Kinetics of reduction of cinnamaldehyde by means of hydrido-and deuteridocobalt tetracarbonyl

Article abstract of DOI:10.1002/(SICI)1097-4601(1997)29:7<473::AID-KIN1>3.0.CO;2-R

The reaction of HCo(CO)₄ (HT) or DCo(CO)₄ (DT) with excess cinnamaldehyde (CA) in methylcyclohexane (RH) at 22.2° and under 1 atm of CO follows pseudo-first-order kinetics in HT or DT with an inverse isotopic effect of 034. Identifie

Full text of DOI:10.1002/(SICI)1097-4601(1997)29:7<473::AID-KIN1>3.0.CO;2-R

Kinetics of Reduction of
Cinnamaldehyde by Means
of Hydrido- and
Deuteridocobalt
Tetracarbonyl
LEROY H. KLEMM,1 MILTON ORCHIN2
1Department of Chemistry, University of Oregon, Eugene, Oregon 97403-1253
2Department of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221
Received 25 July 1996; accepted 14 November 1996
ABSTRACT: The reaction of HCo(CO)4 (HT) or DCo(CO)4 (DT) with excess cinnamaldehyde
(CA) in methylcyclohexane (RH) at 22.2° and under 1 atm of CO follows pseudo-first-order ki-
netics in HT or DT with an inverse isotopic effect of 0.54. Identified products of the reaction
are hydrocinnamaldehyde (HCA) and styrene (STY). The STY is believed to be an artifact of
the thermal decomposition of the true product PhCH2CH2C( ϭ O)Co(CO)4 (X) or its isomer.
Reduction of the carbon-carbon double bond in CA is effected by hydrogen from both the
cobalt compound and RH. It is proposed that the reaction involves a free-radical chain mech-
n
anism in which the rate of the slow step is proportional to [CA]0 , the initial molar concentra-
tion of CA raised to a power of 1.5–1.8. Additionally the rate of conversion of CA to HCA and
X meets the criteria of a homocompetitive reaction with [CA], [HCA], and [STY] simple func-
tions of t0.5 (where t is reaction time) for use of DT or (in a single case) a function of (t0.5 ϩ t)
for use of HT. ©1997 John Wiley & Sons, Inc. Int J Chem Kinet 29: 473–481, 1997.
tions came from HT. The reaction, conveniently fol-
lowed by titration,
INTRODUCTION
Problems associated with the catalytic hydrogenation
of cinnamaldehyde (CA) to form hydrocinnamalde-
hyde (HCA) rather than cinnamyl alcohol, hydrocin-
namyl alcohol, or mixtures of various reduction prod-
ucts have been summarized by Rylander [1]. In
contrast, Goetz and Orchin [2] reported that
HCo(CO)4 (HT), used in excess, readily effects re-
duction of the carbon-carbon double bond of CA in
the mixed solvent hexane-benzene at 25° and atmo-
spheric pressure to give HCA in 97% yield. It was as-
sumed that both hydrogen atoms involved in the reac-
O
ϩ2H
CH"CH9CH
CA
O
CH2CH2CH (1)
HCA
was pseudo-first-order in HT, but gave some accom-
Correspondence to: L. H. Klemm
©1997 John Wiley & Sons, Inc.
CCC 0538-8066/97/070473-09
panying evolution of gas, presumably CO from de-

474
KLEMM AND ORCHIN
composition of HT. We now report a more elaborate
kinetic study of this reaction in unmixed solvents,
i.e., mainly in methylcyclohexane at 22.2 Ϯ 0.1° or
to a lesser extent in toluene at 3.2 Ϯ 0.1°, in an effort
to ascertain the mechanism of the reaction. Also the
isotopic effect of using DCo(CO)4 (DT) in place of
HT is reported.
0.7 M) freshly distilled CA in the appropriate solvent
and a CO atmosphere was thermally equilibrated
(with stirring) in a reaction flask held in the bath.
At time t ϭ 0 the reaction was started by injecting
a measured volume of stock solution HT into the re-
action flask containing CA. Periodically a sample
(usually 1.00 ml) of the reaction mixture was trans-
ferred to one of the NaOH samples at elapsed time t.
