Photoinduced proton transfer promoted by peripheral subunits for some Hantzsch esters

Article abstract of DOI:10.1021/jp5078246

It is noted that, for a small series of 3,5-diacetyl-1,4-dihydrolutidine (DDL) derivatives and the corresponding Hantzsch esters, the presence of methyl groups at the 2,6-positions serves to extinguish fluorescence in solution but not in the solid state. Emission is weakly activated and affected by changes in solvent polarity. The latter situation arises because the optical transition involves intramolecular charge transfer. Calculations, both semiempirical and DFT, indicate that, in all cases, rotation of the carbonyl function is facile and that the dihydropyridine ring is planar. These calculations also indicate that the 2,6-methyl groups do not affect the generic structure of the molecule. It is proposed that illumination increases the molecular dipole moment and pushes electron density toward the carbonyl oxygen atom. Proton transfer can now occur from one of the methyl groups, leading to formation of a relatively low-energy, neutral intermediate, followed by a second proton transfer step that forms the enol. Reaction profiles computed for the ground-state species indicate that this route is highly favored relative to hydrogen transfer from the 4-position. The barriers for light-induced proton transfer are greatly reduced relative to the ground-state process but such large-scale structural transformations are hindered in the solid state. A rigid analogue that cannot form an enol is highly emissive in solution, supporting the conclusion that proton transfer is in competition to fluorescence in solution. (Figure Presented).

Full text of DOI:10.1021/jp5078246

Subscriber access provided by Georgia Tech Library
Article
Photo-Induced Proton Transfer Promoted by
Peripheral Subunits for Some Hantzsch Esters
Sébastien Azizi, Gilles Ulrich, Maud Guglielmino, Stéphane Le
Calvé, Jerry Hagon, Anthony Harriman, and Raymond Ziessel
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/jp5078246 • Publication Date (Web): 04 Dec 2014
Downloaded from http://pubs.acs.org on December 10, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry A is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 32
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Photoꢀinduced Proton Transfer Promoted by Peripheral
Subunits for Some Hantzsch Esters
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Sébastien Azizi,a Gilles Ulrich,a Maud Guglielmino,b Stéphane le Calvé,b Jerry P. Hagon,c Anthony
Harriman,*c and Raymond Ziessel*a
(a) Institut de Chimie et Procédés pour l’Energie, l’Environnement et la Santé (ICPEES), Laboratoire de
Chimie Moléculaire et Spectroscopies Avancées (ICPEESꢀLCOSA), UMR 7515 au CNRS, Ecole
Européenne de Chimie, Polymères et Matériaux, 25 rue Becquerel, 67087 Strasbourg Cedex 02, France.
(b) ICPEES ꢀ UMR 7515 CNRS ꢀ UdS. Institut de Chimie, Procédés, Catalyse, Energie, Environnement
et Santé, Physicoꢀchimie de l'atmosphère, 1 rue Blessig, 67084 Strasbourg Cedex 02, France.
(c) Molecular Photonics Laboratory, School of Chemistry, Bedson Building, Newcastle University,
Newcastle upon Tyne, NE1 7RU, UK
AUTHOR EMAIL ADDRESS: anthony.harriman@ncl.ac.uk
RECEIVED DATE
TITLE RUNNING HEAD: Longꢀrange Proton Transfer
CORRESPONDING AUTHOR FOOTNOTE: anthony.harriman@ncl.ac.uk: ziessel@unistra.fr
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 2 of 32
ABSTRACT: It is noted that, for a small series of 3,5ꢀdiacetylꢀ1,4ꢀdihydrolutidine (DDL) derivatives
and the corresponding Hantzsch esters, the presence of methyl groups at the 2,6ꢀpositions serves to
extinguish fluorescence in solution but not in the solid state. Emission is weakly activated and affected
by changes in solvent polarity. The latter situation arises because the optical transition involves
intramolecular charge transfer. Calculations, both semiꢀempirical and DFT, indicate that, in all cases,
rotation of the carbonyl function is facile and that the dihydropyridine ring is planar. These calculations
also indicate that the 2,6ꢀmethyl groups do not affect the generic structure of the molecule. It is proposed
that illumination increases the molecular dipole moment and pushes electron density towards the
carbonyl oxygen atom. Proton transfer can now occur from one of the methyl groups, leading to
formation of a relatively lowꢀenergy, neutral intermediate, followed by a second proton transfer step that
forms the enol. Reaction profiles computed for the groundꢀstate species indicate that this route is highly
favored relative to hydrogen transfer from the 4ꢀposition. The barriers for lightꢀinduced proton transfer
are greatly reduced relative to the groundꢀstate process but such largeꢀscale structural transformations
