N[1,3]-sigmatropic shift in the benzidine rearrangement: Experimental and theoretical investigation

Article abstract of DOI:10.1039/c4ob00080c

The N[1,3]-sigmatropic shift in the benzidine rearrangement has been studied in depth experimentally with the aid of density functional theory (DFT) calculations. The designed substituted N,N′-diaryl hydrazines rearrange exclusively to the expected o/p-semidines and diphenylines. Intercrossing experiments support the intramolecular rearrangement process. Radical trapping experiments exclude the intermediacy of biradicals in the rearrangements. Computational results demonstrate that the o-semidine rearrangement involves a novel N[1,3]-sigmatropic shift and the p-semidine rearrangement proceeds via tandem N[1,3]/N[1,3]-sigmatropic shifts, while the diphenyline rearrangement occurs through cascade N[1,3]/[3,3]-sigmatropic shifts. The proposed mechanism involving the key N[1,3]-sigmatropic shift as the rate-limiting step is in good agreement with reported kinetic isotope measurements. The combined methods provide new insight into the formation mechanism of o/p-semidines and diphenylines in the benzidine rearrangement and support the unprecedented suprafacial symmetry allowed N[1,3]-sigmatropic shift with an inversion of the configuration in the migrating nitrogen atom. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ob00080c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
N[1,3]-sigmatropic shift in the benzidine
rearrangement: experimental and theoretical
investigation†
Cite this: Org. Biomol. Chem., 2014,
12, 4952
Shili Hou,‡ Xinyao Li‡ and Jiaxi Xu*
The N[1,3]-sigmatropic shift in the benzidine rearrangement has been studied in depth experimentally
with the aid of density functional theory (DFT) calculations. The designed substituted N,N’-diaryl hydra-
zines rearrange exclusively to the expected o/p-semidines and diphenylines. Intercrossing experiments
support the intramolecular rearrangement process. Radical trapping experiments exclude the intermedi-
acy of biradicals in the rearrangements. Computational results demonstrate that the o-semidine
rearrangement involves a novel N[1,3]-sigmatropic shift and the p-semidine rearrangement proceeds via
tandem N[1,3]/N[1,3]-sigmatropic shifts, while the diphenyline rearrangement occurs through cascade
N[1,3]/[3,3]-sigmatropic shifts. The proposed mechanism involving the key N[1,3]-sigmatropic shift as the
rate-limiting step is in good agreement with reported kinetic isotope measurements. The combined
methods provide new insight into the formation mechanism of o/p-semidines and diphenylines in the
benzidine rearrangement and support the unprecedented suprafacial symmetry allowed N[1,3]-sigma-
tropic shift with an inversion of the configuration in the migrating nitrogen atom.
Received 11th January 2014,
Accepted 17th April 2014
DOI: 10.1039/c4ob00080c
www.rsc.org/obc
Introduction
The [1,3]-sigmatropic shift,
a typical thermal pericyclic
rearrangement, is a powerful strategy for the construction of
biologically active molecules in synthetic organic chemistry.1–5
The structure of the transition state and the configuration of
the products from the [1,3]-sigmatropic shift have been pre-
dicted by the Woodward–Hoffmann selection rule through the
suprafacial symmetry of the frontier molecular orbital
approach with the inversion of the configuration in the
migrating groups (Fig. 1).6 Among them, the C[1,3]-sigma-
tropic shift has been widely explored experimentally and theoreti-
cally,2 while the O/N[1,3]-sigmatropic shift has been rarely
reported. Recently, we have offered mechanistic insight into
the O[1,3]-sigmatropic shift in the abnormal Claisen rearrange-
ment.7 Now we are particularly interested in the N[1,3]-sigma-
tropic shift.
Fig. 1 Concerted [1,3]-sigmatropic shift with inversion of the configur-
ation in the migrating group.
Acid-catalyzed benzidine rearrangements have been studied
extensively over more than 150 years,8 in which the parent N,
Scheme 1 Benzidine rearrangement of N,N’-diphenylhydrazine (1).
N′-diphenyl hydrazine (1) gives p-benzidine (2, 70%) and
diphenyline (3, 30%)9,10 as the main products and some other
secondary products such as o-benzidine (4), p-semidine (5),
and o-semidine (6) (Scheme 1).11 In some cases, o-benzidine,
p-semidine, and o-semidine type products were obtained
in considerable yields from certain substituted N,N′-diaryl
hydrazines.
State Key Laboratory of Chemical Resource Engineering, Department of Organic
Chemistry, Faculty of Science, Beijing University of Chemical Technology, Beijing
100029, People’s Republic of China. E-mail: jxxu@mail.buct.edu.cn;
Tel: +86 10 64435565
†Electronic supplementary information (ESI) available: Details of rearrangement
reactions, copies of 1H NMR and 13C NMR spectra of products, copies of LC-MS
diagrams, and computational details. See DOI: 10.1039/c4ob00080c
‡These two authors contributed equally.
4952 | Org. Biomol. Chem., 2014, 12, 4952–4963
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
A large amount of work has been devoted to the mechanis-
tic investigation of the benzidine rearrangements, where the
controversies were concentrated on the polar transition state
theory (one concerted step) and Dewar’s π complex theory (two
stepwise steps).12 The question as to whether the rearrange-
ment occurred by
a stepwise or concerted mechanism
remained unresolved until measurements of heavy-atom
kinetic isotope effects (KIEs) were performed by Shine and co-
workers.13–15 A concerted [5,5]-sigmatropic shift was proposed
on the basis of KIE results on nitrogen and carbon atoms for
the formation of 2. Furthermore, an inverse secondary deuter-
ium isotopic effect for the disappearance of 1 supported the
conclusion drawn from the nitrogen and carbon KIE
results.13b,c In contrast, the formation of diphenyline (3) was
characterized via a substantial KIE for the N atom but with a
slight 2,2′,6,6′-13C4 KIE, in accord with an intramolecular and
nonconcerted mechanism.13c In addition, KIE results for the
formation of o-benzidines from N,N′-di(2-naphthyl)hydrazine
and N-2-naphthyl-N′-phenylhydrazine were clearly indicative of
Scheme 2 Acid-catalyzed rearrangement of 2,4’,6-trisubstituted
N,N’-diarylhydrazines 7.
a [3,3]-sigmatropic rearrangement.14 Nitrogen and carbon KIEs we designed N,N′-diaryl hydrazines with 2,4′,6-substituents in
for the conversion of N-4-methoxyphenyl-N′-phenylhydrazine order to prevent the formation of p-benzidine and o-benzidine
to the corresponding p-semidine and o-semidine were products, simplifying the separation and determination of
observed and the p-semidine rearrangement was assumed to the rearrangement products. N,N′-Diaryl hydrazines 7 with
be likely a concerted [1,5]-sigmatropic shift, whereas there was different substituents were synthesized from 2,6-disubstituted
a
slight 2,2′,6,6′-13C4 KIE for the o-semidine rearrange- N′-Boc-N-aryl hydrazines and 4-substituted aryl halides by a
ment.15a–d Accordingly, the π-complex theory was ruled out by Cu(I)-catalyzed coupling reaction.19
Shine’s kinetic experiments, but it has been revived by recent
We envisioned that the 2,4′,6-trisubstituted N,N′-diaryl
hydrazines 7 would give rise to semidines and diphenylines
calculations.16,17
To date, the mechanisms for the formation of p-benzidines (Scheme 2). N,N′-Diaryl hydrazine 7a was first examined upon
and o-benzidines in benzidine rearrangements have been veri- reflux in 95% ethanol for 2 h in the presence of concentrated
fied clearly as [5,5]- and [3,3]-sigmatropic shifts, respectively. HCl. After workup, we obtained the expected diphenyline 8a in
However, to the best of our knowledge, the formation of 5% yield and p-semidine 9a in 5% yield, concomitant with
diphenylines, p-semidines, and o-semidines seem to undergo the disproportionation products such as azobenzene 11a
ambiguous pathways.18 Further unraveling the mechanisms and corresponding arylamines 12a and 13a. However, no
remain highly desirable in organic chemistry. After analyzing o-semidine type product 10a was observed. Other two nitro
the structures of diphenylines, p-semidines, and o-semidines substituted N,N′-diaryl hydrazines 7b–c gave similar results.
and considering the existence of the C[1,3] and O[1,3]-sigma- With a trifluoromethyl substituent, N,N′-diaryl hydrazines 7d–f
tropic shifts, we propose that the N[1,3]-sigmatropic shift may underwent acid-catalyzed rearrangement to provide better
be involved in the formation of diphenylines, p-semidines, and results, affording 10–21% yields of the diphenylines 8d–f and
o-semidines in the benzidine rearrangement. Herein, we 16–18% yields of p-semidines 9d–f with the corresponding
present our detailed experimental and computational studies disproportionation products (see ESI† for details). Moreover,
on the N[1,3]-sigmatropic shift in the formation of semidines all reactions were subjected to LC-MS analysis without the
and diphenyline in the benzidine rearrangement. We believe observation of the o-semidines 10.
that our in-depth mechanistic insight into the N[1,3]-sigma-
To obtain o-semidine type products, we designed 2,4,4′,6-
tropic shift in the benzidine rearrangement is critical not only tetrasubstituted N,N′-diaryl hydrazines 14, which would sup-
to understand the benzidine rearrangement completely, but press the formation of diphenyline and p-semidine products.
also to enrich the theory of heteroatom [1,3]-sigmatropic shifts.