As time permitted each basic sample mixture was
back-titrated with standard (0.01 M) HCl to deter-
mine the amount of HT used. The data were cor-
rected for spontaneous decomposition of HT, as mea-
sured by means of an occasional sample withdrawn
from the stock solution and titrated in the same way.
A stock solution of DT was obtained by mixing 90
ml of the HT stock solution successively with 10, 5,
and 5 ml of 99.8 atom % D2O followed by the afore-
mentioned freezing and drying procedure. A mass
spectrum of the stock solution showed it was 97.2%
isomerically pure DT. Kinetic experiments with DT
were conducted in an identical manner to those with
HT. No correction was made for the presence of HT
in the DT solution.
Typical pseudo-first-order plots for HT and DT re-
actions are shown in Figure 1 and data for all of the
experiments are presented in Table I. These plots
were made of log[(Vϱ Ϫ Vt)/Vϱ] vs. t, where t is the
reaction time in minutes and Vt and Vϱ are volumes
of standard HCl (in ml) used for back titration at
times t and “infinity,” respectively, [5], by means of a
Mathcad computer program [6] which determines the
least-squares linear fit of the data and calculates the
slope kЈ and the correlation coefficient (0.995 to
0.999) for it. For most experiments, linear log plots
were obtained over 2.4–4.4 half lives. However, the
three slowest reactions (Expts. 1, 10, 11) were fol-
lowed for only 300–500 min.
EXPERIMENTAL
Unless otherwise noted, all liquids and solutions were
contained in rubber-stoppled flasks and transferred by
means of syringes. For each kinetic experiment a se-
ries of flasks, flushed with CO at 1 atm pressure, was
charged with 20 ml (excess) of standard (0.01 M)
NaOH solution each. Also for each experiment a
stock solution of HT was prepared as follows. Typi-
cally, a mixture of 7.0 g of Co2(CO)8 [3] and 70 ml of
reagent-grade solvent (methylcyclohexane or toluene)
was stirred magnetically in an atmosphere of CO in a
flask bearing an attached gasometer and a stoppled
side neck. Twenty ml (excess) of dimethylformamide
was added slowly while gas evolution (CO) was mon-
itored. The next day the mixture of two liquid layers
was cooled to 0–5° and treated dropwise with 30 ml
of concentrated HCl (12 M). The blue aqueous layer
(CoCl2) was removed and the yellow hydrocarbon
layer was washed three times in situ with 25-ml por-
tions of ice-cold, CO-purged, distilled water [4]. The
gasometer was disconnected and the flask was im-
mersed in a bath of dry ice-EtOH to freeze remaining
H2O and dry the HT solution. Sufficient HT solution
(usually 30 ml) for the experiment was now placed in
a clean, dry flask (preflushed with CO) held in a con-
stant temperature bath. After 30 min the equilibrated
HT stock solution was ready for use. Meanwhile, a
measured volume (usually an excess) of standard (ca.
In Expts. 4 (with HT) and 7–9 (with DT) the or-
Figure 1 Typical pseudo-first-order plots of log[(Vϱ Ϫ Vt)/Vϱ ] vs. reaction time t (in minutes) for disappearance of
HCo(CO)4 in Expts. 2 (Fig. a) and 5 (Fig. b) and of DCo(CO)4 in Expt. 9 (Fig. c). Vϱ ϭ 17.98, 15.9, and 18.14 ml, respec-
tively.

REDUCTION OF CINNAMALDEHYDE
475
Table I Experimental Data for the Kinetics of Reaction of HCo(CO)4 (HT) and DCo(CO)4 (DT) with
Cinnamaldehyde (CA)
Initial Molar
Reaction Conditions
Concentrations
Cobalt
reagent
used
First-order rate
Half-life of
HT ␱r DT
(min)
Correctedf
rate
constant, k
solvent,b
temp.
[HT]0 or
[DT]0
No. of data
points
constant kЈ
c
d
1 e
Expt. no.a
[CA]0
(104 minϪ
)
1
2
3
4
5
6
7
8
9
MeC, 22.2°
ditto
HT
HT
HT
HT
HT
HT
DT
DT
DT
HT
HT
0.152
0.129
0.140
0.133
0.151
0.116
0.118
0.123
0.141
0.184
0.150
0.0965
0.181
0.239
0.245
0.322
0.358
0.160
0.229
0.283
0.190
0.570
19
18
11
13
12
8
35.8
93.7
194.
