Rate constants and Arrhenius functions for ring opening of a cyclobutylcarbinyl radical clock and for hydrogen atom transfer from the Et 3B-MeOH complex

Article abstract of DOI:10.1021/jo800500e

(Chemical Equation Presented) Kinetics of ring opening of the 1-cyclobutyldodecyl radical (1) were studied, and an Arrhenius function over the temperature range -20 to 47°C was determined. The radical clock reaction has kinetics described by log k = 13.2-13.5/2.313RT (in kcal/mol), and k = 1.5 × 10³ s^-1 at 20°C. Previous kinetic studies of hydrogen atom transfer trapping of radical 1 by the triethylborane-methanol complex at variable temperatures (J. Org. Chem. 2007, 72, 5098) were analyzed with the newly obtained kinetic data.

Full text of DOI:10.1021/jo800500e

SCHEME 1
Rate Constants and Arrhenius Functions for
Ring Opening of a Cyclobutylcarbinyl Radical
Clock and for Hydrogen Atom Transfer From
the Et3B-MeOH Complex
Jing Jin and Martin Newcomb*
Department of Chemistry, UniVersity of Illinois at Chicago,
845 West Taylor Street, Chicago, Illinois 60607
radicals,1 but only limited kinetic studies6,7 and computational
studies8 of cyclobutylcarbinyl radical ring openings are available.
Recently, we used the ring-opening reaction of the 1-cyclobu-
tyldodecyl radical (1) to the trans-4-hexadecenyl radical (2)
(Scheme 1)9 in kinetic studies of hydrogen atom transfer
trappings by water and methanol complexes of triethylborane.10
In that work,9 radical 1 was produced from xanthate ester radical
precursor 3, and the rate constant for ring opening of 1 at
ambient temperature was estimated from previous competition
kinetic studies of the cyclobutylcarbinyl radical ring opening.6
In this contribution, we report production of radical 1 from a
second radical precursor, PTOC ester 4,11,12 and kinetic calibra-
tion of the ring opening of 1 over a temperature range of -30
to 40 °C.
PTOC esters are incompatible with Et3B, which precluded
the use of radical precursor 4 in our earlier study.9 Nonetheless,
PTOC ester radical precursors are generally more versatile than
xanthate esters. The xanthates are limited for practical purposes
to precursors for secondary alkyl radicals, and even then can
react inefficiently. Moreover, radical chain initiation with PTOC
esters can be accomplished with visible light irradiation if
desired, which is a significant advantage for low-temperature
studies that we exploited in this work.12b Thus, as a tool for
most radical kinetic studies, PTOC ester 4 will be superior to
xanthate ester 3.
men@uic.edu
ReceiVed March 6, 2008
Kinetics of ring opening of the 1-cyclobutyldodecyl radical
(1) were studied, and an Arrhenius function over the
temperature range -20 to 47 °C was determined. The radical
clock reaction has kinetics described by log k ) 13.2 - 13.5/
2.313RT (in kcal/mol), and k ) 1.5 × 103 s-1 at 20 °C.
Previous kinetic studies of hydrogen atom transfer trapping
of radical 1 by the triethylborane-methanol complex at
variable temperatures (J. Org. Chem. 2007, 72, 5098) were
analyzed with the newly obtained kinetic data.
Synthetic applications of radical chemistry generally involve
chain reactions where the radical conversions of interest compete
with other radical processes, and knowledge of rate constants
for radical reactions is essential for synthetic planning. Although
absolute rate constants for several radical reactions have been
measured directly, most of the radical kinetics information has
been determined via indirect kinetic studies,1 and a major portion
of that data involves competition reactions with ”radical clocks”,
which are calibrated unimolecular radical rearrangements.2
Carbon-centered radical clocks span a wide range of reactivity,
with clock lifetimes from picoseconds to seconds.1 The most
popular clocks are based on the 5-hexenyl radical cyclization
(k ) 2 × 105 s-1 at 20 °C)3,4 and the cyclopropylcarbinyl radical
ring opening (k ) 7 × 107 s-1 at 20 °C).5
The preparation of radical precursor 4 followed standard
methods. In brief, cyclobutylmethanol was converted to its
mesylate derivative, which was used for the preparation of
cyclobutylacetonitrile. Alkylation of this nitrile with 1-bro-
moundecane gave 2-cyclobutyltridecanecarbonitrile. Hydrolysis
(6) Beckwith, A. L. J.; Moad, G. J. Chem. Soc., Perkin Trans. 2 1980, 1083–
1092.