Similarly, N,N′-diaryl hydrazines 14 were synthesized from
2,4,6-trisubstituted N′-Boc-N-aryl hydrazines and 4-substituted
aryl halides via a Cu(I)-catalyzed coupling reaction. To our
delight, the substrates 14a and 14b underwent acid-catalyzed
rearrangement to provide the designed o-semidine type pro-
ducts 15a and 15b in 35% and 16% yields, respectively
(Scheme 3), concomitant with the disproportionation products
Results and discussion
Experimental investigation on the acid-catalyzed semidine and
diphenyline rearrangements
Since the benzidine rearrangements can undergo a concerted 13, 16 and 17 (see ESI† for details).
[5,5]-sigmatropic rearrangement to produce p-benzidines as major
Therefore, 2,4′,6-trisubstituted N,N′-diaryl hydrazines under-
products, or a [3,3]-sigmatropic shift to yield o-benzidines, went acid-catalyzed rearrangement to produce diphenyline
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 4952–4963 | 4953

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 Acid-catalyzed rearrangement of 2,4,4’,6-tetrasubstituted
N,N’-diarylhydrazines 14.
type products (up to 21% yield) and p-semidine type products
(up to 18% yield), while 2,4,4′,6-tetrasubstituted N,N′-diaryl
hydrazines gave rise to o-semidine type products (up to 35%
yield). In all the acid-catalyzed rearrangements, competitive
disproportionation reactions were inevitable.
Scheme 5 Radical trapping experiments in the formation of o-semi-
dine, p-semidine, and diphenyline type products.
Control experiments
standard conditions. The radical trapper has no significant
effect on the conversion of the rearrangement, excluding a
radical mechanism in a solvent cage. This is consistent with
the radical-free process reported by Shine.15 The results indi-
cate that even if biradicals were generated, they formed dispro-
portionation products rather than semidines and diphenylines.
It is unclear whether the formation of semidines and dipheny-
lines are intramolecular or intermolecular processes, although
it is well-known that the benzidine rearrangement is an intra-
molecular reaction for the formation of o- and p-benzidines.20
We performed intercrossing experiments to clarify the mech-
anism (Scheme 4). A mixture of equimolar amounts of N,N′-
diaryl hydrazines 14a and 14b was treated under the standard
conditions. Only two o-semidines 15a and 15b were detected
without the intercrossing o-semidines, as determined by
LC-MS analysis (see ESI† for details). Likewise, the intercross-
ing experiment with equimolar amounts of the N,N′-diaryl
hydrazines 7a and 7e was conducted and produced two
diphenylines 8a and 8e, as well as two p-semidines 9a and 9e
without any intercrossing products (see ESI† for details). The
formation of no intercrossing products indicates that all the
rearrangements are intramolecular processes.
Computational studies
DFT calculations21 using the B3LYP/6-311++G(d,p) level were
employed to locate all the stationary points involved.22 Fre-
quency calculations at the same level were performed to
confirm each stationary point to be either an intermediate or a
transition state structure. The free energies in solution were
computed by a self-consistent reaction field (SCRF) using the
conductor polarizable continuum model (CPCM) method in
ethanol at the same level.23
o-Semidine rearrangement. Our intercrossing and radical
trapping experiments have excluded ionic and radical mecha-
nisms. Thus, the formation of o-semidine products should be
an intramolecular process for the acid-catalyzed benzidine
rearrangement. We propose two possible intramolecular
mechanisms for the o-semidine rearrangement: (1) a concerted
N[1,3]-sigmatropic shift with a configuration inversion of the
nitrogen atom, which is orbital symmetry allowed (Scheme 6,
pathway A); (2) a tandem [3,3]- and C[1,3]-sigmatropic shifts
process with a configuration inversion of the carbon atom
(Scheme 6, pathway B).
To rule out a solvent-caged biradical mechanism for the
rearrangements, radical trapping experiments were performed
as well (Scheme 5). Treatment of hydrazine 7d under the stan-
dard conditions with TEMPO gave rise to diphenyline 8d in
22% yield and p-semidine 9d in 17% yield. Hydrazine 14a
afforded o-semidine 15a in 33% yield with TEMPO under the
To understand the above possible mechanisms, mono-pro-
tonated hydrazine 14a-H as a representative model system was
examined by DFT calculations (Fig. 2). Diprotonated hydrazine
14a-2H was also taken into account, whereas the scission of
the N–N bond took place spontaneously possibly due to the
unstable vicinal dicationic structure. Another mono-proto-
nated hydrazine 14a-H′ is more unstable than 14a-H by
1.6 kcal mol−1 in terms of Gibbs free energy due to protona-
tion on the weaker basic nitrogen atom. 14a-H undergoes a
concerted N[1,3]-sigmatropic shift through a transition state
o-TS1 with an activation free energy of 11.0 kcal mol−1 to
afford a stable intermediate o-Int1 (pathway A). In o-TS1, the
Scheme 4 Intercrossing experiments for the formation of o-semidines,
p-semidines, and diphenylines.
4954 | Org. Biomol. Chem., 2014, 12, 4952–4963
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 3 Structures and distortion/interaction analysis of transition states
for the o-semidine rearrangement. Distances of interest are reported in
angstroms.
Scheme 6 Proposed mechanism of the o-semidine rearrangement.
In addition, we attempted to calculate the potential profiles
from 14a-H′ for the sake of comparison. However, unfortu-
nately, its o-TS1 cannot be located. On the other hand, the two
stereoisomers of the o-TS1 were also considered in the calcu-
lation (Fig. 2). On optimization, the potential energy of the
o-TS1 with exo-H decreased continuously till the o-TS1 with
exo-H was converted to the o-TS1 with endo-H because the
endo-H o-TS1 shows less steric hindrance than the exo-H one.
Additionally, a weak H–π interaction exists in endo-H o-TS1
due to a distance between the H and the benzene ring of
approximately 2.6 Å (see ESI† for details) and without the
repelling interaction between the lone pair of electrons on the
nitrogen and the π electron cloud of benzene ring (or called
cyclohexadiene part). Both steric and electronic effects indicate
that the transition state endo-H o-TS1 is more stable than exo-
H o-TS1. A similar phenomenon was observed in o-TS3. The
results indicate that steric hindrance plays an important role
in the stabilization of the endo-H transition states in both N
and C[1,3] sigmatropic shifts.
Distortion/interaction analysis,24 which is a powerful tool to
understand the factors that stabilize the transition states, was
employed to allow for deep understanding of the main reasons
why o-TS1 is lower in energy than o-TS2 (Fig. 3). The activation
energy (ΔE‡) can be mainly separated into the distortion
energy of anilines (ΔE‡dist) and the interaction energy between
the two distorted fragments (ΔE‡int). In o-TS1, the interaction
energy between the two fragments is very small, but both frag-
Fig. 2 Free energy profiles for the o-semidine rearrangement.
computed distances of the N–N bond breaking and the N–C ments are hardly distorted from initial equilibrium geome-
bond making are 2.75 and 2.95 Å, respectively (Fig. 3). The tries. In contrast, there is much more distortion of the
alternative pathway B involves a [3,3]-sigmatropic shift via a fragments in o-TS2, and this is only partially compensated for
transition state o-TS2 with an activation free energy of by more effective interaction. Thus, the pathway through o-TS1
20.4 kcal mol−1, leading to an intermediate o-Int2. The dis- is favorable for the o-semidine rearrangement, which is domi-
tances of the N–N bond cleavage and the C–C bond formation nantly attributed to low distortion in the transition state.
in o-TS2 are 2.83 and 2.13 Å, respectively (Fig. 3). The following
In terms of the o-semidine rearrangement, Shine and
C[1,3]-sigmatropic shift via a transition state o-TS3 is almost co-workers have reported KIE for the rearrangements of N-4-
barrierless to form o-Int1. In the o-TS3, the computed C–C and methoxyphenyl-N′-phenylhydrazine (18)15a and N,N′-di(4-chloro-
C–N bonds in the four-membered ring transition state are phenyl)hydrazine (21) (Scheme 7).15b–e The formation of
almost dissociated due to the rigid ring. Finally, the assistance o-semidine 19 from [15N,15N′]-18 resulted in an average KIE of
of a water molecule facilitates the tautomerization from the 1.074, while the generation of o-semidine 22 from [14N,15N′]-21
o-Int1 to 15a-H.
and [2,2′,6,6′-13C4]-21 furnished KIEs of 1.0155 and 0.9963,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 4952–4963 | 4955

View Article Online
Paper
Organic & Biomolecular Chemistry
but stepwise pathways: tandem N[1,3]/[3,3]-sigmatropic shifts
(pathway C1) and cascade [3,3]/C[1,3]-sigmatropic shifts
(pathway D1). For the formation of p-semidines, we also
propose two possible intramolecular but stepwise pathways:
tandem N[1,3]/N[1,3]-sigmatropic shifts (pathway C2) and
cascade [3,3]/[3,3]-sigmatropic shifts (pathway D2) (Scheme 8).