74.0
55.3
36.6
23.6
17.6
28.3
16.3
12.0
825.
137.
0.218
0.189
(0.155)g
0.224
0.216
0.239
0.398
0.403
0.396
0.013
0.013
ditto
ditto
ditto
ditto
ditto
ditto
ditto
T, 3.2°
ditto
125.3
189.3
294.0
393.3
244.5
426.4
579.1
8.4
9
12
10
15
16
10
11
50.5
a Listed in order of increasing [CA]0 for each subset of conditions and reagents, as given in columns 2 and 3. Not listed in chronological
order.
b MeC is methylcyclohexane; T is toluene.
c Obtained from the relationship [HT]0 ϭ V0M, where M is the titrated molarity of the standard HCl used and V0 is the volume (in ml) ob-
tained by extrapolation of the plot for log (VϱϪVt) to t ϭ 0. Likewise for [DT]0 .
d Concentration of prepared standard solution used, adjusted for the volume of HT (or DT) solution added.
e From the slope of the Mathcad plot noted before.
f For experiments 1–6, k ϭ kЈ/[CA]01.758; for experiments 7–9, k ϭ kЈ/[CA]01.523; for experiments 10–11, k ϭ kЈ/[CA]0
g Omitted in calculation of average, i.e., k ϭ 0.217, for experiments 1–6.
1.63
.
ganic layer from each neutralized kinetic sample was
analyzed by gas chromatography using disk integra-
tion and a column packed with Ucon Polar held at
180°. Three main components, viz. recovered CA,
HCA, and styrene (STY), were identified by compar-
ison of retention times with those of authentic sam-
ples. Time-dependent analytical data for the organic
product composition from Expts. 4 and 8 are pre-
sented in Tables II and III and plotted in Figures 2
and 3.
It was assumed that the apparent first-order rate
constant kЈ (Table I) can be represented by the rela-
n
tionship kЈ ϭ k[CA]0 , where k is a corrected first-or-
der rate constant. The numerical values of n for Ex-
pts. 1–6 (with HT) and for Expts. 7–9 (with DT) are
obtained from Figures 4 and 5. There log kЈ is plotted
vs. log [CA]0 to give values for n of 1.758 and 1.523,
respectively. From these n values one obtains an aver-
age rate constant for kH of 0.217 and for kD of 0.399,
i.e., an isotopic effect of kH/kD ϭ 0.54. This value
compares favorably with that of 0.45 reported for re-
action of styrene with HT [7,8]. Although only two
experiments (10 and 11) were conducted with HT in
toluene at 3.2°, one obtains values for n of 1.63 and
kH of 0.013 for them.
The occurrence of nonintegral values for n in ki-
netic expressions has been noted by Shoemaker et al.
[9], who suggest that two or more simultaneous
mechanisms may be in competition or that no single
reaction step is effectively rate controlling. More
specifically, nonintegral (especially 1/2 or 3/2) values
for n are associated with radical chain reactions [10].
These generalizations are considered in our mecha-
nistic proposals. Also, we adopt the concept of transi-
tion state theory that the activated complex involves
RESULTS AND DISCUSSION
Examination of Table I and Figure I shows three gen-
eral relationships, which occur, viz. (1) the disappear-
ance of either HT or DT follows first-order kinetics
closely, (2) the rate of reaction with DT is larger than
with HT, i.e., there is an inverse isotopic effect, and
(3) with either HT or DT the rate constant kЈ in-
creases monotonically with the increasing initial con-
centration of CA used. A fourth pertinent relationship
is found in the stoichiometry of the reaction, as exem-
plified by the data in Tables II and III. This result is
considered later.

