Stabilization of long-chain intermediates in solution. octyl radicals and cations

Article abstract of DOI:10.1016/j.molstruc.2013.02.020

The rearrangements of 1-octyl, 1-decyl and 1-tridecyl intermediates obtained from thermal lead(IV) acetate (LTA) decarboxylation of nonanoic, undecanoic and tetradecanoic acid were investigated experimentally through analysis and distribution of the products. The relationships between 1,5-, 1,6- and possibly existing 1,7-homolytic hydrogen transfer in 1-octyl-radical, as well as successive 1,2-hydride shift in corresponding cation have been computed via Monte-Carlo method. Taking into account that ratios of 1,5-/1,6-homolytic rearrangements in 1-octyl- and 1-tridecyl radical are approximately the same, the simulation shows very low involvement of 1,7-hydrogen rearrangement (1,5-/1,6-/1,7-hydrogen rearrangement = 85:31:1) in 1-octyl radical.

Full text of DOI:10.1016/j.molstruc.2013.02.020

Journal of Molecular Structure 1040 (2013) 19–24
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Stabilization of long-chain intermediates in solution. Octyl radicals and cations
⇑
´
´
´
Aleksandar V. Teodorovic , Dalibor M. Badjuk, Nenad Stevanovic, Radoslav Z. Pavlovic
Faculty of Science, University of Kragujevac, Radoja Domanovica 12, P.O. Box 60, 34000 Kragujevac, Serbia
g r a p h i c a l a b s t r a c t
h i g h l i g h t s
"
"
"
Rearrangements products
distribution study of the long-chain
radicals and cations.
Monte Carlo simulation of the
stabilization processes of 1-octyl
radical.
Ratio of homolitical rearrangements
determined by Monte Carlo
simulation.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 August 2012
Received in revised form 24 October 2012
Accepted 8 February 2013
Available online 27 February 2013
The rearrangements of 1-octyl, 1-decyl and 1-tridecyl intermediates obtained from thermal lead(IV) ace-
tate (LTA) decarboxylation of nonanoic, undecanoic and tetradecanoic acid were investigated experimen-
tally through analysis and distribution of the products. The relationships between 1,5-, 1,6- and possibly
existing 1,7-homolytic hydrogen transfer in 1-octyl-radical, as well as successive 1,2-hydride shift in cor-
responding cation have been computed via Monte-Carlo method. Taking into account that ratios of 1,5-/
1,6-homolytic rearrangements in 1-octyl- and 1-tridecyl radical are approximately the same, the simula-
tion shows very low involvement of 1,7-hydrogen rearrangement (1,5-/1,6-/1,7-hydrogen rearrange-
ment = 85:31:1) in 1-octyl radical.
Keywords:
Carbocations
Radicals
Rearrangements
Alkanoic acids
Lead(IV) acetate
Monte Carlo simulation
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
rearrangements of long-chain primary alkyl radical, higher dis-
tance transfer of 1,10- and 1,11-type were proposed in mechanism
The 1,5- and 1,6-intramolecular hydrogen abstractions are the
important stabilization pathways of primary flexible alkyl radical
chain [1]. 1,5-Transposition of the radical center from carbon to
the d-carbon atom involves 1,5-hydrogen transfer, passing through
a chair-like six-membered cyclic transition state I (Scheme 1), and
seven-membered cyclic transition state in the case of 1,6-hydrogen
atom abstraction [2].
1,4-Hydrogen migration, involving a five-membered cyclic tran-
sition state, is rare [1,3] but it was theoretically considered [1,4].
