Molecular Complexation and Kinetics for Diels-Alder Condensation of Naphthylalkenes with Tetracyanoethylene

Article abstract of DOI:10.1021/jo9804902

The charge-transfer spectra of 1:1 molecular complexes of seven 1-alkyl-1-(2-naphthyl)ethenes (where R=alkyl group: H,Me,Et,n-Pr,i-Pr,t-Bu,Neopent) and 1-vinylnaphthalene with TCNE were determined in ClCH2CH2Cl as solvent at 27.1 deg C.Equilibrium constants, K; for these complexes vary with R, as based on two opposing factors, viz. (a) the polar substituent factor ?* and (b) the angle-of-twist θ between the planes of the naphthyl and vinyl groups.For R=Neopent (?* dominant) K is 50 times as large as for R=t-Bu (θ dominant).Except for R=t-Bu, kinetics of reaction conform with the equation D+TCNE <*> (K) D*TCNE <*> (k₁/k_-1) P, where D is the donor alkene, P is the Diels-Alder 1,4-cycloaddition product, and k₁ and k_-1 are first-order reaction rate constants.Values of k₁ vary from 0.71 min^-1 (R=H) to 55.5 min^-1 (R=Neopent) and the corresponding relative second-order rate constants k₂ (or k₁K) from 1 to 4000.The rate constant k_-1 was measured only for 1b (R=Me, 0.0017 min^-1) in the solvent mixture p-xylene/ClCH2CH2Cl.Formation of 1b*TCNE complex gives ΔH = 10.0 kcal/mol and ΔS = 38.4 eu, and conversion to P shows an Arrhenius activation energy of E_a = 7.24 kcal/mol.It is proposed that the preferred conformation of a naphthylalkene for complexation has the R and vinyl groups projecting outward from opposite sides of the plane of the naphthalene ring.The TCNE molecule then aligns parallel to the naphthalene ring on the vinyl side where (except for R=t-Bu) it can slide into the geometry of the transition state to form P.

Full text of DOI:10.1021/jo9804902

J . Org. Chem. 1998, 63, 6503-6510
6503
Molecu la r Com p lexa tion a n d Kin etics for Diels-Ald er
Con d en sa tion of Na p h th yla lk en es w ith Tetr a cya n oeth ylen e
†
LeRoy H. Klemm,* Wayne C. Solomon, and Amir P. Tamiz
Department of Chemistry, University of Oregon, Eugene, Oregon 97403-1253
Received March 16, 1998
The charge-transfer spectra of 1:1 molecular complexes of seven 1-alkyl-1-(2-naphthyl)ethenes
(
where R ) alkyl group: H, Me, Et, n-Pr, i-Pr, t-Bu, Neopent) and 1-vinylnaphthalene with TCNE
were determined in ClCH CH Cl as solvent at 27.1 °C. Equilibrium constants, K, for these
2
2
complexes vary with R, as based on two opposing factors, viz. (a) the polar substituent factor σ*
and (b) the angle-of-twist θ between the planes of the naphthyl and vinyl groups. For R ) Neopent
(
σ* dominant) K is 50 times as large as for R ) t-Bu (θ dominant). Except for R ) t-Bu, kinetics
of reaction conform with the equation D + TCNE a (K) D‚TCNE a (k
1
/k-1) P, where D is the
donor alkene, P is the Diels-Alder 1,4-cycloaddition product, and k
1
and k-1 are first-order reaction
vary from 0.71 min (R ) H) to 55.5 min (R ) Neopent) and the
-
1
-1
rate constants. Values of k
1
corresponding relative second-order rate constants k
2
(or k
1
K) from 1 to 4000. The rate constant
-1 was measured only for 1b (R ) Me, 0.0017 min ) in the solvent mixture p-xylene/ClCH CH
Cl. Formation of 1b‚TCNE complex gives ∆H ) 10.0 kcal/mol and ∆S ) 38.4 eu, and conversion
to P shows an Arrhenius activation energy of E ) 7.24 kcal/mol. It is proposed that the preferred
-
1
k
2
2
-
a
conformation of a naphthylalkene for complexation has the R and vinyl groups projecting outward
from opposite sides of the plane of the naphthalene ring. The TCNE molecule then aligns parallel
to the naphthalene ring on the vinyl side where (except for R ) t-Bu) it can slide into the geometry
of the transition state to form P.
In tr od u ction
lenes 1b-f, respectively.7 However, the lack of NMR
spectral facilities at that time did not permit definitive
establishment of the proposed structural formulas. We
now present spectral data to confirm some representative
structures from that array of compounds as well as for
the additional 2-alkenylnaphthalene 1g (R ) neopentyl)
and its TCNE adduct 2g.
There have been numerous studies of the molecular
complexation of tetracyanoethylene (TCNE) with ali-
phatic and aromatic hydrocarbons. Extensive data are
presented for these charge-transfer (CT) complexes by
Frey and co-workers.1 Particularly pertinent to the
present investigation are the results for styrene, R-meth-
ylstyrene, and 1- and 2-vinylnaphthalenes, which not
only form CT complexes but also undergo Diels-Alder
cycloaddition with TCNE. Thus, J apanese workers3
found that the equilibrium constant for complexation of
styrene with TCNE is larger than that for R-methylsty-
,2
-6
3
rene in CHCl at 25 °C but that the latter hydrocarbon
undergoes thermal 1,4-cycloaddition faster than styrene
does under the same conditions. Their adducts were not
isolated, but the structures were ascertained by spectral
investigation on solutions at elevated pressure. Consid-
erably earlier, we reported the formation of isolable
crystalline 1:1 adducts (2a , 4) of the vinylnaphthalenes
(1a , 3), as well as those (2b-f) of the 2-alkenylnaphtha-
*
Tel.: (541) 346-1695. Fax: (541) 346-0487. E-mail: lklemm@oregon.
uoregon.edu.
†
Present address: Department of Aeronautical and Astronautical
Engineering, 101 Transportation Bldg., University of Illinois, Urbana,
IL 61801-2997.