476
KLEMM AND ORCHIN
Table II Measured Data for Kinetic Experiment Number 4. Reaction Conditions: Solvent
Methylcyclohexane, Temperature 22.2°. Reaction of HCo(CO)4 (HT) with Cinnamaldehyde
(CA) to give Hydrocinnamaldehyde (HCA) plus Styrene (STY)
Molar Concentrations in Organic Product
Mixturec
Time t
(min)
HCl used
Vt (ml)a
[HT]tb
M
[CA]t
[HCA]t
[STY]t
0
(3.7)d
4.76
5.71
6.70
8.75
0.133
0.122
0.112
0.102
0.080
0.245e
0.232
0.227
0.212
0.210g
0.218g
0.214
0.190
0.159g
0.162g
0.160
0.153
0.141
0.129
0.121
(0.000)f
0.009
0.011
0.021
0.024g
0.020g
0.028
0.043
0.062g
0.059g
0.061
0.060
0.076
0.083
0.088
(0.000)f
0.004
0.006
0.013
0.009g
0.008g
0.003
0.011
0.022g
0.022g
0.025
0.033
0.028
0.033
0.035
3.26
8.65
14.63
28.53
36.40
46.29
56.18
9.75
10.71
11.94
0.071
0.061
0.048
66.57
77.68
88.15
122.33
144.36
(ϱ)h
12.44
13.34
14.27
15.27
0.043
0.034
0.024
0.014
0.009
(0.000)f
(16.60)i
a Corrected for spontaneous decomposition of HT.
b [HT]t ϭ M(V Ϫ Vt)/1000. See text of experimental part.
c Based on theϱassumption that [CA]t ϩ [HCA]t ϩ [STY]t ϭ [CA]0 throughout the experiment and that
molarities are proportional to the integrated chromatographic areas found in each sample.
d Obtained from extrapolation of the first-order kinetic plot to t ϭ 0.
e See footnote d, Table I.
f Not measured experimentally.
g Duplicate chromatographic data from the same neutralized sample.
h Hypothetical infinity.
i From extrapolation of log plot.
CA:0.088 ϩ 0.035 ϭ 0.123 M phenyl groups in
products. One can, therefore, propose balanced eq.
(2) for Expt. 4, where the styrene found is believed to
be a decomposition product of X and/or its isomer X؅
[4] (vide infra). It seems likely that styrene is accom-
panied by ethylbenzene due to hydrogenolysis of X
or X؅, but we did not measure PhEt independently.
Formation of Co2(CO)8 has been established by oth-
ers [2, 7, 13–16] and R—R is a simplified represen-
tation of expected hydrocarbon byproducts (including
likely ones from disproportionation) from the interac-
tion of R free radicals. No attempt was made to iso-
late or identify these byproducts. Compounds X and
X؅ were proposed as products from hydroformylation
of styrene with HT and CO [4,7]. However, in our re-
action we propose that X and X؅ form from CA
rather than from styrene. Since an equilibrium should
exist between X and X؅ in the reaction mixture we
confine further discussion only to X [4]. It should be
noted that linear acylcobalt tetracarbonyls are well-
1 molecule of HT or DT plus 1.75 or 1.5 (respec-
tively) molecules of CA [11].
Meanwhile, the data in Tables II and III show that
the stoichiometry of our reaction with either HT or
DT does not conform to a simple 2:1 molar ratio of
Co reagent to CA, as had been suggested earlier [2].
In fact, this discrepancy was also noted by Terapane
[12]. In particular, our data show that the solvent
methylcyclohexane (shown here as RH) is not chemi-
cally inert. Rather, it furnishes some of the hydrogen
atoms involved in the transformation CA:HCA.
