Experimental survey of the kinetics of acene Diels-Alder reactions

Article abstract of DOI:10.1002/poc.3463

The rate profiles for pseudo-first-order reactions between tetracene and selected dienophiles. The reactivity ranged roughly two orders of magnitude, from the slowest reacting dienophile (2,3-dichloromaleic anhydride, light blue) to the fastest dienophile (N-methylmaleimide, orange). Other utilized dienophiles included, in order of reactivity, maleic anhydride (navy), tetrafluoro-1,4-benzoquinone (purple), and p-benzoquinone (green).

Full text of DOI:10.1002/poc.3463

Research article
Received: 7 November 2014,
Revised: 28 February 2015,
Accepted: 5 May 2015,
Published online in Wiley Online Library: 26 June 2015
(wileyonlinelibrary.com) DOI: 10.1002/poc.3463
Experimental survey of the kinetics of acene
Diels–Alder reactions
Brittni A. Qualizza and Jacob W. Ciszek *
a
a
Second-order rate constants were gathered for solution Diels–Alder reactions of substituted and unsubstituted acenes, with
the intention of ascertaining ideal diene–dienophile combinations. Particular focus was placed on the larger ring systems
namely tetracene, pentacene, and rubrene. The rate constants between the acenes ranged roughly six orders of magnitude,
from the slowest reacting diene, rubrene, to the fastest diene, pentacene. The utilized dienophiles covered a large range
of reactivity from 2,3-dichloromaleic anhydride to tetracyanoethylene. To aid in the interpretation of acene reactivity,
constants were compared to the extensive body of Diels–Alder literature with well-studied dienes such as anthracene and
trans-1-methoxy-1,3-butadiene serving as points of reference. Complex reaction kinetics for the addition of MeTAD and
rubrene was found: initial fast consumption generated an intermediate, followed by dramatically slower product formation.
The kinetic data creates a foundation for the analysis of prior and future reactions between organic semiconductor acene
materials with volatized dienophiles, a surface functionalization technique for enhancing these electronic materials. Copyright ©
2015 John Wiley & Sons, Ltd.
Keywords: rubrene; 4-methyl-1,2,4-triazoline-3,5-dione; UV–vis; stopped flow; steric hindrance; rate constants; stereoisomers
and expand the data library of acene/dienophile pairs, but an
even greater priority was placed on the applicability of these
INTRODUCTION
For 85 years the Diels–Alder reaction has provided chemists with
the ability to form C―C bonds and has continuously evolved to
match modern scientists’ needs and interests. These interests
include a remarkably regio-selective reaction within aqueous
molecules to later use for vapor/solid reaction on organic semi-
conductor surfaces. For example, several of the halogenated
dienophiles were picked as they provide ideal diagnostic signals
for surface analysis, like XPS, ToF-SIMS, or mass spectrometry.
We compare these systems with the well-studied trans-1-
methoxy-1,3-butadiene and anthracene, which serves as a point
of reference for systems of interest.
[
1]
molecular hosts exhibiting enzymatic behavior, a versatile
means for linking functionalizable organic molecules directly to
[
2,3]
semiconductor surfaces like silicon,
or acting as the model
reaction in deciphering RNA catalyzed C―C bond formation
[4]
in vitro, with each example providing a new facet for this
classic reaction. One additional example is our recent communi-
cation, where the reaction is applied to the surface of tetracene
EXPERIMENTAL
[
5]
and rubrene single crystals via vapor dosing of the dienophiles.
NMR kinetic measurements
When these organic semiconductors have their surface composi-
tion modified, it alters many of the properties, including device
General kinetic experiments were performed at 85 °C in toluene-d8
and monitored directly via H NMR. Dienes were prepared as a stock
solution of deuterated toluene at a concentration of approximately
1
[6]
performance. From a mechanistic standpoint, this system
demonstrated unusual steric effects; the reaction of one face of
the tetracene crystal is virtually inert, while another face is facile.
The dienophiles’ steric bulk is also expected to play a critical role
for these confined systems. However, analysis of surface data has
been hindered by the relative lack of corresponding solution
kinetic data.