(1) Frey, J . E.; Andrews, A. M.; Ankoviac, D. G.; Beaman, D. N.;
DuPont, L. E.; Elsner, T. E.; Lang, S. R.; Zwart, M. A. O.; Seagle, R.
E.; Torreano, L. A. J . Org. Chem. 1990, 55, 606.
(
2) Frey, J . E.; Andrews, A. M.; Combs, S. D.; Edens, S. P.; Puckett,
J . J .; Seagle, R. E.; Torreano, L. A. J . Org. Chem. 1992, 57, 6460.
3) Nakahara, M.; Uosaki, Y.; Sasaki, M.; Osugi, J . Bull. Chem. Soc.
J pn. 1980, 53, 3395.
4) Uosaki, Y.; Nakahara, M.; Sasaki, M., Osugi, J . Chem. Lett. 1979,
27; Bull. Chem. Soc. J pn. 1981, 54, 2893.
5) Uosaki, Y.; Nakahara, M.; Osugi, J . Bull. Chem. Soc. J pn. 1981,
4, 2569, 3681; 1982, 55, 41.
6) Tsuzuki, H.; Uosaki, Y.; Nakahara, M.; Sasaki, M.; Osugi, J . Bull.
Chem. Soc. J pn. 1982, 55, 1348.
(
(
This paper concerns primarily the effect of changing
the R group (1) on the value of the equilibrium constant
7
(
5
(
(7) Klemm, L. H.; Solomon, W. C.; and Kohlik, A. J . J . Org. Chem.
1962, 27, 2777.
S0022-3263(98)00490-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/01/1998

6
504 J . Org. Chem., Vol. 63, No. 19, 1998
Klemm et al.
K for 1:1 molecular complexation between 1 and TCNE
and (2) on the kinetics of the Diels-Alder reaction to give
This situation implies that 1c, 1d , and 1g have closely
7
similar angles of twist (θ) in all three compounds.
cold equimolar mixture of 1g and TCNE in CHCl
A
-
2
. For comparison purposes, 1-vinylnaphthalene (3) is
3
also included in this study.
EtOAc precipitated crystalline adduct 2g (60%). Typical
5
,6
1
As based on investigations in the styrene and an-
thracene8 systems, the 1:1 CT complex has been shown
to be an intermediate in the 1,4-cycloaddition process,
as indicated by eq 1, where D is the hydrocarbon donor,
structures corroborated by H NMR spectra are those of
,9
the 2-alkenylnaphthalenes 1b, 1e, and 1g and the Diels-
Alder adducts 2a , 2f, and 4. Mass spectral data are also
presented for 1a , 1b , 1g, 2a , 2f, 2g, and 4. As noted
7
previously, the primary adduct 2a (mp 170.5-171.5 °C)
K
k1
undergoes base-catalyzed isomerization to the disubsti-
D + TCNE {
\
} D‚TCNE {k-1} P
(1)
tuted naphthalene 6 (mp 201-202 °C). This structural
1
transformation is now confirmed by H NMR in that 2a
P is the Diels-Alder product, K is the equilibrium
constant for CT complexation, and k and k-1 are first-
has four protons that resonate at δ >7 ppm, while 6 has
six protons in this aromatic region. Both 2a and 6
undergo the retro-Diels-Alder reaction on electron im-
pact, but 2a forms pertinent fragments at m/z 154 (for
1
order rate constants for the conversion of D‚TCNE into
P, and its reverse process, respectively. Three separate
investigations were made in order to corroborate and
quantify eq 1 for the alkenylnaphthalenes, viz. (1)
measurements of K and the attendant extinction coef-
+
+
1
a ) and 128 (for TCNE ), while 6 forms the most
+
abundant peak at 204 (for C14
8 2
H N ), with apparent loss
of CH )C(CN) . The isomerization of 2a to 6 is also
2
2
ficients ꢀ in ClCH
2 2
CH Cl at 27.1 °C, (2) direct determi-
consistent with a change in ultraviolet absorption spec-
nation of k by following the decrease in concentration
1
trum. Adduct 2f, like 2a , reverts to its ionized hydro-
of D‚TCNE as a function of time t, and (3) direct
determination of k-1 by following the equilibration of P
with p-xylene, as in eq 2. Comparison of the results from
+
carbon precursor (1f ), m/z 210, in the mass spectrometer.
It should be noted that there is no evidence that linear
condensation of 1 to form adduct 8 occurs.
k
-
1
P + p-xylene { } p-xylene‚TCNE + D
(2)
Ch a r ge-Tr a n sfer Da ta
k1
(excess)
(colored)
Charge-transfer data for D‚TCNE complexes (Table 1)
were measured in 1,2-dichloroethane, rather than in
CH Cl2 or CHCl3 (as used by many other research
groups), because of its lower volatility and because of the
procedures (1) and (2) then provides justification for the
intermediacy of D‚TCNE (except for D ) 1f), as shown
in eq 1. Direct comparison of the relative rates of
2
reaction of D and TCNE to form P is then given by the
relatively high solubility (about 0.11 mol/L at room
temperature) of TCNE therein. The colored complex
forms as rapidly as solutions of D and TCNE can be
mixed.
, as calculated from eq 3.5
2
,10a
second-order rate constant k
k ) k K
(3)
2
1
Values of K and ꢀ were obtained from Scott plots5,11 in
In our previous paper,7 we presented data on (1)
ultraviolet absorption spectra in absolute ethanol, (2)
polarographic half-wave reduction potentials in aqueous
dioxane, and (3) gas-liquid chromatographic retention
times in Carbowax 20M (at a constant temperature) for
alkenes 1a -f. The data indicated that these compounds
exist in varying conformations, i.e., in preferred angles
of twist (θ) between the planes of the naphthalene ring
and of the vinyl double bond. This order of twist, as given
in terms of the R group, is H (θ = 0°) < Me e Et ) n-Pr
the form of eq 4, where [TCNE]
concentrations of TCNE and D (always the larger of the
two), respectively, A and A are initial and final absor-
bances, and l is the path length (in cm) through the
0 0
and [D] are the initial
0
∞
absorption cell. From a plot of [TCNE]
0
/(A
0
- A
∞
) versus
[
TCNE]0
1
1
)
+
(4)
(A - A ) Kꢀl[D]0 ꢀl
0
∞
<
i-Pr , t-Bu (θ near 90°). Despite these different
1
0
/[D] one determines ꢀ from the intercept and, thence,
inherent conformations, all of the alkenes were found to
undergo Diels-Alder reactions with TCNE. The goal of
the current study is further delineation of how the
various angles of twist effect interaction between the
alkene and TCNE.