Thus, for Expt. 4 (Table II) a change in [HT] from
0.133 M to 0.009 M (i.e., ⌬M ϭ Ϫ 0.124) is accom-
panied by a change in [CA] from 0.245 M to 0.121 M
(i.e., also ⌬M ϭ Ϫ0.124, instead of Ϫ 0.062 ex-
pected if two HT molecules had reacted with one CA
molecule). Over the same reaction period, the con-
centrations of HCA and STY changed from 0 M to
0.088 and 0.035, respectively. Consistent with the
conservation of phenyl groups, one has 0.124 M

REDUCTION OF CINNAMALDEHYDE
477
Table III Measured Data for Kinetic Experiment Number 8. Reaction Conditions as in Table
II, but with DCo(Co)4 (DT) used instead of HT
Molar Concentrations in Organic Product
Mixturec
Time t
(min)
HCl used
Vt (ml)a
[DT]tb
M
[CA]t
[HCA]t
[STY]t
0
3.50
7.45
(6.0)d
7.20
9.20
0.123
0.110
0.090
0.078
0.069
0.053
0.047
0.041
0.027
0.022
0.013
0.007
0.002
0.000
0.229e
0.207
0.193
0.181
0.173
0.161
0.136
0.133
0.121
0.119
0.091
0.084
0.084
0.086
(0.000)f
0.013
0.024
0.036
0.039
0.049
0.062
0.070
0.075
0.082
0.096
0.096
0.102
0.102
(0.000)f
0.011
0.012
0.013
0.016
0.018
0.031
0.026
0.033
0.028
0.042
0.049
0.044
0.042
11.58
15.28
19.73
25.60
30.31
37.32
45.45
55.18
65.37
79.96
103.35g
10.37
11.27
12.84
13.36
13.96
15.34
15.76
16.68
17.22
17.76
17.93
a,b Same as footnotes in Table II, but for DT instead of HT.
c See footnote c, Table II. Products were not analyzed for deuterium content.
d – f Same as footnotes in Table II.
g Hypothetical infinity.
known compounds, stable in an
Equation (2) shows that only a 72% yield of HCA
is produced in Expt. 4. However, in the study by
Goetz and Orchin [2], where excess HT was used,
any X formed should react further as in eq. (3) to give
potentially a quantitative yield of HCA [19,20], per-
haps containing isomeric 2-phenylpropanal.
O
HCo(CO)4 ϩ PhCH"CHCH ϩ 0.72 RH
(HT)
(CA)
O
X ϩ HCo(CO)4
0.72 PhCH2CH2CH ϩ 0.36 Co2(CO)8
O
(HCA)
PhCH2CH2CH ϩ Co2(CO)8 (3)
O
(HCA)
ϩ 0.28 {PhCH2CH2CCo(CO)4 and/or
(X)
O
PhCHCCo(CO)4} ϩ (0.36 R—R) (2)
CH3
(X؅)
atmosphere of CO for hours or days in dilute solution
(0.1 M or less) but decomposing in the vapor state
[17–21]. Based on the observations of Heck and
Breslow for alkyl- and acylcobalt tetracarbonyls, X
should be converted into styrene upon injection into a
gas chromatograph at elevated temperature, as occurs
in our work-up [22].
Figure 2 Plot of the transformation of cinnamaldehyde
(circles) into hydrocinnamaldehyde (upper squares) and
styrene (lower squares) for Expt. 4, with HCo(CO)4. See
text for algebraic equations of curves.

478
KLEMM AND ORCHIN
While only Expt. 4 of those using HT was checked
for stoichiometry, all three of those using DT (Expts.
7–9) were investigated in this manner. Typically,
Expt. 8 shows these correlations for the complete re-
action of DT: ⌬M ϭ Ϫ 0.123 for DT, Ϫ 0.143 for
CA, ϩ 0.102 for HCA, and ϩ0.042 for STY. Again
phenyl groups are conserved and one obtains partially
balanced eq. (4). We did not ascertain the actual loca-
tions of the D atoms in the products. Equation (4)
merely illustrates one simple possible distribution.
0.86 DT ϩ CA ϩ 0.85 RH
Figure 4 Plot of log kЈ vs. log [CA]0 to obtain values for
n and k in the relationship kЈ ϭ k[CA]0 for Expts. 1–6,
O
n
0.71 PhCH2CHDCH
with HCo(CO)4 .
(HCA)
O
and, presumably, the remainder from the aldehyde
group.
In an attempt to clarify the kinetics of the conver-
sion CA:HCA ϩ X (i.e., STY) we assumed that
the time-dependent concentration of each of these
components is given by an expression of the form
0.15 PhCH2CHDCCo(CO)4
0.14 PhCH2CH2CCo(CO)4
ϩ
ϩ ... (4)
O
(X)
Analogously, partial equations (5) and (6) show the
stoichiometries for Expts. 7 and 9, respectively.