2
mM with the internal standard, 1,2,4,5-tetramethylbenzene, added
at 10 times the concentration of the diene. Neat dienophile was
added directly to the NMR tube (already at 85 °C), with the
concentration at a 10-fold excess relative to the diene, and out-
side of initial hand agitation, no stirring was used, and reactions
were kept in the dark. Consumption of the diene was monitored
as a function of time to determine the pseudo-first order rate
constant, kobs, and these values were periodically checked with
the values for product formation to ensure that starting material
was not being oxidized, degraded, precipitated, etc.
While the rate of anthracene’s reaction has been studied
[
7–10]
extensively with a range of dienophiles
and tetracene/
(with limited
[
11,12]
pentacene has been studied theoretically
[
13]
experimental reports),
we seek to generate an expansive
report to aid in future interpretations of acene systems. Herein,
we report the kinetic data for rubrene, tetracene, and pentacene
with common dienophiles (N-methylmaleimide, fumarodinitrile,
* Correspondence to: Jacob Ciszek, Department of Chemistry and Biochemistry,
Loyola University, Chicago, 1032 W. Sheridan Rd, Chicago, IL 60660, USA.
E-mail: jciszek@luc.edu
4
-methyl-1,2,4-triazoline-3,5-dione, tetracyanoethylene, and maleic
anhydride) and specialized molecules used in surface reactions
2,3-dichloromaleic anhydride, p-benzoquinone, and tetrafluoro-
,4-benzoquinone) (Fig. 1). The molecules selected for these
experiments were chosen to encompass a range of reactivity
(
1
a B. A. Qualizza, J. W. Ciszek
Department of Chemistry and Biochemistry, Loyola University Chicago, 1032
W. Sheridan Rd, Chicago, IL 60660, USA
J. Phys. Org. Chem. 2015, 28 629–634
Copyright © 2015 John Wiley & Sons, Ltd.

B. A. QUALIZZA AND J. W. CISZEK
Figure 1. The reaction pathway of an acene type diene with generic R1–4 substituted dienophiles. Acene systems for these experiments consist of
three or more linearly fused benzene rings: anthracene, tetracene, pentacene, and rubrene. Dienophiles spanned a range of reactivity and consisted
of substituted anhydrides, MeTAD, a substituted maleimide, quinones, and cyano substituted olefins. Structures of acenes and dienophiles are also
listed in Fig. 3
From these general procedures, several nuances must be
reported. The long reaction time for rubrene required that air
free techniques be utilized (glove box and schlenk flask) with
aliquots removed periodically. Similarly, the sensitivity of trans-
facilitate comparison of the kinetics at elevated and room tem-
perature, the reaction between tetracene and N-methylmaleimide
was performed at both temperatures utilizing the respective
instrumentation. Data plots for UV–vis experiments can be found
in the Supporting Information.
1
-methoxy-1,3-butadiene meant that samples were prepared in
the glove box. Additionally, rubrene’s characteristic signals
could not be used to monitor the rate of reaction. As a result,
production formation was used to determine kinetics rather than
starting material consumption.
RESULTS AND DISCUSSION
Determination of second-order rate constants
The extent of reaction was monitored over time for a period of
anywhere from 20 min to >35 days. In the case of reactions
completing in less than 30 min, measurements were initially
taken every 30 s, then every minute after t = 5 min and then
every 2 min after t = 10 min. Transient acquisition took approxi-
mately 16 s (two transients), and triplicate measurements of
starting materials gave integrations within ±3ꢀ. For reactions
reaching completion in more than an hour but less than a day,
measurements were taken every 5 min until t = 30 min, then
every 10–60 min thereafter (~60 s acquisition time, eight
transients). For slower reactions, taking 24 to 72 h, the initial data
point was acquired within the first hour of the reaction initiation
and was acquired thereafter every 4 to 16 h (~103 s acquisition
time, 16 transients). The rubrene kinetic run was taken approxi-
mately every 7 days and displayed nearly equimolar formation
of oxidized rubrene and adduct. In every other instance no
discernible side products were seen in the final spectra, and
when examined, consumed starting material directly correlated
to product formation. Raw NMR data files (FID) are contained
as a separate data file in the Supporting Information.