1,7-Hydrogen atom abstraction by silylmethyl radicals in aryl sulf-
ones substrate is one stage of Fuchs’ synthesis of alkenes [5]. Be-
side the most common 1,5- and 1,6-intramolecular homolytical
for thermal peroxy-alkanoic acids decarboxylation [6]. The oxida-
tive decarboxylations of saturated carboxylic acids, RCOOH, with
lead(IV) acetate (LTA) (under thermal or UV-photolytic conditions,
in different solvents) have been the subject of numerous studies
which have shown that the major fragmentation products obtained
in these oxidations were, usually, esters (i.e. acetoxyalkanes, ROAc,
and carboxylates, ROCOR) and alkenes, R(-H), in addition to satu-
rated hydrocarbons, RH, and phenylalkanes, RPh (when benzene
was used as a solvent) [7,8]. Moreover, it was established that all
these decarboxylation products are derived from the initial carbon
centered radical fragment Rꢀ (which is generated in a free-radical
chain process [8]) and the corresponding, subsequently formed
(by one-electron oxidation), carbocation (R+), whereby the relative
product distribution is dependent on structural features of the
starting acid and reaction conditions.
⇑
Corresponding author. Tel.: +381 34 335 039; fax: +381 34 335 040.
E-mail address: alek@kg.ac.rs (A.V. Teodorovic´).
0022-2860/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2013.02.020

´
20
A.V. Teodorovic et al. / Journal of Molecular Structure 1040 (2013) 19–24
Scheme 1. 1,5-Hydrogen atom abstraction.
The formation of rearranged products can also be explained via
acetoxyoctanes: 100 m long capillary glass column packed with
CW20M, at 130 °C under 0.25 MPa H2 and 100 m long capilllary
glass column packed with CW20M, at 100 °C under 0.25 MPa H2;
phenyloctanes: 60 m long capillary glass column packed with
CW20M, at 120 °C under 0.18 MPa H2; acetoxydecanes: 100 m long
capillary glass column packed with Carbowax 20M (CW20M), at
140 °C under 0.25 MPa H2; decenes and decane: 100 m long capil-
lary glass column packed with O-4-n-pentyloxybenzoyl oxime
(PBO) liquid crystal in chloroform [20]; phenyldecane: capillary
glass column packed with 4-methoxy-40-ethoxyazoxybenzene
(MEAB), at 130 °C under 0.18 MPa H2, 20 m long; tridecene and tri-
decane: 100 m long capillary glass column packed with PBO liquid
crystal, at 82 °C under 0.18 MPa H2 [16]; acetoxytridecanes: 100 m
long capilllary glass column packed with CW20M, at 57 °C under
0.25 MPa H2.
IR spectra: Perkin–Elmer FTIR spectrophotometer model Spec-
trum One.
NMR spectra: Varian Gemini 2000 NMR spectrometer at
200 MHz. Samples was analyzed as CDCl3 solutions using tetra-
methylsilane (TMS) as internal standard.
GCMS analysis: the gas chromatograph (Varian Gas-Chromato-
graph Model 340, column packed with CW20M) was connected via
an open split interface and a fused silica capillary (at 250 °C) to the
ion source of a Finnigan MAT 8230 spectrometer; EIMS: ion source,
170 °C, 70 eV.
Reaction mixtures were first fractionated by preparative gas
chromatography. Every fraction was analyzed by analytical gas
chromatography and GCMS analysis. Reaction product (acetates,
alkenes, phenylalkanes, etc.) were known compounds and were
characterized and identified on the basis of the spectral data and/
or by comparison with authentic samples synthesized by indepen-
dent routes.
1,2-hydride shifts, known as successive rearrangement in the alkyl
cations. Higher order rearrangements are not excluded (particu-
larly 1,3-hydride shift) [9]. However, their possible participations
do not influence significantly the product distribution.
Monte Carlo (MC) simulation is powerful method for determi-
nation of many parameters and for study of complicated processes,
such as chain dimension evaluation [10], calculation of low-energy
structures [11], simulation of consecutive metastable fragmenta-
tions [12], carbonium ion rearrangements [13] and optimization
in the refinement of molecular structure parameters from gas-
phase electron diffraction data [14]. In order to better model mag-
netic field effects on radical pairs in homogenous solutions Monte
Carlo approaches have recently been employed [15–18]. This paper
deals with transfer of hydrogen and hydride in long chain alkyl
intermediates such as 1-octyl, 1-decyl and 1-tridecyl radicals, ob-
tained by LTA reaction of corresponding alkanoic acids [19]. The
pathways of 1-octyl radical stabilization were analyzed by Monte
Carlo simulation method. The yields of the products obtained
experimentally were used as input data for Monte Carlo
simulation.