K from the slope of the straight line, with an average
correlation coefficient of 0.998 for the nine hydrocarbons
in Table 1 (see Figure 1).
The donor compounds are listed in Table 1 in order of
increasing experimental value of the equilibrium con-
stant K. Though irregularities occur, one notes a general
increase in λmax and a corresponding decrease in ꢀ with
increasing values of K for the series naphthalene (7),
New Syn th eses a n d Str u ctu r a l Stu d ies
Friedel-Crafts condensation of naphthalene (7) with
1
-vinylnaphthalene (3), and 1a -g. The inverse relation-
3
5
,3-dimethylbutanoyl chloride gave a 64% yield of ketone
ship between ꢀ and K is also seen in Table 2, where
1
, identified as the 2-naphthyl derivative by H NMR
measurements were extended to temperatures other than
(singlet at 8.47 ppm) and by oxidation to 2-naphthoic
2
7.1 °C for three members of this series. Such general
acid. Wittig reaction of 5 produced 1g (67%), which
showed an ultraviolet absorption spectrum essentially
superimposable in shape on those reported for 1c (R )
Et) and 1d (R ) n-Pr), albeit not identical in log ꢀ values.
correlations were noted years ago by Orgel and Mul-
liken12 and more recently for the TCNE complexes of
5
styrene and its 4-methyl derivative. In comparing
1
styrenes with their methylbenzene analogues, Frey et al.
observed that the former have much larger K values,
(8) Kiselev, V. D.; Miller, J . G. J . Am. Chem. Soc. 1975, 97, 4036.
(9) Lofti, M.; Roberts, R. M. G. Tetrahedron 1979, 35, 2131.
(10) Atkins, P. W. Physical Chemistry, 5th ed.; W. H. Freeman: New
(11) Scott, R. L. Rec. Trav. Chim. Pays-Bas 1956, 75, 787.
York, 1994; (a) pp 887-888; (b) pp 877-878.
(12) Orgel, L. E.; Mulliken, R. S. J . Am. Chem. Soc. 1957, 79, 4839.

Diels-Alder Condensation of Naphthylalkenes with TCNE
J . Org. Chem., Vol. 63, No. 19, 1998 6505
Ta ble 1. Ch a r ge-Tr a n sfer Da ta for Som e Su bstitu ted Na p h th a len es w ith TCNE in 1,2-Dich lor oeth a n e a t 27.1 °C
long wavelength band
exptl equil constantc
calcd equil constantf
donor compd substituent on
short wavelength λmaxa molar extinction no. of data
angle of twist adjusted
K
a
coeff, ꢀb
pointsd
K (L/mol)e
θ (deg)g
h
(D) no.
naphthalene
band λmax (nm) (nm)
σ* value (L/mol)
7i
nonej
2-(t-Bu-vinyl)
2-vinyl
1-vinyl
2-(i-Pr-vinyl)
2-(n-Pr-vinyl)
2-(Et-vinyl)
2-(Me-vinyl)
2-(Neopent-vinyl)
426
455
472.5
425
480
487.5
(500)
(506)
475
547
570
577
599
580
600
600
594
610
3060
2980
3320
1720
820
375
155
100
39
7
8
8
9
6
7
10
6
0.20k
1
1
3
1
1
1
1
1
f
0.23 ( 0.05
0.27 ( 0.04
0.32 ( 0.12
1.09 ( 0.26
5.6 ( 2.3
7.5 ( 2.0
10.5 ( 0.7
13.3 ( 2.9
75.6
18.6
-0.79
0.00
0.23
i
l
m
a
i
n
e
d
c
b
g
57.3
-0.68
-0.61
-0.59
-0.49
-0.65
1.29
9.1
3.6
18.5
9.5
40.7o
49.6o
36.7
p
p
5
39.4o
a
Determined from a 1:1 molar mixture of donor and TCNE. Determined from Scott eq 4. c Determined from eqs 4 and 7 for long-
b
d
e
wavelength bands only. Used in eq 4. Averaged values obtained from eqs 4 and 7, where the ( variation shown represents half of the
difference between the two numbers obtained and not experimental accuracy in K. Donors are listed in order of increasing value of K.
f
Calculated by empirical eq 5 with F* ) 1.018 and a ) 2.75 to give a least-squares fit of the data for 1b-g combined. See: Bennet, C. A.;
Franklin, N. L. Statistical Analysis in Chemistry and Chemical Industry; J ohn Wiley: New York, 1954; pp 222-226. g Calculated for the
parametric quantum mechanical model AM1: Dewar, M. J . S.; Zoebisch, E. G.; Healey, E. F; Stewart, J . J . P. J . Am. Chem. Soc. 1985,
h
1
)
07, 3902. Ballinger, P.; Long, F. A. J . Am. Chem. Soc. 1960, 82, 795. σ* is adjusted algebraically to give R ) H ) 0.00, rather than R
i
j
Me ) 0.00. Reference 2 lists λmax and K values for these compounds in CH2Cl2 at 22 °C, as given in footnotes j, l, and n. Reported
k
data: 550-560 and 429-433 nm; K ) 0.6-1.19. Determined only by means of eq 4. Since 7 does not undergo Diels-Alder reaction, K
l
m
cannot be calculated from eq 7 for comparison with this value. Reported data: 580 and 465 nm; K ) 0.68 ( 0.02. Previous experiments
n
o
indicate that this is effectively 0° (ref 7). Reported data: 603 and 432 nm, K not given. Ultraviolet absorption spectra of these three
compounds indicate that θ is effectively the same in all of them. p Shoulder.