All reactions, except for the reaction between trans-1-methoxy-1,3-
butadiene with maleic anhydride and rubrene with 4-methyl-1.2.4-
triazoline-3,5-dione, took place cleanly and gave linear ln[acene]
vs. time plots. Reactions were treated as pseudo-first-order, and
known dienophile concentrations allowed determination of the
À1 À1
second-order rate constant, k (M
s ). Rate constants from data
S
collected at 85 °C and room temperature are listed in Tables 1 and
2, respectively (as 1000 × k ). To associate the rate constant
S
obtained at 85 °C to the room temperature measurement, data
on tetracene’s reaction with N-methylmaleimide was collected
at both temperatures. This produced one second-order constant
À1 À1 À1
at 85 °C (1.2 × 10
(1.5 × 10
M s ) and another at room temperature
À2 À1 À1
M s ). Furthermore, the rate constant for the
tetracene/maleic anhydride reaction was measured three times,
producing rate constant values within a 17ꢀ agreement of one
Table 1. Rate constants for all elevated temperature reactions
(85 °C)
À1 À1
Diene
Dienophile
k
S
× 1000 (M s )
UV–vis kinetic measurements
Rubrene
N-Methylmaleimide
0.004
1.4
The final class of reactions is those demonstrating rapid kinetics.
Because of their rate, reactions were performed at room
temperature, at 10 to 100-fold lower concentrations, and with a
UV–vis. These reactions extended anywhere from 1 s to 24 h.
For the fast reactions between tetracene and tetracyanoethylene
or 4-methyl-1,2,4-triazoline-3,5-dione, a stopped-flow UV–vis was
used to acquire the data at a rapid rate. Additionally, for these
reactions, chloroform was used as the solvent because of a
charge transfer complex between tetracyanoethylene and
trans-1-Methoxy-1, N-Methylmaleimide
-butadiene
trans-1-Methoxy-1, Maleic anhydride
-butadiene
3
6.4
3
Anthracene
Anthracene
Tetracene
Tetracene
Tetracene
N-Methylmaleimide
Maleic anhydride
N-Methylmaleimide
Maleic anhydride
Tetrafluoro-1,
1.7
0.21
120
19
5.6
[14]
toluene.
As pentacene has been reported to readily oxidize,
4-benzoquinone
[
15,16]
reaction solutions were degassed before analysis.
As a
Tetracene
Tetracene
Tetracene
ρ-Benzoquinone
Fumarodinitrile
2,3-Dichloromaleic
anhydride
0.91
0.35
0.29
manner of consistency, all other UV–vis samples were prepared
in a similar manner. Diene concentration was monitored by
using absorbance of the lowest energy band, and analysis was
performed in a manner analogous to the NMR experiments. To
wileyonlinelibrary.com/journal/poc
Copyright © 2015 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2015, 28 629–634

EXPERIMENTAL SURVEY OF THE KINETICS OF ACENE DIELS–ALDER REACTIONS
multiple products were present in the reaction mixture having
chemical shifts commensurate of bridgehead protons (4.92–
.68 ppm) suggesting both B and C ring reaction. These
Table 2. Rate constants for all room temperature reactions
20–25 °C)
(
4
À1 À1
components were separated via HPLC, and UV detection was
used to distinguish the anthracene moiety of the B ring isomer
from the naphthalene moieties of the major C ring isomer.
Diene
Dienophile
S
k × 1000 (M s )
Tetracene
Tetracene
Tetracyanoethylene
4-Methyl-1,2,
5 000 000
690 000
1
According to H NMR, the ratio of C to B isomers was nearly
4
-triazoline-3,5-dione
equivalent to the proposed 6:1 ratio (when a 1:1 pentacene:N-
methylmaleimide mixture was heated to 110 °C for several
minutes, an 8.5:1 ratio occurred). Additional experimental details
and results are contained in the supporting information, includ-
ing HPLC chromatograms, NMR integrations, and a UV spectrum
of the two separated components from the reaction mixture.
The data above also allows us to briefly comment on the
applicability of aromaticity surrogates, such as NICS (nucleus-
independent chemical shifts). NICS is one of the most widely
employed indicators of aromaticity and was proposed by
Tetracene
N-Methylmaleimide
15
110
—
Pentacene N-Methylmaleimide
Rubrene
Tetracyanoethylene
À1 À1
Rubrene
4-Methyl-1,2,
4
k1 (M
s
) = 22
À1
-triazoline-3,5-dione k × 1000 (M s ) = 0.019
2
another (standard deviation), and the errors were assumed to
be of similar magnitude for the other reactions. Tetracene,
pentacene, and rubrene generate more than one diastereoisomer,
Schleyer et al. to be used as an assessment of reactivity based
[
17]
[19]
termed endo and exo,
for each reaction ring because of the
on the magnetic properties of a structure.