2. Experimental
2.1. General remarks
The benzene, 1-phenyloctane, 1-phenyldecane, 1-phenyltridec-
ane, 7-tridecanone,1-, 2-, 3-, 4-octanol, 1-, 2-, 3-, 4-, 5-decanol, 1-,
2-, 3-, 4-, 5-, 6-tridecanol, nonanoic, undecanoic and tetradecanoic
acid were commercial p.a. reagents (Sigma Aldrich, Fluka), dried
and distilled just before use by conventional techniques. Purity of
carboxylic acids was checked by GC analysis of their methyl esters,
which were obtained by reaction of these acids with methanol in
the presence of sulfuric acid as a catalyst.
2.2. Oxidative lead(IV) acetate decarboxylation of alkanoic acids
The synthesis of 7-tridecanol was performed by reduction of 7-
tridecanone with LiAlH4 in dry Et2O.
The thermal LTA decarboxylation in benzene was performed as
described previously [21]. Mixture of 10 mmol of LTA and 10 mmol
of alkanoic acids (nonanoic, undecanoic or tetradecanoic acid) was
dissolved in 60 cm3 of benzene. Before starting the oxidative
decomposition slow stream of purified Ar was introduced into stir-
red mixture of acid and LTA in benzene for 30 min at room temper-
ature and in the absence of light. The oxygen-free mixture
resulting from the above described procedure [21] was stirred
and heated under reflux (without light protection) until comple-
tion (i.e. until disappearance of tetravalent lead which was moni-
tored by potassium iodide/strach paper tracks) and then worked
up.
After completion of the LTA oxidation, the mixture was cooled,
treated with 50 cm3 of diethyl ether and filtered off. The filtrate
was washed with water, diluted HCl (1:1), aqueous Na2CO3 (5%)
and water. After drying (CaSO4), the solvents were removed under
reduced pressure and the products in the residue were analyzed by
analytical gas chromatography and separated by preparative gas
chromatography. The results of all runs are given in Tables 1, 3
and 4.
All acetoxyoctanes, acetoxydecanes and acetoxytridecanes
were prepared from corresponding alcohols by the reaction with
acetic anhydride in pyridine (the yields were 90–97%). Acetox-
yalkanes were purified by preparative GC and their purity was
checked by analytical GC and characterized by IR, 1H and 13C
NMR spectra.
Lead(IV) acetate (Sigma Aldrich) was recrystallized twice from
glacial acetic acid containing 3% acetic anhydride, sucked dry in a
Büchner funnel, and dried in a vacuum desiccators over potassium
hydroxide.
Preparative GC: Varian Aerograph 920 gas chromatograph; the
column consisted of CW20M; hydrogen was used as a carrier gas
at inlet pressure of 0.18 MPa. Analytical gas chromatography
(GC): Varian Model 3400 gas chromatograph and Perkin–Elmer
F-11 gas chromatograph equipped with a flame ionization detector
was used to measure the retention characteristics. Hydrogen was
used as a carrier gas at different inlet pressure. The conditions
for the separation of octenes and octane: 200 m long capillary glass
column packed with Squalane (SQ), at 42 °C under 0.18 MPa H2;

´
21
A.V. Teodorovic et al. / Journal of Molecular Structure 1040 (2013) 19–24
Table 1
2.3. Monte Carlo simulation
Product distribution (%) in LTA reaction with nonanoic (n = 8), undecanoic (n = 10)
and tetradecanoic (n = 13) acid.a,b,c,d
Fortran 90 generated the numbers uniformly distributed be-
tween 0 and 1 by using an intrinsic function called RANDOM_NUM-
BER (gamma). The optional name ‘‘gamma’’ is defined by the user.