A
representative model system with mono-protonated
hydrazine 7a-H was examined by DFT calculations to under-
stand the proposed pathways (Fig. 4). 7a-H could undergo the
N[1,3]-sigmatropic shift via a transition state dp-TS1 with an
activation free energy of 17.0 kcal mol−1. This step of the reac-
tion is exergonic by 15.8 kcal mol−1, giving a stable intermedi-
ate dp-Int1 (pathway C). Once the intermediate dp-Int1 is
formed, two pathways can be followed: (i) the [3,3]-sigmatropic
Scheme 7 Acid-catalyzed rearrangement of N,N’-diarylhydrazines
18 and 21.
respectively. The 15N KIE is more obvious than the 13C KIE, shift of dp-Int1 via transition state d-TS2 requires an activation
and even an inverse 13C kinetic isotope effect was observed energy of 31.5 kcal mol−1 (but only 15.7 kcal mol−1 in terms of
close to unity. The KIE results indicate that the transition state the Gibbs free energy) to yield an unstable intermediate d-Int2,
in the rate-determining step should be an early (reactant-like) followed by tautomerization to deliver mono-protonated diphe-
transition state rather than a late transition state, consistent nyline 8a-H (pathway C1); (ii) dp-Int1 undergoes another
with the calculated results because o-TS1 in the N[1,3] sigma- N[1,3]-sigmatropic shift via a transition state p-TS2, requiring an
tropic shift is an early transition state, while o-TS2 in the [3,3] activation energy of 36.3 kcal mol−1 (20.5 kcal mol−1 in terms of
sigmatropic shift is a late transition state. Thus, Shine’s KIE the Gibbs free energy) to give rise to a stable intermediate
results support the N[1,3]-sigmatropic shift mechanism for the p-Int2 (pathway C2). Alternatively, 7a-H could also undergo a
formation of o-semidine.
[3,3]-sigmatropic shift via transition state dp-TS3 with an acti-
Diphenyline and p-semidine rearrangements. For the diphenyl- vation energy of 24.2 kcal mol−1 to yield an unstable inter-
ine rearrangement, we put forward two possible intramolecular mediate dp-Int3 (pathway D), which can further undergo two
Scheme 8 Proposed mechanism for diphenyline and p-semidine rearrangements.
4956 | Org. Biomol. Chem., 2014, 12, 4952–4963
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 Potential energy profiles of the diphenyline and p-semidine rearrangements.
different rearrangements: (i) the consequent C[1,3]-sigma- rate-limiting step for the formation of diphenyline 8a, while
tropic shift of dp-Int3 through d-TS4 requires 32.9 kcal mol−1 the second N[1,3] shift is both the rate-limiting and the rate-
in terms of the Gibbs free energy (the activation energy of determining step for the generation of p-semidine 9a. The
16.2 kcal mol−1), leading to 8a-H (pathway D1); (ii) the [3,3]- computed distances of all transition states are in a reasonable
sigmatropic shift of dp-Int3 results in the formation of 9a-H range except for those of d-TS4 which reach the extent of scis-
with an activation energy of 5.0 kcal mol−1 (21.7 kcal mol−1 in sion (Fig. 5).
terms of the Gibbs free energy) (pathway D2).
Shine and co-workers have reported the KIEs for the
Therefore, pathway C is favored over pathway D by 7.2 kcal rearrangement of N,N′-diphenylhydrazine (1) into diphenyline
mol−1 in the first step of the tandem processes and predomi- (3).13e,14d The KIEs for [15N,15N′]-1 and [2,2′,6,6′-13C4]-1 are
nant in each of the second steps in terms of the Gibbs free 1.0367 and 0.9953, respectively. For [4,4′-13C2]-1, no carbon
energy. Although the two second steps in pathway D could KIE was observed. These findings imply that the cleavage of
occur with relative lower activation energies (16.2 kcal mol−1 the N–N bond and the formation of the N–C2 bond are in the
and 5.0 kcal mol−1, respectively) than those in pathway C, it is rate-limiting step. Similarly, a more obvious 15N KIE (variation
very difficult for the first step reaction in pathway D due to its magnitude) was observed than 13C KIE, an inverse 13C KIE.
higher activation energy (24.2 kcal mol−1) and endergonic The KIE results reveal that the transition state in the rate-limit-
process. However, the intermediate dp-Int1 is more stable than ing step should be an early (reactant-like) transition state
dp-Int3 by 32.5 kcal mol−1. In the consequent step from rather than a late transition state, consistent with our calcu-
dp-Int1, two different pathways can correspond to mono-proto- lated tandem N[1,3]/[3,3] sigmatropic shift mechanism with
nated diphenyline 8a-H (pathway C1) and mono-protonated the first N[1,3] sigmatropic shift as the rate-limiting step
p-semidine 9a-H (pathway C2). Eventually, our calculated because the dp-TS1 in the first N[1,3] sigmatropic shift is
results draw the conclusion that the first N[1,3] shift is the an early transition state with a higher potential energy of
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 4952–4963 | 4957

View Article Online
Paper
Organic & Biomolecular Chemistry
respectively.15b–e The results indicate that the rate-limiting step
involves the nitrogen atom, ortho and para carbon atoms, and
also matched with our proposed tandem N[1,3]/N[1,3]-sigma-
tropic shift mechanism with the second N[1,3]-sigmatropic
shift as the rate-limiting step for the formation of p-semidine.
From a viewpoint of energy, it is reasonable to consider the
formation mechanism of p-semidines as a tandem N[1,3]/
N[1,3]-sigmatropic shift mechanism with the second N[1,3]-sig-
matropic shift as the rate-limiting step. However, the carbon
KIEs for both ortho and para carbon atoms are very small,
unlike the transition states o-TS1 and dp-TS1, no obvious
relationship between their variation magnitudes and the tran-
sition state structure of p-TS2 is observed due to the experi-
mental determination precision. The small carbon KIEs can
possibly be attributed to the rigid benzene ring involved in the
transition state p-TS2. Small carbon KIEs were observed in
several cyclic transition states previously.25–27
In the rearrangements of diphenylines, o- and p-semidines,
the inverse carbon KIE is generally observed. Unlike deuterium
KIE, the heavy atom primary inverse KIEs have seldom been
observed previously.28 They were assumed to generate due to
nonlinearity in the transition states, causing bending modes
in addition to stretching modes in vibrations. In our investi-
gated rearrangements, all the transition states in the rate-limit-
ing steps are four-membered ring ones. Thus, the inverse
carbon KIE can be attributed to the nonlinear transition states
in the rate-limiting steps.
Influence of substituents on transition states and KIEs. The
substituents can change the transition states (early or late tran-
sition states), resulting in KIE changes, even normal to inverse
or inverse to normal.28 Our investigated N,N′-diarylhydrazines
are different from those in the KIE experiments. To verify the
impact of the substituents on the phenyl group(s) on the tran-
sition states in the semidine and diphenyline rearrangements,
we further calculated the potential energy profiles for the for-
mation of semidines and diphenylines from N,N′-diarylhydra-
zines 1, 18, and 21 (Table 1). The results indicated that all the
transition states o-TS1 for N,N′-diarylhydrazines 14a, 1, 18, and
21 are early transition states (Table 1, columns 1 and 2), indi-
cating that these diarylhydrazines should show similar 15N
and 13C KIEs in their o-semidine rearrangements. That is, the
reported KIEs of 18 and 21 can represent those in our studied
system of N,N′-diarylhydrazine 14a.
Fig. 5 Structures of transition states for the p-semidine and diphenyl-
ine rearrangements. Distances of interest are reported in angstroms.
17.0 kcal mol−1, while the d-TS2 in the second [3,3] sigma-
tropic shift is a late transition state. On the other hand, the
transition state dp-TS3 is a late transition state and locates at a
higher potential energy of 24.2 cal mol−1 as well. However
d-TS4 is also an early transition state, but with the highest poten-
tial energy of 32.9 cal mol−1. After this analysis, we can con-
clude that our proposed tandem N[1,3]/[3,3]-sigmatropic shifts
and the first N[1,3] sigmatropic shift as the rate-limiting step
for the formation mechanism of the diphenylines are consist-
ent with Shine’s KIE observation.
The averaged KIEs for the formation of p-semidine 20 from
[15N,15N′]-18 and [4′-14C]-18 were measured as 1.0296 and
1.039, respectively.15a,e On the basis of the KIE results, a con-
certed [1,5]-sigmatropic shift through a six-membered ring
transition state from the protonated 18 was proposed by Shine.
However, the proposed transition state for the concerted [1,5]-
sigmatropic shift seems to have a large distorted energy with
the rigid benzene ring. It is not a reasonable process. However,
the KIEs support our tandem N[1,3]/N[1,3] sigmatropic shift
process with the second one as the rate-limiting step.
