Electrophilic reactivities of azodicarboxylates

Article abstract of DOI:10.1002/chem.201001598

The kinetics of the reactions of the azodicarboxylates 1 with the enamines 2 have been studied in CH₃CN at 20 °C. The reactions follow a secondorder rate law and can be described by the linear free energy relationship logk₂(20 °C) =

Full text of DOI:10.1002/chem.201001598

DOI: 10.1002/chem.201001598
Electrophilic Reactivities of ACHUTNGRNENUG zodicarboxylates
Tanja Kanzian and Herbert Mayr*[a]
Dedicated to Professor Gꢀnter Schꢁfer on the occasion of his 75th birthday
Abstract: The kinetics of the reactions
reactivities of 1 turned out to be com-
parable to those of a,b-unsaturated
iminium ions, amino-substituted benz-
hydrylium ions, and ordinary Michael
acceptors. While the E parameters of
the azodicarboxylates 1 determined in
this work also hold for their reactions
of the azodicarboxylates 1 with the en-
amines 2 have been studied in CH3CN
at 208C. The reactions follow a second-
order rate law and can be described by
the linear free energy relationship
with triarylphosphines, they cannot be
used for estimating rate constants for
their reactions with amines. Compari-
son of experimental and calculated rate
constants for cycloadditions and ene
reactions of azodicarboxylates provides
information on the concertedness of
these reactions.
logk2ACHTUNGTRENNUNG(208C)=sACHTUNGTRENNUNG(N+E) (E=electrophi-
licity parameter, N=nucleophilicity pa-
rameter, and s=nucleophile-specific
slope parameter). With E parameters
from ꢁ12.2 to ꢁ8.9, the electrophilic
Keywords: electrophilic amination ·
kinetics · organocatalysis · pericyclic
reaction · reaction mechanisms
Introduction
compounds,[5] we have now quantified the electrophilic reac-
tivities of the four azodicarboxylates 1a–d using the linear
free energy relationship [Eq. (1)].
Azodicarboxylates 1 have been used as strong electrophiles
in numerous synthetic transformations,[1] for example, elec-
trophilic aminations of carbonyl compounds,[2] which have
first been described in 1954 by Huisgen and Jakob.[2a] They
have been reported to be highly reactive dienophiles and
enophiles in Diels–Alder and ene reactions,[3] and their reac-
tion with phosphines represents the first step of the syntheti-
cally important Mitsunobu reaction.[4] In recent years, orga-
nocatalytic enantioselective a-aminations of aldehydes and
ketones with azodicarboxylates attracted great attention,[5]
in which the electrophiles 1 are attacked by intermediate en-
amines. Electrophiles used for enamine-activated reactions
of carbonyl compounds should react fast with the enamine
intermediates but not with the enols, which are present in
equilibrium with the tautomeric carbonyl compounds. As
the reactions of azodicarboxylates with enamines represent
the key step of organocatalytic a-aminations of carbonyl
log kð20 ꢀCÞ ¼ sðN þ EÞ
ð1Þ
In numerous investigations we have shown that the reac-
tions of carbocations and Michael acceptors with s, n, and p
nucleophiles follow Equation (1), in which electrophiles are
described by E (electrophilicity parameter) and nucleophiles
are described by N (nucleophilicity parameter) and s (nucle-
ophile-specific slope parameter).[6]
In this way, we were able to set up comprehensive electro-
philicity and nucleophilicity scales, which allow us to predict
the rates of polar organic reactions with an accuracy of
better than a factor of 102 in a reactivity range of 40 orders
of magnitude.[7]
The nucleophile-specific parameters N and s of the enam-
ines 2a–e have recently been derived from the rates of their
reactions with benzhydrylium ions (Table 1).[7a,8] These en-
[a] T. Kanzian, Prof. Dr. H. Mayr
Department Chemie, Ludwig-Maximilians-Universitꢀt Mꢁnchen
Butenandtstrasse 5-13 (Haus F), 81377 Mꢁnchen (Germany)
Fax : (+49)89-2180-77717
E-mail: herbert.mayr@cup.uni-muenchen.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001598. It contains prepara-
tive procedures, product characterization, and details of the individu-
al runs of the kinetic experiments.
11670
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11670 – 11677

FULL PAPER
Table 1. Enamines 2 used as reference nucleophiles in this study and
their reactivity parameters N and s in dichloromethane.
following the decrease of the absorbances of the electro-
philes 1 at 405 nm. From the observed monoexponential
decays the first-order rate constants, kobs, were obtained
(Figure 1). Details are given in the Supporting Information.
As shown for the reaction of diethyl azodicarboxylate
(1b) with morpholinocyclopentene (2c) in Figure 1, plots of
kobs versus the enamine concentrations were linear as re-
quired by Equation (2):
Enamine
N (s)[a]
2a
2b
2c
2d
2e
11.40 (0.83)
13.36 (0.81)
13.41 (0.82)
14.91 (0.86)
15.06 (0.82)
kobs ¼ k2½2ꢂ
ð2Þ
The slopes of these plots yielded the second-order rate
constants k2, which are listed in Table 2.
[a] References [7a] and [8].
amines have now been em-
ployed as reference nucleo-
philes for determining the elec-
trophilicities E of 1a–d.
