C-H bond functionalization with the formation of a C-C bond: A free radical condensation reaction based on the phthalimido-N-oxyl radical

Article abstract of DOI:10.1002/ejoc.201301530

The development of a new chemical process that effects the conversion RH + C=C-C-X → R-C-C=C + HX, in which X is the phthalimido-N-oxyl radical (PINO^·), is reported. The reaction yields are high, mass balances are excellent, and C-H bond functionalization and C-C bond formation are achieved in a single transformation. The byproduct of the reaction, N-hydroxyphthalimide, precipitates from solution and can be easily removed by simple filtration (and recycled). The kinetic chain lengths are shorter and the reaction times are longer (relative to those of the analogous reactions of allyl bromides), most likely because PINO^· is a less-reactive hydrogen-atom abstractor. There appears to be no significant difference in efficiency in the addition-elimination steps. Competition experiments reveal that Br^· and PINO^· are comparable in leaving group ability. The introduction of a new chain carrier, the phthalimido-N-oxyl radical (PINO^·), leads to an improved chain reaction. This chain reaction is successful and high reaction yields are reported for the functionalization of hydrocarbons. Kinetic studies reveal that this reaction is an efficient chain process, and the leaving group ability of PINO ^· is comparable to that of Br^·. Copyright

Full text of DOI:10.1002/ejoc.201301530

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201301530
C–H Bond Functionalization with the Formation of a C–C Bond: A Free
Radical Condensation Reaction Based on the Phthalimido-N-oxyl Radical
Shradha Patil,[a] Liang Chen,[a] and James M. Tanko*[a]
Keywords: Radicals / C–H functionalization / Kinetics / Allylation / Condensation reactions
The development of a new chemical process that effects the
conversion RH + C=C–C–XǞR–C–C=C + HX, in which X
is the phthalimido-N-oxyl radical (PINO·), is reported. The
reaction yields are high, mass balances are excellent, and C–
H bond functionalization and C–C bond formation are
achieved in a single transformation. The byproduct of the re-
action, N-hydroxyphthalimide, precipitates from solution and
can be easily removed by simple filtration (and recycled).
The kinetic chain lengths are shorter and the reaction times
are longer (relative to those of the analogous reactions of allyl
bromides), most likely because PINO· is a less-reactive hy-
drogen-atom abstractor. There appears to be no significant
difference in efficiency in the addition–elimination steps.
Competition experiments reveal that Br· and PINO· are com-
parable in leaving group ability.
Introduction
C–H bond functionalization catalyzed by N-hydroxy-
phthalimide has received a great deal of attention in the
past decade as a simple, environmentally benign method to
oxidize organic compounds.[1] The key player in this chem-
istry is the phthalimido-N-oxyl radical (PINO·, 1), which
serves as a hydrogen-atom abstractor in these systems.[2]
PINO· is more reactive than dialkyl nitroxyl radicals such
as 2,2,6,6-tetramethylpiperidine N-oxide (TEMPO), be-
cause the lone pair of electrons on the nitrogen atom is tied
up in resonance with the carbonyl groups, and this dimin-
ishes the contribution of resonance form 2 (Figure 1) and
effectively localizes the spin density on the oxygen atom.[3]
As a hydrogen-atom abstractor, PINO· exhibits low reac-
tivity and high selectivity. For example, the relative reacti-
vity of the benzylic C–H bonds of cumene, ethyl benzene,
and toluene towards PINO· are 3° (26) Ͼ 2° (9) Ͼ 1° (1)
on a per hydrogen basis.[4] These values are remarkably sim-
ilar to the relative reactivity trend of these substrates
towards Br·: 3° (59) Ͼ 2° (25) Ͼ 1° (1).[5] It was this similar-
ity to Br· that led us to wonder whether the chemistry of
PINO· could be extended in such a way to achieve C–H
bond functionalization and C–C bond formation in a single
chemical process.
