Synthesis, characterization, and computational study of N,N′-bis (2′,4′-dihydroxyphenyl)-1.4-quinonediimine, a hydroxyl-capped three-ring quinonediimine with sterically hindered substituent on outer rings

Article abstract of DOI:10.1016/j.molstruc.2010.11.010

Density functional calculations were performed on N,N′-bis (2′,4′-dihydroxyphenyl)-1.4-quinonediimine, a hydroxyl-terminated quinonediimine with an additional meta-hydroxyl substituent on each outer ring. Results of the calculations were compared to exper

Full text of DOI:10.1016/j.molstruc.2010.11.010

Journal of Molecular Structure 985 (2011) 299–306
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Synthesis, characterization, and computational study of N,N0-bis
(20,40-dihydroxyphenyl)-1.4-quinonediimine, a hydroxyl-capped three-ring
quinonediimine with sterically hindered substituent on outer rings
⇑
Lawrence T. Sein Jr. , Justine M. Pagillo
Department of Chemical and Physical Sciences, Cedar Crest College, Allentown, PA 18104, USA
a r t i c l e i n f o
a b s t r a c t
Density functional calculations were performed on N,N0-bis (20,40-dihydroxyphenyl)-1.4-quinonediimine,
a hydroxyl-terminated quinonediimine with an additional meta-hydroxyl substituent on each outer ring.
Results of the calculations were compared to experimental properties of the herein synthesized com-
pound. The calculated LUMO levels for isomers of the title compound range from ꢀ2.764 to ꢀ3.279 eV.
The calculated electron affinities range from 1.398 to 1.989 eV. Both the LUMO levels and electron affin-
ities are greatest in magnitude for the syn, anti isomer. The HOMO levels, on the other hand, range from
ꢀ5.383 eV (for the anti, anti isomer) to ꢀ5.541 eV (syn, anti). The predicted electronic transitions for the
molecule in ethanol solvent correspond closely to those seen at 580 and 254 nm. The isomers vary widely
in energy, leading to a marked preference for the syn, syn (outer) isomer – a direct result of the steric
effect of the hydroxyl group at the 20 position.
Article history:
Received 19 September 2010
Received in revised form 3 November 2010
Accepted 3 November 2010
Available online 19 November 2010
Keywords:
Resorcinol
DFT
Conducting polymers
Polyaniline
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
A variety of modifications can been made to the outer rings of
the oligomer by replacing aniline or a substituted aniline by any
Conducting polymers derived from polyaniline (PAni) have
been widely investigated [1]. Unfortunately, PAni has low solubil-
ity in most common organic solvents, leading to interest in aniline
oligomers [2,3]. Aniline oligomers that are amino-terminated at
each end share many properties with PAni. They can be easily
interconverted between multiple oxidation states just as can PAni,
and the same names are commonly assigned to these redox states
– leucoemeraldine for the most reduced state, emeraldine for the
intermediate oxidation state, and pernigraniline for the most oxi-
dized state. The oligomers are effective at corrosion control, and
they are most effective when doped by mineral acids when in
the emeraldine form, just as polyaniline is most effective at corro-
sion control when doped with mineral acids in the emeraldine
form.
aromatic compound possessing a sufficiently strong activating
group for electrophilic aromatic substitution and an open para po-
sition for attack by the p-phenylenediamine. A novel quinonedii-
mine (Fig. 1) has been synthesized from resorcinol, the
properties of which are herein investigated. The synthesis utilizes
resorcinol to generate meta-substituted hydroxyl groups on the
outer rings of the oligomer. Previous reports have described
mono-substituted [2i] or ortho-substituted hydroxyl-capped [2j]
quinonediimines. Resorcinols (which feature meta-substituted hy-
droxyl groups) are an interesting class of organic molecule, some
members of which are important natural products [4]. In addition,
resorcinols have found use in synthetic and crystal engineering
applications [5].
The smallest amino-terminated oligomer of aniline that pos-
sesses three distinct oxidation states is the one with three aromatic
rings. This oligomer can be easily synthesized by reaction of p-phe-
nylene diamine with (substituted) anilines, with the p-phenylene
diamine molecule becoming the central quinoid ring, and the
substituted anilines forming the outer benzenoid rings. A substi-
tuted quinonediimine is thus formed, which is analogous to an
amino-capped aniline oligomer in the emeraldine redox state.
2. Experimental methods
All reagents employed were purchased from Sigma–Aldrich and
used without further purification. The resorcinol terminated qui-
nonediimine N,N0-bis (20,40-dihydroxyphenyl)-1,4-quinonediimine
(hereafter resorcinol quinonediimine, RQD) was synthesized [2] as
follows. 1.00 g (0.00925 mol) of 1,4-para–phenylenediamine was
dissolved in 200 ml of 2 M hydrochloric acid using magnetic stir-
ring at room temperature. The solution was then cooled to approx-
imately ꢀ5 °C by immersion of the reaction beaker in an ice/salt
⇑
Corresponding author. Tel.: +1 610 606 4666x3615.