It works according to the next algorithm:
CnH2n+1COOH ? CnH2n+1X + CnH2n
Yield %
X = H
X = Ph
X = OAc
CnH2n
n = 8
n = 10
n = 13
17.33
7.91
12.70
39.80
55.53
50.67
16.90
21.72
12.37
15.84
4.44
11.61
1. Assume the transition probability pij for all intermediates. They
have values between 0 and 1.
2. Then generate a start element 1a.
3. By using intrinsic function RANDOM_NUMBER (gamma), the
probability of formation any element is given.
a
b
c
Acid:LTA molar ratio = 1:1.
Solvent: benzene; temperature: 81 °C.
Yields calculated from analytical liquid–gas chromatograms.
d
For isomers see Tables 3 and 4.
4. It was assumed that the element i was created if the condition
P
i
was satisfied k¼1pk 6
c
.
5. Points 3 and 4 are repeated until the final product is created.
6. The reaction is monitored with the probability assignment
scheme.
Table 2
Homolytical rearrangement of long-chain 1-alkyl radicals.
7. The final outcome is the occurrence of any of the final products,
Ai.
CH3(CH2)(p+q+1).CH ? CH3(CH2)pꢀCH(CH2)qCH3
Radicals
Type of hydrogen atom abstraction
8. When it comes to the final product, element 1a is generated
again and so on with N replications, and the final product is
counted, Ai = Ai + 1.
1,5-
1,6-
1,7-
1-tridecyl
1-decyla
1-octyl
p = 7
q = 3
p = 4
q = 3
p = 2
q = 3
p = 6
q = 4
p = 3
q = 4
p = 1
q = 4
p = 5
q = 5
p = 2
q = 5
p = 0
q = 5
After N repetitions valid response is
RAi = N. Fractions of the fi-
nal products of fi = Ai/N are compared with our experimental data
(Experimental Section 2.2) and thus for all possible sets of proba-
bilities pij. Finally, the probability pij is chosen so that the best frac-
tions obtained agree with experimental data.
a
5-Decyl radical is formed by 1,5- and also by 1,6-hydrogen rearrangements of 1-
decyl radical.
3. Results and discussion
Table 3
Oxidative degradation of alkanoic acids by means of LTA under
thermal conditions (81 °C) in benzene gives products as shown in
Table 1. Beside these products, in this reaction carboxylates are
also obtained. However, they are not analyzed by gas chromatogra-
phy because of their long retention time. Distribution of the ob-
tained alkanes, as well as acetoxyalkanes and alkenes, indicates
the significance of homolytical rearrangements of 1-alkyl radical
(Table 2).
Tetradecanoic acid gives 1-tridecyl radical as intermediate. Pri-
mary 1-trydecyl radical isomerizes by 1,5- and 1,6-homolytical
rearrangement to more stable secondary 5- and 6-tridecyl radicals.
1-Tridecyl radical can be oxidized to 1-tridecyl cation, which can
undergo 1,2-hydride shift isomerisation [22]. Beside oxidation of
1-tridecyl radical, there is a possibility that the rearranged tridecyl
radicals can undergo oxidation and hydride shift isomerisation.
One stabilization pathway of unrearranged tridecyl radical, used
here as example, is shown in Scheme 2.
Relative yields (%) of acetoxy alkane products obtained by oxidative thermal LTA
decarboxylation of nonanoic (n = 8), undecanoic (n = 10) and tetradecanoic (n = 13)
acid.
CnH2n+1OAca
n = 8
n = 10
n = 13
1-
2-
3-
4-
5-
6- + 7-
17.40
9.00
19.70
53.90
7.83
1.11
0.41
1.10
89.55
6.00
0.60
0.70
1.80
65.60
25.30
a
These numbers represent normalized values to 100%, overall yields of isomers
are given in Table 1.
