Verdazyl radicals as substrates for organic synthesis: Unique access to tetrahydropyrazolotriazinones, pyrazolotriazinones and dihydrotetrazinylacrylonitriles

Article abstract of DOI:10.1002/ejoc.201000789

1,5-Dimethyl-3-phenyl-6-oxoverdazyl radical was recently shown to undergo a disproportionation reaction to form an azomethine imine, which reacted with a series of dipolarophiles, to form novel tetrahydropyrazolotetrazinone heterocyclic structures, demonstrating for the first time the use of verdazyl radicals as substrates for organic synthesis. Herein, we report on the chemistry of this verdazyl radical with captodative olefins and show that the reactions that occur, either an addition reaction or a hydrogen abstraction reaction followed by a cycloaddition reaction, are defined by the captodative olefin. Also highlighted is an intramolecular rearrangement of initially formed tetrahydropyrazolotetrazinone cycloadducts to either pyrazolotriazinone or tetrahydropyrazolotriazinone structures. While the rearrangement falls under the very general definition of the Dimroth rearrangement, the mechanism that is proposed is distinct and provides the opportunity to go from one particular structural motif to another under mild conditions, expanding the usefulness of this chemistry. The use of a verdazyl radical as a precursor for organic synthesis is demonstrated. The use of captodative olefins as the dipolarophiles leads to a variety of unexpected products, in a few of the cases as a result of a rearrangement reminiscent of the Dimroth rearrangement but actually proceeding by a unique mechanism. A proposed mechanism for the rearrangement is suggested.

Full text of DOI:10.1002/ejoc.201000789

FULL PAPER
DOI: 10.1002/ejoc.201000789
Verdazyl Radicals as Substrates for Organic Synthesis: Unique Access to
Tetrahydropyrazolotriazinones, Pyrazolotriazinones and
Dihydrotetrazinylacrylonitriles
Eric K. Y. Chen,[a] Matthew Bancerz,[a] Gordon K. Hamer,[a] and Michael K. Georges*[a]
Keywords: Radicals / Verdazyl radical / Cycloaddition / Rearrangement / Nitrogen heterocycles
1,5-Dimethyl-3-phenyl-6-oxoverdazyl radical was recently
shown to undergo a disproportionation reaction to form an
the captodative olefin. Also highlighted is an intramolecular
rearrangement of initially formed tetrahydropyrazolotetrazi-
none cycloadducts to either pyrazolotriazinone or tetrahydro-
pyrazolotriazinone structures. While the rearrangement falls
under the very general definition of the Dimroth rearrange-
ment, the mechanism that is proposed is distinct and pro-
vides the opportunity to go from one particular structural mo-
tif to another under mild conditions, expanding the useful-
ness of this chemistry.
azomethine imine, which reacted with
a series of di-
polarophiles, to form novel tetrahydropyrazolotetrazinone
heterocyclic structures, demonstrating for the first time the
use of verdazyl radicals as substrates for organic synthesis.
Herein, we report on the chemistry of this verdazyl radical
with captodative olefins and show that the reactions that oc-
cur, either an addition reaction or a hydrogen abstraction re-
action followed by a cycloaddition reaction, are defined by
tain structures, a feature highlighted in this submission.
Since most synthetic drugs are nitrogen containing hetero-
cyclic compounds,[2] the development of new strategies for
the synthesis of heterocyclic structures offers the potential
for new drug discovery, the anticipated endgame of this
work. Of particular interest are (i) the recent discovery that
flat heterocyclic structures can be used to disrupt Myc/Max
dimerization, a process implicated in virulent forms of
breast cancer[6] and (ii) the use of pyrazolotriazinone com-
pounds in dynamic combinatorial chemistry (DCC).[7,8]
We recently reported that 1,5-dimethyl-3-phenyl-6-oxov-
erdazyl radical[9] (1), where R = Ph, can serve as a precursor
to the azomethine imine 1a. In the presence of electron-
poor dipolarophiles, for example acrylates, methacrylates
and acrylonitrile, as well as weakly electron-rich di-
polarophiles, such as styrene and its derivatives, 1a un-
dergoes a 1,3-dipolar cycloaddition reaction to give the tet-
rapyrazolotetrazinone structure 2 (Scheme 1).[10] Herein, we
report on the chemistry of verdazyl radicals with the cap-
todative olefins 1-chloroacrylonitrile, 1-acetoxyacrylonitrile,
2-phenyl-1-thiomethylacrylonitrile and methyl α-acetoxyac-
rylate. In the process, we show that the reaction that occurs,
either an addition reaction or a hydrogen abstraction reac-
tion followed by a cycloaddition reaction, is defined by the
captodative olefin. Also highlighted is an intramolecular re-
arrangement of initially formed tetra- or di-hydropyrazolot-
etrazinone cycloadducts to either tetrahydropyrazolotriazi-
none or pyrazolotriazinone structures, respectively. While
the rearrangement falls under the very general definition
of the Dimroth rearrangement,[11,12] it would appear to be
mechanistically unique.
Introduction
We recently proposed a new synthetic strategy for the
synthesis of heterocyclic compounds based on the use of
verdazyl radicals, a family of stable radicals that can be iso-
lated and stored for extended periods without extensive de-
composition.[1] One of the requirements of diversity-ori-
ented synthesis (DOS) is the development of new synthetic
strategies that are flexible enough to provide combinatorial
libraries of diverse small molecules for biological screen-
ing.[2,3] We believe the use of verdazyl radicals will provide
this opportunity once the chemistry of these interesting mo-
lecules, and the molecular motifs they can provide, has been
elucidated.