In addition, the formation of p-semidine 23 from [15N,15N′]-
21, [14N,15N′]-21, [4,4′-13C2]-21, [4-14C]-21, and [2,2′,6,6′-13C4]-21
furnished KIEs of 1.0282, 1.0162, 0.9934, 1.0029, and 0.9973,
For p-semidine rearrangements, except for N,N′-diphenyl-
hydrazine 1, for which p-TS2 cannot be located in its calculation
Table 1 The calculated energies of transition states and intermediates in the formation of diphenylines and semidines from different diarylhydra-
zines in pathways A and C (ΔG and ΔG≠ in kcal mol−1
)
Hydrazine
ΔG≠ (o-TS1)
ΔG (o-Int1)
ΔG≠ (dp-TS1)
ΔG (dp-Int1)
ΔG≠ (d-TS2)
ΔG (d-Int2)
ΔG≠ (p-TS2)
ΔG (p-Int2)
14a
7a
1
18
21
11.0
—
14.7
14.2
12.8
−11.0
—
−13.0
−8.9
−10.8
—
—
—
—
—
20.5
NL
16.4
13.3
—
−22.1
NC
−12.4
−19.1
17.0
14.7
14.8
12.8
−15.0
−13.0
1.7
15.7
15.0
NL
4.0
4.7
NC
NC
−10.8
NL
NL = not located. NC = not calculated.
4958 | Org. Biomol. Chem., 2014, 12, 4952–4963
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
(consistent with trace p-semidine 5 generated in the exper-
iment), the transition states p-TS2 for both N,N′-diarylhydra-
zines 18 and 21 show higher potential energy than the
Experimental section
General information
corresponding dp-TS1 and are early transition states, indicat- Melting points were obtained on a melting point apparatus
ing that the second N[1,3] sigmatropic shift is the rate-limiting and are uncorrected. 1H NMR and 13C NMR spectra were
step in the formation of the corresponding p-semidines 20 and recorded on a 300 MHz or 400 MHz spectrometer with TMS as
23 as in the formation of p-semidine 9a. The hydrazine 21 an internal standard in CDCl3 solution. IR spectra were taken
shows a similar potential energy profile to that of the hydra- on a FT-IR spectrometer in KBr. HRMS data were obtained
zine 7a in the p-semidine rearrangement. Thus, the KIEs of 21 with an LC/MSD TOF mass spectrometer. Purification of reac-
should represent those of 7a. However, the hydrazine 21 pos- tion products was carried out by column chromatography
sesses lower energy barriers than 7a in both N[1,3] sigmatropic using silica gel (200–300 mesh). TLC separations were per-
shifts. This is the reason why 7a yields 9a in a low yield of 5%, formed on silica gel G plates with petroleum ether–ethyl
while 21 generated 23 in a relatively high yield of 12%.
acetate, and the plates were visualized with UV light.
Considering the diphenyline rearrangements, transition
states d-TS2 are not located in the calculation for hydrazines
18 and 21, in agreement with no observation of the corres-
General procedure for the synthesis of N-Boc-N,N′-diaryl
hydrazines 7 and 14
ponding diphenylines experimentally. Thus, only N,N′-diphenyl- A round bottom flask was charged with N′-Boc-N-aryl hydrazine
hydrazine 1 was compared with hydrazine 7a. However, (24, 48 mmol), 4-substituted iodobenzene (40 mmol), CuI
unexpected results were obtained. For hydrazine 7a, its tran- (0.78 g, 4 mmol), 1,10-phenanthroline (1.44 g, 8 mmol),
sition state d-TS2 is slightly higher (0.3 kcal mol−1) than its Cs2CO3 (15.64 g, 48 mmol) and 40 mL of dry DMF at room
dp-TS1 in terms of the Gibbs free energy, but a less obvious temperature. The reaction mixture was degassed, charged with
difference is seen from the viewpoint of calculation. Importantly, N2 gas and heated to 80 °C. After 4–5 h, the resulting mixture
the dp-TS1 of hydrazine 1 is an early transition state as in was cooled to room temperature, diluted with ethyl acetate
hydrazine 7a. It is matched with the variation magnitudes of (100 mL), and filtered. The filtrate was then washed twice with
its 15N and 13C KIEs. The KIEs of hydrazine 1 could reflect a brine (2 × 100 mL). The organic layer was dried over anhydrous
feature of the diphenyline rearrangements. Although it is magnesium sulfate, filtered, and concentrated under reduced
difficult to provide an undoubted mechanism for the for- pressure. The residue was purified by flash chromatography
mation of diphenylines in the benzidine rearrangement on the with a mixture of petroleum ether and ethyl acetate as an
basis of the current information, our proposed mechanism is eluent to afford the desired product, which was recrystallized
a more reasonable one than proposed until now and our from a mixture of petroleum ether and ethyl acetate to give
current investigation provides a further new insight and com- crystals 7 or 14.
prehensive understanding on the o- and p-semidine and
diphenyline rearrangements.
tert-Butyl 2-(2,6-dimethylphenyl)-1-(4-nitrophenyl)hydrazi-
necarboxylate (7a). Orange crystals, 2.57 g, yield 18%,
m.p. 153–154 °C, H NMR (300 MHz, CDCl3) δ: 1.27 (s, 9 H),
1
2.17 (s, 6 H), 6.27 (s, 1H), 6.82 (t, J = 7.4 Hz, 1 H), 6.97 (d, J =
7.4 Hz, 2 H), 8.07–8.12 (m, 2 H), 8.19–8.25 (m, 2 H). 13C NMR
(75 MHz, CDCl3) δ: 153.3, 149.3, 143.3, 142.8, 129.7, 124.8,
Conclusion
In summary, to search for the existence of the N[1,3]-sigma- 124.1, 121.9, 119.7, 83.7, 27.6, 19.0. IR (KBr) ν (cm−1): 3359,
tropic shift and to further elucidate the formation mechanisms 2978, 2932, 1721, 1590, 1514, 1476, 1308. HRMS (ESI) calcd for
of o/p-semidines and diphenylines, we have investigated the C19H23N3O4 [M + H]+ m/z: 358.1761, found 358.1769.
mechanisms of the acid-catalyzed semidines and diphenyline
tert-Butyl 2-(2-ethyl-6-methylphenyl)-1-(4-nitrophenyl)hydra-
rearrangements with designed N,N′-diarylhydrazines. After a zinecarboxylate (7b). Orange crystals, 2.38 g, yield 16%,
systematic investigation of experiments and theoretical calcu- m.p. 118–119 °C, 1H NMR (300 MHz, CDCl3) δ: 1.19 (t, J =
lations, it is reasonable to consider the acid-catalyzed o-semi- 7.5 Hz, 3 H), 1.26 (s, 9 H), 2.15 (s, 3 H), 2.56 (q, J = 7.5 Hz,
dine rearrangement as a N[1,3]-sigmatropic shift, p-semidine 2 H), 6.34 (s, 1H), 6.85–7.03 (m, 3 H), 8.06–8.12 (m, 2 H),
rearrangement as tandem N[1,3]/N[1,3]-sigmatropic shifts, and 8.20–8.25 (m, 2 H). 13C NMR (50 MHz, CDCl3) δ: 153.4, 149.4,
diphenyline rearrangement as cascade N[1,3]/[3,3]-sigmatropic 143.0, 142.8, 131.2, 129.8, 127.3, 125.3, 124.1, 122.1, 119.8,
shifts. The N[1,3]-sigmatropic shift is orbital suprafacial sym- 83.7, 27.6, 24.7, 19.5, 14.2. IR (KBr) ν (cm−1): 3359, 2974, 2932,
metry allowed with an inversion of the migrating nitrogen 2875, 1716, 1590, 1515, 1469, 1342, 1113. HRMS (ESI) calcd for
atom. The proposed intramolecular processes are supported C20H25N3O4 [M + H]+ m/z: 372.1918, found 372.1914.
by intercrossing experiments, radical trapping experiments,
tert-Butyl 2-(2,6-diethylphenyl)-1-(4-nitrophenyl)hydrazine-
and KIE observations measured by Shine. The current results carboxylate (7c). Orange crystals, 2.00 g, yield 13%,
not only provide a comprehensive understanding of the for- m.p. 164–165 °C, 1H NMR (300 MHz, CDCl3) δ: 1.15 (t, J =
mation of o/p-semidines and diphenylines in the benzidine 7.5 Hz, 6 H), 1.26 (s, 9 H), 2.52 (q, J = 7.5 Hz, 4 H), 6.37
rearrangement, but also disclose a novel N[1,3]-sigmatropic (s, 1 H), 6.94–7.04 (m, 3 H), 8.07–8.12 (m, 2 H), 8.21–8.26
shift that has potential mechanistic possibility in other reactions. (m, 2 H). 13C NMR (50 MHz, CDCl3) δ: 153.5, 149.4, 143.2,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 4952–4963 | 4959

View Article Online
Paper
Organic & Biomolecular Chemistry
142.3, 131.9, 127.3, 124.1, 122.6, 120.1, 83.8, 27.7, 24.9, 14.4. General procedure for the acid-catalyzed rearrangements of N,
IR (KBr) ν (cm−1): 3349, 2983, 1721, 1589, 1492, 1441, 1339. N′-diaryl hydrazines 7 and 14
HRMS (ESI) calcd for C21H27N3O4 [M + H]+ m/z: 386.2074,
A round bottom flask was charged with N,N′-diaryl hydrazine
found 386.2089.
tert-Butyl 2-(2,6-dimethylphenyl)-1-(4-(trifluoromethyl)phenyl)-
hydrazinecarboxylate (7d). Colorless crystals, 5.63 g, yield
(7 or 14, 1 mmol), 95% ethanol (10 mL), and conc. HCl
(0.5 mL) under nitrogen at room temperature. The reaction
mixture was refluxed for 2 h, then cooled to room temperature,
neutralized with solid NaHCO3, filtered, and concentrated.