In this paper, particular attention was paid to the product distri-
bution of isomeric alkenes and acetoxy alkanes in order to deter-
mine the involvements of 1,5- and 1,6-homolytic rearrangement
of hydrogen in 1-tridecyl, 1-decyl and specially 1-octyl radical (Ta-
bles 3 and 4). Besides, it was interesting to estimate also the
involvement of 1,7-rearrangement in these 1-alkyl radicals.
It has been considered that formation of acetoxy alkanes (Ta-
ble 3) goes through acetylation of rearranged and unrearranged al-
kyl cations [22]. The formation of 7-tridecyl cation goes mainly
through 6,7-hydride shift in 6-tridecyl cation. An interesting
question here is if 7-tridecyl radical obtained by 1,7-homolytical
rearrangement of 1-tridecyl radical is also a precursor of 7-tride-
cyl-cation. It has to be pointed out that formation of the 7-radical
happens with less probability due to enthropic factor, in compari-
son with 1,5- and 1,6-hydrogen atom abstraction processes.
In Table 3 yields of 6- and 7-acetoxytridecane are given overall
because chromatographic conditions for the separation of 6- and 7-
isomers unfortunately have not been found.
Table 4
Relative yields (%) of alkene products obtained by oxidative thermal LTA decarbox-
ylation of nonanoic (n = 8), undecanoic (n = 10) and tetradecanoic (n = 13) acid.
a
CnH2n
n = 8
n = 10
n = 13
1-
2.80
4.83
18.59
44.42
8.90
0.83
1.39
0.17
0.29
11.40
33.37
10.01
33.65
–
16.74
0.26
0.51
0.43
0.86
7.21
22.40
9.45
31.16
2.49
8.50
(Z)-2-
(E)-2-
(Z)-3-b
(E)-3-b
(Z)-4-
(E)-4-
(Z)-5-
(E)-5-
(Z)-6-
(E)-6-
6.51
22.86
–
–
–
–
–
a
These numbers represent normalized values to 100%, overall yields of isomers
are given in Table 1.
b
(Z)- and (E)-3-octenes are partially separated. Roughly calculated ratio: 1:3.

´
22
A.V. Teodorovic et al. / Journal of Molecular Structure 1040 (2013) 19–24
Scheme 2. One stabilization pathway: oxidation of unrearranged and 1,5-rearranged tridecyl radical to corresponding cation and its hydride shift isomerisation.
Accordingly, 1-decyl radical undergoes the similar way of sta-
that numbering of carbon chain is started from the primary carbon
atom previously carrying radical center. Unfortunately, the system
is not labeled. Identical products from different pathways of stabil-
ization are hardly distinguished, e.g. 5-octyl cation 5b gives
4-acetoxyoctane 4e in an unlabeled system.
Since in all LTA reactions toluene is present in reaction mixtures
as a result of decarboxylation of acetic acid itself, methyl radical
might contribute ratio of octyl radicals. Although, its contribution
is expected to be in less extent, still it was taking into account
(see Scheme 3, path presented by interrupted line).
bilization, giving isomeric acetoxydecanes (Table 3) and isomeric
decenes (Table 4). The positions 5 and 6 are equivalent in unla-
beled decyl system. Consequently, 1,5- and 1,6-rearrangement of
primary decyl radical are not distinguishable. Therefore, the per-
centage yield of 5-acetoxydecane is comparable to the sum of per-
centage yield of 5-, 6- (and possibly existing 7-acetoxytridecane).
The formation of 4-acetoxydecane is interesting, concerning the
possibility of 1,7-homolytical rearrangement of 1-decyl radical, be-
side the process of hydride-type rearrangement (from 5-decyl cat-
ion mainly, and to a less extent from 1-decyl cation by successive
1,2-, 2,3-, 3,4-Hꢁ transfers).
Decarboxylation of nonanoic acid gives 1-octyl radical 1a
(Scheme 3) which partially undergoes to corresponding cation
1b. The radical isomerizes to form more stable secondary 5- and
6-octyl radicals 5a and 6a by 1,5- and 1,6-hydrogen atom abstrac-
tion, respectively (5aꢂ4a and 6aꢂ3a in our system which is
unlabeled).