E-mail address: ltsein@cedarcrest.edu (L.T. Sein).
0022-2860/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2010.11.010

300
L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
Fig. 1. Redox states of N,N0-(20,40-dihydroxyphenyl)-1,4-quinonediimine. R1@OH for leucoemeraldine and emeraldine states, and O for pernigraniline.
water solution. 4.22 g (0.0185 mol) of ammonium persulfate were
added slowly. Upon color change of the solution from clear to light
brown, 2.04 g (0.0185 mol) of resorcinol were added slowly. The
stirring continued for 1 h. The solution was allowed to reach room
temperature, and excess 2 M ammonium hydroxide solution was
added with magnetic stirring. The solution was then vacuumed fil-
tered to recover the emeraldine base. The crude amine base prod-
uct (mixed with a considerable remainder of the hydrochloride
salt) was dissolved in absolute ethanol. Excess sodium carbonate
was added with occasional stoppering of the flask and shaking, un-
til there was no further evolution of carbon dioxide from reaction
of the acid with the carbonate. More highly purified amine free
base was removed from this mixture by Soxhlet extraction with
ethanol, which was then evaporated in a fume hood. Ethanol was
used in preference to acetonitrile, the resorcinol quinonediimine
being more soluble in ethanol than in acetonitrile.
ettes on a Beckmann–Coulter DU-600 spectrophotometer. The
solution concentrations were in the range of 10ꢀ5–10ꢀ4 M.
IR spectra were collected on a Thermo Scientific ‘‘Smart Per-
former’’ Attenuated Total Reflectance (ATR) attachment (diamond
crystal) to
a
Nicolet 380 FT-IR. IR N,N0-bis (20,40-dihydroxy-
phenyl)-1,4-quinonediimine (ATR, cmꢀ1): mmax 1078, 800, 968,
1279, 1416, 1497, 1574, 2808, 3026, 3151, 3336. Vibrational fre-
quencies were calculated at B3LYP/6-31Gꢁ on B3LYP/6-31Gꢁ opti-
mized geometries. Frequencies were then scaled using the widely
used 0.9945 scale factor [6].
The melting point was impossible to determine, being greater
than 350 °C.
3. Computational methods
UV–vis spectra were collected in acetonitrile, chloroform,
dimethylsulfoxide, ethanol, or methanol using 1 cm quartz cuv-
Optimized structures for each isomer of RQD in the emeraldine
redox state (Fig. 2) were computed using the 6-31Gꢁ [7] Pople-style

L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
301
energies was the B3LYP hybrid density functional [8] within the
Gaussian 03 suite of programs [9]. Tight convergence criteria were
selected by use of the SCF = Tight keyword option. This option is
standard when diffuse functions are utilized.
Different basis sets were employed in this work for geometry
optimization [6-31Gꢁ] and single point energy [6-311 + G(2d,p)]
calculations, since experience has shown that use of large basis
sets is much less advantageous for geometry optimization than
for energy calculation. Calculated thermochemical properties of a
set of molecules (the ‘‘G2’’ set) for which reliable experimental val-
ues are available have been compared with the experimental val-
ues [11]. These comparisons allow an estimation of the accuracy
of various density functional and basis set combinations. Results
for
B3LYP/6-31Gꢁ
optimization
followed
by
B3LYP/6-
311 + G(2d,p) single point energy calculation had a mean absolute
deviation from experiment of 3.2 kcal molꢀ1, whereas results for
B3LYP/6-311 + G(2d,p) optimization followed by B3LYP/6-
311 + G(2d,p) single point energy calculation had a mean absolute
deviation from experiment of 3.1 kcal molꢀ1 – a minimal improve-
ment in accuracy at best, at a huge increase in computational cost.
Enthalpies were calculated for the lowest energy singlet and
triplet states of each isomer. Triplet energies for the quinonedii-
mine isomers range from 12.71 kcal molꢀ1 for the syn, syn (outer)
isomer to 20.58 kcal molꢀ1 for the syn, syn (inner) isomer. For each
isomer, S0 was lower in energy than T1. For the dihydrochloride
salts, the triplet energies were calculated in range from
6.79 kcal molꢀ1 for the syn, syn (outer) isomer to 17.37 kcal molꢀ1
for the syn, syn (inner) isomer, with the triplets (T1) being again
higher in energy than S0, but lower than S1. The splitting between
S0 and T1 was consistently smaller for the dihydrochloride salts
than for the emeraldine bases, but the same two isomers in each
case exhibited the maxima and minima. S1 energies were predicted
by CIS, TD-DFT, ZINDO/S, and
DSCF.
Electron affinities were computed on the neutral minimum
geometries (thereby giving the vertical electron affinities) using
the 6-311 + G(2d,p) basis and the B3LYP functional as imple-
mented by Gaussian 03 [9] by difference of the computed energies
(single point energy plus zero-point correction) of the neutral mol-
ecule and the anion.