are hindered in the solid state. A rigid analogue that cannot form an enol is highly emissive in solution,
supporting the conclusion that proton transfer is in competition to fluorescence in solution.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: Hantzsch esters / fluorescence / transition state / hydrogen transfer / ketoꢀenol
INTRODUCTION
Formaldehyde is one of the more harmful air pollutants1,2 due to its relatively high atmospheric
concentration in urban environments and its deleterious impact on human health. Indeed, it has been
classified as carcinogenic by IARC.3 Such concerns mean that there is a need for reliable but
straightforward analytical protocols to enable the routine monitoring of formaldehyde vapors; in
practical terms, such analyses are most conveniently performed after extraction of gaseous formaldehyde
into water.1,4ꢀ7 In seeking to develop sensitive ways to monitor the presence of formaldehyde in aqueous
solution, we were drawn to the in-situ formation of fluorescent 3,5ꢀdiacetylꢀ1,4ꢀdihydrolutidine (DDL)
2
ACS Paragon Plus Environment

Page 3 of 32
The Journal of Physical Chemistry
derivatives.8ꢀ16 Although this earlier work focused mostly on identification of waterꢀsoluble DDL
derivatives, it was noted that the fluorescence properties of these compounds were highly sensitive to
the nature of substituents in close proximity to the ring nitrogen atom. We set out now to examine the
origin of this observation.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
R
O
R
O
O
EtO
OEt
Hantzsch esters
R = H, Me, phenyl, tolyl
N
DDL
N
H
H
Chart I. Generic formulae for the DDLꢀtype derivatives and the corresponding Hantzsch esters
described herein.
In so doing we note that these DDL derivatives, being members of the 1,4ꢀdihydropyridine family, are
closely related to the soꢀcalled Hantzsch esters, which are already known to display solventꢀdependent
fluorescence spectral properties.17 Both DDL and Hantzsch esters (Chart I) can be formed by reactions
of formaldehyde under rather mild conditions and might be adapted as fluorescent markers for the
presence of this analyte in solution.18 To aid the mechanistic studies, a small series of 1,4ꢀ
dihydropyridine derivatives has been prepared and fully characterized. Included within this series is a
rigid derivative that inhibits internal rotation of the carbonyl functions. It is evident that the Hantzsch
esters exhibit the same relationship between fluorescence quantum yield and molecular topology as
observed for the DDL derivatives. Specifically, the presence of methyl groups at the 2,6ꢀpositions
extinguishes emission in solution but not in the solid state. There is no obvious structural reason for this
observation.
RESULTS and DISCUSSION
The target compounds are depicted in Chart II and make use of complementary synthetic approaches to
prepare the DDL derivatives and the corresponding Hantzsch esters. In particular, a new synthetic
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 4 of 32
procedure was devised, starting from propargylic esters, for the synthesis of compounds 1 and 2. A more
conventional procedure was adapted to prepare the 1,4ꢀdihydropyridine derivatives 3ꢀ6. The former
method makes use of a condensation reaction involving the propargylic ester with paraꢀformaldehyde
and ammonium acetate, as is illustrated by way of Scheme 1. It might be noted that this latter reaction,
which is performed under mild conditions, has particular relevance for the introduction of an analytical
protocol for detection of formaldehyde. The intention is to develop this routine as a means by which to
establish viable fluorescenceꢀbased tests for the presence of formaldehyde.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart II. Chemical formulae and reference numbers for the compounds examined in this contribution.
It should be noted that the pivotal propargylic ester was prepared according to Scheme 2. Here, the
tolyl ester A requires deprotonation of the acetylenic unit, followed by reaction with ethyl
chloroformate. Preparation of the 1,4ꢀdihydropyridine derivatives follows a standard procedure based on
condensation of the appropriate acetylacetonate derivative with paraꢀformaldehyde and ammonium
4
ACS Paragon Plus Environment

Page 5 of 32
The Journal of Physical Chemistry
acetate, as is outlined by way of Scheme 3. Reaction yields are high and isolation of the final product is
straightforward.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. General synthetic protocol used for preparation of compounds 1 and 2.
O
OEt
i) nBuLi, -78°C
ii) ethylchloroformate, -78°C
A
Scheme 2. Outline of the methodology employed for the preparation of precursor A.