Results and Discussion
Reactions of azodicarboxylates
1 with enamines 2: Treatment
of the azodicarboxylates 1b–d
with the enamines 2c and 2d
yielded the a-hydrazino-substi-
tuted ketones 3[2a,9] after aque-
ous workup (Scheme 1). Initial
attack of the nucleophilic
carbon of the enamine 2 at the
azo functionality of 1 leads to
Figure 1. Exponential decay of the absorbance at 405 nm during the reaction of 1b with 2c ([1b]=6.52ꢃ
10ꢁ3 m; [2c]=4.81ꢃ10ꢁ2 m; kobs =2.32ꢃ101 sꢁ1). Inset: Determination of the second-order rate constant k2 (k2 =
4.48ꢃ102 mꢁ1 sꢁ1) as the slope of the correlation between the first-order rate constants kobs and the concentra-
tion of the enamine 2c.
zwitterionic
intermediates,
which are hydrolyzed with for-
mation of the final products 3.
The rates of the reactions of the azodicarboxylates 1 with
the enamines 2 were determined photometrically in CH3CN
at 208C. For the kinetic studies, the nucleophiles 2 were
used in high excess over the electrophiles 1 to achieve first-
order conditions. The reaction progress was monitored by
Table 2. Second-order rate constants k2 for the reactions of the azodicar-
boxylates 1a–d with the enamines 2a–e in acetonitrile at 208C.
exptl
calcd
[a]
exptl
Azodicarboxylate (E) Enamine k2
[mꢁ1 sꢁ1
]
k2
[mꢁ1 sꢁ1
]
k2
k2
/
calcd
1a (ꢁ8.89)
2a
2b
2c
2a
2b
2c
2e
2a
2b
2c
2d
2e
2a
2b
2c
2d
2e
1.72ꢃ102
3.02ꢃ103
4.88ꢃ103
1.84ꢃ101
3.04ꢃ102
4.48ꢃ102
8.52ꢃ103
4.98
1.21ꢃ102
4.17ꢃ103
5.08ꢃ103
1.09ꢃ101
3.97ꢃ102
4.70ꢃ102
1.06ꢃ104
3.73
1.4
0.72
0.96
1.7
0.76
0.95
0.80
1.3
0.81
0.90
1.5
0.69
1.4
1b (ꢁ10.15)
1c (ꢁ10.71)
1d (ꢁ12.23)
1.13ꢃ102
1.47ꢃ102
5.95ꢃ103
2.54ꢃ103
2.86ꢃ10ꢁ1
5.07
1.40ꢃ102
1.63ꢃ102
4.08ꢃ103
3.68ꢃ103
2.06ꢃ10ꢁ1
8.28
0.61
1.1
1.5
1.02ꢃ101
2.99ꢃ102
1.50ꢃ102
9.34
2.03ꢃ102
2.11ꢃ102
0.71
[a] Calculated by using Equation (1), N and s parameters from Table 1,
and E from this table.
Scheme 1. Reactions of 1b–d with the enamines 2c and 2d.
Chem. Eur. J. 2010, 16, 11670 – 11677
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11671

H. Mayr and T. Kanzian
Application of Equation (1) implies slopes of 1.0 for the
plots of (logk2)/s versus N. Figure 2 shows that this condition
is roughly fulfilled for the reactions of 1 with the enamines
Figure 2. Correlation of (logk2)/s against the corresponding nucleophilici-
ty parameters N of the enamines 2a–e for the reactions of azodicarboxy-
lates 1a–d with 2 [the slopes are fixed to 1.0 as required by Eq. (1)].
Figure 3. Comparison of the electrophilicity parameters of azodicarboxy-
lates 1 with other electrophiles.
between ꢁ9 and ꢁ12, azodicarboxylates are not attacked by
typical enols, but react fast with enamines, comparable to
other Michael acceptors, which proved to be suitable for en-
amine-activated reactions of carbonyl compounds.
To compare these reactivities with rate constants deter-
mined in dichloromethane, we have also studied the influ-
ence of solvent on some of the reactions of 1 with 2. Table 3
shows that the second-order rate constants for the reactions
of 1 with 2 in acetonitrile (dielectric constant e=35.94)[14]
and dichloromethane (e=8.93)[14] differ by less than a factor
of 3. It is, therefore, not necessary to consider the effect of
ꢁ
2, in which a new C N bond is formed. Therefore, electro-
philicity parameters, E, of 1 were determined by a least-
squares fit, that is, by minimization of D2 =
ꢀ
(logk2ꢁs
(N+E))2. The last column of Table 2 shows that
calculated and experimental rate constants agree within a
factor of 2, which is within the confidence limit of Equa-
tion (1). It is remarkable that the N and s parameters of 2a–
e, which have been derived from their reactivities toward
benzhydrylium ions, also hold for their reactions with the ni-
trogen electrophiles 1. In line with this interpretation, we
have previously reported that the N and s parameters of
carbon nucleophiles can also be used to calculate rates of re-
actions with diazonium ions.[10]
The electrophilicities of the azodicarboxylates 1a–d cover
three orders of magnitude. They are only slightly less reac-
tive than arenediazonium ions,[10] comparable to a,b-unsatu-
rated iminium ions,[11] amino-substituted benzhydrylium
ions,[7a] and ordinary Michael acceptors (Figure 3).[7b,12] The
slightly less electrophilic, structurally related 1,2-diaza-1,3-
dienes react with nucleophiles at the alkenyl and not at the
azo group.[13]
The origin of the remarkably large effect of the different
ester groups on the electrophilicities of the azodicarboxy-
lates in the order of Bn>Et>iPr>tBu, as shown in Fig-
ures 2 and 3, is not known. With electrophilicity parameters
Table 3. Comparison of second-order rate constants k2 for the reactions
of the azodicarboxylates 1a–d with the enamines 2a–e in acetonitrile and
dichloromethane at 208C.