Our interest in verdazyl radicals began with an investiga-
tion of their use as mediators for living-radical polymeriza-
tions[4] but that focus shifted to exploring whether they
could be used as substrates in organic synthesis. In spite of
being discovered over 45 years ago,[1a] surprisingly little is
known about the chemistry of verdazyl radicals. However,
since they are structurally diverse and contain a wealth of
heteroatoms,[5] they would appear to be ideal substrates for
preparing novel heterocyclic compounds. From a synthetic
chemistry perspective the value of many heterocyclic com-
pounds lies in their propensity to undergo intramolecular
rearrangement reactions leading to otherwise hard to ob-
[a] Department of Chemical and Physical Sciences, University of
Toronto at Mississauga,
3359 Mississauga Road, Mississauga, Ontario, L5L 1C6,
Canada
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000789.
View this journal online at
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 5681–5687
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5681

E. K. Y. Chen, M. Bancerz, G. K. Hamer, M. K. Georges
FULL PAPER
Scheme 1. Formation of tetrahydropyrazolotetrazinone cycloadducts 2 from verdazyl 1.
respectively, with 7 providing 3 upon silica gel purification,
the product that was initially expected from the reaction
Results and Discussion
In an effort to introduce different functionality into the
structures produced by the reaction of 1a with di-
polarophiles, 1 was allowed to react in the presence of 1-
chloroacrylonitrile, a known ketene equiv.,[13] in anticipa-
tion of forming 3 (Scheme 2). However, the reaction yielded
less than 5% of 3. The major product was 4, with the E-
configuration as confirmed by single-crystal X-ray diffrac-
tion (the crystal X-ray structure of 4 is provided in the Sup-
porting Information). The formation of 4 occurs, we sug-
gest, by the addition of 1 to the double bond of 1-chloro-
acrylonitrile, giving radical 5, which is subsequently trapped
by another molecule of 1, affording 6.[14] Loss of HCl from
6 affords 4.
with 1-chloroacrylonitrile (Scheme 3).
The addition of 1 to the double bond of 1-chloroacrylo-
nitrile was somewhat surprising, since in all our previous
work with verdazyl radicals as mediators for living-radical
polymerizations,[4] and in our initial cycloaddition stud-
ies,[10] this type of reaction was not noticed. In hindsight,
given that 5 is a captodative radical, the verdazyl radical
Scheme 3. Cycloaddition reactions between 1a and 1-acetoxyacry-
lonitrile 7 and (E)-2-(methylthio)-3-phenylacrylonitrile 8.
Methyl α-acetoxyacrylate (MAA),[15] interestingly, gave
addition reaction might have been expected. However, neither the anticipated cycloadduct 9 nor the verdazyl radi-
methyl methacrylate and methacrylonitrile could conceiva- cal addition product when stirred with 1 at room temp. for
1
bly also form captodative radicals but preferred to undergo 24 h (Scheme 4). Instead, H and 13C NMR suggested that
the cycloaddition reaction in high yield.[10] In these cases pyrazolotriazinone 11 was the major product. Single-crystal
the methyl group is a weak electron donor. The captodative X-ray diffraction confirmed the structure and showed the
radical that would result from these substrates would be less pyrazolotriazinone moiety to be planar.
stable than 5 and, therefore, less likely to form.
Repeating the reaction of 1 with MAA for only 3 h en-
Two other captodative olefins, 1-acetoxyacrylonitrile 7 abled the isolation of 9 in low yield. As expected 9 was
and (E)-2-(methylthio)-3-phenylacrylonitrile 8, underwent unstable and upon standing in CDCl3 in an NMR tube at
the cycloaddition reaction in modest yields, 43% and 42%, room temp. it eliminated acetic acid, evident in the 1H
Scheme 2. Addition of verdazyl 1 to 1-chloroacrylonitrile.
5682
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5681–5687

Verdazyl Radicals as Substrates for Organic Synthesis
Scheme 4. Reaction of verdazyl 1 with MAA and MP; intermediates and final product.
Scheme 5. Reaction of DMADC in the presence of 1 to give 13 via 12.
NMR spectrum of the reaction mixture, to give 10. Heating
either 9 or 10 in refluxing ethyl acetate gave 11 in 89% yield
(Scheme 4). The addition of excess NaH to a solution of 10
in THF resulted in the formation of 11 in 76% yield in 5 h
at room temp. (the crystal X-ray structure of 11 is provided
in the Supporting Information).
In parallel with the work with the captodative olefins,
substituted alkynes were also being investigated as sub-
strates for the cycloaddition reaction. The reaction of
methyl propiolate (MP) in the presence of 1 (Scheme 4) gave
a small amount of 11 and the α,β-unsaturated ester 10 as
the major product, which again was readily converted into
11 in refluxing EtOAc in 76% yield. Using a disubstituted
alkyne provided the opportunity to add functionality to the
pyrazolotriazinone structure. Thus, the reaction of dimethyl
acetylenedicarboxylate (DMADC) in the presence of 1 gave
12, which subsequently underwent rearrangement upon
heating to give 13 in 50% yield (Scheme 5).