Scheme 3. Outline of the Hantzsch synthesis adapted to prepare the target compounds, starting from
either the acetylacetone derivative or the dimedone substrate.
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 6 of 32
In the absence of Xꢀray crystal structural data, quantum chemical calculations (DFT/B3LYP/6ꢀ
311G**/CPCM) were performed in order to establish lowestꢀenergy conformations for these
compounds. The first important feature relates to the conclusion19,20 that the sixꢀmember ring, in each
compound, is in essence planar (Figure 1). This observation, which is in excellent agreement with an
earlier Xꢀray crystallographic study,21,22 is consistent with the inclusion of dipolar resonance forms as
part of the overall electronic signature of the molecular structure. It might be noted that the presence of
aryl groups at the 4ꢀposition causes the ring to buckle.23,24 Molecular planarity of our compounds is
supported by the observation that the C4ꢀNꢀH angle is very nearly 1800 in each case. Other informative
structural outputs include the NꢀC2, CꢀO and C2ꢀC3 bond lengths as compiled in Table 1, and the
corresponding dihedral angles between the ring and the carbonyl residues.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Energyꢀminimized structure computed (DFT/B3LYP/6ꢀ311G**/CPCM) for 4 (N blue, O red,
C gray and H white).
We can begin a discussion of these findings by reference to the DDLꢀlike analogue 4. Here, the
lowestꢀenergy conformation has the two carbonyl groups directed towards the lower rim of the planar
dihydrolutidine ring (Figure 1); again, this observation is in excellent agreement with a prior crystal
structure determination.21,22 The dihedral angles between the ring and carbonyl groups are close to 00.
However, internal rotation, calculated25 by DFT (B3LYP/ccꢀpVTZ0) methods, is facile, and the
structure with one carbonyl group pointing away from the ring is destabilized by only ca. 1 kJ/mol
(Figure 2). This “instability” is caused by slight steric crowding due to the presence of hydrogen atoms
6
ACS Paragon Plus Environment

Page 7 of 32
The Journal of Physical Chemistry
on the proximal methyl groups. The computed structure having both carbonyl groups pointing away
from the ring is only marginally (i.e., <4 kJ/mol) less stable. As the corresponding dihedral angle
increases, there is a concomitant increase in the molecular dipole moment, calculated26 by DFT
(B3LYP/augꢀccꢀpVTZ) methods, from 2.1D for the lowestꢀenergy conformer to 9.4D with one twisted
carbonyl group and 17.8D with two twisted carbonyl units. The ease of rotation means there will be a
wide distribution of geometries present at any given moment. The rigid DDLꢀlike compound, 6, has a
large dipole moment of 14.8D since the carbonyl groups are directed away from the dihydropyridine ring
(Table 1).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Effect of changing dihedral angle on the total energy (open circles) and molecular dipole
moment (filled circles) for rotation of one of the carbonyl functions present in compound 4.
Identical (DFT/B3LYP/ccꢀpVTZ0) calculations26 made for the 3,5ꢀdiethylesterꢀ1,4ꢀdihydropyridine
derivatives confirm that the carbonyl groups tend to align with the dihydropyridine ring, having no real
preference for pointing towards or away from the ring. The energy barriers for full rotation rotation of
the carbonyl groups are modest (e.g., 8.8 kJ/mol for 1 and 8.0 kJ/mol for 3). The phenyl analogue, 5,
positions the aryl rings at a dihedral angle of ca. 750 with respect to the dihydropyridine ring and there is
a small (i.e., 7 kJ/mol) barrier to rotation of the ester unit. Steric crowding between the ethyl group and
the phenyl ring limits full rotation of these subunits and the carbonyl groups show a slight preference for
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 8 of 32
pointing towards the ring. The closest hydrogenꢀbonding type interactions are found for the ester O atom
and the C1ꢀmethyl group in 3, where the shortest distance is ca. 2.1 Å. For 1, this closest approach is
relaxed slightly to 2.3 Å.