The residue was purified by flash column chromatography.
3′,5′-Dimethyl-5-nitro-1,1′-biphenyl-2,4′-diamine (8a). Yellowish
crystals, 13 mg, yield 5%, m.p. 186–188 °C, 1H NMR (300 MHz,
CDCl3) δ: 2.23 (s, 6 H), 3.73 (s, 2 H), 4.54 (s, 2 H), 6.67 (q, J1 =
2.6 Hz, 1 H), 7.01 (s, 2 H), 8.01 (dd, J1 = 2.6 Hz, J2 = 2.3 Hz,
2 H). 13C NMR (126 MHz, CDCl3) δ: 150.1, 142.9, 139.2,
128.6, 127.1, 126.7, 126.2, 124.5, 122.3, 113.6, 17.7. IR (KBr)
ν (cm−1): 3433, 3340, 2962, 2873, 1284. HRMS (ESI) calcd for
C14H15N3O2 [M + H]+ m/z: 258.1237, found 258.1245.
1
37%, m.p. 105–106 °C, H NMR (300 MHz, CDCl3) δ: 1.28 (s,
9 H), 2.19 (s, 6 H), 6.23 (s, 1 H), 6.80 (t, J = 7.5 Hz, 1 H), 6.95
(d, J = 7.5 Hz, 2 H), 7.59 (d, J = 8.7 Hz, 2 H), 7.95 (d, J = 8.7 Hz,
2 H). 13C NMR (50 MHz, CDCl3) δ: 153.8, 146.7, 143.6, 129.6,
125.3 (q, J1 = 3.8 Hz), 125.2 (q, J2 = 32.5 Hz), 125.0, 124.3 (q,
J3 = 270 Hz), 121.6, 120.4, 82.8, 27.6, 18.9. IR (KBr) ν (cm−1):
3340, 2982, 1697, 1618, 1525, 1474, 1323. HRMS (ESI) calcd for
C20H23F3N2O2 [M + H]+ m/z: 381.1784, found 381.1799.
tert-Butyl 2-(2-ethyl-6-methylphenyl)-1-(4-(trifluoromethyl)-
phenyl)hydrazinecarboxylate (7e). Colorless crystals, 7.10 g,
yield 45%, m.p. 121–121.5 °C, 1H NMR (300 MHz, CDCl3) δ:
1.18 (t, J = 7.5 Hz, 3 H), 1.28 (s, 9 H), 2.17 (s, 3 H), 2.56 (q, J =
7.5 Hz, 2 H), 6.30 (s, 1H), 6.84–7.02 (m, 3 H), 7.58–7.61 (m,
2 H), 7.93–7.96 (m, 2 H). 13C NMR (50 MHz, CDCl3) δ: 153.9,
146.6, 143.0, 131.4, 129.7, 127.2, 125.5, 125.5 (q, J2 = 32.5 Hz),
125.4 (q, J3 = 3.7 Hz), 124.3 (q, J1 = 270 Hz), 122.0, 120.7, 83.0,
27.7, 24.7, 19.5, 14.3. IR (KBr) ν (cm−1): 3367, 2975, 2932, 2869,
1720, 1615, 1469, 1322, 1160, 1115. HRMS (ESI) calcd for
C21H25F3N2O2 [M + H]+ m/z: 395.1941, found 395.1937.
3′-Ethyl-5′-methyl-5-nitro-1,1′-biphenyl-2,4′-diamine (8b). Brown
1
crystals, 8 mg, yield 3%, m.p. 131–133 °C, H NMR (300 MHz,
CDCl3) δ: 1.29 (t, J1 = 7.5 Hz, 3 H), 2.34 (s, 3 H), 2.58 (q, J1 =
7.5 Hz, 2 H), 3.77 (s, 2 H), 4.55 (s, 2 H), 6.68 (dt, J2 = 1.6 Hz,
J3 = 2.7 Hz, 1 H), 7.02 (s, 2 H), 8.00 (d, J3 = 2.7 Hz, 1 H), 8.03 (d,
J2 = 1.6 Hz, 2 H). 13C NMR (126 MHz, CDCl3) δ: 150.3, 142.3,
138.9, 128.4, 128.0, 127.1, 126.7, 126.5, 126.2, 124.4, 122.5,
113.5, 24.2, 17.7, 12.9. IR (KBr) ν (cm−1): 3482, 3376, 2966,
2873, 1307. HRMS (ESI) calcd for C15H17N3O2 [M + H]+ m/z:
272.1394, found 272.1410.
N,N′-(3′,5′-Diethyl-5-nitro-1,1′-biphenyl-2,4′-diyl)diacetamide
(8c). A mixture (48 mg) of diphenyline and p-nitroaniline in
5 mL of (Ac)2O was stirred at room temperature for 12 h. The
resulting mixture was diluted with water (50 mL), and extracted
with ethyl acetate (2 × 50 mL). The organic layer was washed
with saturated NaHCO3 (50 mL), brine (50 mL), dried over
anhydrous Na2SO4, filtered, concentrated under reduced
pressure and purified by flash chromatography with a mixture
of petroleum ether and ethyl acetate as an eluent to afford 8c.
Yellowish solid, 15 mg, yield 4%, m.p. 230–232 °C, 1H NMR
(300 MHz, DMSO-d6) δ: 1.17 (t, J1 = 7.5 Hz, 6 H), 2.03 (s, 3 H),
2.10 (s, 3 H), 2.59 (q, J1 = 7.5 Hz, 4 H), 7.23 (s, 2 H), 7.99 (d,
J3 = 8.9 Hz, 1H), 8.12 (d, J2 = 2.6 Hz, 1H), 8.22 (dd, J2 = 2.6 Hz,
J3 = 8.9 Hz, 1H), 9.34 (s, 1 H), 9.68 (s, 1 H). 13C NMR (126 MHz,
DMSO-d6) δ: 169.1, 168.8, 143.9, 142.0, 141.3, 135.2, 135.1,
134.5, 126.4, 126.0, 125.3, 122.8, 24.4, 23.4, 22.6, 14.8. IR (KBr)
ν (cm−1): 3250, 3246, 2962, 2928, 2866, 2847, 1654, 1508, 1350,
1274. HRMS (ESI) calcd for C20H23N3O4 [M + H]+ m/z:
370.1761, found 370.1768.
tert-Butyl 2-(2,6-diethylphenyl)-1-(4-(trifluoromethyl)phenyl)-
hydrazinecarboxylate (7f). Colorless crystals, 4.41 g, yield 27%,
m.p. 83–84 °C, 1H NMR (300 MHz, CDCl3) δ: 1.15 (t, J = 7.5 Hz,
6 H), 1.27 (s, 9 H), 2.54 (q, J = 7.5 Hz, 4 H), 6.34 (s, 1H),
6.90–7.03 (m, 3 H), 7.58–7.61 (m, 2 H), 7.92–7.95 (m, 2 H).
13C NMR (50 MHz, CDCl3) δ: 154.1, 146.7, 142.5, 132.1, 127.3,
125.7 (q, J2 = 32.3 Hz), 125.4 (q, J3 = 3.7 Hz), 124.3 (q, J1
=
270 Hz), 122.4, 120.9, 83.0, 27.8, 24.9, 14.5. IR (KBr) ν (cm−1):
3365, 2971, 2935, 2876, 1720, 1615, 1456, 1322, 1160, 1123.
HRMS (ESI) calcd for C22H27F3N2O2 [M + H]+ m/z: 409.2097,
found 409.2086.
tert-Butyl 2-(4-cyano-2,6-dimethylphenyl)-1-(p-tolyl)hydra-
zinecarboxylate (14a). Colorless crystals, 4.50 g, yield 32%,
m.p. 157–159 °C, H NMR (300 MHz, CDCl3) δ: 1.33 (s, 9 H),
2.23 (s, 6 H), 2.32 (s, 3 H), 6.43 (s, 1 H), 7.14 (m, 2 H), 7.21 (s,
2 H), 7.47 (m, 2 H). 13C NMR (75 MHz, CDCl3) δ: 154.1, 148.0,
140.4, 134.3, 133.2, 128.9, 125.6, 121.6, 119.6, 103.7, 82.5, 27.9,
20.7, 19.0. IR (KBr) ν (cm−1): 3356, 2977, 2926, 2218, 1717.
HRMS (ESI) calcd for C21H25N3O2 [M + H]+ m/z: 352.2020,
found 352.2031.