Scheme 6. NaH induced rearrangement of 14 and subsequent de-
carboxylation.
While the saturated analogue of 10, compound 14, is
stable in refluxing EtOAc, it readily rearranges to 15 in 82%
yield when treated with 2 equiv. of NaH at room temp.
(Scheme 6). Compound 14 is also readily converted into 15
upon treatment with lithium diisopropylamide at 0 °C for
1 h, followed by addition of water, suggesting that the for-
mation of the anion is vital to the success of the reaction.
Scheme 7 outlines a proposed mechanism for the conver-
sion of 14 to 15. One of the keys to this mechanism is the
flexibility of the bicyclic ring system, due to the facile inver-
sion of the ring-junction nitrogen centers allowing the fused
Scheme 7. Proposed mechanism for the rearrangement of 14 to 15.
Energy levels calculated for the transitions states and in-
rings to approach each other not unlike a butterfly folding termediate, which would occur with the proposed mecha-
it wings, enabling the carbanion to approach and interact nism, are provided in (Figure 2).
with the carbonyl carbon center. DFT calculations show
Possibly more interesting at this point is that treatment
that as the carbanion centre approaches the carbonyl center of 14 or 15 with excess NaH gives the decarboxylated prod-
to form a four atom ring intermediate (int) (Figure 1), the uct 16 in 92% and 76% yields, respectively (Scheme 7).
C–N bond of the ring lengthens, facilitating it breakage.
Thus, aromatization of the five-membered ring is not the
Eur. J. Org. Chem. 2010, 5681–5687
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5683

E. K. Y. Chen, M. Bancerz, G. K. Hamer, M. K. Georges
FULL PAPER
Scheme 8. KOtBu-induced rearrangement of 17 to 18.
Conclusions
While verdazyl radicals are considered stable relative to
radicals in general, they are still reactive and, as has been
demonstrated herein, can lead to a variety of unique nitro-
gen-containing small molecules. There is sufficient struc-
tural variety in both the verdazyl radical and dipolarophiles
to provide a library of diverse compounds enabling this new
synthetic strategy to be ideal for DOS. In addition, pres-
ently under investigation are two other unique intermo-
lecular rearrangements of our pyrazolotetrazinone struc-
tures, dictated by substituents in the molecule and reaction
conditions, leading to other structurally interesting motifs.
These various rearrangements point to the strength of
working with structurally flexible heterocyclic molecules
where the nucleophilic heteroatoms are well positioned to
react intramolecularly with electrophilic substituents to give
strained-ring intermediates that readily collapse to other
interesting scaffolds.
Figure 1. DFT-generated lowest-energy conformation of a four-
membered intermediate (int) on the pathway from 14 to 15. The
four-membered ring is highlighted with dashed-line bonds.
Experimental Section
Figure 2. Energy levels to the two transitions states, ts1 and ts2 and
the four-membered intermediate (int) determined by DFT calcula-
tions. The calculations were carried out at the B3LYP/6-
31G+(d,p)//B3LYP/6-31G(d) level using the Gaussian G03W pro-
gram.[16] Minima on the 14a Ǟ 15a reaction pathway potential
energy surface were located with the opt = gdiis option and transi-
tion states using the QST2 procedure. The nature of the calculated
stationary points was tested by frequency analysis and the closed-
shell nature of all species determined by wavefunction stability cal-
culations. Relative energies (kcalmol–1) are expressed relative to the
energy of 14a. ts1,ts2 are transitions states 1 and 2, respectively,
while int is the four-membered intermediate depicted in Figure 1.
General: All reagents and solvents were purchased from Sigma–
Aldrich or Fluka unless otherwise stated. The inhibitor hydroqui-
none monomethyl ether was removed from acrylates by passing
them through a short column packed with inhibitor remover resin
purchased from Sigma–Aldrich. Silica gel chromatography was per-
formed with Silica Gel 60 (particle size 40–63 µm) obtained from
EMD. Thin-layer chromatrography (TLC) plates were obtained
from EMD. Verdazyl radical 1[18a] and methyl 2-acetoxyacrylate
(MAA)[18b] were synthesized according to published procedures.
NMR spectra were recorded on a Varian Unity INOVA spectrome-
ter at 20 °C, operating at 500 MHz for 1H NMR and 125 MHz for
13C NMR or a Bruker Avance III spectrometer at 23 °C, operating
at 400 MHz for 1H NMR and 100 MHz for 13C NMR spec-
troscopy. Chemical sifts (δ) are reported in parts per million (ppm)
referenced to tetramethylsilane (δ = 0 ppm) for 1H NMR spectra
and CDCl3 (δ = 77.0 ppm) for 13C NMR spectroscopy. Coupling
constants (J) are reported in Hertz (Hz).
driving force for this facile decarboxylation reaction, as was
first suspected based on previous work showing a decarb-
oxylation reaction leading to aromatic heterocycles.[17] The
mild conditions for the decarboxylation leading to satu-
rated heterocycles are interesting, and to the best of our
knowledge unprecedented. Further investigation is ongoing
with other substrates to determine whether the reaction is
general or unique to the substrates used in this work. It
would appear that the decarboxylation is not occurring due
to NaOH being present in the reaction mixture resulting
from NaH reacting over time with trace water since no reac-
tion is observed when 14 is treated with NaOH in THF at
room temp. for 24 h.