The enthalpy of the gas phase reaction between one mole of
each quinonediimine in the emeraldine oxidation state (EB) and
two moles of hydrogen chloride to give one mole of the emeraldine
dihydrochloride salt (ES2)
EB þ 2HCl ! ES2
ð1Þ
Fig. 2. Calculated molecular structures (from top) of syn-, syn- (inner) – N,N0-(20,40-
dihydroxyphenyl)-1,4-quinonediimine; syn-, syn- (outer) – N,N0-(20,40-dihydroxy-
phenyl)-1,4-quinonediimine; syn-, anti- N,N0-(20,40-dihydroxyphenyl)-1,4-quinone-
diimine; anti-, syn- N,N0-(20,40-dihydroxyphenyl)-1,4-quinonediimine; and anti-,
anti- N,N0-(20,40-dihydroxyphenyl)-1,4-quinonediimine, optimized at B3LYP/6-
31Gꢁ.
was calculated using single point energy calculations at the B3LYP/
6-311 + G(2d,p) level of theory on B3LYP/6-31Gꢁ optimized geome-
tries (Fig. 3). Zero-point energy and enthalpic corrections to
298.15 K were calculated using frequency calculations at B3LYP/6-
31Gꢁ [11b].
Wavelengths of electronic transitions were calculated using the
D
SCF method [12], which has already demonstrated its effective-
Gaussian basis set (double f quality for the valence orbitals, plus
one set of polarization functions on heavy atoms) and Becke’s hy-
brid density functional B3LYP [8] within the Gaussian 03 suite of
programs [9]. All electrons were included in the computations.
The structures were then confirmed as true minima by frequency
calculations at the B3LYP/6-31Gꢁ level, zero imaginary frequencies
being detected. Zero-point energies were computed at B3LYP/6-
31Gꢁ with no scaling factors.
Single point energies for each structure were computed using
the 6-311 + G(2d,p) Pople-style Gaussian basis set [10]. This basis
is triple f quality on the valence orbitals, supplemented by one
set of diffuse functions and two additional d polarization functions
on heavy atoms, and one set of p polarization functions on hydro-
gen. The computational method employed for the single point
ness in predicting electronic transitions for a number of oligoani-
lines [2]. For comparison purposes, the transition energies were
also calculated using the Configuration Interaction, Singles (CIS)
method [13], Time-dependent density functional theory (TD-DFT)
[14], and the ZINDO/S formalism [15] on B3LYP/6-31Gꢁ optimized
geometries. All methods were employed on each structure.
The effects of solvation on molecular structure and the elec-
tronic spectra were assessed by use of the Self-Consistent Isoden-
sity Polarizability Continuum Model (SCIPCM) [16], with a model
acetonitrile solvent (Fig. 4). This solvent was chosen to allow com-
parison with the experimental UV–vis spectra, which were taken in
acetonitrile. The quinonediimines were geometry optimized using
B3LYP/6-31Gꢁ chemistry in the model solvent (acetonitrile), and
the electronic spectra predicted using the
DSCF method with the

302
L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
Center. Visualizations of molecular structures and orbitals were
generated by ChemCraft [17].
4. Results and discussion
Calculations (Table 1) find the syn, syn (outer) isomer of RQD to
be lowest in enthalpy. Unlike other quinonediimines investigated
[2], the differences in enthalpies in the gas phase among isomers
is very large, up to nearly 15 kcal molꢀ1. In acetonitrile solvent,
the calculated enthalpy differences are somewhat smaller, still
favoring the syn, syn (outer) structure. However, only two struc-
tures could be successfully calculated. The SCIPCM algorithm has
difficulties with irregularly shaped molecules. The presence of
the hydroxyl group at the 20 position causes the solvent cavity cre-
ated by the method to be sufficiently ill-shaped to cause the meth-
od to abort. Assuming the isomers are Boltzmann distributed, the
population of RQD in the gas phase at 298.15 K should be more
than 99.99% syn, syn outer (sso), with syn, syn inner (ssi); syn, anti
(sa); anti, syn (as); and anti, anti (aa) each less than 0.01% each.
The non-linearity and non-planarity of the three rings in the
quinonediimine lead to important structural effects. If there are
no substituents at the 20, 30, 50, and 60 positions, there are only
two isomers, one with the rings on the same side of the central
quinoid ring (syn), or with the rings on opposite sides (anti). The
syn conformer belongs to point group C2v, and the anti isomer be-
longs to Ci. With a substituent at the 20 position (as in RQD), there
are five isomers. The ssi and sso isomers belong to the point group
C2v, whereas the aa isomer belongs to Ci. The two remaining iso-
mers belong to group C1.