MeCN
CH2Cl2
CH2Cl2
k2MeCN/k2
1
2
k2
[mꢁ1 sꢁ1
]
k2
[mꢁ1 sꢁ1
]
1a
1b
2c
2a
2c
2a
2c
2a
2b
2c
2e
4.88ꢃ103
1.84ꢃ101
4.48ꢃ102
4.98
1.81ꢃ103
1.08ꢃ101
2.38ꢃ102
2.21
2.7
1.7
1.9
2.3
1.8
2.3
2.0
2.6
1.5
1c
1d
1.47ꢃ102
2.86ꢃ10ꢁ1
5.07
8.36ꢃ101
1.24ꢃ10ꢁ1
2.48
1.02ꢃ101
1.50ꢃ102
3.98
1.03ꢃ102
11672
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11670 – 11677

Azodicarboxylates
FULL PAPER
solvent on E when applying Equation (1) on reactions of 1
with nucleophiles in dichloromethane.
by Equation (1). This agreement is surprising: Since E pa-
rameters are generally derived from reactions with a series
of C-centered nucleophiles (E of 1 from reactions with en-
amines, Table 2, Figure 2) and N and s parameters are gen-
erally derived from the rates of reactions with a series of C-
centered electrophiles (N and s of 4 from reactions with
benzhydrylium ions), Equation (1) can only be expected to
Reactions with triarylphosphines: To check whether the
electrophilicity parameters E of 1a–d (Figure 3) derived
from the rates of their reactions with enamines 2 (carbon
nucleophiles) are also suitable for the prediction of the rates
of the reactions of azodicarboxylates with heteronucleo-
philes, we studied the kinetics of the reactions of 1a–d with
the triarylphosphines 4 and the amines 5, the N and s pa-
rameters of which have previously been derived from the
rates of their reactions with benzhydrylium ions.[15,16]
ꢁ
hold for the formation of C X bonds, that is, for reactions
in which at least one of the reaction centers in the electro-
phile or nucleophile is carbon.[15] The unexpected observa-
ꢁ
tion that Equation (1) also holds for the formation of an N
P bond reminds one of the observation that Equation (1), as
well as the Ritchie Equation, were also found to work for
the combinations of diazonium ions with several heteronu-
Nucleophilic additions of the triarylphosphines 4 to di-
ACHTUNGTRENNUNGalkyl azodicarboxylates 1, which correspond to the first step
cleophiles.[10,18] The comparable magnitudes of the P N
(290 kJmol ) and C N (305 kJmol ) bond energies may
ꢁ
of the Mitsunobu reaction, yield the so-called Huisgen zwit-
terions, as depicted in Scheme 2.[17]
ꢁ1
ꢁ1
ꢁ
account for this finding.[19]
Reactions with amines: The reactions of azodicarboxylates 1
with amines 5 have been reported to yield triazanes 6 in pe-
troleum ether.[20] Scheme 3 shows that triazanes 6 are also
formed in acetonitrile, that is, the solvent in which the kinet-
ꢁ
ic investigations were performed. Though N N single bonds
Scheme 2. Reactions of 1 with phosphines 4.
are generally rather weak, the triazanes 6 are stabilized by
two electron-withdrawing ester groups and can be isolated
without problem.[20a]
The kinetics of the reactions of 1 with triarylphosphines
in CH2Cl2 were determined photometrically, as described
above for the reactions of 1 with enamines 2. Linear plots of
kobs versus [PAr3] confirmed second-order rate laws for
these reactions, and the resulting second-order rate con-
stants for the attack of PAr3 at the N=N unit are summar-
ized in Table 4.
As shown in the last column of Table 4, the experimental
rate constants for the reactions of 1 with the phosphines 4
generally agree within a factor of 10 with those calculated
Scheme 3. Reactions of 1c and 1d with amines 5d–f.
Table 4. Experimental and calculated second-order rate constants for the
reactions of the azodicarboxylates 1 with the phosphines 4 in dichlorome-
thane at 208C.
Nevertheless, the thermodynamic driving force for these
additions seems to be rather low. As shown below, the rela-
tive reactivities of the azodicarboxylates 1 toward amines
are the same as toward enamines 2 and phosphines 4. For
that reason, one can expect that 1a should even react faster
with amines than the azodicarboxylates 1b–d. The fact, that
no conversion was observed when 1a was combined with
morpholine 5d and pyrrolidine 5 f must, therefore, be due to
unfavorable thermodynamics (fast reverse reactions).