Mass spectrometry was performed on an AB/Sciex QStar mass
spectrometer with an ESI source, MS/MS and accurate mass capa-
bilities, associated with an Agilent 1100 capillary LC system.
X-ray data were collected on a Bruker–Nonius Kappa-CCD dif-
fractometer using monochromated Mo-Kα radiation, and measure-
ments were made using a combination of Φ and ω scans with κ
offsets to fill the Ewald sphere. The data were processed using the
Denzo-SMN package. Absorption corrections were carried out
using SORTAV. The structure was solved and refined using
SHELXTL V6.1 for full-matrix least-squares refinement that was
based on F2. All H atoms were included in the calculated positions
and allowed to refine in the riding-motion approximation with Uiso
tied to the carrier atom.
In testing the generality of this base induced rearrange-
ment reaction, cycloadduct 17 formed from N,N-dimethyl-
acrylamide was treated with potassium tert-butoxide to give
18 (Scheme 8).
5684
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5681–5687

Verdazyl Radicals as Substrates for Organic Synthesis
CCDC-766562 (for 4) and CCDC -766561 (for 11) contain the sup-
plementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
δ = 159.0, 146.1, 134.2, 131.7, 130.4, 129.4, 128.6, 127.1, 113.3,
52.0, 51.3, 36.8 ppm. HRMS (ESI): calcd. for C14H14N4O3 [M +
H]+ 287.1125; found 287.1138.
α,β-Unsaturated Ester Cycloadduct 10 via Cycloaddition of 1a and
Methyl Propiolate: Verdazyl radical 1 (300 mg, 1.5 mmol) was dis-
solved in methyl propiolate (0.65 mL, 7.5 mmol). The reaction was
oxygenated and stirred at room temp. for 48 h. The excess mono-
mer was removed in vacuo. Purification by silica gel chromatog-
raphy (toluene/ethyl acetate, 17:1) gave 10 (52 mg, 18%) as a color-
less liquid.[18d]
General Procedure for Cycloaddition Reactions: The neat olefin
(20 mmol) was placed in a three-neck round-bottomed flask
equipped with a septum and an adaptor that was connected to a
gas bubbler. Oxygen was bubbled into the reaction flask for 15 min
at ambient temperature via a syringe needle pierced through the
septum. 1,5-Dimethyl-3-phenyl-6-oxoverdazyl (203 mg, 1 mmol)
was added and the reaction solution was stirred at room temp. for
24 h under an atmosphere of O2. Excess olefin was removed in
vacuo and the products were purified by flash silica gel chromatog-
raphy.
5-Methyl-7-phenylpyrazolo[1,5-d][1,2,4]triazin-4(5H)-one (11): The
general procedure was slightly modified such that less MAA
(1.065 g, 7.5 mmol) was used relative to 1,5-dimethyl-3-phenyl-6-
oxoverdazyl 1 (203 mg, 1.0ϫ10–3 mol). The excess monomer was
removed in vacuo. Purification by silica gel chromatography (hex-
anes/ethyl acetate, 3:1) gave the product as a white solid (96 mg,
32%). 1H NMR (500 MHz, CDCl3, 20 °C): δ = 8.15–8.05 (m, 2 H,
PhH), 8.015 (d, J = 1.94 Hz, 1 H, pyrH), 7.60–7.50 (m, 3 H, PhH),
7.226 (d, J = 1.94 Hz, 1 H, pyrH), 3.83 (s, 3 H, CH3) ppm. 13C
NMR (125 MHz, CDCl3): δ = 153.7, 143.7, 138.5, 134.6, 131.0,
129.4, 128.8, 128.5, 106.5, 37.9 ppm. HRMS (ESI): calcd. for
C12H11N4O [M + H]+ 227.0929; found 227.0927.; m.p. 119–121 °C.
Vinyl Cyanoacetate Cycloadduct 3, Post Hydrolysis: Compound 3
was prepared according to the general procedure. Hydrolysis of the
initially formed cycloaddition product appeared to occur during
purification by silica gel column chromatography (1:9 ethyl acetate/
dichloromethane) to give a white solid (110 mg, 45%); m.p. 108–
1
111 °C. H NMR (500 MHz, CDCl3, 20 °C): δ = 7.75–7.65 (m, 2
H, PhH), 7.50–7.35 (m, 3 H, PhH), 4.18 (t, J = 8.14 Hz, 2 H, CH2),
3.43 (s, 3 H, CH3), 2.71 (t, J = 8.14 Hz, 2 H, CH2) ppm. 13C NMR
(125 MHz, CDCl3): δ = 169.3, 156.9, 141.4, 131.3, 128.7, 128.6,
127.5, 42.5, 37.5, 30.2 ppm. HRMS (EI): calcd. for C12H12N4O2
[M]+ 244.0960; found 244.0960.
Conversion of 10 to 11 via Heating: Product 10 (20 mg, 0.07 mmol)
was refluxed in 5 mL of ethyl acetate for 30 min. The solvent was
removed in vacuo to give 11 (14 mg, 89%).