Enthalpies of activation for isomerization of RQD were calcu-
lated at the B3LYP/6-311 + G(2d,p) level after identifying the tran-
sition state for inversion at the imine nitrogen (transition state
optimized and zero-point corrected to 298.15 K at B3LYP/6-
31Gꢁ). The value (17.365 kcal molꢀ1) is less for the enthalpy of acti-
vation for isomerization calculated for the unsubstituted EB
(19.622 kcal molꢀ1). Similarly, enthalpies of activation for rotation
of RQD between syn and anti isomers (relative to the hydroxyl
group nearest the imine nitrogen, the 20 position) were calculated
after identifying the transition state for rotation of the outer amine
rings (same level of theory as for isomerization). The value
(0.741 kcal molꢀ1) is again less than the enthalpy of activation
for rotation calculated for the unsubstituted EB (1.876 kcal molꢀ1).
Calculations reveal characteristic structural effects of the hydro-
xyl group at the 20 position. In the analogous phenol quinonedii-
mine without
a
carboxyl substitution, i.e., N,N0-bis (40-
hydroxyphenyl)-1,4-quinonediimine [‘‘phenol quinonediimine’’],
the tilt of the outer benzenoid rings with respect to the inner quin-
oid ring (CACAN@C) is 43.82° for the syn isomer, and 43.03° for
the anti; for the corresponding dihydrochloride salt (ES2) without
the hydroxyl group at the 20 position, the ring tilt is the somewhat
smaller 36.08° for the syn isomer, and 33.22° for the anti. The
greater conductivity of the doped emeraldine salt forms relative
to the emeraldine base is often attributed in part to their greater
(near) planarity, and we see this pattern for the phenol emeraldine.
For the RQD, the ring tilts are calculated at 50.29° (ssi), 31.11°
(sso), 48.94° (sa), 49.67° (as), and 29.85° (aa). In other words, the
presence of the hydroxyl group at the 20 position on each outer ring
has a very large effect on the calculated ground state geometry in
the gas phase. The effect is greatest in precisely those isomers
where the hydroxyl group is in the greatest proximity to the imine
nitrogen atom. Possible geometric/steric effects of the hydroxyl on
neighboring chains in the solid state were not addressed.
Fig. 3. Calculated molecular structures (from top) of syn-, syn- (inner) – N,N0-(20,40-
dihydroxyphenyl)-1,4-quinonediimine dihydrochloride; syn-, syn- (outer) – N,N0-
(20,40-dihydroxyphenyl)-1,4-quinonediimine dihydrochloride; syn-, anti- N,N0-
(20,40-dihydroxyphenyl)-1,4-quinonediimine dihydrochloride; anti-, syn- N,N0-
(20,40-dihydroxyphenyl)-1,4-quinonediimine dihydrochloride; and anti-, anti-N,N0-
(20,40-dihydroxyphenyl)-1,4-quinonediimine dihydrochloride, optimized at B3LYP/
6-31Gꢁ.
orbital energies generated by a single-point calculation at B3LYP/6-
311 + G(2d,p) on the B3LYP/6-31Gꢁ solvent optimized geometries.
The computations were performed on the National Science
Foundation HP GS1280 system at the Pittsburgh Supercomputing
Calculated ring tilts for the resorcinol emeraldine salt (RES2) are
much large than those of the unsubstituted phenol quinonediimine
– 44.66° (ssi), 36.66° (sso), 44.29° (sa), 44.13° (as), 43.78° (aa). If we

L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
303
Fig. 4. Calculated molecular structures (from top) of syn-, syn- (outer) – N,N0-(20,40-dihydroxyphenyl)-1,4-quinonediimine dihydrochloride and anti-, syn- N,N0-(20,40-
dihydroxyphenyl)-1,4-quinonediimine dihydrochloride in acetonitrile solvent, optimized at B3LYP/6-31Gꢁ using the SCIPCM method.
Table 1
consider a ring tilt of 36.08° to be typical for hydroxyl-terminated
quinonediimines (i.e., those with an hydroxyl group at the 40 posi-
Relative enthalpies of RQD (with respect to lowest energy isomer) calculated at
B3LYP/6-311 + G(2d,p) on B3LYP/6-31Gꢁ optimized geometries, with zero-point and
tion, but no other ring substitutions, as in phenol quinonediimine),
then the steric effect of the hydroxyl substituent at the 20 position
accounts for approximately 8° of the tilt.
For the RES2 dihydrochlorides (Table 2) in the gas phase, calcu-
lated enthalpies show the sso isomer as most stable. Next highest
are the sa and aa isomers. The ssi and as isomers are very much
enthalpic corrections to 298.15 K at B3LYP/6–31Gꢁ. Solvent effects in acetonitrile
modeled using SCIPCM method.
Gas phase (kcal molꢀ1
)
Solvent (kcal molꢀ1
)
Syn, syn (inner)
Syn, syn (outer)
Syn, anti
Anti, syn
Anti, anti
14.247
0.000
6.225
14.847
8.156
N/A
0.000
N/A
10.041
N/A
Table 4
Calculated and experimental ATR-IR vibrational frequencies of syn, syn (outer) RQD
on B3LYP/6-31Gꢁ optimized geometries. Calculated frequencies were scaled by
0.9945 [6].