Ta ble 2. Tem p er a tu r e Dep en d en ce of TCNE
Com p lexa tion w ith Na p h th a len e a n d Tw o
2
-Alk en yln a p h th a len es in 1,2 -Dich lor oeth a n e
donor compd
no. of
molar extinction
Ka,b
(L/mol)
(D) no.
T °C measurements
coeff, ꢀa
7
7
7
1
1
1
1
1
1
-29.6
+0.5
27.1
27.1
47.6
-29.6
+0.5
8.7
5
7
7
8
6
9
9
7
6
3370
2340
3060
3320
1720
5500
820
0.346
0.376
c
0.20
0.27
c
a
a
b
b
b
b
0.538
0.24
1.51
6.1
525
100
c
27.1
10.5
a
Average value for the correlation coefficient of the nine
experiments is 0.998. Data refer to the long-wavelength CT bands.
b
A plot of ln K versus 1/T (where T is temperature on the Kelvin
scale) for naphthalene (7) did not yield a straight line. This
nonlinearity may be ascribed to experimental error in values of ꢀ
as determined by eq 4 or to an inherent property of the system.
On the contrary, 1b did give a reasonably straight line (Daniels,
F.; Alberty, R. A. Physical Chemistry, 4th ed.; J ohn Wiley: New
York, 1975; pp 170-173). c See Table 1.
gated), and 1f (effectively nonconjugated) have nearly
identical K values. This situation, plus the order R )
i-Pr < n-Pr e Et < Me e Neopent for increasing stability
of the complex, implies that electronic charge donation
to the naphthalene ring by the group R operates through
the vinyl group and will be altered by the angle of twist
θ. In an attempt to rationalize the experimental values
of K in a semiquantitative manner, however, we assume,
to a first approximation, that K is a function of both σ*,
the Taft polar substituent constant, and θ, acting inde-
pendently of one another. The linear free-energy rela-
tionship, eq 5, is used for this calculation. In this
F igu r e 1. Scott plots (eq 4) of 1000[TCNE]
0
/(A
0
- A
∞
) versus
2
CH -
1
00/[D] for three 1-alkyl-1-(2-naphthyl)ethenes in ClCH
0
2
Cl at 27.1 °C. Calculated values of K and ꢀ are included in
Table 1. Concentrations (M).
2
log K ) F*σ* + a cos θ
(5)
while the latter have larger absorbances. In general,
however, all alkyl substituents on benzene or naphtha-
lene tend to increase the value of K.
equation, F* and a are constants to be determined, and
calculation is not applied to 1a since σ* is taken as zero
in that case. The function cos θ satisfies the boundary
2
2
It is noteworthy from Table 1 that naphthalene (which
does not undergo Diels-Alder reaction with TCNE),
-vinylnaphthalene (nearly coplanar and highly conju-
conditions of 1 for θ ) 0° and 0 for θ ) 90°. Also, it is
consistent with the simple concepts that the CT band of
longest wavelength involves electron transfer from the
2

6
506 J . Org. Chem., Vol. 63, No. 19, 1998
Klemm et al.
sandwich structure. This complexation will be more
facile as conjugation by the vinyl group increases (for
smaller θ) and as increasing π-electronic charge is driven
from the alkyl side of the ring toward the acceptor TCNE
molecule on the other side.
Mea su r em en t a n d Use of Ra te Con sta n t k
1
Kinetics of the reaction of the complex D‚TCNE to give
P were conducted on mixtures of D (used in large excess,
such that [D]
t
0 2 2
= [D] ) and TCNE in ClCH CH Cl at
constant temperature. Disappearance of the colored
complex was followed spectrophotometrically as a func-
tion of time, t. Three typical linear plots of log absor-
bance (A) versus t, as shown in Figure 3, indicate that
the reaction is first order in the concentration of the
F igu r e 2. Proposed preferred conformations of 1b (right-hand
diagram) and 1f (left-hand diagram) for complexation with
TCNE. In these molecular mechanical (AM1) drawings, the
naphthalene ring is viewed edgewise and the TCNE molecule
is expected to align itself in a plane parallel to that of
naphthalene and on the vinyl side.
complex. For quantitative determination of k
eq 6, which assumes that k-1 is negligibly small, as
compared to k Then if one lets kobs represent the
1
we used
1
.
-2.303 log(A - A ) )
∞
k1[D]0
HOMO of D to the LUMO of TCNE13-15 and that this
transition can be regarded as occurring in a one-
t - 2.303 log(A - A ) (6)
0
∞
(
)
K[D] + 1
0
dimensional potential box of length y, where λmax is
directly proportional to the square of y.16 Data used for
coefficient of t in eq 6 one obtains eq 7, from which one
calculates k
1
and a second value of K as derived from a
(a modified Guggenheim
plot6 ). Illustrative data for one series of experiments
θ and σ*, the procedure for calculations, and the calcu-
lated values of K obtained are included in Table 1. It is
apparent that calculated values have the correct order
of magnitude, with good agreement for large values of θ
linear plot of 1/ kobs versus 1/[D]
0
,19
1
kobs
1
k1
1
[D]0
K
k1
(i.e., for 1e and 1f) and larger deviations from the
)
+
(7)
(
)
experimental results for θ < 50°. It seems clear that
2
some modification in the term a cos θ could improve the
fit. However, one arrives at the overall conclusion that
the stability of the complex D‚TCNE increases with
electronic donation by the R group and decreases with
increasing angle of twist θ.
From Table 2, it is notable that naphthalene shows
little change in K over a temperature range of 56.7 °C,
while K increases 40-fold for 1b over this same range and
with donor 1b are presented in Table 3 and typical
kinetic plots are shown in Figure 4. In every case, good
linearity was obtained over at least 2 half-lives, usually
much longer. Comparative first-order rate constants k
and second-order rate constants k are listed in Table 4.
One notes that 1- and 2-vinylnaphthalenes have com-
parable K, k , and k values. While the tert-butyl
1
2
1
2
compound 1f has similar stability in its complex, steric
hindrance appears to slow its Diels-Alder addition by a
factor of at least 20 times at 27.1 °C. At that standard
doubles for 1a over a shorter range of 20.5 °C.