Conversion of 10 to 11 Using NaH: Product 10 (20 mg, 0.07 mmol)
was dissolved in 3 mL of dry THF. Excess solid sodium hydride
(20 mg) was added and the reaction was allowed to proceed for 4 h
at room temp. The reaction mixture was cooled in an ice bath and
quenched with methanol. The THF was removed in vacuo. The
reaction mixture was taken up in ethyl acetate (10 mL) and washed
with a cold brine solution (10 mL). The ethyl acetate solution was
dried with Na2SO4 and evaporated to give 11 (12 mg, 76%) with
no starting material remaining according to TLC. Repeating the
reaction with no sodium hydride showed only a very faint spot for
11 on a TLC plate when a large amount of the reaction solution
was used.[18e]
Verdazyl Radical Addition Product 4: Compound 4 was prepared
according to the general procedure and purified by silica gel col-
umn chromatography (1:1 ethyl acetate/hexanes) to give a white
solid (160 mg, 35%); m.p. 172–174 °C. 1H NMR (500 MHz,
CDCl3, 20 °C): δ = 7.70–7.35 (m, 10 H, PhH), 6.80 (s, 1 H, C=CH),
3.32 (s, 3 H, CH3), 3.26 (s, 3 H, CH3), 3.19 (s, 3 H, CH3), 3.15 (s,
3 H, CH3) ppm. 13C NMR (125 MHz, CDCl3): δ = 156.3, 155.2,
142.3, 141.6, 131.8, 130.7, 129.4, 128.8, 127.4, 127.2, 114.3, 91.5,
39.1, 37.5, 36.8, 34.4 ppm. HRMS (EI): calcd. for C23H23N9O2 [M]
+
457.2000; found 457.1975.
2-Phenyl-1-(methylthio)acrylonitrile Cycloadduct 8a: Compound 8a
was prepared according to the general procedure and purified by
silica gel column chromatography (1:4 ethyl acetate/hexanes) to
give a white solid (150 mg, 42%); m.p. 144–146 °C. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 7.97–7.84 (m, 2 H, PhH), 7.48–7.36
(m, 8 H, PhH), 4.79–4.71 (m, 2 H, CH2), 3.70–3.55 (m, 2 H, CH2),
3.45 (s, 3 H, CH3), 2.07 (s, 3 H, CH3) ppm. 13C NMR (100 MHz,
CDCl3): δ = 155.2, 143.5, 134.3, 131.4, 129.9, 129.0, 128.9, 128.7,
128.5, 128.3, 113.1, 73.8, 58.3, 50.5, 37.1, 29.6, 16.0 ppm. HRMS
(EI): calcd. for C20H19N5OS [M]+, 377.1310; found 377.1309.
Cycloadduct of 1a with Dimethyl Acetylenedicarboxylate (12): Verd-
azyl 1 (300 mg, 1.48 mmol) was dissolved in dimethyl acetylened-
icarboxylate (0.9 g, 7.4 mmol) and stirred for 3 d at room temp.
under an oxygen atmosphere. The reaction mixture was purified by
silica gel column chromatography (1:3 ethyl acetate/hexanes) to
give 12 as a yellow oil (209 mg, 41%). 1H NMR (400 MHz, CDCl3,
20 °C): δ = 7.60–7.63 (dd, 2 H, PhH), 7.46–7.50 (tt, 1 H, PhH),
7.39–7.43 (tt, 2 H, PhH), 4.84 (s, 2 H, CH2), 3.72 (s, 3 H, CH3),
3.39 (s, 3 H, CH3), 3.25 (s, 3 H, CH3) ppm. 13C NMR (100 MHz,
CDCl3): δ = 162.0, 159.5, 156.5, 141.9, 140.4, 131.7, 129.4, 128.8,
128.2, 104.8, 53.0, 52.0, 51.5, 37.5 ppm. HRMS (EI): calcd. for
C16H16N4O5 [M]+ 284.0909; found 284.0903.
MAA Cycloadduct 9: Compound 9 was prepared according to the
general procedure with the modification that the reaction was
stopped after 3 h. Purification by silica gel column chromatography
(hexanes/ethyl acetate, 3:1) gave 9 as a colorless liquid. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 7.65–7.70 (m, 2 H, PhH), 7.45–7.35
(m, 3 H, PhH), 4.37–4.28 (m, 1 H, CH2), 3.61–3.51 (m, 1 H, CH2),
3.43 (s, 3 H), 3.21–3.10 (m, 1 H, CH2), 3.17 (s, 3 H, CH3), 2.49–
2.40 (m, 1 H, CH2), 2.03 (s, 3 H, CH3) ppm.
Rearrangement of 12 to 13: Cycloadduct 12 (110 mg, 0.32 mmol)
was heated at 150 °C neat for 2 d. The reaction mixture was puri-
fied by silica gel column (1:3 ethyl acetate/hexanes) to give 13 as
a white solid (45 mg, 50%); m.p. 100–103 °C. Unreacted starting
material (47 mg, 42%) was recovered. 1H NMR (400 MHz, CDCl3,
20 °C): δ = 8.37 (s, 1 H, pyrH), 8.03–8.05 (m, 2 H, PhH), 7.52–7.57
(m, 3 H, PhH), 3.98 (s, 3 H, CH3), 3.84 (s, 3 H, CH3) ppm. 13C
NMR (100 MHz, CDCl3): δ = 161.5, 151.9, 145.7, 138.2, 133.2,
131.2, 129.6, 128.5, 128.3, 114.8, 52.5, 38.5 ppm. HRMS (EI):
calcd. for C14H12N4O3 [M]+ 345.1193; found 345.1203.