[Y]t ϭ a0 ϩ bt0.5 ϩ ct
where a0 is the initial concentration of mixture com-
ponent Y (i.e., 0 for Y is HCA or STY, cf., Table I
for Y is CA) and b and c are constants. Moreover, al-
gebraically one should have [CA]t ϩ [HCA]t ϩ
[STY]t #[CA]0 . The data from Tables II and III
plus Expts. 7 and 9 were then surveyed for plausibil-
ity of fit to these equations by means of a Jandel
Table Curve 2-D program [23]. For Expts. 7–9 with
DT, the data fit well (except for some discrepancy in
a0 ) to the simpler form where c ϭ 0, as given in eqs.
0.77 DT ϩ CA ϩ 0.94 RH 9:
0.71 HCA ϩ 0.29 X ϩ . . . (5)
0.80 DT ϩ CA ϩ 0.87 RH 9:
0.67 HCA ϩ 0.33 X ϩ . . . (6)
Despite the variations in Expts. 4 and 7–9 the yields
of HCA and X (average values of 70% and 30%) re-
main essentially constant, as based on the CA con-
sumed. Of the two g-atoms of H added to the CA car-
bon-carbon double bond only 0.77–1 arises from the
Co starting material; 0.72–0.94, from the solvent;
Figure 5 Plot of log kЈ vs. log [CA]0 to obtain values for
n
Figure 3 Same legend as for Figure 2, but for Expt. 8,
n and k in the relationship kЈ ϭ k[CA]0 for Expts. 7–9,
with DCo(CO)4. See text for algebraic equations of curves.
with DCo(CO)4 .

REDUCTION OF CINNAMALDEHYDE
479
[CA]t ϭ 0.248 Ϫ 0.00657t 0 . 5 Ϫ 0.00039t,
[HCA]t ϭ Ϫ0.003 ϩ 0.00515t0.5 ϩ 0.00024t, and
[STY]t ϭ 0.000 ϩ 0.00138t 0 . 5 ϩ 0.00015t;
average r2 ϭ 0.914 (Fig. 2). Rate equations then take
the form (17).
(7)–(15). Conformance of experimental data to the
For Expt. 7:
[CA]t ϭ 0.173 Ϫ 0.0140t0.5
[HCA]t ϭ Ϫ 0.006 ϩ 0.0093t0.5
[STY]t ϭ Ϫ 0.006 ϩ 0.0047t0.5
average r2 ϭ 0.948
(7)
(8)
(9)
d[Y]/dt ϭ b/2t0.5 ϩ c
(17)
We are unable to devise a mechanistic scheme which
accounts quantitatively for all of the preceding obser-
vations. However, Scheme I, incorporates qualita-
tively the crucial aspects of such a mechanism for the
three experiments with DT.
For Expt. 8:
[CA]t ϭ 0.238 Ϫ 0.0183t0.5
[HCA]t ϭ Ϫ 0.006 ϩ 0.0128t0.5
[STY]t ϭ Ϫ 0.003 ϩ 0.0057t0.5
average r2 ϭ 0.957
(10)
(11)
(12)
Initiation steps:
kЈ
DT ϩ 1.5 CA
[DT--CA1.5]Ϯ
rate
controlling
O
и
и
[PhCHCHDCH ϩ T--CA0.5]
For Expt. 9:
(A )
(Z )
и
и
[CA]t ϭ 0.287 Ϫ 0.0227t0.5
[HCA]t ϭ Ϫ 0.008 ϩ 0.0164t0.5
[STY]t ϭ 0.004 ϩ 0.0063t0.5
average r2 ϭ 0.967
(13)
(14)
(15)
k1
k2
T ϩ 0.5 CA
и
[Z ]Ϯ
и
O
и
0.5[PhCHCH CCo(CO) ] ϩ 0.5 T
и
2
4
(B )
и
Propagation steps:
equations for Expt. 8 can be seen in Figure 3. A
check on these equations for Expt. 8 was also made
by obtaining linear plots of concentrations vs. t0.5 by
the Mathcad program with corresponding values for
intercept, slope, and correlation coefficient r for the
three compounds of 0.237, Ϫ0.0189, Ϫ 0.990;
Ϫ 0.006, 0.0129, 0.991; and Ϫ 0.002, 0.0055, 0.963,
respectively. One notes that the conversions of CA
into HCA and X (or STY) are homocompetitive [24],
i.e., they have identical kinetic equations, except for
the constants a0 and b, and the ratios of b values (e.g.,
0.0128/0.0183 ϭ 0.70 and 0.0057/0.0183 ϭ 0.31 in
Expt. 8) are consistent with the observed yields of
HCA (70%) and STY (30%). The rationale for these
relationships is that both HCA and X are derived
from an effectively common intermediate (vide in-
fra). Differentiation of eqs. (7)–(15) leads to rate
equations of the type (16).