As a qualitative
asymmetry produced in the adducts. However these isomers are
relatively consistent between acenes (generally about 1:2) and
do not display unusual kinetics, so the two are combined here.
means of examining ring reactivity in polyacene systems, more
negative NICS values are used to predict the location of reaction
(i.e. reaction will occur on the given ring where a more negative
NICS value exists). When examining our structural isomer data,
NICS values (more precisely the NICS(0)π values given by
Established theory of Diels–Alder reactivity
[18]
Schleyer)
having a difference of 3 or more generated only
In this section we aim to introduce much of the experimental or
theoretical precedence of the Diels–Alder on acenes and
highlight how our data contributes to this area. Each subsequent
paragraph within this section first describes consensus in
literature and concludes with our data and analysis. To begin,
unsubstituted acenes are reasonably facile partners in the
Diels–Alder reaction as the fused aromatic core provides several
electron rich dienes, all of which are a viable site for the reaction.
Of these dienes, the centermost ring is primarily attacked during
reaction (Fig. 2); here the loss of aromaticity in the product is less
the central ring isomer. For example, anthracene and tetracene
differ in NICS values between their B and A rings by 3.1. In the
case of pentacene, the difference between C and A is 4.5 (no
detectable amount of A ring isomers were generated in our
experiments). Additionally, small differences in NICS(0)π values
(the difference between the C and B ring in pentacene is only
0.9) generated the aforementioned mixture of structural isomers.
It would be interesting to study a more extended system in the
future, such as hexacene, where the NICS(0)π difference is 1.5.
[
18]
Theory also predicts that each successive linear addition of a
benzene ring to an acene system causes an increase in reactivity,
[
18]
substantial than for reaction at outlying rings.
calculations, of reaction between acenes and acetylene, predict
00ꢀ reactivity on the B rings for anthracene and the B rings
Theoretical
[
13]
a result borne out in previous experiments.
If the observed
1
trend is described using the aromaticity of the central ring,
opposite rates of reaction would be expected; however, such
assessment neglects the nature of the formed product and its
1
of tetracene. When examined experimentally using H NMR
spectra, these systems contain splitting patterns which would
allow us to distinguish these structural isomers. In all instances,
the data supports the adduct formation exclusively on the inner
most ring of anthracene and tetracene.
[
18]
substantial contribution to the reaction pathway and rate.
In
our data the trend along the acenes is similar (Fig. 3): when
N-methylmaleimide reacts with tetracene as opposed to anthra-
cene, the observed rate constant increased by 68 times (Table 1).
This is consistent with theoretical calculations which predict rate
differences of 38 times from anthracene to tetracene with
In contrast to tetracene and anthracene, pentacene should be
the exception, with a mere 6.7 kJ per mole difference between
the activation energies on the C and B rings which corresponds
to a roughly 6:1 ratio of products (when one considers the
[
18]
acetylene,
and is comparable to experimental results on
[
18]
[13]
duplicity of reaction sites for the B ring).
In our experiments,
similar systems.
This trend continued with pentacene as the
reaction with N-methylmaleimide
was seven times faster compared
to the same reaction with tetracene
(Table 2).
Literature precedence also pro-
vides a predicted trend in dienophile
reactivity, one expected to carry over
to our acene systems. The reactants
commonly utilized in Diels–Alder
reactions (tetracyanoethylene, 4-
methyl-1,2,4-triazoline-3,5-dione,
N-methylmaleimide, maleic anhy-
Figure 2. The labeling of the rings in the acenes studied in this work. Acene systems consisted of anthracene,
tetracene, pentacene, and rubrene, respectively, shown above. Substantial difference in reactivity exists dride, and fumarodinitrile; listed in
between rings in the same acene: electronically, A is the least reactive and C is the most
order of reactivity) thus provide a
J. Phys. Org. Chem. 2015, 28 629–634
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

B. A. QUALIZZA AND J. W. CISZEK
Figure 3. Dienophiles and dienes used to test the reactivity of acene class materials in the Diels–Alder reaction and trends in reactivity
reference point which will allow the relative kinetics of the second,
surface specific, group to be referenced to a greater body of litera-
ture. For the established group, the trend in our data matched that
which was anticipated, e.g. tetracyanoethylene displayed the
fastest reaction time, 4-methyl-1,2,4-triazoline-3,5-dione was more
facile than the anhydride, while lower reactivity was observed with
in Fig. 3. Molecules like 2,3-dichloromaleic anhydride, a doubly
chlorinated maleic anhydride derivative, and tetrafluoro-1,4-
benzoquinone, a perfluorinated benzoquinone, are expected to
participate in the Diels–Alder reaction and have sufficient
volatility, making the surface/vapor reaction viable. We seek
not to only understand their base reactivity in comparison to
their unsubstituted versions, but to tease out information on
electronic and steric effects using the slight variance in halogen
electronegativity, halogen size, and ring size.