1
2
3
4
5
6
7
8
9
Table 1. Structural data computed (DFT/B3LYP/6ꢀ311G**/CPCM) for the various target compounds
embedded in a solvent reservoir of dielectric constant = 78.a,b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Property
NꢀH / Å
4
1
6
5
1.027
179.9
118.8
118.8
1.224
1.225
1.366
1.367
6.8
1.023
179.3
119.8
119.9
1.225
1.225
1.361
1.361
0.00
1.024
179.7
119.5
119.5
1.226
1.227
1.363
1.364
176.5
177.5
1.027
179.7
118.6
119.0
1.230
1.230
1.368
1.369
0.05
C3ꢀNꢀH / 0
C1ꢀNꢀH / 0
C5ꢀNꢀH / 0
C=O / Å
C1=C2 / Å
C4=C5 / Å
C1ꢀC2ꢀCꢀO / 0
C5ꢀC4ꢀCꢀO / 0
3.6
1.3
0.2
aFor atom labeling see compound 1 in Chart II. Uncertainty in bond lengths ±0.001 Å and in bond
angles 0.05°.
b
The absorption and emission spectra of the target compounds presented in Chart II were recorded in
polar solvents, such as propanꢀ2ꢀol, ethanol, water or mixtures thereof, and an example is given as
Figure 3 (see Figures S1 to S5 for other examples). The absorption spectra are comparable along the
series and show several unstructured maxima, with the higherꢀenergy band seen at about 270 nm being
assigned to the π→π* transition of the α,β unsaturated ester or ketone.27 In some cases, this transition
lies closer to 250 nm. The characteristic featureless absorption band appearing within the 350 to 450 nm
range is possibly due to a chargeꢀtransfer transition between the NH fragment and the conjugated
electronꢀwithdrawing acetyl or ester groups.17 Bathochromic shifts of 5 and 13 nm are found for this
transition, respectively, for compounds 4 and 6 on moving from ethanol to water, but the bandꢀshape
8
ACS Paragon Plus Environment

Page 9 of 32
The Journal of Physical Chemistry
remains unchanged. Similar shifts were found throughout the series. It might be noted that the apparent
εMAX derived for the DDL derivative 4 decreases from 7,500 Mꢀ1 cmꢀ1 in propanꢀ2ꢀol to 4,200 Mꢀ1 cmꢀ1
in water, probably indicating that this compound has limited solubility in water. This is not so for the
other compounds but none of the compounds are expected to be highly soluble in water.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10000
8000
6000
4000
2000
0
250
350
450
550
650
λ / nm
Figure 3. Representative example of absorption (2.0 x 10ꢀ4 M; black curve at higher energy),
fluorescence (2.0 x 10ꢀ5 M; black curve at lower energy) and excitation (2.0 x 10ꢀ6 M) circles) spectra
recorded for 1 in propanꢀ2ꢀol at room temperature.
Compounds 1 and 6 are intensely fluorescent in solution and weakly in the solid state (Table 2). The
other compounds exhibit rather weak fluorescence in propanꢀ2ꢀol solution at room temperature but
emission is readily detected in the solid state (see Figures S6 to S11). In each case, the emission spectral
profile is fairly broad, lacking in fine structure, and subjected to a considerable Stokes’ shift (SS).
Indeed, the latter parameters range from 3,700 to 4,500 cmꢀ1 in propanꢀ2ꢀol when the respective peak
maxima, values for λABS and λFLU being given in Table 2, are used. These spectral shifts should be
considered as exaggerations of the true situation, however, since both absorption and emission spectra
are complex compilations of overlapping bands. Indeed, spectral deconstruction of the absorption profile
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 10 of 32
recorded for the various Hantzsch esters in propanꢀ2ꢀol indicates that the (0,0) transition is located at ca.
25,000 cmꢀ1 while that for 4 is redꢀshifted to about 23,500 cmꢀ1 (Table 2). The rigid DDL derivative, 6,
gives an intermediate value of 24,660 cmꢀ1. The most satisfactory spectral deconstruction indicates that a
mediumꢀfrequency stretching vibration of 1,450 and 1,610 cmꢀ1, respectively, accompanies the
absorption process for the esters and DDLꢀlike compounds. A representative example of this spectral
deconstruction is given as part of the Supporting Information (Figures S12ꢀS17).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Summary of spectroscopic properties recorded for the target compounds at room temperature.