The reactions of 1 with the secondary amines 5a–f were
studied in CH3CN by following the decay of the electro-
philesꢄ absorbances. When the amines were used in large
excess, first-order kinetics were observed with exponential
decays of the absorbances of the azodicarboxylates 1. How-
ever, linearity of the kobs versus [5] plots was only observed
for the reactions of 1c with 5c and 5e. For all other reac-
tions of the azodicarboxylates 1 with the secondary amines
5a–f, the plots of the kobs values versus the amine concentra-
tions showed concave curvatures (Figure 4), indicating that
exptl
calcd
calcd
1
Phosphine 4 (N/s)[a]
k2
[mꢁ1 sꢁ1
k2
[mꢁ1 sꢁ1
]
k2
k2
/
[b]
exptl
G
]
U
1a 4a ((4-ClC6H4)3P) (12.58/0.65)
1a 4b (Ph3P) (14.33/0.65)
6.41ꢃ101
4.83ꢃ102
2.42ꢃ103
2.50ꢃ102
3.44ꢃ103
1.56ꢃ104
3.26ꢃ104
5.21ꢃ102
2.43ꢃ103
5.40ꢃ103
1.64ꢃ101
2.25ꢃ102
1.06ꢃ103
2.43ꢃ103
3.9
7.1
6.4
3.0
3.6
3.8
2.3
3.8
5.0
4.9
3.3
3.5
11
1a 4c ((4-MeC6H4)3P) (15.44/0.64)
1a 4d ((4-OMeC6H4)3P) (16.17/0.62) 1.07ꢃ104
1b 4b (14.33/0.65)
1b 4c (15.44/0.64)
1b 4d (16.17/0.62)
1c 4a (12.58/0.65)
1c 4b (14.33/0.65)
1c 4c (15.44/0.64)
1c 4d (16.17/0.62)
1d 4a (12.58/0.65)
1d 4b (14.33/0.65)
1d 4c (15.44/0.64)
1.45ꢃ102
6.13ꢃ102
2.36ꢃ103
4.28
4.51ꢃ101
2.16ꢃ102
7.25ꢃ102
4.83ꢃ10ꢁ1 1.69
2.14
2.32ꢃ101
1.13ꢃ102
1.11ꢃ101
10
[a] N/s-parameters from reference [15]. [b] Calculated by using Equa-
tion (1), N and s parameters from reference [15], and the E parameters
from Table 2.
Chem. Eur. J. 2010, 16, 11670 – 11677
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11673

H. Mayr and T. Kanzian
½5ꢂ=kobs ¼ 1=k2 þ kꢁ2=ðk2½5ꢂkaÞ
kobs ¼ k2½5ꢂ
ð7Þ
ð8Þ
If kꢁ2 !ka[5], Equation (6) is transformed into Equa-
tion (8), that is, a second-order reaction [Eq. (2)] with rate-
ꢁ
determining formation of the N N bond. As mentioned
before, this situation was only observed for the reaction of
1c with 5c and 5e. At high amine concentrations, the
second-order rate law [Eq. (8)] also holds for the reactions
of 1b with 5b and 5c, and for 1d with 5b. As exemplified in
Figure 5a for the reaction of 1d with 5b, the kobs/[5b] value
Figure 4. Plots of kobs versus [5d] for the reaction of 5d with the azodicar-
boxylate 1b (CH3CN, 208C).
a second molecule of amine is involved in the rate-determin-
ing step, which acts as a base catalyst (Scheme 4).
Scheme 4. Reactions of the azodicarboxylates 1 with the amines 5.
The assumption that the formation of the intermediate I1
is a reversible process is in line with the observation that ter-
Figure 5. Plots of k
for the reaction of 5b with the azodicarboxylate 1d.
obs/ACHTUNGTREN[UNG 5b] versus [5b] (a) and [5b]/kobs versus 1/ACHTUNGTRENNUNG[5b] (b)
tiary amines [for example, 1,8-diazabicycloACHTUNGTRENUN[NG 5.4.0]undec-7-
ene (DBU) or 1,4-diazabicyclo[2.2.2]octane (DABCO)] do
AHCTUNGTRENNUNG
not react with the azodicarboxylates 1.
According to Scheme 4, the change of the concentration
of the betaine intermediate I1 can be expressed by Equa-
tion (3).
increases with increasing amine concentration [5b], because
the proton transfer (ka[5b]) is getting faster. At high amine
concentrations, where kobs/[5b] is constant, the attack of the
amine at the azodicarboxylate is rate limiting and kobs/[5b]=
k2 [Eq. (8)]; the second-order rate constant equals the inter-
cept on the y axis. The whole range of amine concentrations
can be evaluated by using Equation (7). The linear plot of
[5]/kobs against 1/[5], as shown in Figure 5b, provides 1/k2 as
the intercept. The second-order rate constants for the reac-
tion of 1c with 5d, determined by the different evaluations,
are thus in fair agreement. Equation (7) has analogously
been used to determine the second-order rate constants k2
for the first step of the reactions of 1b with 5b and 5c.
d½I1ꢂ=dt ¼ k2½1ꢂ½5ꢂꢁkꢁ2½I1ꢂꢁka½I1ꢂ½5ꢂꢁkp½I1ꢂ
ð3Þ
With the assumption of a steady-state concentration for
the intermediate I1 (d[I1]/dt=0), the rate law can be ex-
pressed by Equations (4) and (5).
ꢁd½1ꢂ=dt ¼ k2½1ꢂ½5ꢂ ðka½5ꢂ þ kpÞ=ðkꢁ2 þ ka½5ꢂ þ kpÞ
kobs ¼ k2½5ꢂ ðka½5ꢂ þ kpÞ=ðkꢁ2 þ ka½5ꢂ þ kpÞ
ð4Þ
ð5Þ
ꢁ
In all other combinations of 1 and 5, rate-limiting N N
bond formation was never observed, not even at high amine
concentrations. The second-order rate constants for the first
step of the reactions were also determined by plotting
[5]/kobs against 1/[5]. Deviations from these linear plots
occur only at very low amine concentrations, probably due
to the operation of kp. If the kobs values at low amine con-
centrations are neglected (open circles in Figure 6), the
Let us first neglect the direct proton transfer (kp) from
NH+ to Nꢁ in the zwitterionic intermediate I1. Equation (5)
is then reduced to Equation (6), which can be transformed
into Equation (7).