The estimation of involvement of 1,7-hydrogen rearrangement
in unlabeled substrate is complicated because the product of this
process is 7-octyl radical 7a which may be easily oxidized to
7-octyl cation 7b (7bꢂ2b), but also could be formed via 1,2-hy-
dride shift of 1b which is analog with process shown in Scheme 2.
All octyl radicals are also subjected to oxidation to give isomeric
octyl cations (Table 2).
Octyl radicals react also with benzene resulting in formation of
phenyloctanes. The formation of 1-phenyloctane might be ex-
plained by oxidation of free radical intermediate II [8] (Scheme 4).
From other phenyloctanes only 2-phenyloctane is detected (ꢃ2%
compared to 1-phenyloctane).
It is interesting that the percentage yield of 2-acetoxyoctane 2e
as well as 2-octenes 2f, formed under thermolysis of lead(IV) non-
anoate, is notably higher in comparison with analogous products
obtained from undecanoic and tetradecanoic substrates. At first
sight, it seems that 1,7-hydrogen atom abstraction in 1a has
significant contribution to product distribution. More acceptable
explanation is as follows: 6-octyl radical 6a (6aꢂ3a), obtained by
1,6-hydrogen rearrangement, is primarily oxidized to 6-octyl cat-
ion 6b (6bꢂ3b), following with 3,2-hydride shift and resulting in
formation of more stable 7b (7bꢂ2b) (Scheme 3). This mechanistic
proposal is supported by higher expected ratio of 4-acetoxy/3-
acetoxyoctane than the one of 5-acetoxy/6-acetoxytridecane.
The stabilization pathway of 1a is shown in Scheme 3. Interme-
diates and stable products nomenclature were taken in such way
Monte Carlo method is useful for quantitative estimation of the
products and intermediates which is not distinguishable by an
experiment (e.g. 2e and 7e, 3e and 6e, 4e and 5e).
Algorithm and program in this paper are easily adaptable for
different carbon chain lengths, thus including all the processes in
the given reaction. Using MC simulation it is possible to efficiently
and quickly (much faster than the time needed for isotopic marked
substrates preparation and analyze more complex mixture of the
products) investigate the impact of some factors on the concentra-
tion of intermediates and final products, as well as to evaluate the
stability and relationship between all the products.
To estimate relative rates of 1,5-, 1,6- and possibly existing 1,7-
homolytic transfer of hydrogen in 1a, the product analysis and
product distribution were performed by MC simulation method.
The final yields of the stable products or their fractions were deter-
mined experimentally. As the carbon skeleton was not labeled,
some intermediates and products are the same, such as 2b and
7b, 3b and 6b, 4b and 5b.
Scheme 3 presents 1a stabilization processes. All intermediates
branch to the next, previous one (except to 1a) or end product.
These branches of intermediates have fractions. The condition that
the sum of all fractions from an intermediate equals to 1 must be
satisfied.
Therefore, these fractions can be treated as probability of form-
ing an intermediate from another one, or some final product.
Let us separate and analyze ‘‘element 3b’’ and present it in
Scheme 5. From Schemes 3 and 5 it can be seen that element 3b
arises from elements 2b, 4b and 3a, while the considered element
forms elements 2f, 3f, 4b, 3e and 2b.
Input arrows show forming of ‘‘element 3b’’, while output ar-
rows show forming of its products. Corresponding forming proba-
bilities are presented with pij.
If element 3b was occurred, then elements 2f, 3f, 4b, 3e and 2b
are formed with corresponding probabilities. The probability that

´
23
A.V. Teodorovic et al. / Journal of Molecular Structure 1040 (2013) 19–24
Scheme 3. Monte Carlo (MC) simulation of octyl radical stabilization processes.
Scheme 4. Formation of 1-phenyloctane.
element 2f arises from element 3b is p3b2f. All probabilities which
describe element 3b were presented in Scheme 5, where following
expression is valid p3b2f + p3b3f + p3b2b + p3b3e + p3b4b = 1.