Figure 1. Some of the resonance contributors in the phthalimido-
N-oxyl radical.
leaving groups (halogen, SnR3, SiR3, SR, etc.) through an
addition–elimination process.[6] (These and related allyl
compounds are also useful as chain-transfer agents in free
radical polymerizations.)[7] In small molecule synthesis,
there has been particular interest, and success, in de-
veloping synthetically useful radical-based allyl transfer re-
actions that do not involve toxic metals such as tin and
mercury.[6c,8]
In 1999, we reported a bromine atom based method that
achieves hydrocarbon functionalization and allyl transfer in
a single step: R–H + CH2=C(Z)CH2BrǞRCH2C(Z)=CH2
+ HBr.[9] The mechanism for this reaction is summarized in
Scheme 1 (X = Br). The bromine atom abstracts a hydrogen
atom from a hydrocarbon such as toluene (which has rela-
tively weak, benzylic C–H bonds) to generate an intermedi-
ate benzyl radical R·. Addition of R· to the alkene generates
β-bromo radical 5, which undergoes β-cleavage to form the
final product with the regeneration of X·. The bromine
atom is the key to this reaction, because Br· is sufficiently
reactive that it can abstract a hydrogen atom with high
Carbon-centered radicals, typically generated from or-
ganic halides or organometallic compounds, can be all-
ylated by using allyl compounds with suitable homolytic
[a] Department of Chemistry, Virginia Polytechnic Institute and
State University,
Blacksburg, VA 24060, USA
E-mail: jtanko@vt.edu
https://www.chem.vt.edu
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201301530.
502
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 502–505

C–H Bond Functionalization with the Formation of a C–C Bond
selectivity yet sufficiently stable that β-bromo radicals read-
ily undergo fragmentation. This balance is not achieved
with other halogen atoms: Cl· and F· are not selective in H-
atom abstractions; I· is a good leaving group for β-fragmen-
tation, but it does not abstract hydrogen at a reasonable
rate. Consequently, because of the parallels between Br· and
PINO·, it seemed possible that a similar allyl transfer reac-
tion could be developed on the basis of PINO· chemistry, if
PINO· was a suitable leaving group for β-cleavage.
Figure 2. Substituted allyl-PINO substrates.
Table 1. Results for the reactions of 6a and 6b with hydrocarbons.[a]
[b]
PhCHR2
Z
Time [h]
% 7a or 7b % 6a or 6b
PhCH3
PhCH3
PhCH2CH3
PhCH2CH3
PhCH(CH3)2 Ph
Ph
CO2Et
Ph
24
42
29
17
29
24
77
48, 43[c]
89
85
76
23
21, 30[c]
9
7
22
9
CO2Et
Scheme 1. Propagation steps in the allyl transfer reaction.
PhCH(CH3)2 CO2Et
91
[a] Reactions performed at 120 °C by using di-tert-butyl peroxide
(DTBPO, 20 mol-%) as the initiator. [b] Neat. [c] Time: 24 h.
Overall, the reactions utilizing the allyl-PINO substrates
were considerably cleaner than the analogous reactions with
allyl bromides. This is illustrated by the results summarized
in Table 2. For each of these experiments, the reaction con-
ditions were identical with regard to time, temperature, and
so on; the difference was the use of an allyl bromide versus
an allyl-PINO substrate. Although the allyl bromides
tended to react faster under comparable conditions, the
mass balances were lower and undesirable side products
were formed.
Results and Discussion
In this manuscript, we report a newly developed free rad-
ical reaction that achieves all of these objectives, C–H bond
functionalization and C–C bond formation in a single reac-
tion on the basis of PINO· chemistry through a process that
proceeds by the same mechanism as that shown in
Scheme 1. This chemistry is compared and contrasted to
the analogous chemistry of Br· in terms of reaction yields,
mass balance, kinetic chain lengths, and leaving group ef-
fects (Br· vs. PINO·).
Table 2. Comparison of the reactions of allyl bromides and allyl-
PINO substrates with hydrocarbons.[a]
Generally, the “allyl-PINO” compounds that would be
needed as substrates for this new reaction would be synthe-
sized directly through the condensation of N-hydroxy-
phthalimide with the corresponding allyl alcohols.[10] For
our work, however, we elected to synthesize the allyl-PINO
substrates from the corresponding allyl bromides, because
the latter were also needed for comparison purposes in this
study. (Many of the allyl bromides are hygroscopic and thus
more difficult to handle.)