1
tert-Butyl 2-(4-cyano-2-ethyl-6-methylphenyl)-1-(4-ethylphenyl)-
hydrazinecarboxylate (14b). Colorless crystals, 5.62 g, yield
37%, m.p. 164–164.5 °C, 1H NMR (300 MHz, CDCl3) δ: 1.20
(t, J = 7.5 Hz, 3 H), 1.23 (t, J = 7.5 Hz, 3 H), 1.33 (s, 9 H), 2.24
(s, 3 H), 2.59 (q, J = 7.5 Hz, 2 H), 2.64 (q, J = 7.5 Hz, 2 H), 6.45
(d, J = 4.1 Hz, 1 H), 7.17 (m, 2 H), 7.23 (s, 1 H), 7.28 (s, 1 H),
7.49 (m, 2 H). 13C NMR (75 MHz, CDCl3) δ: 154.1, 147.6, 140.5,
140.5, 133.1, 131.5, 130.8, 127.6, 126.0, 121.6, 119.7, 103.8,
82.3, 28.0, 27.8, 24.3, 19.4, 15.3, 13.6, 8.7. IR (KBr) ν (cm−1):
3343, 2964, 2929, 2219, 1718. HRMS (ESI) calcd for
C23H29N3O2 [M + H]+ m/z: 380.2333, found 380.2346.
3′,5′-Dimethyl-5-trifluoromethyl-1,1′-biphenyl-2,4′-diamine
(8d). Colorless crystals, 59 mg, yield 21%, m.p. 74–75 °C,
1H NMR (300 MHz, CDCl3) δ: 2.23 (s, 6 H), 3.70 (brs, 2 H), 4.06
(brs, 2 H), 6.73 (d, J = 9.0 Hz, 1 H), 7.02 (s, 2 H), 7.32 (m, 2 H).
13C NMR (126 MHz, CDCl3) δ: 146.8, 142.4, 128.6, 127.5, 127.5
(q, J = 3.6 Hz), 127.3, 124.9 (q, J = 270.6 Hz), 124.7 (q, J =
3.6 Hz), 122.11, 120.0 (q, J = 32.4 Hz), 114.4, 17.6. IR (KBr)
ν (cm−1): 3481, 3386, 2933, 2857, 1108. HRMS (ESI) calcd for
C15H15F3N2 [M + H]+ m/z: 281.1260, found 281.1276.
3′-Ethyl-5′-methyl-5-trifluoromethyl-1,1′-biphenyl-2,4′-diamine
(8e). Yellowish oil, 29 mg, yield 10%, 1H NMR (300 MHz,
4960 | Org. Biomol. Chem., 2014, 12, 4952–4963
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
CDCl3) δ: 1.28 (t, J1 = 7.5 Hz, 3 H), 2.23 (s, 3 H), 2.58 (q, J1 = (brs, 2 H), 5.63 (brs, 1 H), 6.79 (d, J = 8.4 Hz, 2 H), 6.80 (s, 2 H),
7.5 Hz, 2 H), 3.72 (brs, 2 H), 4.08 (s, 2 H), 6.75 (d, J2 = 8.4 Hz, 7.38 (d, J = 8.4 Hz, 2 H). 13C NMR (126 MHz, CDCl3) δ: 149.4,
1 H), 7.03 (s, 2 H), 7.33 (d, J2 = 8.4 Hz, 1H), 7.34 (s, 1H). 139.9, 130.9, 126.5 (q, J = 3.5 Hz), 124.9 (q, J = 270.4 Hz), 124.0,
13C NMR (126 MHz, CDCl3) δ: 146.8, 141.8, 128.4, 127.8, 127.6, 122.8, 119.6 (q, J = 32.3 Hz), 113.3, 17.7. IR (KBr) ν (cm−1):
127.4 (q, J = 3.6 Hz), 127.4, 126.5, 124.9 (d, J = 270.7 Hz), 124.7 3382, 2929, 2853, 1109. HRMS (ESI) calcd for C15H15F3N2
(q, J = 3.7 Hz), 122.3, 119.8 (q, J = 32.3 Hz), 114.4, 24.2, 17.6, [M + H]+ m/z: 281.1260, found 281.1274.
12.9. IR (KBr) ν (cm−1): 3481, 3386, 2966, 2873, 1108. HRMS
(ESI) calcd for C16H17F3N2 [M + H]+ m/z: 295.1417, found diamine (9e). Yellowish oil, 53 mg, yield 18%, 1H NMR
295.1414. (300 MHz, CDCl3) δ: 1.25 (t, J1 = 7.5 Hz, 3 H), 2.18 (s, 3 H), 2.52
3-Ethyl-5-methyl-N1-(4-(trifluoromethyl)phenyl)benzene-1,4-
3′,5′-Diethyl-5-trifluoromethyl-1,1′-biphenyl-2,4′-diamine (8f). (q, J1 = 7.5 Hz, 2 H), 3.60 (brs, 2 H), 5.65 (s, 1 H), 6.80 (d, J2 =
Yellowish oil, 43 mg, yield 14%, 1H NMR (300 MHz, CDCl3) 8.4 Hz, 2 H), 6.81 (s, 2 H), 7.38 (d, J2 = 8.4 Hz, 2H). 13C NMR
δ: 1.29 (t, J1 = 7.5 Hz, 6 H), 2.58 (q, J1 = 7.5 Hz, 4 H), 3.77 (brs, (126 MHz, CDCl3) δ: 149.4, 139.3, 131.1, 128.6, 126.5 (d, J =
2 H), 4.09 (brs, 2 H), 6.75 (d, J2 = 8.1 Hz, 1 H), 7.04 (s, 2 H), 3.7 Hz), 124.9 (d, J = 270.4 Hz), 123.8, 123.1, 121.9, 119.6 (q, J =
7.33 (d, J2 = 8.1 Hz, 1 H), 7.35 (s, 1 H). 13C NMR (126 MHz, 32.4 Hz), 113.3, 24.2, 17.8, 13.0. IR (KBr) ν (cm−1): 3385, 2934,
CDCl3) δ 146.8, 141.3, 128.1, 127.8, 127.6, 127.5 (q, J = 3.7 Hz), 2873, 1110. HRMS (ESI) calcd for C16H17F3N2 [M + H]+ m/z:
126.4, 124.9 (q, J = 270.7 Hz), 124.8 (q, J = 3.7 Hz), 120.0 (q, J = 295.1417, found 295.1411.
32.4 Hz), 114.5, 24.3, 13.0. IR (KBr) ν (cm−1): 3483, 3389, 2966,
3,5-Diethyl-N1-(4-(trifluoromethyl)phenyl)benzene-1,4-diamine
1
2874, 1108. HRMS (ESI) calcd for C17H19F3N2 [M + H]+ m/z: (9f). Yellow oil, 49 mg, yield 16%, H NMR (300 MHz, CDCl3)
309.1573, found 309.1577.
δ: 1.26 (t, J1 = 7.5 Hz, 6 H), 2.54 (q, J1 = 7.5 Hz, 4 H), 3.63 (brs,
3,5-Dimethyl-N1-(4-nitrophenyl)benzene-1,4-diamine (9a). 2 H), 5.69 (s, 1 H), 6.82 (d, J2 = 8.4 Hz, 2 H), 6.84 (s, 2 H), 7.39
Orange crystals, 13 mg, yield 5%, m.p. 193–195 °C, 1H NMR (d, J2 = 8.4 Hz, 2 H). 13C NMR (126 MHz, CDCl3) δ: 149.4,
(300 MHz, CDCl3) δ: 2.20 (s, 6 H), 3.63 (s, 2 H), 6.04 (s, 1 H), 138.7, 131.3, 128.9, 126.6 (q, J = 3.5 Hz), 124.9 (q, J = 270.3 Hz),
6.71 (d, J = 9.2 Hz, 2 H), 6.83 (s, 2 H), 8.07 (d, J = 9.2 Hz, 2 H). 121.7, 119.7 (q, J = 33.1 Hz), 113.3, 24.3, 13.0. IR (KBr) ν
13C NMR (126 MHz, CDCl3) δ: 152.4, 141.0, 138.6, 129.0, 126.4, (cm−1): 3385, 2965, 2874, 1111. HRMS (ESI) calcd for
124.8, 122.8, 112.3, 17.7. IR (KBr) ν (cm−1): 3363, 2962, 1301. C17H19F3N2 [M + H]+ m/z: 309.1573, found 309.1572.
HRMS (ESI) calcd for C14H15N3O2 [M + H]+ m/z: 258.1237,
4-[(2-Amino-5-methylphenyl)amino]-3,5-dimethylbenzonitrile
found 258.1249.