[
5]
[20]
fumarodinitrile.
Experimental rate constant data reported by
Biermann and Schmidt, for reaction between tetracene and
maleic anhydride (trichlorobenzene, 91.5 °C), differs by just a
factor of five which is easily attributed to the slight difference in
solvent and temperature.
Chemical intuition suggests that, based purely on electronic
effects, electron-withdrawing halogen substituents should
higher reactivity in the Diels–Alder reaction. This is supported
by theory which predicts that these substituents result in a
stabilized LUMOdienophile allowing for a more overlap with the
Of further interest was the impressive similarity in literature
rate constants of N-methylmaleimide and maleic anhydride with
[
21]
dienes spanning reactivity of over a million fold.
It was
[
22]
because of this trend that comparable reactivity between these
dienophiles with the acenes was anticipated. Our results do
deviate from this slightly. When the two are reacted with trans-
HOMOdiene
.
Detailed analysis of almost one hundred diene/
[
11]
dienophile combinations allowed Kiselev and Konovalov
to
develop an empirical model of the energy of the system based
on three thermodynamic factors. The first two are extent of
frontier orbital interaction: the HOMO/LUMO energy difference
and the interatomic distance between the diene’s reactive
1
-methoxy-1,3-butadiene, maleic anhydride occurred 4.5 times
faster than with N-methylmaleimide, notably outside the 0.5–2
[21]
range reported.
This ordering is inverted for the acenes,
where the reaction of maleic anhydride occurred at a rate 0.1
and 0.17 times that of N-methylmaleimide for both anthracene
and tetracene. Although slightly different from reported results,
this trend was not investigated further with our other diene–
dienophile pairings.
carbons (C
coefficients.
1
, C
4
) which have a strong influence on the overlap
In addition they add a factor pertaining to the
[
23]
energy of bond cleavage/formation. Interestingly though, their
discerned relationship (much like chemical intuition) predicts
faster reactivity for halogenated dienophiles, but experimental
evidence (including that same work) shows reduced reaction rates
when the substitution is on the double bond in the dienophile.
In fact, virtually all instances of Diels–Alder reactions show
inhibited reaction for chloro- or bromo-substituted dienophiles,
signifying substantial effects brought on by size of the substitu-
Incorporation of halogens for XPS measurements: electronic
and steric effects
The primary motivation for this work is to provide reference data
for reactions we wish to study on surfaces. Measurement at
surfaces often needs tagged molecules, whereby the addition
of halide(s) is convenient. Typically halogens are not found as
contaminants either inherently in the organic acenes or within
common atmospheric adsorbates, making them an appropriate
group for proving definite reactivity. Furthermore, many halogen
substituted dienophiles are readily available for purchase, and
give us the potential to tag many of the other compounds listed
[
10,24]
ent which counters its ability to withdraw electron density.
In a report by Andrews and Keefer, a single chlorine substitution
caused a decrease in rate of 2–8 times, and the decelerating
effect was considered to be caused by more double bond
character existing in the C―Cl bond of the dienophile, thereby
reducing the C―C double bond character. The exception to this
is fluorinated species, which have been reported to accelerate
[
25]
Diels–Alder reactions.
We too find analogous reactivity in
wileyonlinelibrary.com/journal/poc
Copyright © 2015 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2015, 28 629–634

EXPERIMENTAL SURVEY OF THE KINETICS OF ACENE DIELS–ALDER REACTIONS
the halogenated species: the addition of fluorine to benzoqui-
none sped up the reaction 6-fold, while the addition of chlorine
to maleic anhydride caused a 65-fold decrease in the rate con-
stant. It would be interesting to see if this effect is amplified at
the surface where access to the diene is dramatically more hin-
dered than in solution. This will be reexamined once surface ki-
netic experiments are determined.