α,β-Unsaturated Ester Cycloadduct 10: Elimination of acetic acid
from 9[18c] occurred in quantitative yield at room temp. over 48 h
to give 10 as a colorless liquid. 1H NMR (400 MHz, CDCl3, 25 °C):
δ = 7.58–7.54 (m, 2 H, PhH), 7.43–7.36 (m, 3 H, PhH), 5.90 (t, J
= 2.78 Hz, 1 H, C=CH), 4.70 (d, J = 2.78 Hz, 2 H, CH2), 3.34 (s,
Methyl Acrylate Cycloadduct (14): Compound 14 was prepared ac-
3 H, CH3), 3.30 (s, 3 H, CH3) ppm. 13C NMR (125 MHz, CDCl3): cording to the general procedure and purified by silica gel column
Eur. J. Org. Chem. 2010, 5681–5687
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5685

E. K. Y. Chen, M. Bancerz, G. K. Hamer, M. K. Georges
FULL PAPER
chromatography (1:1 ethyl acetate/hexanes) to give a white solid
was dried with Na2SO4 and the solvents evaporated in vacuo to
(179 mg, 62%); m.p. 115–117 °C. 1H NMR (500 MHz, CDCl3, give 16 (106 mg, 92%) as a pale yellow oil. 1H NMR (400 MHz,
20 °C): δ = 7.67–7.62 (m, 2 H, PhH), 7.47–7.42 (m, 1 H, PhH), CDCl3, 25 °C): δ = 7.60–7.65 (m, 2 H, PhH), 7.38–7.44 (m, 3 H,
7.42–7.37 (m, 2 H, PhH), 4.24 (dd, J = 3.9, 9.0 Hz, 1 H, CH), 4.22–
PhH), 4.26 (br., s, 1 H, NH), 4.18–4.24 (m, 1 H, CH2), 3.40 (br.,
4.17 (m, 1 H, CH2), 3.56 (s, 3 H, CH3), 3.52–3.46 (m, 1 H, CH2), d, J = 0.72 Hz, 3 H, CH3), 3.09 (br., t, J = 7.45 Hz, 1 H, CH),
3.37 (s, 3 H, CH3), 2.48–2.39 (m, 1 H, CH2), 2.27–2.20 (m, 1 H,
2.47–2.70 (m, 3 H, CH2) ppm. 13C NMR (100 MHz, CDCl3): δ =
161.2, 149.1, 132.9, 129.9, 128.3, 128.0, 58.4, 46.3, 36.4, 35.2 ppm.
CH2) ppm. 13C NMR (125 MHz, CDCl3): δ = 171.3, 154.2, 146.0,
130.9, 130.9, 128.7, 127.5, 62.1, 52.4, 44.1, 36.7, 29.8 ppm. HRMS HRMS (EI): calcd. for C12H14N4O [M]+ 230.1168; found 230.1168.
(EI): calcd. for C14H16N4O3 [M]+ 288.1222; found 288.1218. FTIR
Conversion of 15 to 16 with NaH: Product 15 (40 mg, 0.14 mmol)
was dissolved in 5 mL of dry THF. Excess solid sodium hydride
(KBr): 3041, 2993, 2956, 1738, 1676 cm–1.
Rearranged Methyl Acrylate Cycloadduct Pre-decarboxylation (15):
Cycloadduct 14 (140 mg, 0.5 mmol) was dissolved in 15 mL of dry
THF in a dry 3-neck round-bottomed flask equipped with a stir
bar, cooled to 0 °C and degassed with N2 for 30 min. Sodium hy-
dride (20 mg, 1.0 mmol) was added. The reaction mixture was
warmed to room temp. and then left at that temperature for an
additional 0.5 h. Three drops of methanol was added to quench
the remaining sodium hydride and the THF was removed in vacuo.
The resulting oil was dissolved in ethyl acetate and washed with
brine. The organic layer was dried with Na2SO4 and the solvents
(50 mg) was added and the reaction was allowed to proceed for 4 h
at room temp. The reaction mixture was cooled in an ice bath and
quenched with methanol. The THF was removed in vacuo. The
reaction mixture was taken up in ethyl acetate (10 mL) and washed
with a cold brine solution (10 mL). The organic layer was dried
with Na2SO4 and evaporated to give 16 (28 mg, 76%) with no start-
ing material remaining according to TLC.
N,N-Dimethylacrylamide Cycloadduct 17: Compound 17 was pre-
pared according to the general procedure and purified by
recrystallization with EtOAc to give a dark yellow solid (253 mg,
1
evaporated in vacuo to give 15 (118 mg, 82%) as an oil. H NMR
1
84%); m.p. 141–143 °C. H NMR (400 MHz, CDCl3, 25 °C): δ =
(400 MHz, CDCl3, 25 °C): δ = 7.70–7.62 (m, 2 H, PhH), 7.40–7.32
(m, 3 H, PhH), 5.40 (br., 1 H, NH), 3.81 (s, 3 H, CH3), 3.40–3.30
(m, 1 H, CH2), 3.21 (dt, J = 13, 3 Hz, 1 H, NCH2), 2.85 (s, 3 H,
CH3), 2.86–2.80 (m, 1 H, CH2), 2.09 (dt, J = 13, 5 Hz, 1 H, NCH2)
ppm. 13C NMR (100 MHz, CDCl3): δ = 168.6, 150.2, 147.6, 129.5,
127.9, 127.8, 127.6, 85.6, 52.6, 36.9, 36.7, 27.4 ppm. HRMS (ESI):
calcd. for C14H17N4O3 [M + H]+ 289.1295; found 289.1297.