Table 2
Enthalpies of RQD dihydrochlorides calculated at B3LYP/6-
311 + G(2d,p) on B3LYP/6-31Gꢁ optimized geometries, with
zero-point and enthalpic corrections to 298.15 K at B3LYP/6-
31Gꢁ, relative to lowest enthalpy syn, syn (outer) isomer.
Calculated
Experimental
Assignment
(cmꢀ1
)
(cmꢀ1
)
806
800
1078
1279
1416
1497
1574
2808
3026
Quinoid stretch
CAH wagging
1125
1302
1420
1523
1596
N/A
H (kcal molꢀ1
)
CAH wagging, 20-OH wagging
CAH wagging on benzenoid rings
C@N symmetric stretch
C@N asymmetric stretch
CAH stretch
D
Syn, syn (inner)
Syn, syn (outer)
Syn, anti
Anti, syn
Anti, anti
18.226
0.000
10.086
18.408
10.467
3178
CAH stretch for between OH
groups
3217
3461
3151
3336
CAH stretch
20-OH stretch
Table 3
Enthalpies of RQD in the pernigraniline redox state calculated
at B3LYP/6-311 + G(2d,p) on B3LYP/6-31Gꢁ optimized geom-
etries, with zero-point and enthalpic corrections to 298.15 K at
B3LYP/6-31Gꢁ, relative to anti, anti isomer.
Table 5
Vertical electron affinities of RQD calculated at B3LYP/6-311 + G(2d,p) on B3LYP/6-
31Gꢁ optimized geometries, with zero-point and enthalpic corrections to 298.15 K at
B3LYP/6-31Gꢁ.
D
H (kcal molꢀ1
)
Base (eV)
Dihydrochloride (eV)
Syn, syn (inner)
Syn, syn (outer)
Syn, anti
Anti, syn
Anti, anti
17.691
0.000
7.232
15.203
6.430
Syn, syn (inner)
Syn, syn (outer)
Syn, anti
Anti, syn
Anti, anti
1.398
1.989
1.971
1.425
1.758
3.298
3.697
3.713
3.324
3.546

304
L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
Table 6
HOMO/LUMO levels of RQD calculated at B3LYP/6-311 + G(2d,p) on B3LYP/6-31Gꢁ optimized geometries, with zero-point and enthalpic corrections to 298.15 K at B3LYP/6-31Gꢁ.
HOMO base (eV)
HOMO dihydrochloride (eV)
LUMO base (eV)
LUMO dihydrochloride (eV)
Syn, syn (inner)
Syn, syn (outer)
Syn, anti
Anti, syn
Anti, anti
ꢀ5.466
ꢀ5.402
ꢀ5.541
ꢀ5.389
ꢀ5.383
ꢀ6.221
ꢀ6.512
ꢀ6.452
ꢀ6.142
ꢀ6.212
ꢀ2.764
ꢀ3.265
ꢀ3.279
ꢀ2.774
ꢀ3.057
ꢀ4.545
ꢀ4.911
ꢀ4.930
ꢀ4.556
ꢀ4.760
Table 7
Predicted and experimental electronic transitions of various isomers of RQD in nanometers. Since it is impossible to determine the relative populations of the isomers in solution,
the same ‘‘experimental’’ wavelengths are listed for each isomer.
Isomer
Experimental
Transition
D
SCF
TD-DFT
CIS
ZINDO/S
Syn, syn (inner)
580
N/A
254
H ? L
459
408
253
544
329
282
476
453
H ꢀ 1 ? L
H ? L + 1
239
Syn, syn (outer)
Syn, anti
580
N/A
254
H ? L
580
410
276
566
404
295
541
458, 377
H ꢀ 1 ? L
396
237
H ? L + 1
580
N/A
254
H ? L
548
434
266
613
500
282
519
466, 367
H ꢀ 1 ? L
H ? L + 1
Anti, syn
580
N/A
254
H ? L
474
405
261
552
328
281
487
443
H ꢀ 1 ? L
377
240
H ? L + 1
Anti, anti
580
N/A
254
H ? L
533
410
271
567
482
336
520
448
H ꢀ 1 ? L
370
240
H ? L + 1
the highest in enthalpy. The enthalpies of reaction of RQD with two
moles of hydrogen chloride to yield the corresponding dihydro-
chloride salt were computed. For each isomer, the reaction was
clearly exothermic. The most favorable (lowest) enthalpy of reac-
tion was found for the sso isomer, at ꢀ14.55 kcal molꢀ1. The aa iso-
mer had the next most favorable enthalpy of reaction at
ꢀ12.24 kcal molꢀ1, with the remaining three isomers clustered
around ꢀ10.5 kcal molꢀ1. The equivalent reaction for phenol qui-
nonediimines has enthalpy of reaction of ꢀ13.85 kcal molꢀ1 (syn)
and ꢀ13.72 kcal molꢀ1 (anti). For the 30,40-dihydroxy quinonedii-
mine (the one with ortho hydroxyl groups on the outer rings),
the enthalpies of reaction range from ꢀ9.71 kcal molꢀ1 to
ꢀ13.07 kcal molꢀ1. The favorable enthalpy for reaction for the sso
isomer of RQD with HCl can be attributed to two factors. First,
the sso isomer both the base and dihydrochloride salts of RQD
has its hydroxyl substituent (20 position) optimally located to min-
Fig. 5. UV–vis spectra of N,N0-(20,40-dihydroxyphenyl)-1,4-quinonediimine in acetonitrile.