Thus, the idea of applying MC method for simulation the reac-
tion of octyl radical or octyl cation isomerization was clear. For that
reason, a program code was written to simulate the given transfer
probabilities of all ‘‘intermediates’’ in the proposed simple model
reaction.
products (e.g. 2e plus 7e = 2.39 what is in very good agreement
with experimental value of 2.32 for 2e, but not in the case of 2f;
2f plus 7f = 3.33 while experimental value is only 1.93).
First, 1,5- and 1,6-homolytic rearrangements appeared to be the
most interesting process. 5- and 6-Octyl radicals, 5a and 6a, gener-
ated in this way undergo further oxidation, giving 5- and 6-octyl
cations 5b and 6b. 5- and 6-Acetoxyoctanes (5e and 6e), as well
as (Z/E)-4-, 5- and 6-octenes 4f, 5f and 6f are the products of their
final stabilization (appeared as 4f, 3f and 2f). It is supposed that the
ratio of 1,5- and 1,6-transposition in 1a is very similar to corre-
sponding rearrangement in 1-tridecyl radical.
Considering the values obtained by MC simulation, it is impor-
tant to point out that ever experimental value has to be equal to
the sum of MC simulated values concerning the pseudoisomeric

´
A.V. Teodorovic et al. / Journal of Molecular Structure 1040 (2013) 19–24
24
The 1-octyl radical represents an interesting intermediate, par-
ticularly due to the increased possibility for its transformation into
relative stable 2-octyl systems, in comparison with corresponding
1-decyl and 1-tridecyl system. The relationships between 1,5-, 1,6-
and possibly existing 1,7-homolytic hydrogen transfer in 1-octyl-
radical, as well as successive 1,2-hydride shift in corresponding
cation have been computed via Monte-Carlo method. Taking into
account that ratios of 1,5-/1,6-homolytic rearrangements in
1-octyl- and 1-tridecyl radical are approximately the same, the
simulation shows very low involvement of 1,7-hydrogen
rearrangement (1,5-/1,6-/1,7-hydrogen rearrangement = 85:31:1)
in 1-octyl radical.
Acknowledgements
The research was supported by the Ministry of Education and
Science of the Republic of Serbia (Project No. 171021).
Scheme 5. Element 3b.
We wish to thank Professor Ladislav Soják and Professor Ivan
´
Ostrovsky from Chemical Institute, Commenius University for their
contribution for GC analysis.
Octane 1d is also a product of LTA oxidative decarboxylation of
nonanoic acid, rationalized in terms of intermolecular hydrogen
abstraction. This scheme is simplified and does not include the pro-
cess of 1a disproportionation because of the restriction imposed by
the nature of MC simulation applied in this experiment in order to
distinguish different pathways in obtaining the above mentioned
products.
References
[1] E.T. Denisov, G.T. Denisova, Russ. Chem. Rev. 73 (2004) 1091.
ˇ
ˇ
´
[2] Z. Cekovic, J. Serb. Chem. Soc. 70 (2005) 287.
[3] (a) M. Journet, M. Malacria, Tetrahedron Lett. 33 (1992) 1893;
(b) M. Gulea, J.M. López-Romero, L. Fensterbank, M. Malacria, Org. Lett. 2
(2000) 2591.
Primary 1a is also the subject of oxidative transformation to ob-
tain the 1b, which is partially decomposed by 1,2-successive hy-
dride shift. Besides, 1e and 1f which belong to non-rearranged
products, there are also additional acetoxy alkanes (2e and 3e)
and alkenes ((Z/E)-2f, (Z/E)-3f) as rearranged products.
It is very interesting to analyze the existing ‘‘7-octyl radical’’ 7a
resulting from 1,7-homolytical rearrangement of 1a. Unfortu-
nately, 1,2-hydride shift in the 1-octyl cation is probably the main
contributor for its formations. Therefore, determination of contri-
bution of 1,7-homolytic rearrangement is difficult task in this
length of the carbon chain. However, MC simulation of stabiliza-
tion of the 1-octyl radical enables estimation of relative ratios of
1,5-, 1,6- and 1,7-hydrogen abstraction (1,5-/1,6-/1,7-hydrogen
abstraction = 85:31:1).