Encumbered rubrene substrates and surface viable reactions
Rubrene is one of the most attention-grabbing materials
currently being tested for organic electronics because of its high
carrier mobility and increased device performance when
[6]
functionalized. Interestingly, exhaustive literature searches for
Diels–Alder reactions of rubrene show only one report of an
[
26]
adduct (formed with N-methylmaleimide)
and no reports of
the kinetic aspect of this reaction. This is most likely because of
[
27]
the fact that the promising device behavior is rather recent.
Figure 4. UV–vis stopped flow was used to monitor the reaction
between Rubrene and MeTAD at 432 nm for nearly 17 min. The inset
displays the natural log of the data points in the first 1.2 s of reaction
versus time, with a linear fit used to calculate the pseudo-first-order rate
constant
Overall, our results for rubrene were quite interesting.
Tetracyanoethylene (TCNE), one of the fastest dienophiles in litera-
ture, showed no reactivity towards rubrene at room temperature
24 h). In fact even when heating the reaction to 125 °C for 13 days,
no appreciable signal was found in the region from 6.5 to 3.5 ppm,
[11]
(
[5]
which is normally associated with these adducts. For these
measurements, the smallest resolvable signal in this region
corresponded to 0.5ꢀ of the starting material, and this signal cannot
be conclusively assigned this to product. This is in contrast with the
reaction of N-methylmaleimide (a dramatically slower dienophile)
which was 4ꢀ complete at this time, even at lower temperatures.
The TCNE rubrene reaction contrasts nicely with our other fast
dienophile, 4-methyl-1,2,4-triazoline-3,5-dione (MeTAD), which
although generally slower than tetracyanoethylene in the acene
the adduct protons at 2.9 ppm (corresponding to the N-methyl
protons) and 5.4 ppm (corresponding to the bridgehead
protons) were absent. Formation of an intermediate could be
suggested by the transient peak at 2.6 ppm, which after 40 h
1
no longer persists in the H NMR spectrum but is replaced by
the presence of peaks from the isolated adduct. We report two
À1 À1
rates for this reaction, k and k . The first, k equal to 22 M
s ,
1
2
1
was calculated assuming pseudo-first-order reactivity in the ini-
tial rate (1.2 s of data, Fig. 4 inset) before k becomes significant.
[
28]
À1
Diels–Alder reactions (Table 2),
reacted completely with
The rapid kinetics leads to a standard deviation of 22ꢀ for this
rubrene in ~1-h time. This result is not entirely surprising; the
large phenyl substituents on the central rubrene ring are ex-
pected to make this substrate particularly sensitive to the sterics
of the dienophile. TCNE’s substitution on both ends of the olefin
makes it particularly slow in this regard, while the 1,2 positions of
MeTAD are unsubstituted rendering the reaction more facile.
It was also interesting that in the case of rubrene with MeTAD,
the reaction occurred but did not follow pseudo-first-order
reactivity. Within 40 s of reacting, the rubrene concentration
appeared to decrease by half (from 0.2 mM to ~0.1mM) followed
by dramatically slower kinetics. To confirm the influence of two dis-
tinct rates, the reaction was monitored via stopped flow UV–vis.
Figure 4 shows an absorbance vs. time plot over the course of
value. It was also interesting that the data for k appeared to
2
correspond to zero order reactivity (Fig. 4), and its value was
À5
À1
equal to 1.9 × 10 M s
.
In addition to displaying interesting kinetics, rubrene also
allows us to comment on the relative reactivity of each ring in
tetracene. From a reactivity standpoint, rubrene should behave
as a substituted version of tetracene with phenyl groups on
the reactive B rings (Fig. 2). Based off of assignments of products
in our kinetic run, any reactivity of the B rings in rubrene is well
below the detection limit of the experiment, leaving only the A
rings to participate in cycloaddition. Unsurprisingly, the rate of
rubrene was very slow, taking 5 weeks to generate 10ꢀ product
adduct (exclusively one diastereomer). As previously mentioned,
this reduced rate constant for the A ring is because of a decrease
in the enthalpy of reaction and increase in activation barrier, in
comparison to the B ring. More importantly, the gathered rate
constant data for the reaction between N-methylmaleimide
and rubrene gave us the ability to comment on reactivity of
the hindered diene relative to the reactivity of its unhindered
counterpart, tetracene.