A
и
B
и
DT
k3
DT
k5
k7
RH
k4
T ϩ HCA
и
RH
k6
X ϩ T
и
HCA
X
ϩ R
ϩ R
и
и
Termination steps:
k8
k9
2 R
R—R
X ϩ CA
Scheme I
2 T
T2
и
и
In Scheme I brackets indicate assemblages contained
in a solvent cage, the symbol Ϯ designates an inter-
mediate complex or transition state, and the double
hyphen (--) implies association of chemical entities.
d[Y]/dt ϭ b/2t0.5
(16)
The formation of the geminate radical pair [A
ϩ
и
Z ] from DT and CA may be considered to be a
и
modification of a Michaelis–Menten mechanism, but
The chromatographic data for Expt. 4 (Table II) show
considerably more scatter than for Expts. 7–9. They
do not fit a t0.5 relationship well, but do fit the more
extended equations
where both DT and CA are present in appreciable
concentrations [25]. Partial dissociation of Z into
и
T
and CA leads to a subsequent caged radical pair
и

480
KLEMM AND ORCHIN
to the latter institution for awarding him a Visiting Profes-
sorship concurrently. He also acknowledges helpful com-
ments from Professor Paul Engelking of the University of
Oregon in interpretation of the experimental results.
[A ϩ B ] [26], but not in equimolar amounts for
и и
the two components. Disassociated free radical T is
и
shown as located outside of the solvent cage. The rate
of formation of caged radical B is expected to alter
и
the main rate-controlling step. After migration from
the cage, A and B undergo analogous propagation
и и
steps to abstract deuterium from other DT or hydro-
gen from RH to yield corresponding stable products
HCA and X, respectively. Termination of the chain
mechanism results from dimerization of chain carri-
ers R and T , or of hydrogen transfer from A to
BIBLIOGRAPHY
1. P. Rylander, Catalytic Hydrogenation in Organic Syn-
theses, Academic Press, New York, 1979, pp. 78–80.
2. R. W. Goetz and M. Orchin, J. Am. Chem. Soc., 85,
2782 (1963).
3. I. Wender, H. W. Sternberg, S. Metlin, and M. Orchin,
Inorg. Synth., 5, 190 (1957).
4. M. Orchin and W. Rupilius, Catalysis Rev., 6, 85
(1972).
5. C. A. Bennett and N. L. Franklin, Statistical Analysis
in Chemistry and the Chemical Industry, Wiley, New
York, 1954, pp. 234–240.
6. Mathcad-3.1, MathSoft, Inc., 201 Broadway, Cam-
bridge, Massachusetts 02139.
7. F. Ungvary and L. Marko, Organometallics, 1, 1120
(1982).
8. R. M. Bullock in Transition Metal Hydrides, A.
Dedieu, Ed., VCH Publishers, Inc., New York, 1992,
Chap. 8.
9. D. P. Shoemaker, C. W. Garland, and J. I. Steinfeld,
Experiments in Physical Chemistry, 3rd ed, McGraw-
Hill, New York, 1974, p. 275.
10. J. Fossey, D. Lefort, and J. Sorba, Free Radicals in Or-
ganic Chemistry, Wiley, New York, 1995, Chap. 7.
11. Z. G. Szabo in Comprehensive Chemical Kinetics,
C. H. Bamford and C. F. H. Tipper, Eds., Elsevier, New
York, 1969, Vol. 2, p. 1.
12. J. F. Terapane, Diss. Abstr. Int., 30, 4985-B (1970).
13. T. E. Nalesnik, J. H. Freudenberger, and M. Orchin, J.
Organometal. Chem., 221, 193 (1981); 236, 95 (1982).