(15a). Pink crystals, 88 mg, yield 35%, m.p. 133–134 °C,
N-(2-Ethyl-6-methyl-4-((4-nitrophenyl)amino)phenyl)acet- 1H NMR (300 MHz, CDCl3) δ: 2.13 (s, 9 H), 3.65 (s, 2 H), 5.02
amide (9b). A mixture (36 mg) of p-semidine and p-nitro- (s, 1 H), 6.17 (s, 1 H), 6.70 (s, 2 H), 7.36 (m, 2H). 13C NMR
aniline in 5 mL of (Ac)2O was stirred at room temperature for (75 MHz, CDCl3) δ: 145.5, 135.4, 132.5, 131.3, 131.2, 129.0,
12 h. The resulting mixture was diluted with water (50 mL), 123.7, 119.6, 119.3, 116.3, 105.4, 20.6, 18.5. IR (KBr) ν (cm−1):
and extracted with ethyl acetate (2 × 50 mL). The organic layer 3459, 3365, 2924, 2855, 2220, 1601. HRMS (ESI) calcd for
was washed with saturated NaHCO3 (50 mL), brine (50 mL), C16H17N3 [M + H]+ m/z: 252.1495, found 252.1502.
dried over anhydrous Na2SO4, filtered, concentrated under
4-[(2-Amino-5-ethylphenyl)amino]-3-ethyl-5-methylbenzo-nitrile
1
reduced pressure and purified by flash chromatography with a (15b). Pink oil, 45 mg, yield 16%, H NMR (300 MHz, CDCl3)
mixture of petroleum ether and ethyl acetate as an eluent to δ: 1.07 (t, J1 = 7.6 Hz, 3H), 1.17 (t, J2 = 7.5 Hz, 3H), 2.08 (s, 3 H),
afford 9b. Yellow solid, 9 mg, yield 3%, m.p. 256–259 °C, 2.40 (q, J1 = 7.6 Hz, 2H), 2.53 (q, J2 = 7.5 Hz, 2H), 3.64 (s, 2 H),
1H NMR (300 MHz, DMSO-d6) δ: 1.10 (t, J = 7.5 Hz, 3 H), 2.04 5.10 (s, 1 H), 6.15 (s, 1 H), 6.68–6.75 (m, 2 H), 7.35 (s, 1 H),
(s, 3 H), 2.12 (s, 3 H), 2.52 (q, J = 7.5 Hz, 2 H), 6.93 (s, 1 H), 7.39 (s, 1 H). 13C NMR (75 MHz, CDCl3) δ: 144.8, 137.3, 135.8,
6.96 (s, 1 H), 7.04 (m, 2 H), 8.08 (m, 2 H), 9.14 (s, 1 H), 9.22 135.2, 132.5, 132.1, 131.9, 130.5, 122.1, 119.7, 117.6, 116.4,
(s, 1 H). 13C NMR (126 MHz, DMSO-d6) δ: 168.5, 151.0, 142.3, 105.9, 28.1, 24.4, 18.6, 16.0, 13.8. IR (KBr) ν (cm−1): 3366, 2964,
138.2, 137.7, 136.9, 130.5, 126.2, 120.0, 118.5, 113.2, 24.4, 2917, 2871, 2849, 2221, 1599. HRMS (ESI) calcd for C18H21N3
22.5, 18.2, 14.4. IR (KBr) ν (cm−1): 3445, 2961, 2920, 1654, [M + H]+ m/z: 280.1808, found 280.1814.
1581, 1312. HRMS (ESI) calcd for C17H19N3O3 [M + H]+ m/z:
Typical procedure for the intercrossing experiments
314.1499, found 314.1496.
3,5-Diethyl-N1-(4-nitrophenyl)benzene-1,4-diamine (9c). Red A solution of N,N′-diaryl hydrazine 14a (35 mg, 0.1 mmol), 14b
1
oil, 12 mg, yield 4%, H NMR (300 MHz, CDCl3) δ: 1.26 (t, J1 = (38 mg, 0.1 mmol) [or 7a (36 mg, 0.1 mmol), 7e (39 mg,
7.5 Hz, 6 H), 2.54 (q, J1 = 7.5 Hz, 4 H), 3.69 (s, 2 H), 6.19 0.1 mmol)], and conc. HCl (0.15 mL) in 10 mL of 95% ethanol
(s, 1 H), 6.72 (m, 2 H), 6.85 (s, 2 H), 8.06 (m, 2H). 13C NMR was refluxed for 2 h under nitrogen. The reaction mixture was
(126 MHz, CDCl3) δ: 152.5, 139.7, 138.4, 129.5, 128.8, 126.3, cooled to room temperature, neutralized with solid NaHCO3,
122.3, 112.2, 24.2, 12.8. IR (KBr) ν (cm−1): 3353, 2956, 2866, filtered, concentrated. The residue was subjected to the LC-MS
1306. HRMS (ESI) calcd for C16H19N3O2 [M + H]+ m/z: analysis.
286.1550, found 286.1564.
Typical procedure for the radical trapping experiments
3,5-Dimethyl-N1-(4-(trifluoromethyl)phenyl)benzene-1,4-
diamine (9d). Colorless crystals, 50 mg, yield 18%, m.p. A solution of N,N′-diaryl hydrazine 7d (190 mg, 0.5 mmol)
115–117 °C, 1H NMR (500 MHz, CDCl3) δ: 2.18 (s, 6 H), 3.54 or 14a (176 mg, 0.5 mmol), TEMPO (78 mg, 0.5 mmol, 2,2,6,6-
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 4952–4963 | 4961

View Article Online
Paper
Organic & Biomolecular Chemistry
tetramethyl-1-piperidinyloxy, free radical), and conc. HCl
(0.2 mL) in 10 mL of 95% ethanol was refluxed for 2 h under
nitrogen. The reaction mixture was cooled to room tempera-
ture, neutralized with solid NaHCO3, filtered, concentrated.
The residue was purified by flash column chromatography on
silica gel to afford 8d (22% yield) and 9d (17% yield), respect-
ively, or 41 mg of 15a as pink crystals in 33% yield.
2003, 103, 1197; (r) P. Wipf, D. L. Waller and J. T. Reeves,
J. Org. Chem., 2005, 70, 8096; (s) B. O. Ashburn, R. G. Carter
and L. N. Zakharov, J. Org. Chem., 2008, 73, 7305.
3 Si[1,3]-sigmatropic shift: (a) K. Mach, F. Turecek,
H. Antropiusova and V. Hanus, Organometallics, 1986, 5,
1215; (b) M. T. Pereira, M. Pfeffer and M. A. Rotteveel,
J. Organomet. Chem., 1989, 375, 139; (c) M. Takahashi and
M. Kira, J. Am. Chem. Soc., 1997, 119, 1948; (d) M. Takahashi
and M. Kira, J. Am. Chem. Soc., 1999, 121, 8597.
4 S[1,3]-sigmatropic shift: (a) K. Noerkjaer and A. Senning,
Chem. Ber., 1993, 126, 73; (b) V. Ficeri, P. Kutschy,
M. Dzurilla and J. Imrich, Collect. Czech. Chem. Commun.,
1994, 59, 2650; (c) J. Maddaluno, O. Gaonach, A. Marcual,
L. Toupet and G. P. Claude, J. Org. Chem., 1996, 61,
5290.
Acknowledgements
This work was supported in part by The National Basic
Research Program of China (no. 2013CB328905), the National
Natural Science Foundation of China (Nos. 21372025 and
21172017) and specialized Research Fund for the Doctoral
Program of Higher Education, Ministry of Education of China
(no. 20110010110011). This paper is also supported by
“Chemical Grid Project” of Beijing University of Chemical
Technology.
5 P[1,3]-sigmatropic shift: (a) K. Lammertsma, J. T. Hung,
P. Chand and G. M. Gray, J. Org. Chem., 1992, 57, 6557;
(b) B. Wang, C. H. Lake and K. Lammertsma, J. Am. Chem.
Soc., 1996, 118, 1690.
6 (a) R. B. Woodward and R. Hoffmann, The Conservation
of Orbital Symmetry, Verlag Chemie, Weinheim, 1970;
(b) I. Fleming, Frontier Orbitals and Organic Chemical
Reactions, Wiley-Interscience, New York, 1976.
Notes and references
1 B[1,3]-sigmatropic shift: (a) G. Zweifel and A. Horngy,
Synthesis, 1973, 672; (b) K. G. Hancock and J. D. Kramer,
J. Am. Chem. Soc., 1973, 95, 6463; (c) B. M. Miklailov,
Organomet. Chem. Rev., Sect. A, 1972, 8, 1; (d) G. W. Kramer
and H. C. Brown, J. Organomet. Chem., 1977, 132, 9;
7 S. L. Hou, X. Y. Li and J. X. Xu, J. Org. Chem., 2012, 77,
10856.
8 A. W. Hofmann, Proc. R. Soc. London, 1863, 12, 576.
9 R. B. Carlin, R. G. Nelb and R. C. Odioso, J. Am. Chem. Soc.,
1951, 73, 1002.
(e) M. Biihl, P. R. Schleyer, M. A. Ibrahim and T. Clark, 10 A different product ratio (2, 85% and 3, 15%) was reported
J. Am. Chem. Soc., 1991, 113, 2466; (f) U. Henriksen, by Shine and co-workers. See ref. 11f.
J. P. Snyder and T. A. Halgren, J. Org. Chem., 1981, 46, 3767. 11 (a) M. J. S. Dewar, in Molecular Rearrangement, ed.
2 C[1,3]-sigmatropic shift: (a) R. W. Thies, J. Am. Chem. Soc.,
1972, 94, 7074; (b) G. R. Krow and J. Reilly, J. Am. Chem.