A
suggested geometrical interpretation for these differ-
ences, as well as for the K values in Table 1, is that both
faces of naphthalene are open to unhindered approach
by a TCNE molecule. 2-Vinylnaphthalene may be simi-
temperature, k
1
values fall in the order for R of Neopent
values, in the
>
Me > n-Pr > Et . i-Pr . t-Bu, and k
2
order Neopent > Me > Et > n-Pr > i-Pr . t-Bu. The
larly situated. However, Figure 2, obtained by a molec-
ular mechanics program,17 shows conformations for 1b
order of k values is consistent with a general change in
2
the domination of polar effects (for the neopentyl group)
and 1f that we suggest are preferred during TCNE
complexation. In these cases, the R group (Me or t-Bu,
respectively) protrudes outward to one side of the plane
of the naphthalene ring, while the vinyl group protrudes
to that of domination of steric effects, i.e., hindrance to
attainment of transition state geometry (for the tert-butyl
group). Apparently, the presence of a methylene group
between the naphthylvinyl and the tert-butyl portions of
1g introduces sufficient flexibility in the molecule as to
counteract hindrance by the bulky tert-butyl portion.
outward from the other side of this plane. As proposed
by Mobley et al.18 for complexation of asymmetrically
substituted benzenes and naphthalenes, the TCNE will
align itself in a parallel (or nearly parallel) plane on the
vinyl (less hindered) face of the naphthyl group in a π-π
1
Figure 5 is a plot of log K (Table 1) versus log k (Table
4
) for reaction at 27.1 °C. This linear plot for compounds
1
a -e and 1g with a positive slope implies that the
complex D‚TCNE is an intermediate on the pathway from
(
13) Thompson, C. C.; Giumanini, A. G.; Lepley, A. R. Boll. Chim.
Farm. 1967, 106.
14) Merrifield, R. E.; Phillips, W. D. J . Am. Chem. Soc. 1958, 80,
778.
6
,20
reactants D and TCNE to Diels-Alder product P.
The
(
point for the tert-butyl compound 1f (x, y coordinates at
2
(
15) Mobley, M. J .; Rieckhoff, K. E.; Voigt, E.-M. J . Phys. Chem.
1
977, 81, 809.
(19) Note that kobs in eq 7 is different from kobsd, as used in ref 6.
(
16) Gillam, A. E.; Stern, E. S. An Introduction to Electronic
(20) The positive slope is consistent with the concept that the rate
of Diels-Alder cyclization (as measured by k ) increases as the
1
instantaneous concentration of molecular complex (as measured by K)
increases. If the slope were negative, the rate of cyclization would
decrease with increasing concentration of complex and the kinetics
should be more complicated.
Absorption Spectroscopy in Organic Chemistry, 2nd ed.; Edward
Arnold: London, 1960; p 41.
(
17) See footnote g, Table 1.
(18) Mobley, M. J .; Rieckhoff, K. E.; Voigt, E.-M. J . Phys. Chem.
1
978, 82, 2005.

Diels-Alder Condensation of Naphthylalkenes with TCNE
J . Org. Chem., Vol. 63, No. 19, 1998 6507
F igu r e 4. Plot of 1/kobs (min) versus 1/[1f]
M) (eq 7) for 1f in excess TCNE plus ClCH
different temperatures. Calculated values for K and k
C are included in Tables 1 and 4.
0
(in units of 100/
CH Cl at two
at 27.1
2
2
F igu r e 3. Linear plots of log A (arbitrary units) versus
reaction time t (min) for disappearance of the molecular
1
°
complex D‚TCNE in first-order reaction kinetics in ClCH
Cl plus excess TCNE at 27.1 °C: curve 1 for [3] of 0.1063 M;
curve 2 for [3] of 0.0364 M; curve 3 for [1a ] of 0.0272 M.
2 2
CH -
0
according to the Arrhenius eq 9.10b From the slopes and
0
0
ln k ) ln B - E /RT
(9)
Ta ble 3. Illu str a tive Exp er im en ta l Da ta for th e
a
Disa p p ea r a n ce of 1b‚TCNE a s a F u n ction of Tim e a t 27.1
°
C in ClCH2CH2Cl
the intercepts of these curves one calculates the activa-
tion energies E as 7.24 and 17.8 kcal/mol and the
preexponential factors B as 6.58 × 10 min and 4.31
10 L/mol/min for k ) k and k ) k , respectively.
Mea su r em en t a n d Cor r ela tion of Ra te Con sta n t
-1. Measurement of the rate of dissociation of Diels-
Alder product P was determined in one case, that of 2b
in ClCH CH Cl at 27.1 °C at three different initial
3
a
b
c
10 [TCNE]0
expt
no.
[1b]0
[TCNE]0
1/[1b]0
1/kobs
(min)
2
1
6
-1
a
(10 × M) (10 × M) (10-2 L/mol)
2
4
(A0 - A∞)
1
5
×
1
2
9
9
9
9
9
9
a
b
c
d
e
f
1.03
2.00
3.00
5.00
7.27
7.27
5.00
5.00
5.00
0.971
0.500
0.333
2.02
1.62
1.21
10.87
6.59
3.84
21.0
18.3
13.2
2.78
1.56
1.03
5.34
4.16
3.30
k
0.494
0.618
0.824
2
2
concentrations. Since the solutions of 2b are colorless,
progress of the reaction was followed spectrophotometri-
cally by adding a standard 2b-ClCH CH Cl solution to
2 2
excess p-xylene and obtaining the increase in the con-
centration of colored complex (p-xylene‚TCNE) as a
function of time. A calibration chart for the use of this
procedure is shown in Figure 6. The approach to equi-
librium is given by eq 2, and the rate constant, k-1, is
calculated as the slope of a linear plot of the left side of
eq 10 versus time t,22 where a is the initial concentration
a
Listed in chronological order, as conducted. The subscript zero
b
refers to conditions at t ) 0. A specific example of [D]0 as given
c
in eq 7. This value is the sum of free, complexed, and reacted
TCNE at any time, t.