Allyl-PINO compounds 6a and 6b (Figure 2) were pre-
pared by following modified literature procedures[11] and al-
lowed to react with toluene, ethyl benzene, and cumene un-
der standard conditions (Table 1). Optimized product yields
ranged from 50 to 90%, and as the high mass balances
X = Br
X = PINO
PhCHR2
% 8a % 9a
Mass
balance [%]
% 8b % 9b
Mass
balance [%]
[b]
PhCH3
0
0
35
70
36
35
70
77
21
44
75
48
56
20
69
100
95
[c]
PhCH2CH3
PhCH(CH3)2[c]
41
[a] Reactions performed at 120 °C by using DTBPO (20 mol-%) as
the initiator in neat hydrocarbons. [b] Reaction time: 42 h. [c] Reac-
tion time: 3 h.
To probe this chemistry further, kinetic chain lengths
show, the reactions were exceptionally clean. Moreover, the {i.e., the rate of product formation relative to the rate of
obligatory reaction byproduct, N-hydroxyphthalimide initiator disappearance, –(Ѩ[product]/Ѩt)/(2Ѩ[In2]/Ѩt)} were
(PINOH), precipitated from solution and could be easily determined by following product yields as a function of
separated by filtration at the end of the reaction. (In con- time for Z = CO2Et as described previously.[9a] The initial
trast, the Br-based reaction produces HBr as a byproduct chain lengths for the allyl-PINO compounds were consist-
and requires the use of a base or epoxide as a scavenger.)
ently lower than those for the allyl bromides (Table 3), al-
Eur. J. Org. Chem. 2014, 502–505
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
503

S. Patil, L. Chen, J. M. Tanko
SHORT COMMUNICATION
·
though the reduced chain lengths for the allyl-PINO sub- Table 4. Relative rate constants for the reaction of PhCH2 with
various allyl bromides and allyl-PINOs.
strates did not have a deleterious effect on either the prod-
uct yield or mass balance for the reaction. The longer reac-
tion times and shorter kinetic chain lengths suggest that for
X = PINO, one of the propagation steps (hydrogen abstrac-
tion, addition, or β-fragmentation) may not be as efficient
as is the case for X = Br.
Table 3. Kinetic chain lengths for the reactions of hydrocarbons
with CH2=C(CO2Et)CH2X (X = Br, PINO).[a]
X
PhCH3
PhCH2CH3
PhCH(CH3)2
Br[b]
800
270
–
130
60
12
PINO[c]
[a] Reactions performed at 120 °C by using DTBPO (20 mol-%) as
initiator and neat hydrocarbon. [b] Ref.[7] [c] Current work with
same initiator loading.
Given that the elimination of PINO· by β-cleavage was
unprecedented, experiments were designed to determine the
relative leaving group ability of Br versus PINO. In prin-
ciple, this could be achieved through competition experi-
ments by pitting an allyl bromide versus an allyl-PINO for
the benzyl radical (Figure 3). As both substrates would pro-
duce the same organic product if Z1 = Z2, indirect competi-
tion experiments were undertaken by using various Z1, Z2,
and leaving group combinations. The results are summa-
rized in Table 4.
[a] Measured directly. [b] Calculated from other rate constant ratios
in this table.
DTBPO and conducted to low percent conversion to avoid
further reaction of the products) allowed the leaving group
abilities of PINO and Br to be directly determined:
kPINO/kBr = 3 (Ϯ1).
Figure 3. Competition experiment for the determination of rate
constants for the addition–elimination process in the allyl transfer
reaction. In any given competition Z1 ϶ Z2.
From the data in Table 4, the relative rate constants for
the addition–elimination of benzyl radical CH2=C(Z)CH2X
for the same Z substituent could be derived (last three en-
tries in Table 4). The relative rate constants for the allyl
bromides were only slightly higher than those for the allyl-
PINOs. Notably, the relative rate constants are for ad-
dition–elimination, and it is not known which of these steps
is rate limiting. Rather than differences in leaving group
ability, it may be that the allyl bromides are slightly more
reactive towards addition. Nonetheless, the results do not
suggest any major differences in reactivity between allyl-
Scheme 2. Experiment for relative leaving group abilities of PINO·
versus Br·.