L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
305
imize steric hindrance with the inner quinoid ring. Secondly, the
dihydrochloride salt is partially stabilized by a hydrogen-bonding
interaction between the hydroxyl group and the hydrochloride.
Like other quinonediimines, RQD can be easily oxidized to the
pernigraniline oxidation state by addition of concentrated hydro-
gen peroxide. The calculated isomeric preferences in the gas phase
for this state are exactly the same as for the emeraldine base. The
sso is the lowest enthalpy isomer, followed by the sa and aa rela-
tively close to each other, with ssi and as very much higher in en-
ergy, exactly as for the RQD and the RQD dihydrochloride (Table 3).
(and therefore, corrosion inhibition) is the ‘‘pulling down’’ of the
LUMO.
Unsubstituted phenol quinonediimine dihydrochlorides have
calculated HOMO’s at ꢀ6.048 eV (syn isomer), or ꢀ6.378 eV (anti).
For the RQD dihydrochlorides, the calculated HOMO values range
from ꢀ6.452 eV (for the sa isomer) to ꢀ6.142 (as). The additional
hydroxyl groups on the outer rings lower the HOMO, just as they
lower the LUMO.
For RQD, predicted electronic transitions by several methods
are listed in Table 7, with the experimental spectrum in Fig. 5.
The longest wavelength absorption is predicted exactly by the
D
SCF method for the sso isomer, which is expected to be the pre-
dominate isomer. The next predicted absorption is not seen in
solution. The third longest wavelength predicted by SCF is in
4.1. IR spectra
D
The compound has a benzenoid stretch at 1497 cmꢀ1, and a
quinoid stretch at 1574 cmꢀ1. The presence of two, symmetry dis-
tinct hydroxyl groups at each end of the molecule leads to a very
broad absorption from about 2800 to 3400 cmꢀ1. Assignments in
this region are also complicated by intermolecular hydrogen bond-
ing (Table 4).
good agreement with the actual experimental result. TD-DFT gives
reasonably good predictions of the longest and shortest wave-
lengths, ZINDO/S gives poorer results, and CIS does not even pre-
dict a band longer than 396 nm.
For the unsubstituted aniline quinonediimine, the primary elec-
tronic transitions observed are H (HOMO) ? L (LUMO, 544 nm)
4.2. Electronic spectroscopy
Vertical electron affinities (EA) are a key index of potential util-
ity in corrosion inhibition applications (Table 5). Unsubstituted
phenol quinonediimines have EA’s of 1.615 (syn) and 1.434 (anti)
eV. For the corresponding resorcinol quinonediimines (RQD), the
EA’s range from 1.398 eV for the ssi isomer, to 1.989 eV for the
sso isomer. The range of values among the isomers is noticeably
greater than for other quinonediimines studied. The trend in elec-
tron affinity appears to be inversely related to the enthalpy of the
isomer – the more enthalpically stable the isomer, the greater the
electron affinity. The hydroxyl substituent at the 20 position dis-
torts the planarity of the molecule, and thereby affects the
electronics.
The greater EA of RQD (for those isomers for which the EA is
greater) results from the electron withdrawing effect of the hydro-
xyl groups by resonance. The effects of the electronegativity of the
oxygen atoms of the 20-hydroxyl group on the electron affinity
should be essentially independent of conformation, leaving reso-
nance as the primary cause of the increased EA.
The doping of aniline oligomers generally leads to a very sub-
stantial increase in the electron affinity, attributable to the strong
electron withdrawing effects of hydrochloride, primarily by induc-
tion (the electronegativity of chlorine). The calculated electron
affinities of the RQD dihydrochlorides vary from 3.298 eV (ssi) to
3.713 for the sa isomer. This corresponds to a range of 0.415 eV
among the dihydrochloride isomers. For the related phenol qui-
nonediimine dihydrochlorides, they are 3.022 eV (syn), 3.552 eV
(anti), a range of 0.53 eV. Clearly, though the steric hindrance of
the hydroxyl group has a substantial effect on the EA of the resor-
cinol quinonediimine, it has comparatively little effect on the EA of
the corresponding dihydrochloride salts.
The trend in electron affinities for the RQD dihydrochlorides is
unusual. It does not inversely parallel the trend in enthalpies, as
do the electron affinities of the resorcinol quinonediimine bases.