-
0.34, 0.36) falls far off the straight line, inasmuch as
the rate of cycloaddition of 1f is relatively very slow as
compared to its equilibration. Thus, it seems likely that
the kinetics for 1f are more accurately represented by
eq 8 than by general eq 1. However, for 1a -e, and g, it
Xe
aX + X(a - X )
e
e
k1
K
ln
) k-1
t
(10)
1
f‚TCNE {
\
} 1f + TCNE {k-1} P
(8)
[
]
(
2a - Xe)
a(X - X)
e
of 2b, X is the concentration of p-xylene complex at any
is proposed that the complex (with the TCNE molecule
held on the vinyl side of the naphthalene ring) reacts
further to give 1,4-cycloaddition by movement (sliding
and, perhaps, rotating) of the TCNE addendum into the
required geometry of the transition state.
time, and X
e
is that of the complex when equilibrium is
attained. Figure 7 shows the plot for one of the experi-
(
21) B is used here, rather than A, so as to avoid confusion with
absorbance.
Plots of ln k
1 2
and of ln k (see Table 4) for compound
b versus 1/T (in Kelvin ) give linear relationships,
(
22) Frost, A. A.; Pearson, R. G. Kinetics and Mechanism, 2nd ed.;
J ohn Wiley: New York, 1961; pp 186-187.
-
1
1

6
508 J . Org. Chem., Vol. 63, No. 19, 1998
Klemm et al.
Ta ble 4. Kin etic Resu lts for th e Con ver sion of Don or D a n d TCNE in to th e Diels-Ald er 1,4-Cycloa d d u ct (P ) in
ClCH2CH2Cl
expt
series no.
donor compd
(D) no.
no. of
measurements
k1a
standard
deviation for k1
half-life,
t1/2 (min)
k2b
rel k2 value
for 27.1 °C
T °C
(min-1)
(L/mol/min)
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
3
3
1a
1a
27.1
27.1
27.1
27.1
47.6
-29.6
+0.5
8.7
27.1
46.5
27.1
27.1
27.1
0.5
9
6
8
5
7
8
9
7
6
6
10
7
6
4
1.11
0.01
0.01
0.02
0.01
0.07
0.02
0.25
0.31
1.13
5.81
0.57
0.44
0.02
0.36
0.37
0.19
0.20
1.12
0.46
17.8
127
423
e
170
152
1.9
1.9
0
0
.613
.97
c
}
}
1.15
0.70
0.73
2.08
1.92
d
1
1
c
d
1a
1b
1b
1b
1b
1b
1c
1d
1e
1f
0.333
0.359
0.059
0.033
0.017
0.011
0.031
0.025
0.283
11.8
20.8
40.3
64.5
22.7
27.2
3
2.23 × 10
1
1
1
1
1
1
1
2
8.95 × 10
2
8.00 × 10
1
2.45
2.67
e
1.41 × 10
-
-
2
2
1.22 × 10
56.8
15.2
0.012
-4
(1.05 × 10-2)f
(5.52 × 10 )f
-2
1f
1g
27.1
27.1
12
6
4.56 × 10
6 × 10
3
55.5
2.78
738
3.89 × 10
a
First-order rate constant for the reaction D‚TCNEfP. b Second-order rate constant for the reaction D + TCNE f P (calculated by eq
d
3
). c Duplicate series was conducted separately and at a different time from the previous series. Arbitrarily chosen as a standard of
reference. K was not determined for this series. f It is likely that this datum is not directly comparable with other data in this column
e
(see discussion of Figure 5).
F igu r e 5. Linear plot of log (equilibrium constant K) versus
log (first-order rate constant k ) for donor alkenes 1a -e plus
g to support the intermediacy of molecular complex D‚TCNE
as shown in eq 1. T ) 27.1 °C.
1
1
ments; average correlation coefficient 0.994, k-1 ) 1.72
-
3
-1
(
)
0.13 × 10 min ; t1/2 ) 403 min. The ratio of k
1
/k-1
4
2.34 × 10 confirms that k-1 can be neglected as shown
in eq 6.
F igu r e 6. Calibration curve for absorbance A of the complex
p-xylene‚TCNE versus 1000[TCNE] in excess p-xylene. Used
in conjunction with Figure 7.
Exp er im en ta l Section
Purification of alkenes 1a -f, TCNE, and their condensation
products 2a -f was conducted as described previously.7 Ana-
lytical-grade 1,2-dichloroethane was distilled twice through a
packed fractionating column, stored over a mixture of anhy-
drous CaCl /K CO for 1 week, and finally refractionated, bp
2 2 3
4.0 °C. Chemically pure p-xylene was dried with Na wire
and fractionated, bp 136.4-136.6 °C.
3
NMR spectra were determined on CDCl solutions at 300
MHz; MS spectra were obtained by impact with 70 V electrons.
Coupling constants (J ) are given in Hz. Miscellaneous spectra
on previously reported alkenes and Diels-Alder adducts are
presented here.
7
2-(2-Na p h th yl)-1-eth en e (1a ): IR (CHCl ) 1640, 1600,
3
-
1
1490, 1420, 900 cm ; MS m/z 154 (100), 153 (51), 152 (33),
128 (13).
2-(2-Na p h th yl)-1-p r op en e (1b):7 1H NMR δ 7.8-9.6 (m,
4H), 7.73 (dd, J ) 1.8, 6.9, 1H), 7.45-7.56 (m, 2H), 5.60 and
5.26 (2d, J ) 0.9, 2H), 2.33 (s, 3H); MS m/z 168 (100), 153
(50), 128 (38).
8
3-Meth yl-2-(2-n a p h th yl)-1-bu ten e (1e):7 1H NMR δ 8.01
(m, 4H), 7.64-7.73 (m, 3H), 5.51 and 5.36 (2 split s, 2H), 3.18
(m, 1H), 1.36 (2d, J ) 6-7, 6H).
2,3,4,4a -Tet r a h yd r op h en a n t h r en e-3,3,4,4-t et r a ca r bo-

Diels-Alder Condensation of Naphthylalkenes with TCNE
J . Org. Chem., Vol. 63, No. 19, 1998 6509
1
.12 (s, 9H); MS m/z 226 (31), 170 (71), 155 (100), 127 (71).