14. J. A. Roth, P. Wiseman, and L. Ruszala, J.
Organometal. Chem., 240, 271 (1983).
и и и
to form CA plus X. The considerable symmetry
of Scheme I and the essentially equivalent chemical
B
и
natures of A and B are consistent with the homo-
и и
competition noted previously. Abstraction of an alde-
hyde hydrogen atom, especially by a free radical (T
и
here) bearing a metallic element of variable valence,
is well known [27–29] and is proposed to occur in
the rearrangement of part of Z to yield B (rate con-
и
и
stant k2). In particular, cinnamaldehyde undergoes
such hydrogen abstraction during autoxidation [30].
It should be noted that the complexes [DT--
CA ] and [Z ], as well as the nonisolated product
и
1.5
X, are all hypothetical entities which are proposed in
order to rationalize the observed kinetics, identified
products, and reaction stoichiometry. Unfortunately,
time did not permit us to search for these intermedi-
ates via direct spectral methods, though such a study
would certainly be pertinent.
To accommodate Expt. 4 into the foregoing
scheme one merely changes 1.5 CA to 1.75 CA and
changes subscripts and other coefficients for the initi-
ation steps accordingly. Presumably the ratio of rate
constants k1 and k2 will be altered as well.
15. M. Orchin, Annals N.Y. Acad. Sci., 415, 135 (1983).
16. F. Ungvary and L. Marko, J. Organometal. Chem., 219,
397 (1981).
SUMMARY
17. R. F. Heck in Organic Syntheses via Metal Carbonyls,
I. Wender and P. Pino, Eds., Interscience Publishers,
New York, 1968, Vol. 1, pp. 374–377.
The reaction of cinnamaldehyde with HCo(CO)4 or
DCo(CO)4 in methylcyclohexane at 22.2° has been
shown to yield hydrocinnamaldehyde and styrene—
an artifact believed to arise from decomposition of
PhCH2CH2C(ϭO)Co(CO)4 . Kinetic studies indicate
that a free-radical chain mechanism is involved,
whereby the solvent furnishes much of the hydrogen
used to reduce the carbon-carbon double bond of cin-
namaldehyde. An inverse isotopic effect is found.
18. P. L. Pauson, Organometallic Chemistry, Edward
Arnold, Ltd., London, 1967, pp. 157–161.
19. R. F. Heck, Adv. Organometal. Chem., 4, 243 (1966).
20. R. D. W. Kemmitt and D. R. Russell in Comprehensive
Organometallic Chemistry, G. Wilkinson, Ed., Perga-
mon Press, New York, 1982, Vol. 5, pp. 50–61.
21. R. F. Heck and D. S. Breslow, J. Am. Chem. Soc., 83,
1097 (1961).
22. R. F. Heck and D. S. Breslow, J. Am. Chem. Soc., 83,
4023 (1961).
23. Jandel Scientific, 2591 Kerner Blvd., San Rafael, Cali-
fornia 94901.
One of us (L.H.K.) wishes to thank the University of Ore-
gon for granting him a sabbatical leave to the University of
Cincinnati during the experimental part of this project and

REDUCTION OF CINNAMALDEHYDE
481
24. G. A. Russell in Investigation of Rates and Mecha-
nisms of Reactions, S. L. Friess, E. S. Lewis, and A.
Weissberger, Eds., 2nd ed, Interscience Publishers,
Inc., New York, 1961, Chap. 8.
25. L. Meites, An Introduction to Chemical Equilibrium
and Kinetics, Pergamon Press, New York, 1981, pp.
58–63.
27. D. C. Nonhebel, J. M. Tedder, and J. C. Walton, Radi-
cals, Cambridge University Press, Cambridge, 1979, p.
157.
28. P. I. Abell in Free Radicals, J. K. Kochi, Ed, Wiley,
New York, 1973, Vol. 2, p. 94.
29. K. B. Wiberg and R. Stewart, J. Am. Chem. Soc., 77,
1786 (1955).
26. T. E. Nalesnik and M. Orchin, Organometallics, 1, 222
(1982).
30. A. W. Pound and J. R. Pound, J. Phys. Chem., 38,
1045 (1934).