b
c
e
f
g
f
Cmpd
λABSa (nm)
λFLU
εMAX
SSd
(cmꢀ1)
1900
λFLU
ΦFLU
ΦFLU
τS
(nm)
456
436
433
437
488
455
445
458
437
494
486
488
479
461
415
481
470
447
427
442
435
(Mꢀ1.cmꢀ1)
8600
5200
1600
3800
5600
8200
6400
6900
6200
7500
5050
5100
4400
3700
2050
2300
2600
9100
6100
6300
3900
(nm)
(ns)
7.3
6.3
1
378h
367i
378j
368k
373h
375h
357i
375j
356k
408h
392i
404j
391k
374h
366i
375j
365k
389h
374i
385j
368k
461
0.47
0.51
0.37
0.50
0.02
2
3
2350
2150
500
460
<0.001
0.02
0.08
0.18
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
7.8
9.1
0.006
0.017
0.008
0.02
4
5
6
2560
2280
2020
482
474
448
0.08
0.17
0.02
0.007
0.023
0.007
<0.001
<0.001
<0.001
<0.001
0.59
0.74
0.48
0.20
aMaximum of the lowestꢀenergy absorption band recorded in solution. Maximum of the fluorescence
b
peak recorded in solution. cEffective molar absorption coefficient measured in a certain solvent. dStokes'
e
shift calculated for deconstructed absorption and emission spectra recorded in propanꢀ2ꢀol. Maximum
of the fluorescence peak recorded in a compressed KBr disk. Fluorescence quantum yield or excitedꢀ
f
g
state lifetime determined in a certain solvent. Fluorescence quantum yield determined in a KBr disk.
hPropanꢀ2ꢀol. iDichloromethane. jDimethylsulfoxide. kToluene.
10
ACS Paragon Plus Environment

Page 11 of 32
The Journal of Physical Chemistry
Absorption spectral data collected for the phenyl derivative, 5, and the rigid DDL compound, 6, show
somewhat anomalous εMAX values (Table 2) for the lowestꢀenergy absorption transition. The former
1
2
3
4
5
6
7
8
derivative has a lower than expected εMAX while the latter has a somewhat higher value, although the
spectral profiles and peak energies are not dissimilar to those recorded for the other species. Close
examination of absorption spectra recorded for 5 shows the presence of a lowꢀenergy tail that
corresponds to what might be anticipated for light scattering caused by the presence of particulate
matter.28 We conclude, therefore, that this derivative is poorly soluble in alcoholic solvents.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. KohnꢀSham representations of the HOMO (upper panel) and LUMO (lower panel) computed
(DFT/B3LYP/6ꢀ311G**/water reservoir) for compound 3 at an isoꢀdensity of 0.015.
Molecular orbital calculations29 carried out with the energyꢀminimized structures are fully consistent
with the lowestꢀenergy absorption transition having partial chargeꢀtransfer character. Thus, the HOMO
is localized on the dihydropyridine ring and includes a marked contribution from the N atom. In
contrast, the LUMO is spread over much of dihydropyridine ring, but not the N atom, and includes the
11
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 12 of 32
carbonyl groups (Figure 4). Excitation, therefore, will transfer electronic charge from the N atom to the
carbonyl group. The optimized ground state structures computed in water using the conductorꢀlike
polarizable continuum model30 were used for singleꢀpoint TDꢀDFT (CAMꢀB3LYP) calculations31 to
predict the excitedꢀstate dipole moments and excitation energies, recognizing that there are problems for
such calculations for molecules with substantial chargeꢀtransfer character.32 Indications are that there is
an increase in the molecular dipole moment of about 2D, in each case. This appears to be a small change
but, because the conjugation lengths are short, excitation is likely to lead to a substantial increase in the
contribution of the dipolar resonance forms. Compound 6 has the highest ground state dipole moment
(ꢀGS = 14.8D), but a small increase (ꢁꢀ = 1.7D) upon excitation, which might contribute towards the
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
somewhat enhanced εMAX value. In all cases, the computed excitation energies were underestimated
compared to the experimental values by ca. 10%. The calculated and observed transition energies could
be brought into closer alignment by varying the hybrid functional, as has been demonstrated for a range
of chargeꢀtransfer compounds.32 None of these calculations indicate the likely presence of closeꢀlying
excited states. The amount of chargeꢀtransfer character associated with the groundꢀstate molecules,
based on the orbital composition analysis,33 ranges from 0.32 to 0.65 for compounds 4 and 6,
respectively.
As mentioned above, these compounds fluoresce to varying degrees, both in propanꢀ2ꢀol at room
temperature and after dispersion in a KBr pellet. The emission spectra recorded in dilute solution can be
deconstructed into accompanying vibronic bands, as was done for the corresponding absorption spectra.