Anal. Calcd for C16
H, 7.83.
Oxidation of 5 by means of KO-t-Bu and O
thoic acid, mp 184-185 °C (lit.26 mp 185.5 °C).
,4-Dim et h yl-2-(2-n a p h t h yl)-1-p en t en e (1g). This al-
kene was prepared from 5 by the Wittig procedure as used for
H18O: C, 84.91; H, 8.02. Found: C, 85.04;
2
5
2
gave 2-naph-
4
7
1
e. The crude product from chromatography (mp 44-47 °C,
6
7%) formed needles (mp 49-50 °C) from MeOH: IR (CCl -
4
-
1
2
CS ) 1622, 1599, 1506, 860, 822, 753 cm ; ES (100% EtOH)
λ
2
7
max (log ꢀ) 239 (4.74), 265 sh (3.56), 274.5 (3.68), 284.5 (3.71),
92 sh (3.54), 296 nm (3.55); H NMR δ 7.5-7.9 (m, 4H), 7.2-
.5 (m, 3H), 5.40 and 5.15 (2s, 2H), 2.61 (s, 2H), 1.89 (s, 9H);
1
MS m/z 224 (37), 168 (100), 57 (93). Anal. Calcd for C16
20
H :
C, 91.01; H, 8.99. Found: C, 90.64; H, 8.92.
1
-(2,2-Dim e t h ylp r op yl)-2,3,4,4a -t e t r a h yd r op h e n a n -
th r en e-3,3,4,4-tetr a ca r bon itr ile (2g). An equimolar mix-
ture of alkene 1g (2.24 g) and TCNE (1.28 g) was dissolved in
a 1:1 mixture of CHCl -EtOAc in a vessel protected from the
3
light. When the solution was concentrated and cooled to -10
°
°
C, needles (2.04 g, 60%) of 2g precipitated: mp 143-143.5
C; MS m/z 352 (0.1), 224 (21), 168 (73), 128 (37), 76 (35), 57
(
100). Anal. Calcd for C23
Found: C, 78.59; H, 5.53; N, 15.60.
,2,3,10a -Tetr a h yd r op h en a n th r en e-1,1,2,2-tetr a ca r bo-
n itr ile (4). Following essentially the same procedures as used
20 4
H N : C, 78.38; H, 5.72; N, 15.90.
1
2
7
for 2g, a mixture of purified 1-vinylnaphthalene (3) and
TCNE deposited a 93% yield of adduct 4, obtained as light-
sensitive prisms on recrystallization from CHCl
3
: mp 182-
-
1
1
83 °C; IR (KBr) 2260, 1655, 1570, 1488, 755, 675 cm ; ES
1
(
100% EtOH) λmax (log ꢀ) 245 (4.52), 295 (3.63), 305 (3.59); H
NMR δ 7.43 (d, J ) 7.5, 1H), 7.33 (pentet, J ) 7.5, 2H), 7.14
d, J ) 7.5, 1H), 6.77 (dd, J ) 3, 9.6, 1H), 6.0-6.15 (m, 2H),
.45 (split s, 1H), 3.35-3.58 (dm, 2H); MS m/z 282 (29), 204
26), 154 (100). Anal. Calcd for C18 : C, 76.58; H, 3.57;
(
4
(
10 4
H N
F igu r e 7. Plot of first-order dissociation kinetics of Diels-
Alder adduct 2b as per eq 10. The vertical axis corresponds to
the left-hand side of eq 10 and is a pure number without units.
N, 19.85. Found: C, 76.33; H, 3.95; N, 19.99.
Ch a r ge-tr a n sfer sp ectr a of the TCNE complexes of 1, 3,
and naphthalene were obtained on equimolar mixtures (ca. 1
The horizontal axis is time t (min). The solvent is ClCH
Cl/p-xylene at 27.1 °C.
2
CH
2
-
×
10
-3
M) of the two components in 1,2-dichloroethane
contained in a 10-cm quartz cell, thermostated to 27.1 ( 0.1
C in a Cary 11 spectrophotometer. Solutions of the compo-
°
7
nents were mixed rapidly, and measurements were made by
scanning the region 750-400 nm as rapidly as possible (3-4
min) in order to minimize Diels-Alder reaction. Wavelengths
for the two absorption maxima found in each case are given
in Table 1.
n itr ile (2a ): ES (100% EtOH) λmax (log ꢀ) 231.5 (4.42), 251.5
1
(
4.04), 285 (3.90), 296 (3.96), 313 (3.96), 325 (3.63); H NMR δ
7
1
3
1
.88 (dd, J ) 2.1, 6.3, 1H), 7.38-7.5 (m, 2H), 7.24-7.3 (m,
H), 6.45 (dd, J ) 1, 8.3, 2H), 5.77 (m, 1H), 4.88 (bs, 1H), 3.2-
.6 (m, 2H); MS m/z 282 (19), 154 (100), 153 (50), 152 (30),
28 (26).
Deter m in a tion s of K a n d E for th e lon g-w a velen gth
m a xim u m for ea ch d on or h yd r oca r bon (D) were made
1
-(1,1-Dim eth yleth yl)-2,3,4,4a-tetr ah ydr oph en an th r en e-
7
1
0
from the Scott plots, as shown in eq 4 and Figure 1. [D] , the
3
7
4
1
1
,3,4,4-tetr a ca r bon itr ile (2f): H NMR δ 7.70 (dd, 1H),
.35-7.45 (m, 2H), 7.2-7.3 (m, 1H), 6.70 (dd, J ) 10, 59, 2H),
.68 (bs, 1H), 3.44 (dd, J ) 4, 18.9, 1H), 3.20 (dd, J ) 1.6,
8.9, 1H), 1.34 (s, 9H); MS m/z 338 (8), 210 (100), 195 (57),
53 (97).