7.70–7.63 (m, 2 H, PhH), 7.47–7.35 (m, 3 H, PhH), 4.58–4.54 (m,
1 H, CH), 4.31–4.20 (m, 1 H, CH2), 3.53–3.43 (m, 1 H, CH2), 3.38
(s, 3 H, CH3), 2.71 (s, 3 H, CH3), 2.51 (s, 3 H, CH3), 2.40–2.28 (m,
1 H, CH2), 2.16–2.04 (m, 1 H, CH2) ppm. 13C NMR (100 MHz,
CDCl3): δ = 170.2, 153.6, 146.0, 131.4, 130.5, 128.4, 127.7, 59.5,
44.5, 36.7, 36.4, 35.5, 30.0 ppm. HRMS (ESI): calcd. for
C15H20N5O2 [M + H]+ 302.1616; found 302.1611.
Rearranged Methyl Acrylate Cycloadduct Pre-decarboxylation (15):
Cycloadduct 14 (20 mg, 0.07 mmol) was dissolved in 5 mL of dry
THF in a dry 3-neck round-bottomed flask equipped with a stir
bar, cooled to 0 °C in an ice bath and degassed with N2 for 30 min.
Lithium diisopropylamide (2 , 0.1 mL, 0.2 mmol) was added
dropwise via syringe over 30 seconds. The reaction mixture was
warmed to room temp. after 15 min at 0 °C. The unreacted LDA
was destroyed by the careful addition of water to the reaction mix-
ture. TLC (3:1 dichloromethane/ethyl acetate) of the reaction mix-
ture showed no remaining starting material. The THF was removed
in vacuo. The resulting solid was washed with CDCl3 and the
NMR spectrum of the crude reaction mixture matched the spec-
trum previously obtained for 15.
Rearranged N,N-Dimethylacrylamide Cycloadduct 18: Cycloadduct
17 (30 mg, 0.1 mmol) was dissolved in 3 mL of dry THF in a dry
3-neck round-bottomed flask equipped with a stir bar. Potassium
tert-butoxide (26 mg, 0.2 mmol) was added in small portions. The
reaction mixture was stirred at room temp. for 0.5 h. The THF was
removed in vacuo. 3 mL of ethyl acetate was added to the remain-
ing solid and three drops of saturated ammonium chloride was
added to the suspension. The organic layer was washed twice with
brine, dried with Na2SO4 and the solvents evaporated in vacuo to
give 18 (24 mg, 80%) as an oil. 1H NMR (400 MHz, CDCl3,
25 °C): δ = 7.73–7.65 (m, 2 H, PhH), 7.42–7.32 (m, 3 H, PhH),
5.06 (br., 1 H, NH), 3.43 (dt, J = 12.1, 4.6 Hz, 1 H, CH2), 3.38–
3.30 (m, 1 H, CH2), 3.23 (br., 3 H, CH3), 3.02 (br., 3 H, CH3), 2.79
(s, 3 H, CH3), 2.69 (dd, J = 12.5, 3.5 Hz, 1 H, CH2), 2.11–2.00 (m,
1 H, CH2) ppm. 13C NMR (100 MHz, CDCl3): δ = 168.3, 151.2,
149.1, 129.7, 128.9, 128.0, 127.4, 85.1, 38.1 (br), 37.7 (br), 37.3,
36.5, 25.1 ppm. HRMS (ESI): calcd. for C15H20N5O2 [M + H]+
302.1626; found 302.1611.
Treatment of 14 with NaOH: Cycloadduct 14 (20 mg, 0.07 mmol)
was dissolved in 5 mL of THF in a round-bottomed flask equipped
with a stir bar. Aqueous sodium hydroxide (1 , 1 mL, 1 mmol)
was added. The reaction mixture was stirred at room temp. for
24 h. No change was observed by TLC (3:1 dichloromethane/ethyl
acetate).
Supporting Information (see also the footnote on the first page of
this article): Spectra of compounds 3, 4, 6, 8a, 9–18 and crystal
structures of 4 and 11.
Treatment of 15 with NaOH: Product 15 (20 mg, 0.07 mmol) was
dissolved in 5 mL of THF in a round-bottomed flask equipped
with a stir bar. An aqueous sodium hydroxide solution (0.1 ,
1 mL, 0.1 mmol) was added. The reaction mixture was stirred at
room temp. for 24 h. No change was observed by TLC (3:1 dichlo- Acknowledgments
romethane/ethyl acetate).
Support was provided by the National Science and Engineering
Rearranged Methyl Acrylate Cycloadduct 16, Post Decarboxylation:
Cycloadduct 14 (140 mg, 0.5 mmol) was dissolved in 15 mL of dry
THF in a dry 3-neck round-bottomed flask with a stir bar and
nitrogen gas was introduced for 30 min at 0 °C. Excess solid sodium
hydride (75 mg) was added and the reaction was allowed to proceed
for 1 h at 0 °C and 1 h at room temp. The reaction mixture was
quenched by the slow addition of methanol until effervescence ce-
ased. The THF was removed in vacuo. The resulting oil was dis-
solved in ethyl acetate and washed with brine. The organic layer
Research Council of Canada. M. B. and E. K. Y. C. gratefully
acknowledge a University of Toronto (UT) fellowship and a Helen
Sawyer Hogg Graduate Admission UT Fellowship, respectively.
