca ypt,
t4) Mellor, J. W., “Chcmi eat 8 tatics and Dhmm ics,’' Lung ma #„ Green Jp. , 3#pd #p,
1904.
(6) Van 't Hoff, J. H., " Lectu res on Theoretical and Physical C5 # yittty, putt , pgt
yi##t by mm Pcs,” Edward Arnold, Londo», i8g8.
(6) King, M. C., km 6iz, 28, 70 (t98l).
‹Y Xing, M. C., AmF z, 29, 49 (1982).
t6) WU helmy, L., Hogg. Ann .. 81, 422, 499 (1880).
{9) Be rtheloa, M., hon. C6‹m. Phys., 68 1 sl , I l0 (1862).
(10J Pe r man, E. I*., and Greaves, R. H., free. ftod Snc., A , 80, 3G3 ( 9gg).
t1i} Warder, R. B., Amer When d. , 3, 808 (1881}.
T12) Wander, R. B., Ber. , t't, 136a 1861); see also Warden, R. B., 7. PTiy
(T896—’T); Mr ienr e , 1, ’T4t (188a).
'men., I, t49
(I 3) Urecfi, F'., 8er , l6, ’762 (T BB:3).
(t4) Ureoh, fl., Ber., 20, Z 8fi6 T887).
(Ifi) NJ tech, 5., R‹r , TO, I8fi6 ( l887).
{17} Reieher, L. T., T‹eGig 's A rim. , 232. 103 tl88f›) .
118) Scbwab, L. C., 1)issertation, Univ. ot Amsterdam, 1883.
( I9) Van’t Hoff, J. H., “Etudes de dynamique ehimi9 ue,” F. Muller and Co., Artste raam,
188't.
(20} f’faund ter, L., Ann flys. Chem., 181, 55 t1867).
t$'2J Van t Hoff, J H., and Reicfier, L. T., Z pJiy Chem. , 3. 482 t1889) .
{83 j Buch 6ñc£, G., Z ph y . Chefn ., 2$, 123 t1897 J.
(24 J Wegscheider, R., Ritz d Vie». Aged , 108, iia, 119 i89sl.
(26J Hood, J. 4., PM if log , 80 {6] , 823 (1886}.
(26J Ta**a^ ^, G , Wied. Ann ., 68, 292 (1897 J.
t 27) Rernse u, I., »a mid, E. E., z o«. chem. 2 , 21, 281 i1899).
t88} A rrhenine. S., Z pipe. Cn em. , d, 256 1889 J; a translation of t5# t#q pqp jp tgj yyt
the t deal with temperature dependence is incl uded in Bae k. M. H., and Laid ter, 5
4., '‘S• Iected Readings in Che mica I K ineHes,'' Pergamon 9 q$ t, Oxford , 1867, pp.
J I—35.
(29) Kooij, D. M., Z. pits Chem . , 12, 156 (1893J.
(30) Trauts, M., Z §A ys. Cnern , 66, 496 \1909).
(31) Harcour I, A. V., and Esson, W ., Proc. boy. foe , A14, 470 (1865}.
Harcourt, A. * , ""8 Essor', .. Pfii7 Trocs., i 5G, isz ( i 866); excerpts from this
orticle are re prod need in Back, M. H., and Laidler, K. J., pp, 8-2y, (see ref. (2gj}.
Harcourt, A. V., and Esson, W., PDF I. Are ns., 157, I1'7 {1867).
(34J Harcourt, A. V.. end Esson, W., Phd 1. Tr# n z., 186A, 8iy
{18gs). (35} Veley, V. H ., J. tfi em. Doc , 08, 2122 (190BJ.
Hecht, W., and Donrad, M., Z. phyc. Chem , 4, 450 (1889).
{40) Pendlebury, W. H., and Seward, M., Proc. n«, Soc., z , 4s, .3ss Iie8gi.
B»ga •si , s., z. pays. c ‹o , iz, sas i lso3J.
Bodenstein, M., Z. plys Chem , 29, 296 (1899 J.
498 Journal of Chemical Education
t4a j

(46) Lewis, W. C. McC., d. Chem . Joe. , 118, 471 (1918); excerpts frog thia paper m
included in Back, M. H. and fiaidle r, K. J., pp. 36—40, (see ref (28)}.
Moelwyn Hughes, E. A., “The Kinetics of Reactions in Solution,” 2nd ed., Clarendon
Press, Osford, I947, p. 2,
ISB) King, M. ñ., PM D thesis, 1'he Open U ni verai th,
1960.
T.aid)er, K. J., Cñrm Can «do, 32, V, 3t ( f980).
7'!"gt *-› *- *8"+ Phys . 4. 107 (T9fi5): for an accoun I o* the deve)opment oF
franei- tio n-state theory, ace Rajdie r, K. 4., and King, M. C., J. Plus 35#y ,
gy, 2ply

IS? ) Koh nsturn m, P., and Scheffer, P. E. C., Proc Ronin# I. be d A ccd. Petense Top. ,
I3,
(5:IJ Brandsraa, W. F , Chem Jfceâ 6ted , I9. 318 ( I9B2}.
(54 I Sch effer, F. E. C., and Brand sma, W. F., Dec. TrAu. Him , â5, 52'2 t J
g2g j, t6SJ Brandsma, W. F., Chem Jfeeé6/cd. d7. 9a (1928); ¥8, 1205 ( lgBg).
(fi6} Marcelin, R., 7. ‹ A? m plys , 12, 451 I i 9 la) .
t67 j • •*•*'", R., *" *8*•‹9*•, 3, 120 l i 9J 6); this 9B9em was pu b1›s hed aftet Mareel n
had been k illed in the First W orId War.
158) T'?yring, H., and Polanyi, M., Zeir plys b? e m. , B , l2, 279
(1g3J), (fi9J Peleer, H., and W igne r, F., Zen t. pfiy•. fifiem. , IN , 15,
445 (1982).

(61) I ewis, W. C. Mch?., d. C'Ji en Sr›‹ , 113, 47 I t 1918).

(S21 Hin.she1wor›d , (? N . “The K ineticz of t?hemica1 Change tn C•aseous Systems,” 1st
ed.,
Class mw,Onwdg2r.
is4; LaMer , v. x., .J. r’Ji« . r's,.. , i, Yes t is3«J.
‹key «iasst«»". s., raidi«r, K. u., and Ey i«g, ii., "ih‹ iheo,y or rtaie Processes,"
(66) I aid ler, K. 'J., “’I’heories of (Chemical Reaction Rates, ” McG raw-Hill Book Co., New
Y‹›rk, l969.