-
3
initial molar concentration of D (4 × 10 to 1.6), was always
larger (8-1100 times) than the initial molar concentration of
-
4
-3
TCNE, i.e., [TCNE]
2
0
(1 × 10 to 4 × 10 ). Measurements at
7.1 °C and 47.6 °C were made in a thermostated bath as
above. However, a specially designed Dewar jacketed absorp-
1
,2,3,4-Tetr a h yd r op h en a n th r en e-3,3,4,4-tetr a ca r bon i-
28
tion cell with Pyrex glass windows (cutoff point at 300 nm
7
tr ile (6): ES (100% EtOH) λmax (log ꢀ) 227.5 (4.70), 266 (3.55),
77 (3.74), 286 (3.78), 307 (3.15), 316 (2.17), 322 (2.90); 1
NMR δ 8.38 (d, J ) 9, 1H), 8.04 (t, J ) 9, 2H), 7.87 and 7.74
2t, J ) 7.5, 2H), 7.38 (d, J ) 9, 1H), 3.53 and 2.92 (2t, J )
.3, 4H); MS m/z 282 (41), 204 (100).
in the ultraviolet region) was used for measurements below
room temperature (see Tables 2 and 4). The volume of the
sample compartment was 10 mL, and the light path through
the sample was 2.02 cm. Temperature control was furnished
by adding a coolant (ice-water slurry for 0.5 °C, boiling Freon-
12 for -29.6 °C) to the Dewar flask to a height sufficient to
2
H
(
6
Syn th eses of New Com p ou n d s. 3,3-Dim eth yl-1-(2-
n a p h th yl)bu ta n on e (5). Naphthalene (56 g, 0.44 mol) was
bathe the sample cell. Solutions of D and TCNE in ClCH -
2
2
3
acylated with 3,3-dimethylbutanoyl chloride (59 g, 0.44 mol)
CH Cl were prepared just prior to the experiment. For
2
2
4
by a general Friedel-Crafts procedure. The crude product
bp 115-124 °C/0.11 Torr, 64%) was converted to a monopi-
crate (yellow needles, mp 90.5-91.5 °C), which was dissociated
on a column of alumina with C -CHCl (1:1). Crystalliza-
-CS
675, 858, 825, 746 cm-1; H NMR δ 8.47 (s, 1H), 8.03 (dd, J
experiments above room temperature, the cell compartments
were thermally equilibrated for 2 h and the solutions (in
separate cells of a “quick-mix” vessel) for 20-30 min. Mixing
of the reactants and starting of a stopwatch were conducted
(
6
H
6
3
tion from MeOH gave needles: mp 43-44 °C; IR (CCl
4
2
)
1
1
(
25) von E. Doering, W.; Hains, R. M. J . Am. Chem. Soc. 1954, 76,
82.
(26) Weast, R. C., Ed. Handbook of Chemistry and Physics, 52nd
ed.; Chemical Rubber: Cleveland, 1971; p C-386.
27) Klemm, L. H.; Sprague, J . W.; Ziffer, H. J . Org. Chem. 1955,
20, 200.
(28) Built by H. S. Martin and Son, Evanston, IL.
)
1.4, 8.7, 1H), 7.98 and 7.89 (2d, J ) 7.8, 2H), 7.91 (d, J )
4
8
.7, 1H), 7.62 and 7.57 (2 dt, J ) 1.1, 6.7, 2H), 3.00 (s, 2H),
(
(
(
23) Whitmore, F. C.; Foster, W. S. J . Am. Chem. Soc. 1942, 64, 2966.
24) Buu-Ho ¨ı , N. P.; Cagniant, P. Bull. soc. Chim. 1945, 12 (5), 307.

6
510 J . Org. Chem., Vol. 63, No. 19, 1998
Klemm et al.
simultaneously, and the mixture was rapidly transferred into
the spectrometer. Absorbance, A, was obtained versus time,
t. A was obtained after 8 half-lives (30 min to several hours).
∞
The dissociation of 2b was conducted in a round-bottomed
flask fitted with a water-cooled glass stirrer and a short water-
cooled condenser through which one could withdraw samples.
For experiments below room temperature, the D solution (10
2 2
Into 24 mL of thermally stabilized ClCH CH Cl was washed
mL) was equilibrated in the cell for 20-30 min. Then 100-
74.2 mg of 2b with 1 mL of this solvent. Timing and vigorous
stirring were started. All solid dissolved in 10-20 s. Periodi-
cally, 0.5 mL aliquots were withdrawn and added to 3 mL of
p-xylene in a sample cell, and time and absorbance were
immediately recorded. Data are plotted according to eq 10 (see
Figure 7). The reaction was essentially complete in 30 min.
2
00 µL of TCNE at room temperature was added with vigorous
stirring (magnetic stir bar) and simultaneous start of timing.
Stirring was stopped, and A was measured versus t. Concen-
trations were corrected for change in density of the solvent
with temperature.
Representative plots of log A versus t are shown in Figure
Com p u ta tion s. Calculations were made by means of a
3
(
and of 1/kobs versus 1/[D]
see eqs 6 and 7) and k (see eq 3) are presented in Tables 3
and 4. A log-log plot of K versus k occurs in Figure 5.
0 1
in Figure 4. Data for kobs and k
mainframe computer, and linear plots were obtained from a
29
2
Mathcad program, which provides corresponding values for
1
the intercept, slope, and correlation coefficient for the line.
Deter m in a tion of k-1 was made (as per eq 2) in only one
case, that of the retrogression of 2b to 1b and TCNE (measured
spectrophotometrically as the complex with excess p-xylene)
at 27.1 °C. To a reference cell and to a sample cell were added
Ack n ow led gm en t . We wish to acknowledge help
from Professor Paul Engelking of the University of
Oregon with some mathematical methods used in this
study.
3
2 2
mL of p-xylene and 0.5 mL of ClCH CH Cl and counterbal-
anced for absorbance at 408 nm, the λmax for absorbance by
p-xylene‚TCNE, to obtain a cell correction for the sample cell.
Then into each of several sample cells were placed 3 mL of
p-xylene plus 0.5 mL of a measured quantity of TCNE in
J O9804902
ClCH
2
CH
2
Cl, and the calibration curve of Figure 6 was
(29) Mathcad-3.1, MathSoft, Inc., 201 Broadway, Cambridge, MA
02139.
-
3
-5
determined over a range of [TCNE] of 10 to 10 M.