The resulting vibrational envelopes can be assigned to mediumꢀfrequency stretching modes of 1,400
cmꢀ1 for the Hantzsch esters and 1,600 cmꢀ1 for the DDLꢀlike compound. The spectrally resolved data
allow derivation of corrected Stokes’ shifts (Table 2). Now, it appears that the Hantzsch esters exhibit
Stokes’ shifts of ca. 2,000 cmꢀ1 while the DDLꢀlike derivative 4 shows a more substantial shift of 2,600
cmꢀ1. The rigid DDLꢀlike compound 6 has a less pronounced Stokes’ shift (SS = 2,020 cmꢀ1) than its
more flexible counterpart. This situation seems reasonable since 6, which is the most polar of the
compounds, is less susceptible to largeꢀscale geometry changes on excitation. For 4 and 6, the
12
ACS Paragon Plus Environment

Page 13 of 32
The Journal of Physical Chemistry
fluorescence maxima are red shifted by 18 and 11 nm, respectively, in water relative to propanꢀ2ꢀol, an
observation that is fully consistent with the proposed intramolecular chargeꢀtransfer character.33
The fluorescence quantum yields (ΦFLU) measured in propanꢀ2ꢀol range from 0.02 to 0.59 (Table 2).
This level of variation, taken in conjunction with the minor structural changes, is surprising but in
keeping with previous measurements8ꢀ16 made for DDLꢀlike derivatives. In generic terms, it appears that
substitution at the 2,6 position of the dihydropyridine ring results in a major loss of fluorescence in
solution. However, the rigid derivative, 6, is the most intensely fluorescent compound. In each case,
there is excellent agreement between absorption and excitation spectra. For the strongly emissive
compounds, the excitedꢀstate lifetimes (τS) are around 7 ns (Table 2) and allow estimation of the
corresponding radiative rate constants (kRAD) as being ca. 6.5 x 107 sꢀ1 in solution.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.0
2.0
1.0
0.8
0.0
-1.0
0.0029
0.6
0.4
0.2
0.0
0.0034
-1 -1
0.0039
T / K
450
500
550
600
650
700
750
λ / nm
Figure 5. Temperature dependence recorded for compound 4 in a mixture of C2H5OH and H2O (1:9),
the arrow indicating the direction of increasing temperature. The insert shows Arrheniusꢀtype plots for a
mixture of ethanol and water (1:9) (filled gray circles), water (open circles) and propanolꢀ2ꢀol (filled
black circles).
13
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 14 of 32
Table 3. Parameters derived from an Arrheniusꢀstyle analysis of the temperature effect on the
1
2
3
fluorescence recorded for the various dyes in solution.
4
5
Compound
Propanꢀ2ꢀol
Water
Water/ethanol 9/1
6
7
8
a
a
a
EA : Ab
EA : Ab
EA : Ab
9
6.8 : 1.4
18.0 : 2500
18.0 : 1900
3.8 : 0.25
NA
NA
NA
NA
1
3
4
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5.9 : 50
NA
7.6 : 90
0.7 : 0.09
aActivation energy in units of kJ/mol. bPreꢀexponential factor in units of 109 sꢀ1.
For the two DDLꢀlike compounds, 4 and 6, temperature dependent emission studies were carried out
in both propanꢀ2ꢀol and a 9/1 mixture of water/ethanol (Figure 5 and see also Figures S18ꢀS21).
Although the experimental data are highly restricted, it appears that both compounds display weakly
activated emission in solution. On fitting the data to an Arrheniusꢀstyle model, it turns out that in
water/ethanol 9/1 the activation energies for 4 and 6, respectively, are 7.6 and 0.7 kJ/mol. Somewhat
higher barriers are found in propanꢀ2ꢀol, individual values being 18.0 and 3.8 kJ/mol respectively for 4
and 6, whereas a smaller barrier of only 5.9 kJ/mol is found for 4 in water. The preꢀexponential factors
follow the same trend (Table 3), decreasing as the solvent polarity increases and being markedly higher
for 4 relative to 6. In propanꢀ2ꢀol, activation barriers of 6.8 and 18.0 kJ/mol, respectively, were
determined for 1 and 3, again with quite disparate preꢀexponential factors (Table 3). The more
fluorescent compounds, namely 1 and 6, possess the smaller activation barriers and, more importantly,
the significantly lower preꢀexponential factors.