The X-ray crystal structure determinations were performed by
Alan Lough at the University of Toronto.
[1] a) R. Kuhn, H. Trischmann, Angew. Chem. Int. Ed. Engl. 1963,
2, 155; b) F. A. Neugebauer, Angew. Chem. Int. Ed. Engl. 1973,
5686
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5681–5687

Verdazyl Radicals as Substrates for Organic Synthesis
12, 455–464; c) F. A. Neugebauer, H. Fischer, Angew. Chem.
1980, 92, 766; d) F. A. Kursmitteilung, F. A. Neugebauer, H.
Fischer, Angew. Chem. Int. Ed. Engl. 1980, 19, 724–725; e) F. A.
Neugebauer, H. Fischer, R. Siegel, C. Krieger, Chem. Ber. 1983,
116, 3461–3481.
D. S. Tan, Nature Chem. Bio. 2005, 1, 74–84.
M. D. Burke, S. L. Schreiber, Angew. Chem. Int. Ed. 2004, 43,
46–58.
E. K. Y. Chen, D. Chan-Seng, P. O. Otieno, R. G. Hicks, M. K.
Georges, Macromolecules 2007, 40, 8609–8616.
For a recent reviews see: R. G. Hicks, Org. Biomol. Chem. 2007,
5, 1321–1338.
[11] The Dimroth rearrangement is generally defined as an isomer-
ization process whereby heteroatoms are translocated on a het-
erocyclic ring.[12b]
[12] For two recent reviews, see: a) T. Fujii, T. Itaya, Heterocycles
1998, 48, 359–390; b) E. S. H. El Ashry, Y. El Kilany, N.
Rashed, H. Assafir, Adv. Heterocycl. Chem. 2000, 75, 79–165.
[13] For a review, see: S. Ranganathan, D. Ranganathan, A. K.
Mehrotra, Synthesis 1977, 289–296.
[14] The reaction with 1-chloroacrylonitrile was repeated with tet-
ramethylpypiridinyloxy (TEMPO) radical, a stable nitroxide.
No addition of TEMPO to the 1-chloroacrylonitrile was ob-
served, even not at 70 °C. This would indicate that the verdazyl
radical has a higher reactivity towards captodative olefins than
TEMPO, although more experiments will be necessary to con-
firm this.
[2]
[3]
[4]
[5]
[6] a) H. Wang, D. I. Hammoudeh, A. V. Follis, B. E. Resses, J. S.
Lazo, S. J. Metallo, E. V. Prochownik, Mol. Cancer Ther. 2007,
6, 2399–2408; b) A. Kiessling, B. Sperl, A. Hollis, D. Eick, T.
Berg, Chem. Bio. 2006, 13, 745–751; c) X. Yin, C. Giap, J. S.
Lazo, E. V. Prochownik, Oncogene 2003, 22, 6151–6159; d) T.
Berg, S. B. Cohen, J. Desharnais, C. Sinderegger, D. J. Maslyar,
J. Goldberg, D. L. Boger, P. K. Vogt, Proc. Natl. Acad. Sci.
USA 2002, 99, 3830–3835.
[15] J. Wolinky, R. Novak, R. Vasileff, J. Org. Chem. 1964, 29,
3596–3598.
[16] Gaussian 03, rev. E.01, Gaussian, Inc., Wallingford CT, 2004;
for complete citation see Supporting Information.
[17] T. Sasaki, K. Kanematsu, A. Kakehi, Tetrahedron Lett. 1972,
51, 5245–5248.
[7] A pyrazolotriazinone compound, similar in structure to what
is reported herein, has been used for DCC studies. P. Wipf,
S. G. Mahler, K. Okumura, Org. Lett. 2005, 7, 4483–4486.
[8] For reviews on DCC see: a) S. Ladame, Org. Biomol. Chem.
2008, 6, 219–226; b) P. T. Corbett, J. Leclaire, L. Vial, K. R.
West, J. Wietor, J. K. M. Sanders, S. Otto, Chem. Rev. 2006,
106, 3652–3711.
[9] C. L. Barr, P. A. Chase, R. G. Hicks, M. T. Lemaire, C. L. Ste-
vens, J. Org. Chem. 1999, 64, 8893–8897.
[10] A. Yang, T. Kasahara, E. K. Y. Chen, G. Hamer, M. K.
Georges, Eur. J. Org. Chem. 2008, 27, 4571–4574.
[18] a) F. A. Neugebauer, H. Fischer, R. Siegel, C. Krieger, Chem.
Ber. 1983, 116, 3461–3481; b) J. Wolinky, R. Novak, R. Vasileff,
J. Org. Chem. 1964, 29, 3596–3598; c) Acetic acid was observed
1
by H NMR: δ = 2.10 (s, 3 H), 11.50 (s, 1 H); d) The cycload-
dition reactions with alkynes tend to be very sluggish reactions
resulting in poor yields; e) For all intents and purposes, vir-
tually no reaction occurred although when the TLC plate was
overloaded with the reaction mixture a very faint spot corre-
sponding to 9 was visible.
Received: June 1, 2010
Published Online: August 24, 2010
Eur. J. Org. Chem. 2010, 5681–5687
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5687