The phenol quinonediimine dihydrochloride has calculated LU-
MO’s at ꢀ4.278 eV (syn isomer), or ꢀ4.768 eV (anti). The calculated
LUMO levels (Table 6) for the RQD dihydrochlorides range from
ꢀ4.545 (ssi isomer) to ꢀ4.930 eV (for the sa isomer). The lower
LUMO levels of the RQD dihydrochlorides can be attributed to
the slight additional withdrawal of electron density by the two
additional 20 position hydroxyl groups, largely though inductive ef-
fect of the oxygen atoms. The electronic affinities of the various
RQD isomers closely parallel the calculated LUMO levels, suggest-
ing that the most significant determinant of electron affinities
Fig. 6. Calculated frontier orbitals of the syn-, syn- (outer) N,N0-(20,40-dihydroxy-
phenyl)-1,4-quinonediimine at B3LYP/6-311 + G(2d,p) on the B3LYP/6-31Gꢁ opti-
mized structure: LUMO + 1 (top), LUMO (upper middle), HOMO (lower middle), and
HOMO ꢀ 1.

306
L.T. Sein Jr., J.M. Pagillo / Journal of Molecular Structure 985 (2011) 299–306
2555–2564;
and H ? L + 1 (second lowest unoccupied molecular orbital,
277 nm). Oxygenated amino-terminated quinonediimines, on the
other hand, show primarily a H ꢀ 1 (second highest occupied
molecular orbital) ? L transition. RQD acts more like the unsubsti-
tuted amino-terminated quinonediimine than the hydroxyl-termi-
nated one, even though both RQD and the hydroxyl-terminated
quinonediimine are ‘‘oxygenated’’. The effect of amino and hydro-
xyl substitutions on the character of the most intense electronic
transitions is very subtle, as seen by comparison among such tri-
mers with the substituents slightly rearranged. The frontier orbi-
tals of RQD are displayed in Fig. 6.
(e) M.T. Greiner, M. Festin, P. Kruse, J. Phys. Chem. C 112 (2008) 18991–19004;
(f) Y. Guo, M. Li, A. Mylonakis, J. Han, A.G. MacDiarmid, X. Chen, P.I. Lelkes, Y.
Wei, Biomacromolecules 8 (2007) 3025–3034.
[4] (a) F.E. Dayan, A.M. Rimando, Z. Pan, S.R. Baerson, A.L. Gimsing, S.O. Duke,
Phytochemistry 71 (2010) 1032–1039;
(b) M. Matsuzawa, Y. Katsuyama, N. Funa, S. Horinouchi, Phytochemistry 71
(2010) 1059–1067.
[5] (a) M. Khan, V. Enkelmann, G. Brunklaus, Cryst. Growth Des. 9 (2009) 2354–
2362;
(b) T. Friscic, T.D. Hamilton, G.S. Papaefstathiou, L.R. MacGillivray, J. Chem. Ed.
82 (2005) 1679;
(c) R.L. Halterman, D.E. Martyn, J. Org. Chem. 72 (2007) 7841–7848;
(d) A.N. Sokolov, T. Frišic, L.R. MacGillivray, J. Am. Chem. Soc. 128 (2006) 2806–
2807.
´
[6] A.P. Scott, L. Radom, J. Phys. Chem. 100 (1996) 16502–16513.
[7] (a) R. Ditchfield, W.J. Hehre, J.A. Pople, J. Chem. Phys. 54 (1971) 724;
(b) W.J. Hehre, R. Ditchfield, J.A. Pople, J. Chem. Phys. 56 (1972)
2257;
(c) P.C. Hariharan, J.A. Pople, Theor. Chem. Acc. 28 (1973) 213–222.
[8] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
5. Conclusions
The resorcinol quinonediimine has been synthesized, and its
experimental and theoretical investigation enables a deeper under-
standing of subtle electronic and steric effects in amino- and hy-
droxyl-terminated quinonediimines. The hydroxyl group at the 20
position exerts two distinct effects on the compound. On one hand,
it reinforces the electron withdrawing effect of the hydroxyl group
at the 40 position by electronegativity, while it counteracts its with-
drawal of electron density by resonance. On the other hand, it
destabilizes any isomers in which it closely approaches the imine
nitrogen, so that if the isomers are Boltzmann distributed, at room
temperature all molecules should be in the enthalpically favored sa
isomer. This suggests that structural factors could be used to more
efficiently engineer quinonediimines with desired electronic
properties.
[9] Gaussian 03, Revision C.02, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria,
M.A. Robb, J.R. Cheeseman, J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C.
Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G.
Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J.
Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas,
D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G.
Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong,
C. Gonzalez, J.A. Pople, Gaussian, Inc., Wallingford, CT, 2004.
[10] (a) A.D. McLean, G.S. Chandler, J. Chem. Phys. 72 (1980) 5639–5648;
(b) K. Raghavachari, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem. Phys. 72 (1980)
650–654.
[11] (a) E.R. Davidson, D. Feller, Chem. Rev. 86 (1986) 681–696l;
(b) Exploring Chemistry with Electronic Structure Methods: A Guide to Using
Gaussian, second ed. J.B. Foresman, Æ. Frisch, Gaussian, Pittsburgh, 1996.;
(c) J.-W. Pan, D.W. Rogers, F.J. McLafferty, J. Mol. Struct. (Theochem) 1–2
(1999) 59–66.