Our understanding of the spectroscopic properties of these compounds can now be summarized as
follows: Each of the compounds is bipolar; a conclusion reached from the quantum chemical
calculations and supported by their solubility characteristics. Excitation leads to an increased dipole
moment, as evidenced by the pronounced Stokes’ shifts, solvent dependence and calculations. The
barriers for rotation of the carbonyl groups are very small, except for 6. The temperature dependence
studies indicate the existence of a dark species that is strongly coupled to the ground state and easily
14
ACS Paragon Plus Environment

Page 15 of 32
The Journal of Physical Chemistry
reached from the fluorescent state, except for 1 and 6. Compounds having methyl groups at the 2,6ꢀ
positions of the dihydropyridine ring undergo more rapid nonradiative deactivation than either nonꢀ
substituted Hantzsch esters or the rigid derivative 6. According to the structural calculations, the methyl
group does not prevent the carbonyl group from aligning with the dihydropyridine ring.
In attempting to rationalize these findings it is tempting to invoke formation of an enol34 whereby the
hydrogen attached at the pyridine N atom is transferred to one of the carbonyl oxygen atoms (Chart III).
At the concentrations used throughout this work, bimolecular reactions are most unlikely and, in any
case, emission is restored in the solid state where there is more likelihood for close associations. Enol
formation might be promoted by specific interactions with solvent molecules but this does not explain
the importance of the methyl groups. Instead, we have considered the possibility of intramolecular
proton transfer.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart III. Proposed chemical formulae for the keto/enol equilibrium available for compound 1, with
the dipolar keto species playing a key role in longꢀrange proton transfer.
For 1, enolisation would require an unacceptably longꢀrange protonꢀtunneling step, unless hydrogen
transfer from the C3ꢀposition of the dihydropyridine ring to the carbonyl oxygen atom plays an
intermediary role. With compound 4, it might be that enolisation is promoted by proton transfer from the
methyl group (Scheme 4). Calculations34,35 (DFT/B2LYP/6ꢀ311G**/CPCM) indicate that the enol
15
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 16 of 32
derived from 4 is less stable than the keto form by some 72 kJ/mol in water at the groundꢀstate level.
Identical calculations made for the various 1,4ꢀdihydropyridine derivatives indicate that the enol is less
favorable than the keto form by 101, 107 and 173 kJ/mol, respectively, for 1, 3 and 6. As such,
enolisation is unlikely to take place in the dark. This situation is in good agreement with the structural
characterization studies that clearly indicate the exclusive presence of the keto form. It might be noted in
passing that the enols derived from 1, 3 and 6 are planar but, because of steric clashes between proximal
methyl hydrogen atoms, that derived from 4 is slightly distorted around the dihydropyridine unit (Figure
6).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Proposed scheme for enol formation for compound 3 illustrating the role played by the 2ꢀ
methyl group in proton transfer.
Recognizing the thermodynamic difficulties for enolisation in the ground state, an intrinsic reaction
profile was computed for 4 at the DFT (B3LYP/6ꢀ311G**/CPCM) level (Figure 7). Mutual rotation of
the carbonyl and methyl groups is facile while proton transfer from the C1ꢀmethyl group to the proximal
…
carbonyl oxygen atom (O H separation is ca. 1.95 Å) gives rise to a relatively low energy intermediate
(species A in Figure 7). A wellꢀdefined transition state (TS1 in Figure 7) can be identified for this
reaction and the computed barrier for the groundꢀstate reaction is 103 kJ/mol. For TS1, the relevant
16
ACS Paragon Plus Environment

Page 17 of 32
The Journal of Physical Chemistry
…
O H bond length is 1.203 Å but the overall geometry around the ring is far from planar (Figure 7).
Species A can react further to form the enol. Now, a substantial barrier of 310 kJ/mol is involved in
forming the respective transition state (TS2 in Figure 7). This latter species has a highly distorted
1
2
3
4
5
6
7
8
9
…
…
geometry (Figure 7) with C H and N H bond lengths, respectively, of 1.519 and 1.312 Å. Although
reaction to form species A seems plausible, it is considered unlikely that formation of the final enol
(Scheme 4) will be effective because of the high barrier, unless species A is formed in an excited state.
A similar intrinsic reaction profile is found for compound 3 but the rigid analogue 6 cannot react along
this coordinate. Likewise, compound 1 cannot reach a suitable transition state because the C2 hydrogen
atom is too distant from the carbonyl oxygen atom. The alternative route of utilizing one of the hydrogen
atoms at C4 involves a highꢀenergy barrier (ꢁE = 240 kJ/mol) to reach a dipolar transition state.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Energyꢀminimized structures computed for the enols derived from compounds 1 (upper
panel) and 4 (lower panel).