Soc., 1975, 97, 3837; (c) J. W. Lown, M. H. Akhtar and
W. M. Dadson, J. Org. Chem., 1975, 40, 3363; (d) R. Fusco
and F. Sannicolo, Tetrahedron Lett., 1976, 17, 3991;
(e) R. W. Thies and E. P. Seitz, J. Chem. Soc., Chem.
Commun., 1976, 846; (f) S. Bartlett, R. D. Chambers and
N. M. Kelly, Tetrahedron Lett., 1980, 21, 1891;
(g) J. Barluenga, F. Aznar, R. Liz and M. Bayod, J. Chem.
Soc., Chem. Commun., 1984, 1427; (h) C. Bleasdale and
D. W. Jones, J. Chem. Soc., Chem. Commun., 1985, 1026;
(i) M. E. Jung and S. M. Kaas, Tetrahedron Lett., 1989, 30,
641; ( j) D. Desmaele and N. Champion, Tetrahedron Lett.,
1992, 33, 4447; (k) B. Franzus, M. L. Scheinbaum,
D. L. Waters and H. B. Bowlin, J. Am. Chem. Soc., 1976, 98,
1241; (l) K. Seki, W. Kiyokawa, H. Hashimoto, T. Uyehara,
M. Ueno and T. Sato, Chem. Lett., 1996, 1035;
P. de Mayo, Interscience, New York, 1969, vol. 1,
pp. 323–343; (b) H. J. Shine, in Mechanisms of Molecular
Migrations, ed. B. S. Thyagarajan, Interscience, New York,
1969, vol. 2, pp. 191–247; (c) H. J. Shine, in Aromatic
Rearrangement, Elsevier, New York, 1967, pp. 126–179;
(d) R. A. Cox and E. Buncel, in The Chemistry of the
Hydrazo, Azo, and Azoxy Groups, ed. S. Patai, Wiley,
New York, 1975, pp. 775–859; (e) D. V. Banthorpe,
R. Bramley and J. A. Thomas, J. Chem. Soc., 1964, 2864;
(f) D. V. Banthorp, Chem. Rev., 1970, 70, 295; (g) G. A. Olah,
K. Dunne, D. P. Kelly and Y. K. Mo, J. Am. Chem. Soc., 1972,
94, 7438; (h) J. D. Cheng and H. J. Shine, J. Org. Chem.,
1975, 40, 703; (i) C. A. Bunton and R. J. Rubin, J. Am. Chem.
Soc., 1976, 98, 4236; ( j) A. Zhu-Ohlbach, R. Gleiter,
R. Bleiter, H. L. Schmidt and T. Reda, Eur. J. Org. Chem.,
1998, 2409; (k) E. Buncel, Can. J. Chem., 2000, 78,
1251.
(m) F. X. Talamas, D. B. Smith, A. Cervantes, F. Franco, 12 (a) H. J. Shine, J. Phys. Org. Chem., 1989, 2, 491 (review)
S. T. Cutler, D. G. Loughhead, D. J. Morgans Jr. and (b) J. A. Berson, J. Phys. Org. Chem., 2005, 18, 572.
R. J. Weikert, Tetrahedron Lett., 1997, 38, 4725; 13 (a) H. J. Shine, G. N. Henderson, A. Schmid and P. Cu,
(n) Y. L. Chen, C. K. Chang and N. C. Chang, J. Chin. Chem.
Soc., 1998, 45, 649; (o) P. Bisel, G. Lauktien, E. Weckert and
A. W. Frahm, Tetrahedron: Asymmetry, 1998, 9, 4027;
(p) J. D. Bender, P. A. Leber, R. R. Lirio and R. S. Smith,
J. Org. Chem., 2000, 65, 5396; (q) J. E. Baldwin, Chem. Rev.,
J. Am. Chem. Soc., 1977, 99, 3719; (b) H. J. Shine, H. Zmuda,
K. H. Park, H. Kwart, A. G. Horgan, C. Collins and
B. E. Maxwell, J. Am. Chem. Soc., 1981, 103, 955;
(c) H. J. Shine, H. Zmuda, K. H. Park, H. Kwart,
A. G. Horgan and M. Brechbiel, J. Am. Chem. Soc., 1982,
4962 | Org. Biomol. Chem., 2014, 12, 4952–4963
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
104, 2501; (d) H. J. Shine, K. H. Part, M. L. Brownawell and 22 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) C. Lee,
J. San Filippo Jr., J. Am. Chem. Soc., 1984, 106, 7007;
(e) W. Subotkowski, L. Kupczy-Subotkowska and
H. J. Shine, J. Am. Chem. Soc., 1993, 115, 5073.
W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988,
37, 785.
23 (a) V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102,
1995; (b) M. Cossi, N. Rega, G. Scalmani and V. Barone,
J. Comput. Chem., 2003, 24, 669; (c) Y. Takano and
K. N. Houk, J. Chem. Theory Comput., 2005, 1, 70.
14 (a) H. J. Shine, E. Gruszecka, W. Subotkowksi,
M. Brownawell and J. San Fillipo Jr., J. Am. Chem. Soc.,
1985, 107, 3218; (b) H. J. Shine, L. Kupczyk-Subotkowska
and W. Subotkowski, J. Am. Chem. Soc., 1985, 107, 6674; 24 For recent examples of distortion/interaction analysis, see:
(c) H. J. Shine, L. Kupczyk-Subotkowska and W. Subotkowski,
J. Am. Chem. Soc., 1987, 109, 1286; (d) L. Kupczyk-
Subotkowski, H. J. Shine, W. Subotkowski and J. Zygmunt,
Gazz. Chim. Ital., 1987, 117, 513.
(a) D. H. Ess and K. N. Houk, J. Am. Chem. Soc., 2007, 129,
10646; (b) D. H. Ess and K. N. Houk, J. Am. Chem. Soc.,
2008, 130, 10187; (c) P. H.-Y. Cheong, R. S. Paton,
S. M. Bronner, G.-Y. Im, N. K. Garg and K. N. Houk, J. Am.
Chem. Soc., 2010, 132, 1267; (d) P. Liu, P. McCarren,
P. H. Cheong, T. F. Jamison and K. N. Houk, J. Am. Chem.
Soc., 2010, 132, 2050; (e) Y. Lan and K. N. Houk, J. Am.
Chem. Soc., 2010, 132, 17921, and references cited therein.
15 (a) H. J. Shine, H. Zmuda, H. Kwart, A. G. Horgan and
M. Brechbiel, J. Am. Chem. Soc., 1982, 104, 5181;
(b) E. S. Ree and H. J. Shine, J. Am. Chem. Soc., 1986, 108,
1000; (c) E. S. Ree and H. J. Shine, J. Am. Chem. Soc., 1987,
109, 5052; (d) E. S. Rhee and H. J. Shine, J. Org. Chem., 25 Although slight carbon KIEs were observed in the rate-
1987, 52, 5633; (e) H. J. Shine, in Isotopes in Organic
Chemistry, ed. E. Buncel and W. H. Saunders Jr., Elsevier,
Amsterdam, 1992, ch. 1, vol. 8.
limiting step in the rearrangements: the phenomena have
been found in the concerted 1,3-dipolar cycloaddition of
N,α-diphenylnitrone and styrene with a carbon KIE ranging
from 1.012 to 1.068 depending on which carbon atom was
labelled (ref. 26), and the concerted [2 + 2] cycloaddition of
diphenylketene and styrene with a carbon KIE in the range
of 1.0055–1.08 (ref. 27).
16 S. Yamabe, H. Nakata and S. Yamazaki, Org. Biomol. Chem.,
2009, 7, 4631.
17 G. Ghigo, S. Osella, A. Maranzana and G. Tonachini,
Eur. J. Org. Chem., 2011, 2326.
18 M. J. Smith and J. March, in Advanced Organic Chemistry, 26 B. M. Benjamin and C. J. Collins, J. Am. Chem. Soc., 1973,
John Wiley & Sons, Hoboken, New Jersey, 2007, ch. 18–36,
pp. 1678–1681.
19 K. Y. Kim, J. T. Shin, K. S. Lee and C. G. Cho, Tetrahedron
Lett., 2004, 45, 117.
20 (a) J. Jacobson, Justus Liebigs Ann. Chem., 1922, 428, 76;
(b) C. K. Ingold and H. W. Kidd, J. Chem. Soc., 1933, 984.
21 M. J. Frisch, et al., Gaussian 09, Revision B.01, Gaussian,
Inc., Wallingford, CT, 2010.
95, 6145.
27 C. J. Collins, B. M. Benjamin and G. W. Kabalka, J. Am.
Chem. Soc., 1978, 100, 2570.
28 For reviews, see: (a) S. E. Scheppele, Chem. Rev., 1972, 72,
511; (b) A. Maccoll, Annu.Rep. Prog. Chem., Sect. A: Inorg.
Phys. Chem., 1974, 71, 77 For examples, see; (c) J. Bron and
J. B. Stothers, Can. J. Chem., 1968, 46, 1435; (d) J. Bron and
J. B. Stothers, Can. J. Chem., 1969, 47, 2506.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 4952–4963 | 4963