2
ð6Þ
kobs ¼ k2½5ꢂ ka=ðkꢁ2 þ ka½5ꢂÞ
11674
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11670 – 11677

Azodicarboxylates
FULL PAPER
phile or nucleophile is carbon. Table 5 clearly shows that
ꢁ
Equation (1) cannot be applied to the N N bond-forming
reactions in the first step shown in Scheme 4. The considera-
ꢁ1
ꢁ
bly smaller N N bond energy (159 kJmol ) compared with
ꢁ1
ꢁ
the C N bond energy (305 kJmol ) may account for the
fact, why N parameters of amines that have been derived
from the reactivities of amines with C electrophiles give
values that are far too low in reactions with N electrophiles.
Figure 7 illustrates that acyclic and cyclic secondary
amines 5 are on different correlation lines, considerably
below that for enamines 2, from which the phosphines 4 de-
Figure 6. Plot of [5d]/kobs versus 1/ACTHNUGRTNEUNG[5d] for the reaction of 5d with the
azodicarboxylate 1b (filled circles: included for the determination of k2;
k2 =1/(13.076ms)=7.65ꢃ10ꢁ2 mꢁ1 sꢁ1; open circles: not included for the
evaluation of k2, because at the low concentration of 5d the assumption
that kp !kaACHTUNGTRENNUNG[5d] seems to be invalid).
second-order rate constants k2 can be obtained from the in-
tercepts (1/k2) of the linear correlations [Figure 6 and
Eq. (7)].
Analogous changes of the reaction order of amines have
previously been reported for the reactions of secondary
amines with quinone methides,[16] thiocarbonates,[21] thiono-
benzoates,[22] and activated esters of indole-3-acetic acid.[23]
Table 5 shows the second-order rate constants, k2, for the re-
actions of 1 with 5 obtained by these evaluations.
Table 5. Experimental and calculated second-order rate constants for the
reactions of azodicarboxylates 1 with amines 5 in acetonitrile at 208C.
exptl
calcd
calcd
1
Amine 5 (N/s)[a]
k2
[mꢁ1 sꢁ1
k2
[mꢁ1 sꢁ1
]
k2
k2
/
[b]
exptl
G
]
U
1b 5a (bis(2-methoxy)ethylamine)
(13.24/0.93)
1b 5b (di-n-propylamine) (14.51/
0.80)
1.58ꢃ10ꢁ2
7.48ꢃ102 4.7ꢃ104
3.34ꢃ10ꢁ1
3.08ꢃ103 9.2ꢃ103
Figure 7. Plots of (logk2)/s versus the N parameters of the enamines 2a–c
and 2e (filled circles), the phosphines 4b–d (filled triangles), and the
amines 5a–f (open circles) for the reactions of the azodicarboxylate 1b
with 2, 4, and 5 at 208C. The linear graph refers to the plot of (logk2)/s
against the N parameters of the enamines 2a–c and 2e and the slope is
fixed to unity.
1b 5c (diethylamine) (15.10/0.73)
1b 5d (morpholine) (15.65/0.74)
1b 5e (piperidine) (17.35/0.68)
1b 5 f (pyrrolidine) (18.64/0.60)
1c 5a (13.24/0.93)
1c 5b (14.51/0.80)
1c 5c (15.10/0.73)
1c 5d (15.65/0.74)
1c 5e (17.35/0.68)
1c 5 f (18.64/0.60)
1d 5a (13.24/0.93)
1d 5b (14.51/0.80)
1d 5c (15.10/0.73)
1d 5d (15.65/0.74)
1d 5e (17.35/0.68)
1d 5 f (18.64/0.60)
1.00
4.11ꢃ103 4.1ꢃ103
1.17ꢃ104 1.5ꢃ105
7.87ꢃ104 4.6ꢃ104
1.24ꢃ105 4.2ꢃ103
2.25ꢃ102 4.1ꢃ104
1.10ꢃ103 1.2ꢃ104
7.65ꢃ10ꢁ2
1.72
2.94ꢃ101
5.52ꢃ10ꢁ3
9.52ꢃ10ꢁ2
3.27ꢃ10ꢁ1[c] 1.60ꢃ103 4.9ꢃ103
viate only marginally. Nevertheless, the internal consistency
of the second-order rate constants k2 for the reactions of the
amines 5 with the azodicarboxylates 1b–d is beautifully
demonstrated by Figure 8 in which the reactivities of the
amines 5a–f towards the azodicarboxylates 1c and 1d are
plotted against the corresponding reactivities toward diethyl
azodicarboxylate (1b).