The data in Table 4 also provide information about the
PINOs and allyl bromides in the addition–elimination se- relative rate of the reaction of the benzyl radical with the
quence that leads to product formation. allyl-PINO substrate as a function of Z: CO2Et (3.0) Ͼ CN
To address the intrinsic relative leaving group ability of (1.3) Ͼ Ph (1). These are comparable to allyl bromides: CN
Br· versus PINO· directly, compound 10 (Scheme 2), which (2.8) Ͼ CO2Et (1.6) Ͼ Ph (1.0). As noted in our earlier
·
incorporates both leaving groups into a single substrate, was work, these trends parallel rates of addition of PhCH2 to
synthesized. The reaction of 10 with toluene (initiated with alkenes,[12] which suggests the ordering reflects relative rates
504
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 502–505

C–H Bond Functionalization with the Formation of a C–C Bond
[2] E. Damiani, P. Astolfi, M. Benaglia, A. Alberti, L. Greci, Free
Radical Res. 2004, 38, 67.
of addition. In total, the results suggest that (1) PINO is a
slightly better leaving group than Br and (2) the allyl brom-
ides are slightly more reactive towards radical addition.
(One possibility that explains this is that the β-bromo alkyl
radicals are stabilized by β-bridging interactions).[13] How-
ever, these differences appear minor and for the most part
inconsequential.
It therefore seems most likely that the differences in reac-
tion times and chain lengths for the allyl-PINOs versus allyl
bromides are attributable to the hydrogen-atom abstraction
step. The absolute rate constant for hydrogen abstraction
from toluene by Br is 105 m–1 s–1.[5] In contrast, for PINO,
the absolute rate constant is several orders of magnitude
lower, 0.38 m–1 s–1.[4]
[3] a) R. Amorati, M. Lucarini, V. Mugnaini, G. F. Pedulli, F.
Minisci, F. Recupero, F. Fontana, P. Astolfi, L. Greci, J. Org.
Chem. 2003, 68, 1747; b) Y. Ishii, T. Iwahama, S. Sakaguchi,
K. Nakayama, Y. Nishiyama, J. Org. Chem. 1996, 61, 4520; c)
Y. Ishii, K. Nakayama, M. Takeno, S. Sakaguchi, T. Iwahama,
Y. Nishiyama, J. Org. Chem. 1995, 60, 3934; d) E. G. Rozatsev,
Free Nitroxyl Radicals (Ed.: H. Ulrich), Plenum Press, New
York, 1970; e) G. R. Eduard, V. D. Sholle, Russ. Chem. Rev.
1971, 40, 233; f) G. Moad, E. Rizzardo, S. H. Thang, Aust. J.
Chem. 2005, 58, 379.
[4] N. Koshino, Y. Cai, J. H. Espenson, J. Phys. Chem. A 2003,
107, 4262.
[5] J. M. Tanko, N. K. Suleman, J. F. Blackert, J. Org. Chem. 1991,
56, 6395.
[6] a) G. A. Russell, P. Ngoviwatchai, Y. W. Wu, J. Am. Chem. Soc.
1989, 111, 4921; b) D. L. J. Clive, C. C. Paul, Z. Wang, J. Org.
Chem. 1997, 62, 7028; c) F. L. Guyader, B. Quiclet-Sire, S. Seg-
uin, S. Z. Zard, J. Am. Chem. Soc. 1997, 119, 7410; d) G. F.
Meijs, T. C. Morton, E. Rizzardo, S. H. Thang, Macromole-
cules 1991, 24, 3689; e) Y. Ueno, S. Aoki, M. Okawara, J. Am.
Chem. Soc. 1979, 101, 5414; f) R. T. A. Mayadunne, E. Riz-
zardo, J. Chiefari, J. Krstina, G. Moad, A. Postma, S. H.
Thang, Macromolecules 2000, 33, 243.