1000 s, where rubrene was monitored at 432 nm. Visibly different
rates of reaction occur, neither of which follow pseudo-first-order
reactivity. Many mechanisms common to Diels–Alder reaction of
TADs could provide justification for a non-pseudo-first-order rate
(
zwitterion À,+ or aziridinium ion, AI, or precomplex forma-
[29,30]
tion),
and all mechanisms imply that the starting material is
consumed rapidly to form intermediates before slowing, as equilib-
rium becomes important. Such kinetics should follow Eqn 1.
Similar electron distribution between the A rings of tetracene
and rubrene allows for predictions on the reactivity of rubrene.
Based on this assumption, we can infer that the A ring of rubrene
should have the similar reactivity to the A ring of tetracene.
Theoretical calculations examining approximate differences in
reaction rate between the A and B rings of tetracene with
acetylene (at 85 °C), predict a 9060-fold difference, in favor of
reactivity at the B ring (calculations are without consideration
(1)
1
To confirm the presence of an intermediate in this reaction H
NMR spectroscopy was utilized. Proton spectra indicated that
within the first 15–20 min of reaction, rubrene was consumed
[
18]
(~50ꢀ consumption), but no product was present. Specifically,
of the prefactor). We would then naively expect the reactivity
J. Phys. Org. Chem. 2015, 28 629–634
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

B. A. QUALIZZA AND J. W. CISZEK
of rubrene to be slower than tetracene by roughly this amount,
as it is electronically similar to tetracene but is sterically forced
to react at the A ring as opposed to the B ring. If substantial
differences in the calculated and theoretical rate constants
between tetracene and rubrene exist, it would suggest substan-
tial steric effects induced by the presence of phenyl rings on the
reactivity of the A ring. Rate constant data supports slight steric
influence by the introduction of phenyl substituents to
tetracene, as rubrene was approximately 30 000 times slower.
These same discussions also play a heavy role in reanalyzing
surface data for tetracene (as well as rubrene). Our surface
ability for further functionalization. Overall this work provides
precedence for a range of surface specific studies.
Acknowledgements
The authors would like to thank NSF CAREER award #1056400 for
supporting this work and Adwoa Kankam and Chengeto
Gwengo for preliminary experiments with these systems.
REFERENCES
[
[
1] M. Yoshizawa, M. Tamura, M. Fujita, Science 2006, 312, 251.
2] R. J. Hamers, S. K. Coulter, M. D. Ellison, J. S. Hovis, D. F. Padowitz,
M. P. Schwartz, C. M. Greenlief, J. John, N. Russell, Acc. Chem. Res.
2000, 33, 617.
[5]
experiments showed incomplete reaction at 24 h between
maleic anhydride and tetracene, a result that seems at odds with
the observation of adduct for the dramatically slower rubrene.
The fact that tetracene and rubrene crystals reacted in such
similar amounts of time would make more sense if the reaction
was occurring on same deactivated A ring in both cases. Indeed
the crowded environment of a crystal surface could lead to
sterically demanding reactions, possibly allowing primarily the
exposure of the A rings. This would not be without precedence;
confinement effects have limited the reactivity in a similar
system. Kwon reported that moving a monolayer immobilized
dienophile from the surface to subsurface reduced reactivity by
a factor of 8 and when recessed by an additional ~ 2 Å the rate
[3] J. E. Moses, A. D. Moorhouse, Chem. Soc. Rev. 2007, 36, 1249.
[4] T. M. Tarasow, S. L. Tarasow, B. E. Eaton, Nature 1997, 389, 54.
[5] B. A. Qualizza, S. Prasad, M. P. Chiarelli, J. W. Ciszek, Chem. Comm.
2
013, 49, 4495.
[6] B. Lee, T. Choi, S.-W. Cheong, V. Podzorov, Adv. Funct. Mater. 2009,
9, 3726.
1
[7] H. Olsen, J. P. Snyder, J. Am. Chem. Soc. 1977, 99, 1524.
[8] K. E. Wise, R. A. Wheeler, J. Phys. Chem. A 1999, 103, 8279.
[9] P. Brown, R. C. Cookson, Tetrahedron 1965, 21, 1977.
[10] L. J. Andrews, R. M. Keefer, J. Am. Chem. Soc. 1955, 77, 6284.