1.86ꢃ10ꢁ2
4.52ꢃ103 2.4ꢃ105
4.59ꢃ10ꢁ1[c] 3.27ꢃ104 7.1ꢃ104
1.55ꢃ101
5.73ꢃ104 3.7ꢃ103
8.70
2.7ꢃ104
3.22ꢃ10ꢁ4
7.24ꢃ10ꢁ3
2.60ꢃ10ꢁ2
1.06ꢃ10ꢁ3
3.07ꢃ10ꢁ2
8.09ꢃ10ꢁ1
6.67ꢃ101 9.2ꢃ103
1.24ꢃ102 4.8ꢃ103
3.39ꢃ102 3.2ꢃ105
3.03ꢃ103 9.9ꢃ104
7.01ꢃ103 8.7ꢃ103
Pericyclic reactions: Rate constants for the [2+2] cycloaddi-
tion of 1b with ethyl vinyl ether,[4d] for the Diels–Alder re-
actions of 1b with cyclohexa-1,3-diene[4b] and 2,3-dimethyl-
1,3-butadiene,[4a] and for the ene reaction of diethyl azodi-
carboxylate (1b) with 2-methyl-but-2-ene[4c] have previously
been reported (Table 6). Though the conditions (solvent,
temperature) were not in all cases identical to those for
which the reactivity parameters E, N, and s have been de-
rived, they are similar enough that a comparison between
[a] N/s-parameters from reference [16]. [b] Calculated by using Equa-
tion (1), N and s parameters from reference [16], and the E parameters
from Table 2. [c] Derived from plots of kobs versus [5].
The experimental rate constants are 103 to 106 times
smaller than the values calculated by Equation (1). As de-
scribed above, Equation (1) can only be expected to hold
when at least one of the reaction centers in either electro-
Chem. Eur. J. 2010, 16, 11670 – 11677
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11675

H. Mayr and T. Kanzian
additions proceed stepwise and only one new s bond is
formed in the rate-limiting step, Equation (1) appears to be
applicable.
On the other hand, the Diels–Alder reactions as well as
the ene reaction (entries 2–4 in Table 6) proceed 105 to 108
times faster than calculated by Equation (1). In these cases,
the concerted formation of two new s bonds stabilizes the
transition state and gives rise to a higher reaction rate. In
cases in which pericyclic processes are feasible, Equation (1)
gives only a lower limit for the rate constant. If calculated
and experimental rate constants refer to the same reaction
conditions, the deviation between the two values may even
be considered as a measure for the energy of concerted-
ness.[24]
Conclusion
The p systems of azodicarboxylates and enamines combine
ꢁ
with the formation of only one new C N bond, which
allows us to derive electrophilicity parameters, E, for the
azodicarboxylates 1a–d. They are slightly less electrophilic
than ordinary arenediazonium ions, comparable to a,b-unsa-
turated iminium ions and benzylidenemalononitriles. The E
parameters of the azodicarbox-
Figure 8. Correlation of the rate constants (logk2) for the reactions of 1c
and 1d with the amines 5a–f against the rate constants (logk2) for the re-
action of 1b with the amines 5a–f (in CH3CN, 208C).
ylates determined herein can be
Table 6. Comparison of experimental and calculated second-order rate constants k2 for pericyclic reactions of
1b with different nucleophiles.
used to predict rate constants
not only for the reactions of
these electrophiles with carbon
nucleophiles, but unexpectedly
also for the reactions with phos-
phines, though none of the re-
action centers is carbon, a pre-
requisite for the application of
Equation (1). The reported rate
constants for Diels–Alder and
ene reactions of diethyl azodi-
carboxylate are 105 to 108 times
higher than those calculated by
using Equation (1), indicating
concerted pericyclic reactions.
exptl
[a]
calcd
[b]
calcd
k2exptl/k2
Reaction
k2
[mꢁ1 sꢁ1
]
k2
[mꢁ1 sꢁ1
]
8.38ꢃ10ꢁ5[d]
2.3ꢃ10ꢁ6
36
7.6ꢃ10ꢁ5[e]
3.4ꢃ10ꢁ11
9.6ꢃ10ꢁ10
2.2ꢃ106
9.9ꢃ104
9.46ꢃ10ꢁ5[f]
5.7ꢃ10ꢁ4[h]
6.9ꢃ10ꢁ12
8.3ꢃ107
Experimental Section
General: Azodicarboxylates 1a–d,
phosphines 4a–d, and amines 5a–f
[a] If more than one rate constant was reported for a certain reaction, we selected that which was determined
under conditions most similar to those we used for deriving the reactivity parameters (acetonitrile, 208C,
1013 mbar). [b] Calculated by using Equation (1). [c] Reference [7c]. [d] Reference [3d]. [e] Reference [3b].
[f] Reference [3a]. [g] Reference [7a]. [h] Reference [3c].
were purchased, and the amines
5
were purified by distillation prior to
use. The enamines 2a–e were synthe-
sized by literature procedures.[25]
Reactions of azodicarboxylates 1 with
enamines 2: A solution of the azodicarboxylate 1b–d in dry acetonitrile
was added dropwise to a stirred solution of the enamine 2c or 2d
(1 equiv) in acetonitrile. The mixture was stirred at 208C until the yellow
color of 1 had disappeared. The reaction mixture was quenched by water
and the separated organic layer was dried over MgSO4. The solvent was
removed in vacuo and the crude product was purified by column chroma-
tography.
experimental and calculated rate constants appears to be
justified.