[7] a) G. F. Meijs, E. Rizzardo, S. H. Thang, Polym. Bull. 1990, 24,
501; b) G. F. Meijs, T. C. Morton, E. Rizzardo, S. H. Thang,
Macromolecules 1991, 24, 3689; c) S. Z. Zard, Angew. Chem.
1997, 109, 724; Angew. Chem. Int. Ed. Engl. 1997, 36, 672; d)
Y. K. Chong, J. Krstina, T. P. T. Le, G. Moad, A. Postma, E.
Rizzardo, S. H. Thang, Macromolecules 2003, 36, 2256.
[8] a) J. Xiang, J. Evarts, A. Rivkin, D. P. Curran, P. L. Fuchs,
Tetrahedron Lett. 1998, 39, 4163; b) B. Sire, S. Seguin, S. Z.
Zard, Angew. Chem. 1998, 110, 3056; Angew. Chem. Int. Ed.
1998, 37, 2864; c) G. Ouvry, B. Quiclet-Sire, S. Z. Zard, Angew.
Chem. 2006, 118, 5124; Angew. Chem. Int. Ed. 2006, 45, 5002;
d) N. Charrier, S. Z. Zard, Angew. Chem. 2008, 120, 9585; An-
gew. Chem. Int. Ed. 2008, 47, 9443.
Conclusions
In summary, the results reported herein demonstrate that
allyl-PINO compounds are excellent substrates for the free
radical based allylation of hydrocarbons; this allows C–H
functionalization and C–C bond formation to be achieved
in a single reaction through the mechanism outlined in
Scheme 1. The reaction yields are high and the mass bal-
ances are excellent. Compared to allyl bromides, allyl-PINO
compounds are much easier to handle. Moreover, the by-
product of the reaction, N-hydroxyphthalimide, precipitates
from solution and can be easily removed by simple filtration
(and recycled). Chain lengths are shorter and reaction times
are longer (relative to those of the analogous reactions of
allyl bromides), most likely because PINO is a less-reactive
hydrogen-atom abstractor. There appears to be no signifi-
cant difference in efficiency of the addition–elimination
steps.
[9] a) J. M. Tanko, M. Sadeghipour, Angew. Chem. 1999, 111, 219;
Angew. Chem. Int. Ed. 1999, 38, 159; b) J. A. Struss, S. Mitra,
J. M. Tanko, Tetrahedron Lett. 2009, 50, 2119.
[10] a) C. Deraeve, Z. Guo, R. S. Bon, W. Blankenfeldt, R. DiLuc-
rezia, A. Wolf, S. Menninger, E. A. Stigter, S. Wetzel, A. Cho-
idas, K. Alexandrov, H. Waldmann, R. S. Goody, Y.-W. Wu, J.
Am. Chem. Soc. 2012, 134, 7384; b) C. Pégurier, L. Morellato,
E. Chahed, J. Andrieux, J.-P. Nicolas, J. A. Boutin, C. Benne-
jean, P. Delagrange, M. Langlois, M. Mathé-Allainmat, Bioorg.
Med. Chem. 2003, 11, 789; c) M. Yang, N. Yokokawa, H.
Ohmiya, M. Sawamura, Org. Lett. 2012, 14, 816.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental details, characterization data, and copies of the
1H NMR and 13C NMR spectra of all key intermediates and final
products.
[11] a) S. Albrecht, A. Defoin, C. Tarnus, Synthesis 2006, 1635; b)
H. Pines, H. Alul, M. Kolobielski, J. Org. Chem. 1957, 22,
1113; c) J. Villieras, M. Rambaud, Synthesis 1982, 924.
[12] M. Walbiner, Q. W. Wu, H. Fischer, Helv. Chim. Acta 1995, 78,
910.
Acknowledgments
Acknowledgement is made to the donors of the American Chemi-
cal Society Petroleum Research Fund for support of this research.
[13] P. S. Skell, J. G. Traynham, Acc. Chem. Res. 1984, 17.
Received: October 8, 2013
[1] F. Recupero, C. Punta, Chem. Rev. 2007, 107, 3800.
Published Online: December 9, 2013
Eur. J. Org. Chem. 2014, 502–505
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
505