(7) (a) Ingold, K. U.; Maillard, B.; Walton, J. C. J. Chem. Soc., Perkin Trans.
2 1981, 970–974. (b) Walton, J. C. J. Chem. Soc., Perkin Trans. 2 1989, 173–
177. (c) Newcomb, M.; Horner, J. H.; Emanuel, C. J. J. Am. Chem. Soc. 1997,
119, 7147–7148. (d) Choi, S. Y.; Horner, J. H.; Newcomb, M. J. Org. Chem.
2000, 65, 4447–4449. (e) Emanuel, C. J.; Horner, J. H.; Newcomb, M. J. Phys.
Org. Chem. 2000, 13, 688–692.
Among the slower radical clocks, ring openings of cyclobu-
tylcarbinyl radicals to 4-pentenyl radicals are noteworthy. At
room temperature, these unimolecular reactions are about 2
orders of magnitude less rapid than cyclizations of 5-hexenyl
(8) (a) Baker, J. M.; Dolbier, W. R. J. Org. Chem. 2001, 66, 2662–2666. (b)
Shi, J.; Chong, S.-S.; Fu, Y.; Guo, Q.-X.; Liu, L. J. Org. Chem. 2008, 73, 974–
982.
(9) Jin, J.; Newcomb, M. J. Org. Chem. 2007, 72, 5098–5103.
(10) (a) Spiegel, D. A.; Wiberg, K. B.; Schacherer, L. N.; Medeiros, M. R.;
Wood, J. L. J. Am. Chem. Soc. 2005, 127, 12513–12515. (b) Medeiros, M. R.;
Schacherer, L. N.; Spiegel, D. A.; Wood, J. L. Org. Lett. 2007, 9, 4427–4429.
(11) The acronym PTOC is derived from pyridine-2-thioneoxycarbonyl.
PTOC esters, which are anhydrides of a carboxylic acid and the thiohydroxamic
acid N-hydroxypyridine-2-thione, were developed by Barton’s group for synthetic
applications (see ref 12a).
(12) (a) Barton, D. H. R.; Crich, D.; Motherwell, W. B. Tetrahedron 1985,
41, 3901–3924. (b) Newcomb, M.; Park, S.-U. J. Am. Chem. Soc. 1986, 108,
4132–4134.
(1) Newcomb, M. Tetrahedron 1993, 49, 1151–1176.
(2) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317–323.
(3) Walling, C.; Cioffari, A. J. Am. Chem. Soc. 1972, 94, 6059–6064.
(4) (a) Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J. C. J. Am. Chem. Soc.
1981, 103, 7739–7742. (b) Chatgilialoglu, C.; Newcomb, M. AdV. Organomet.
Chem. 1999, 44, 67–112.
(5) (a) Maillard, B.; Forrest, D.; Ingold, K. U. J. Am. Chem. Soc. 1976, 98,
7024–7026. (b) Newcomb, M.; Glenn, A. G. J. Am. Chem. Soc. 1989, 111, 275–
277. (c) Newcomb, M.; Choi, S. Y.; Horner, J. H. J. Org. Chem. 1999, 64,
1225–1231.
4740 J. Org. Chem. 2008, 73, 4740–4742
10.1021/jo800500e CCC: $40.75 2008 American Chemical Society
Published on Web 05/15/2008

SCHEME 2
FIGURE 1. Kinetics of radical 1 ring openings. (A) Product ratios as
a function of tin hydride concentration for reactions at 18.5 °C. (B)
Relative Arrhenius function for ring opening and tin hydride trapping
of radical 1.
of the latter nitrile gave 2-cyclobutyltridecanoic acid, which was
converted to PTOC ester 4 via the acid chloride that reacted
with the sodium salt of N-hydroxypyrindine-2-thione. Radical
precursor 4 was characterized by NMR spectroscopy and high
resolution mass spectrometry, and the samples employed in
studies conducted at 18.5 °C are shown in Figure 1A. Ring-
opened product 8 was the minor product in all cases and
accounted for less than 2% of the product yield when 0.06 M
Bu3SnH was employed. Nonetheless, the plot of [7]/[8] versus
the initial concentration of Bu3SnH gave a good line. In these
experiments, the ratio of rate constants is given by eq 1,1 and
the slope of the line in Figure 1A is kH/kr ) 860 ( 20 M-1
(1σ). Using a rate constant for reaction of tin hydride with a
secondary alkyl radical at 18.5 °C of kH ) 1.32 × 106 M-1
s-1,4 one calculates a rate constant for ring opening of 1 at 18.5
°C of kr ) 1.5 × 103 s-1. This value is about 1.5 times as great
as the value previously estimated for the ring-opening reaction
of radical 1 at 20 °C.9 One might have thought that the estimated
value would be more accurate, but the literature values for
kinetics of ring opening of cyclobutylcarbinyl radicals are
somewhat varied.14
1
kinetic studies were >98% pure as determined by H NMR
spectroscopy.