Table 6 shows that the reaction of 1b with ethyl vinyl
ether at 308C is only 36 times faster than that calculated for
208C, that is, the rate of the [2+2] cycloaddition is in agree-
ment with the predictions of Equation (1). As [2+2] cyclo-
11676
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11670 – 11677

Azodicarboxylates
FULL PAPER
Reactions of azodicarboxylates
1
with amines 5: The amines 5d–f
Angew. Chem. 2002, 114, 97–102; Angew. Chem. Int. Ed. 2002, 41,
91–95; c) H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem. Res. 2003, 36,
66–77; d) H. Mayr, A. R. Ofial in Carbocation Chemistry (Eds.:
G. A. Olah, G. K. S. Prakash), Wiley, New York, 2004, Chapter 13,
pp. 331–358; e) H. Mayr, A. R. Ofial, Pure Appl. Chem. 2005, 77,
1807–1821; f) H. Mayr, A. R. Ofial, J. Phys. Org. Chem. 2008, 21,
584–595; g) D. Richter, N. Hampel, T. Singer, A. R. Ofial, H. Mayr,
Eur. J. Org. Chem. 2009, 3203–3211.
(2 equiv) was added to a solution of the azodicarboxylate 1c or 1d in
acetonitrile at 208C and the mixture was stirred until the yellow color of
the azodicarboxylate had disappeared. The solvent was evaporated in
vacuo.
Kinetics: The kinetics of the reactions of the azodicarboxylates 1 with
the nucleophiles 2, 4, or 5 were followed by UV/Vis spectroscopy by
using work stations similar to those described previously.[7a,f,26] For slow
reactions (t1/2 >10 s) the UV/Vis spectra were collected at different times
by using a J&M TIDAS diode array spectrophotometer that was connect-
ed to a Hellma 661.502-QX quartz Suprasil immersion probe (5 mm light
path) by fiber optic cables with standard SMA connectors. All kinetic
measurements were carried out in Schlenk glassware under the exclusion
of moisture. The temperature of the solutions during the kinetic studies
was maintained at 208C within ꢃ0.18C by using circulating bath cryostats
and monitored with thermocouple probes that were inserted into the re-
action mixture. Stopped-flow spectrophotometer systems (Applied Pho-
tophysics SX.18MV-R or Hi-Tech SF-61DX2) were used for the investi-
gation of fast reactions of azodicarboxylates with nucleophiles (10 ms<
t1/2 <10 s). The kinetic runs were initiated by mixing equal volumes of
solutions of the azodicarboxylates and the nucleophiles. Concentrations
and rate constants for the individual kinetic experiments are given in the
Supporting Information.
[8] B. Kempf, N. Hampel, A. R. Ofial, H. Mayr, Chem. Eur. J. 2003, 9,
2209–2218.
[9] a) R. M. Moriarty, I. Prakash, Synth. Commun. 1985, 15, 649–655;
b) S. Fioravanti, L. Pellacani, P. A. Tardella, Gazz. Chim. Ital. 1997,
127, 41–44; c) D. Enders, R. Joseph, C. Poiesz, Tetrahedron 1998,
54, 10069–10078; d) S. Fioravanti, L. Olivieri, L. Pellacani, P. A. Tar-
della, J. Chem. Res. Synop. 1998, 338–339; e) N. Kumaragurubaran,
K. Juhl, W. Zhuang, A. Bøgevig, K. A. Jørgensen, J. Am. Chem.
Soc. 2002, 124, 6254–6255; f) G. Dessole, L. Bernardi, B. F. Bonini,
E. Capitꢅ, M. Fochi, R. P. Herrera, A. Ricci, G. Cahiez, J. Org.
ˇ
Chem. 2004, 69, 8525–8528; g) P. Kotrusz, S. Alemayehu, S. Toma,
H.-G. Schmalz, A. Adler, Eur. J. Org. Chem. 2005, 4904–4911; h) C.
Thomassigny, D. Prim, C. Greck, Tetrahedron Lett. 2006, 47, 1117–
1119; i) Y. Hayashi, S. Aratake, Y. Imai, K. Hibino, Q.-Y. Chen, J.
Yamaguchi, T. Uchimaru, Chem. Asian J. 2008, 3, 225–232.
[10] H. Mayr, M. Hartnagel, K. Grimm, Liebigs Ann. 1997, 55–69.
[11] S. Lakhdar, T. Tokuyasu, H. Mayr, Angew. Chem. 2008, 120, 8851–
8854; Angew. Chem. Int. Ed. 2008, 47, 8723–8726.
[12] a) T. Lemek, H. Mayr, J. Org. Chem. 2003, 68, 6880–6886;
b) S. T. A. Berger, F. H. Seeliger, F. Hofbauer, H. Mayr, Org.
Biomol. Chem. 2007, 5, 3020–3026; c) O. Kaumanns, H. Mayr, J.
Org. Chem. 2008, 73, 2738–2745.
[13] T. Kanzian, S. Nicolini, L. De Crescentini, O. A. Attanasi, A. R.
Ofial, H. Mayr, Chem. Eur. J. 2010, DOI: 10.1002/chem.201000828.
[14] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, 3rd
ed., Wiley-VCH, Weinheim, 2003.
Acknowledgements
The authors thank Roland Appel for helpful discussions, and Dr.
Armin R. Ofial and Nicolas Streidl for help during preparation of this
manuscript. We thank the Deutsche Forschungsgemeinschaft (Ma 673/21-
3) for financial support.
[15] B. Kempf, H. Mayr, Chem. Eur. J. 2005, 11, 917–927.
[16] T. Kanzian, T. A. Nigst, A. Maier, S. Pichl, H. Mayr, Eur. J. Org.
Chem. 2009, 6379–6385.
[1] V. Nair, A. T. Biju, S. C. Mathew, B. P. Babu, Chem. Asian J. 2008, 3,
810–820.