[4] G.T. Denisova, E.T. Denisov, Kinet. Catal. 42 (2001) 620.
[5] P.C. Van Dort, P.L. Fuchs, J. Org. Chem. 62 (1997) 7142.
[6] J.Y. Nedelec, D. Lefort, Tetrahedron 31 (1975) 411.
ˇ
ˇ
´
´
[7] M.Lj. Mihailovic, Z. Cekovic, Lj. Lorenc, in: W.J. Mijs, C.R.H.I. De Jonge (Eds.),
Organic Synthesis By Oxidations With Metal Compounds, Plenum Press, New
York and London, 1986, pp. 741–816.
[8] J.K. Kochi, J. Am. Chem. Soc. 87 (1965) 3609.
[9] R.M. Southam, M.C. Whiting, J. Chem. Soc. Perkin Trans. II (1982) 597.
[10] S. Kitamura, T. Minami, Y. Nakamura, H. Isuda, H. Kobayashi, M. Mimura, K.
Kajiwara, S. Ohno, J. Mol. Struct.: THEOCHEM 425 (1997) 395.
[11] L. von Szentpály, I.L. Shamovsky, V.V. Nefedova, V.E. Zubkus, J. Mol. Struct. 308
(1994) 125.
[12] J.L. Holmes, K. Cartledge, A.D. Osborne, Int. J. Mass Spectrom. 29 (1979) 171.
[13] M. Saunders, S.P. Budiansky, Tetrahedron 35 (1979) 929.
[14] K.B. Borisenko, I. Hargittai, J. Mol. Struct. 376 (1996) 195.
[15] A. O’dea, A. Curtis, N. Green, C. Timmel, P. Hore, J. Phys. Chem. A 109 (2005)
869.
[16] T. Miura, H. Murai, J. Phys. Chem. A 112 (2008) 2526.
[17] J. Woodward, C. Vink, Phys. Chem. Chem. Phys. 9 (2007) 6272.
[18] U. Steiner, J.Q. Wu, Chem. Phys. 162 (1992) 53.
4. Conclusions
´
´
´
[19] (a) V. Andrejevic, A. Teodorovic, M. Lj. Mihailovic, Abstract of Papers, 29th
IUPAC CONGRES, Cologne, Germany, June 5–10, 1983, Gesellschaft Deutscher
Chemiker, p. 120.;
By using one simple scheme of stabilization of 1-octyl radical
obtained by LTA decarboxylation of nonanic acid, and MC method,
it was possible to determine the relative ratios of different type of
its stabilization. This methodology differentiates isomeric products
in this unlabeled system where some intermediates as well as
products are the same even they were obtained from different sta-
bilization pathways. Input data for MC simulation were based on
the relative ratios of the specific products obtained experimentally.
(b) A.V. Teodorovic´, D.M. Badjuk, N. Stevanovic´, Delegate Manual, Eleventh
Tetrahedron Symposium, Frontiers of Organic Chemistry, Beijing, China, June
22–25, 2010, ELSEVIER, PSB, p. 78.;
(c) M. Lj. Mihailovic´, V. Andrejevic´, A.V. Teodorovic´, Abstract of Papers, Part II,
7th Yugoslav Congress on Chemistry and Chemical Technology, Novi Sad,
Serbia, 1983, Yugoslav Chemical Society, pp. 54–55.
[20] L. Sojak, I. Ostrovski, J. Chromatogr. 356 (1986) 105.
ˇ
ˇ
´
´
[21] M.Lj. Mihailovic, J. Bošnjak, Z. Cekovic, Helv. Chim. Acta 57 (1974) 1015.
[22] E.J. Corey, J. Casanova, J. Am. Chem. Soc. 85 (1963) 165.