PTOC esters react in radical chain reactions that can be
initiated either thermally or photochemically with visible light
irradiation to give the sequence of reactions shown in Scheme
2.12 Reaction of 4 in either an initiation reaction or a chain
propagation reaction gives acyloxyl radical 5, which rapidly
decarboxylates to give the desired radical 1. Byproduct radical
5 is also formed in the initiation reactions. In the indirect kinetic
study in this work, hydrogen atom trapping of 1 by tin hydride
to give product 7 competes with ring opening, and the ring-
opened radical 2 also will react with tin hydride to give acyclic
product 8; previous studies established that the ring-opening
reaction ultimately gives trans-4-hexadecene as the major
product.9 The stannyl radical formed in the trapping reactions
reacts with another molecule of precursor 4 in a chain propaga-
tion step.
A number of hydrogen atom transfer trapping agents can be
used to trap the alkyl radical when PTOC esters are the
precursors,12 and Bu3SnH provides a well-calibrated rate
constant for the competition reaction.4 PTOC esters also react
in “self-trapping” reactions, reaction of a carbon radical
intermediate with the thione group of a second molecule of
PTOC ester to give an alkyl 2-pyridinyl sulfide product, and
the rate constant at room temperature for this self-trapping
reaction is similar to the rate constant for reaction of an alkyl
radical with Bu3SnH.13 Therefore, when tin hydride is present
in large excess to the PTOC ester radical precursor, the self-
trapping reaction is not important.
[7]/[8] ) (kH/kr)[Bu3SnH]
(1)
(2)
log(kr/kH) ) log{[8][Bu3SnH])/[7]}
An Arrhenius function for reaction of radical 1 was deter-
mined by conducting a series of reactions of precursor 4 (ca.
0.002 M) with Bu3SnH (ca. 0.02 M) at temperatures between
-20 and 47 °C. The expression for analysis of the data is given
in eq 2, which is the logarithmic form of eq 1 after rearrange-
ment. A plot of the log of the kinetic ratio against the inverse
of temperature gave the relative Arrhenius function shown in
Figure 1B, which gives log(kr/kH) ) (4.52 ( 0.12) - (10.02 (
0.16)/2.313RT (errors at 1σ). Combination of these values with
the Arrhenius function for reaction of Bu3SnH with a secondary
alkyl radical15 gives an Arrhenius function for the ring opening
of radical 1 in eq 3, where the errors are at 1σ. This Arrhenius
function predicts a rate constant for ring opening of 1 at 20 °C
In competition kinetic studies, PTOC ester radical precursor
4 (0.002 M) in toluene with excess Bu3SnH was thermally
equilibrated, and reactions were initiated by visible light
irradiation. Following the reactions, an internal standard was
added, and the product mixtures were analyzed by GC to
determine the yields of products 7 and 8. The total yields of 7
and 8 as determined by GC analysis using an internal standard
typically were greater than 90%. The results from a series of
(14) From the original product ratios reported by Beckwith and Moad for
reaction of the cyclobutylcarbinyl radical with tin hydride (ref 6) and the
recommended Arrhenius function for reaction of a primary alkyl radical with
Bu3SnH (ref 4b), one obtains an Arrhenius function for ring opening of the
cyclobutylcarbinyl radical of log k ) 12.9-13.1/2.313RT. When low-temperature
ESR data were included; however, the reported Arrhenius function was log k )
12.6-12.2/2.313RT (ref 7b).
(15) The Arrhenius function for reaction of the 2-propyl radical with Bu3SnH
is log k ) (8.71 ( 0.37)-(3.47 ( 0.49)/2.313RT (see ref 4b).
(13) Newcomb, M.; Kaplan, J. Tetrahedron Lett. 1987, 28, 1615–1618.