[17] a) D. C. Morrison, J. Org. Chem. 1958, 23, 1072–1074; b) E. Brunn,
R. Huisgen, Angew. Chem. 1969, 81, 534–536; Angew. Chem. Int.
Ed. Engl. 1969, 8, 513–515; c) R. D. Guthrie, I. D. Jenkins, Aust. J.
Chem. 1982, 35, 767–774; d) S. Moebs-Sanchez, G. Bouhadir, N.
Saffon, L. Maron, D. Bourissou, Chem. Commun. 2008, 3435–3437.
[18] a) C. D. Ritchie, Acc. Chem. Res. 1972, 5, 348–354; b) C. D. Ritchie,
Can. J. Chem. 1986, 64, 2239–2250.
[19] N. Wiberg, A. F. Holleman, E. Wiberg, Lehrbuch der Anorganischen
Chemie, 102nd ed., Walter de Gruyter, Berlin, 2007, p. 143.
[20] a) K.-H. Linke, H. J. Gçhausen, Chem. Ber. 1971, 104, 301–306;
b) N. Egger, L. Hoesch, A. S. Dreiding, Helv. Chim. Acta 1983, 66,
1416–1426.
[21] a) I.-H. Um, S. Yoon, H.-R. Park, H.-J. Han, Org. Biomol. Chem.
2008, 6, 1618–1624; b) E. A. Castro, Pure Appl. Chem. 2009, 81,
685–696.
[22] I.-H. Um, J.-A. Seok, H.-T. Kim, S.-K. Bae, J. Org. Chem. 2003, 68,
7742–7746.
[23] M. J. Pfeiffer, S. B. Hanna, J. Org. Chem. 1993, 58, 735–740.
[24] a) W. v. E. Doering, W. R. Roth, R. Breuckmann, L. Figge, H.-W.
Lennartz, W.-D. Fessner, H. Prinzbach, Chem. Ber. 1988, 121, 1–9;
b) M. Hartnagel, K. Grimm, H. Mayr, Liebigs Ann. 1997, 71–80;
c) H. Mayr, A. R. Ofial, J. Sauer, B. Schmied, Eur. J. Org. Chem.
2000, 2013–2020.
[25] H. G. O. Becker, Organikum, 19th ed., Verlag der Wissenschaften,
Berlin, 1993.
[26] H. Mayr, R. Schneider, C. Schade, J. Bartl, R. Bederke, J. Am.
Chem. Soc. 1990, 112, 4446–4454.
[2] a) R. Huisgen, F. Jakob, Justus Liebigs Ann. Chem. 1954, 590, 37–
54; b) E. Erdik, Tetrahedron 2004, 60, 8747–8782; c) J. M. Janey,
Angew. Chem. 2005, 117, 4364–4372; Angew. Chem. Int. Ed. 2005,
44, 4292–4300.
[3] a) G. Desimoni, G. Faita, P. P. Righetti, L. Toma, Tetrahedron 1990,
46, 7951–7970; b) G. Jenner, R. B. Salem, J. Chem. Soc. Perkin
Trans. 2 1990, 1961–1964; c) G. Desimoni, G. Faita, P. P. Righetti, A.
Sfulcini, D. Tsyganov, Tetrahedron 1994, 50, 1821–1832; d) M. R.
Gholami, A. Habibi Yangjeh, J. Phys. Org. Chem. 2000, 13, 468–
472; e) A. Chanda, V. V. Fokin, Chem. Rev. 2009, 109, 725–748.
[4] a) O. Mitsunobu, M. Eguchi, Bull. Chem. Soc. Jpn. 1971, 44, 3427–
3430; b) V. Nair, R. S. Menon, A. R. Sreekanth, N. Abhilash, A. T.
Biju, Acc. Chem. Res. 2006, 39, 520–530; c) T. Y. S. But, P. H. Toy,
Chem. Asian J. 2007, 2, 1340–1355; d) K. C. K. Swamy, N. N. B.
Kumar, E. Balaraman, K. V. P. P. Kumar, Chem. Rev. 2009, 109,
2551–2651.
[5] a) B. List, J. Am. Chem. Soc. 2002, 124, 5656–5657; b) A. Bøgevig,
K. Juhl, N. Kumaragurubaran, W. Zhuang, K. A. Jørgensen, Angew.
Chem. 2002, 114, 1868–1871; Angew. Chem. Int. Ed. 2002, 41, 1790–
1793; for reviews: c) A. Berkessel, H. Grçger, Asymmetric Organo-
catalysis, Wiley-VCH, Weinheim, 2005; d) Enantioselective Organo-
catalysis (Ed.: P. I. Dalko), Wiley-VCH, Weinheim, 2007; e) S. Mu-
kherjee, J. W. Yang, S. Hoffman, B. List, Chem. Rev. 2007, 107,
5471–5569.
[6] H. Mayr, M. Patz, Angew. Chem. 1994, 106, 990–1010; Angew.
Chem. Int. Ed. Engl. 1994, 33, 938–957.
[7] a) H. Mayr, T. Bug, M. F. Gotta, N. Hering, B. Irrgang, B. Janker, B.
Kempf, R. Loos, A. R. Ofial, G. Remennikov, H. Schimmel, J. Am.
Chem. Soc. 2001, 123, 9500–9512; b) R. Lucius, R. Loos, H. Mayr,
Received: June 7, 2010
Published online: September 10, 2010
Chem. Eur. J. 2010, 16, 11670 – 11677
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11677