[12] U. Salzner, P.G. Pickup, R.A. Poirier, J.B. Lagowski, J. Phys. Chem. A 102 (1998)
2572.
[13] J.B. Foresman, M. Head-Gordon, J.A. Pople, M.J. Frisch, J. Phys. Chem. 96 (1992)
135.
References
[1] (a) W. Li, H.-L. Wang, J. Am. Chem. Soc. 126 (2004) 2278–2279;
(b) X.-G. Li, M.-R. Huang, W.D.Y.-L. Yang, Chem. Rev. 102 (2002) 2925–3030;
(c) Y. Wei, J. Chem. Ed. 78 (2001) 551;
´
ˇ
´
(d) M. Trchová, I. Šedenková, E.N. Konyushenko, J. Stejskal, P. Holler, G. Ciric-
Marjanovic´, J. Phys. Chem. B 110 (2006) 9461–9468;
(e) H. Sakurai, M. T.S. Ritonga, H. Shibatani, T. Hirao, J. Org. Chem. 70 (2005)
2754–2762;
[14] (a) R. Bauernschmitt, R. Ahlrichs, Chem. Phys. Lett. 256 (1996) 454–464;
(b) M.E. Casida, C. Jamorski, K.C. Casida, D.R. Salahub, J. Chem. Phys. 108 (1998)
4439–4449;
(f) J.P. Foreman, A.P. Monkman, J. Phys. Chem. A 107 (2003) 7604–7610.
[2] (a) S.A. Jansen, T. Duong, A. Major, Y. Wei, L.T. Sein Jr., Synth. Met. 105 (1999)
107;
(c) R.E. Stratmann, G.E. Scuseria, M.J. Frisch, J. Chem. Phys. 109 (1998) 8218–
82124;
(b) L.T. Sein Jr., Y. Wei, S.A. Jansen, Synth. Met. 108 (2000) 101;
(c) L.T. Sein Jr., Y. Wei, S.A. Jansen, J. Phys. Chem. A 104 (2000) 11371;
(d) L.T. Sein Jr., T. Duong, Y. Wei, S.A. Jansen, Synth. Met. 113 (2000) 145;
(e) L.T. Sein Jr., Y. Wei, S.A. Jansen, Compos. Theor. Polym. Sci. 11 (2001) 83;
(f) L.T. Sein Jr., Y. Wei, S.A. Jansen, Synth. Met. 126 (2002) 117;
(g) L.T. Sein Jr., Y. Wei, S.A. Jansen, J. Mol. Struct. (Theochem) 577 (2002) 35;
(h) L.T. Sein Jr., Y. Wei, S.A. Jansen, Synth. Met. 143 (2004) 1;
(i) L.T. Sein Jr., J. Phys. Chem. A 112 (2008) 2598;
(d) C. Van Caillie, R.D. Amos, Chem. Phys. Lett. 308 (1999) 249–255;
(e) C. Van Caillie, R.D. Amos, Chem. Phys. Lett. 317 (2000) 159–164;
(f) F. Furche, R. Ahlrichs, J. Chem. Phys. 117 (2002) 7433–7447;
(g) G. Scalmani, M.J. Frisch, B. Mennucci, J. Tomasi, R. Cammi, V. Barone, J.
Chem. Phys. 124 (2006) 094107: 1–094107: 15.
[15] (a) J.E. Ridley, M.C. Zerner, Theo. Chim. Acta 32 (1973) 111;
(b) A.D. Bacon, M.C. Zerner, Theo. Chim. Acta 53 (1979) 21;
(c) M.C. Zerner, G.H. Lowe, R.F. Kirchner, U.T. Mueller-Westerhoff, J. Am. Chem.
Soc. 102 (1980) 589;
(d) W.P. Anderson, W.D. Edwards, M.C. Zerner, Inorganic Chem. 25 (1986)
2728.
[16] J.B. Foresman, T.A. Keith, K.B. Wiberg, J. Snoonian, M.J. Frisch, J. Phys. Chem.
100 (1996) 16098.
(j) L.T. Sein Jr., A.F. Lashua, Synth. Met. 159 (2009) 1183–1190;
(k) L.T. Sein Jr., A.F. Lashua, J. Mol. Struct. 977 (2010) 220–229.
[3] (a) M.M. Wienk, R.A.J. Janssen, J. Am. Chem. Soc. 119 (1997) 4492–4501;
(b) P.L. Franzen, L. De Boni, D.S. Santos Jr., C.R. Mendonça, S.C. Zílio, J. Phys.
Chem. B 108 (2004) 19180–19183;
(c) A.K. Flatt, B. Chen, P.G. Taylor, M. Chen, J.M. Tour, Chem. Mater. 18 (2006)
4513–4518;
(d) A.N. Ivanova, A.V. Tadjer, N.P. Gospodinova, J. Phys. Chem. B 110 (2006)
[17] <http://www.chemcraftprog.com>.