J. Org. Chem. Vol. 73, No. 12, 2008 4741

of kr ) 1.5 × 103 s-1, in good agreement with the results from
the 18.5 °C studies discussed above.
reactions of Et3B complexes, but one point is seemingly
undeniable. The hydrogen atom transfer reaction of the Et3B-
CH3OH complex is increasingly competitive at reduced tem-
peratures in comparison to “typical” radical reactions. That is,
the H-atom transfer reaction might be “relatively slow” at room
temperature, but it becomes “moderately fast” at low temper-
atures such as -30 °C. This conclusion, if it proves to be general
for hydrogen atom transfer reactions of Et3B-water and
-alcohol complexes, has important synthetic ramifications
because it suggests that product distributions might be dramati-
cally altered by changing the reaction temperature instead of
the reagent concentrations.
log kr ) (13.2 ( 0.4) - (13.5 ( 0.6)/2.313RT
(3)
The log A term in the Arrhenius function for ring opening of
radical 1 is consistent with expectations. Due to the rigidity of
the system, the reaction should display little entropy demand
in the transition state (i.e., ∆S‡ ≈ 0), which would give a
predicted log A term of 13.1 at room temperature. Correcting
for the degenerate ring opening paths for radical 1 gives log A
) 12.9 for cleavage of one bond in radical 1. Also as expected,
both terms in the Arrhenius function for ring opening of radical
1 are similar to those for the cyclobutylcarbinyl radical,14 and
one notes that the more appropriate model, the 1-cyclobutylethyl
radical, was reported to react slightly slower than the cyclobu-
tylcarbinyl radical.6
The ring opening of radical 1 was initially used in a study of
the rate constants for hydrogen atom trapping of the alkyl radical
by the triethylborane complexes of water and methanol,9
reactions representative of a general class of radical trapping
reactions that can be synthetically important.10,16 The kinetic
results of the present work indicate that the rate constants for
the hydrogen atom transfer trapping reactions by these com-
plexes are about 1.5 times as great as previously reported. The
kinetic difference will not affect synthetic planning greatly, but
the more important ramification is that these results provide an
Arrhenius function for ring opening of radical 1 that is reliable
at low temperatures. This, in turn, can be combined with the
relative Arrhenius function for reaction of Et3B-CH3OH with
this radical that was previously determined.9
The resulting Arrhenius function for reaction of the
Et3B-CH3OH complex with a secondary alkyl radical is given
in eq 4. Due to accumulated errors in the previous9 and present
relative Arrhenius function determinations, the precision for a
rate constant at a given temperature calculated from eq 4 might
be poor, with an error of a factor of 2 or even greater, but the
value of the log A term is the more important point. The pre-
exponential factor in the Arrhenius function is quite small,
indicating either a highly organized transition state for the
hydrogen atom transfer reaction or an entropically unfavorable
prior equilibration before the rate-determining step in the
process, which we assume would be the actual hydrogen atom
transfer step.
The ring opening of radical 1 can now be used for calibration
of slow radical reactions using “radical clock” indirect kinetic
methods.1 Synthetic methods for the preparations of two
precursors for radical 1, the xanthate ester 3 and the PTOC ester
4, have been reported, and alkyl halide precursors for 1 are
expected to be readily available. The long alkyl side chain in
the clock results in high molecular weight products that do not
evaporate under typical workup conditions and are well resolved
from solvents when analyzed by GC.
Experimental Section
The preparation of 1-cyclobutyltridecanoic acid and PTOC ester
4 is described in the Supporting Information.
Kinetic Studies. Toluene solutions of PTOC ester 4, ca. 0.002
M, and Bu3SnH at the desired concentration were deoxygenated
by bubbling nitrogen through the solutions and equilibrated in a
temperature-regulated bath while shielded from light. The stirred
solutions were irradiated with visible light to initiate the radical
chain reactions. After ca. 30 min, the yellow color of the initial
mixtures was dissipated, and the mixtures were evaluated by thin
layer chromatography to ensure that the reactions were complete.
An internal standard of dodecane was added. The reaction mixture
was washed with 3 M aqueous sodium hydroxide solution, dried
over MgSO4, and filtered through a short pad of silica gel with
hexane elution. The eluent was analyzed by GC with TCD detection
using a low-polarity column (bonded phase DB-5, 15 m × 0.53
mm). 1-Cyclobutyldodecane and trans-4-hexadecane were identified
by GC and GC-mass spectrometry as previously reported.9
Acknowledgment. This work was supported by a grant from
the National Science Foundation (CHE-0601857).
log kH ) (7.2) - (4.7)/2.3RT
(4)
Supporting Information Available: Synthetic details for
preparation of PTOC ester 4 and NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
More careful studies at variable temperatures are necessary
for detailed mechanistic conclusions about H-atom transfer
(16) Pozzi, D.; Scanlan, E. M.; Renaud, P. J. Am. Chem. Soc. 2005, 127,
14204–14205.
JO800500E
4742 J. Org. Chem. Vol. 73, No. 12, 2008