[11] V. D. Kiselev, A. I. Konovalov, J. Phys. Org. Chem. 2009, 22, 466.
[
12] U. Karama, A. A. El-Azhary, A. I. Almansour, A. A. Al-Kahtani, T. M.
Al-Turki, M. H. Jaafar, Molecules 2011, 16, 6741.
[
31]
declined by an additional factor of 3.5.
Thus this explanation
[
[
13] D. Biermann, W. Schmidt, J. Am. Chem. Soc. 1980, 102, 3163.
14] W. Jarzeba, S. Murata, M. Tachiya, Chem. Phys. Lett. 1999, 301, 347.
seems plausible, but obtaining experimental proof is challenging
as this would require assignment of the crystallographic faces of
[15] E. Clar, Polycyclic Hydrocarbons, Academic Press, New York, 1964.
[
32]
[
[
16] P. Coppo, S. G. Yeates, Adv. Mater. 2005, 17, 3001.
17] R. V. Hoffman, Organic Chemistry: An Intermediate Text, second ed.,
John Wiley & Sons, Inc., Hoboken, 2004.
the single crystal surfaces.
These challenging experiments,
along with rate data on the various faces, would allow for
definitive comment.
[
[
18] P. v. R. Schleyer, M. Manoharan, H. Jiao, F. Stahl, Org. Lett. 2001,
3, 3643.
19] G. Portella, J. Poater, M. Solà, J. Phys. Org. Chem. 2005, 18, 785.
CONCLUSIONS
[20] J. Sauer, H. Weist, A. Mielert, Chem. Ber. 1964, 97, 3183.
[
21] D. Tobia, R. Harrison, B. Phillips, T. L. White, M. DiMare, B. Rickborn,
J. Org. Chem. 1993, 58, 6701.
In closing, these data provide an expansive set of Diels–Alder
acene–dienophile pairs and allows for comparison of reactivity
between acene type dienes with surface-relevant dienophiles.
Observed kinetics are consistent with general notions of the
Diels–Alder; both the size and electronic nature of the dienophiles
influenced rate of reaction. It was also observed that the central
ring was primarily attacked for reaction in each system, that the
reactivity of the acenes was influenced by the number of linearly
fused benzene rings (larger being faster), and the kinetics for
rubrene were reported for the first time. Of most importance to
us, the data has given an ability to loosely predict crystal substrate
reactivity for in progress solid state kinetic experiments. Faster
reacting dienophiles, like tetracyanoethylene or tetrafluoro-1,4-
benzoquinone, are good targets for studying crystal reactivity,
and in addition to their speed they have the potential to
provide spectroscopically interesting signals. Other molecules like
N-methylmaleimide and p-benzoquinone are nominally biologi-
cally interesting surface addendums, producing films with the
[
[
[
22] R. Sustmann, Pure Appl. Chem. 1974, 40, 569.
23] J. Sauer, R. Sustmann, Angew. Chem. Int. Ed. Engl. 1980, 19, 779.
24] W. R. Vaughan, K. M. Milton, J. Org. Chem. 1951, 16, 1748.
[25] M. Essers, C. Mück-Lichtenfeld, G. Haufe, J. Org. Chem. 2002,
67, 4715.
[
[
26] J. Rigaudy, N. K. Cuong, J.-Y. Godard, Bull. Soc. Chim. Fr. 1985, 78.
27] V. Podzorov, E. Menard, A. Borissov, V. Kiryukhin, J. A. Rogers, M. E.
Gershenson, Phys. Rev. Lett. 2004, 93, 086602.
[
28] D. R. McMasters, J. Wirz, J. Am. Chem. Soc. 2001, 123, 238.
[29] F. Jensen, C. S. Foote, J. Am. Chem. Soc. 1987, 109, 6376.
[
[
[
30] K. L. Handoo, Y. Lu, V. D. Parker, J. Am. Chem. Soc. 2003, 125, 9381.
31] Y. Kwon, M. Mrksich, J. Am. Chem. Soc. 2002, 124, 806.
32] R. M. Overney, L. Howald, J. Frommer, E. Meyer, H.-J. Güntherodt,
J. Chem. Phys. 1991, 94, 8441.
SUPPORTING INFORMATION
Additional supporting information may be found in the online
version of this article at the publisher’s web-site.
wileyonlinelibrary.com/journal/poc
Copyright © 2015 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2015, 28 629–634