It will not have escaped attention that those derivatives able to form species A (i.e., 3 and 4 but not 1
or 6) are the compounds that do not fluoresce in solution. In fact, proton transfer from the C1ꢀmethyl
17
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 18 of 32
group to the proximal carbonyl oxygen atom is likely to become more favorable under illumination. This
is because of the increased potential energy available to the excitedꢀsinglet state, relative to the ground
state, and the elevated dipole moment that serves to lengthen the CꢀO bond and raise electron density on
the oxygen atom. The temperatureꢀdependent fluorescence studies indicate substantially lower activation
energies for the excited state and attempts to confirm this situation were made by computational
approaches.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Intrinsic reaction pathway computed (DFT/B3LYP/6ꢀ311G**/CPCM) for the ground state of
compound 4, with the relative electronic energies listed in units of kJ/mol. The embedded structures are
calculated geometries for the two transition states; N.B. structures for the keto and enol forms are given
in Figures 1 and 6, respectively, while that for species “A” is given in the ESI.
Starting from the optimized (TDꢀDFT/B3LYP/6ꢀ311G**/CPCM) geometry for the excitedꢀsinglet
state of 4, an intrinsic reaction coordinate analysis was performed for the conversion to species A. In
contrast to the ground state, the barrier for reaching the TS is reduced from 102.8 kJ/mol to only 17.0
kJ/mol. The same analysis performed for 3 gave a reduced barrier of 20 kJ/mol for the excitedꢀsinglet
state. Although higher than the experimental values, the computed barriers indicate that excitedꢀstate
proton transfer is easier than for the corresponding groundꢀstate species. A similar conclusion has been
reached for prototropic ketoꢀenol tautomerism of certain boron dipyrromethene dyes.35 As such,
18
ACS Paragon Plus Environment

Page 19 of 32
The Journal of Physical Chemistry
formation of species A is expected to compete favorably with fluorescence for both 3 and 4. Such
reaction will form species A as a vibrationally hot ground state and therefore the barrier to formation of
the final enol remains very high. The most likely fate for species A is a return to the ground state of the
keto form, for which the activation barriers are 35 and 20 kJ/mol respectively for 3 and 4. These
structural changes are inhibited in the solid state where the molecules are able to fluoresce under
illumination. Proton transfer will not compete with fluorescence for compounds 1 and 6.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUDING REMARKS
With the aid of quantum chemical calculations, it has been possible to rationalize the intriguing
photophysical properties of some DDLꢀlike derivatives8ꢀ16 and their corresponding Hantzsch esters. The
key supposition involves proton transfer from a proximal methyl group to the carbonyl oxygen atom.
This process is facilitated by easy rotation of the carbonyl group that brings the reacting units into
reasonably close proximity. Although tempting to invoke enol formation via a second proton transfer
step, this additional reaction is not needed to explain the fluorescence yields. The first proton transfer
step is fast and favored by internal charge transfer that raises electron density on the carbonyl oxygen
atom. The reaction is inhibited in the solid state but further promoted in more polar solvents.
It is interesting to compare the computed structures of the stable species and their respective transition
states. With specific reference to 4, we note that the keto NꢀH bond and the corresponding enol OꢀH
bond have lengths of 1.007 and 0.962 Å respectively; N.B. the reproducibility of these various bond
lengths is on the order of ±0.001 Å. These values are typical for such species36 and are very similar to
those calculated for the Hantzsch esters. The key intermediate, species A, has NꢀH and OꢀH bond
lengths, respectively, of 1.008 and 0.962 Å. The ring is nonꢀplanar and the two C4 hydrogen atoms
reside in quite disparate environments. For TS1, the NꢀH bond retains a comparable length (i.e., 1.009
Å) to that of the keto form but important structural changes are evident elsewhere. In particular, the
…
…
O H length is relatively long at 1.203 Å while the C H bond is 1.420 Å. The ring is buckled and the
C4 hydrogen atoms are clearly nonꢀequivalent. Moving to TS2, we note that the OꢀH bond takes on the
19
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 20 of 32
…
…
characteristic features of a stable structure, the bond length being 0.963 Å. The N H and C H bond
lengths are 1.312 and 1.519 Å, respectively. These long bonds must make substantial contributions to
the energy barrier.
1
2
3
4
5
6
7
8
9
The ability of the C1ꢀmethyl group to promote nonradiative decay of the excitedꢀsinglet state is both
unusual and unexpected. In biological processes, there is evidence to support the idea of proton transfer
between remote sites37ꢀ40 but the mechanism is not at all clear41 and the role of the solvent has not been
fully explored. In certain photoꢀinduced ketoꢀenol tautomerizations (e.g., with 7ꢀhydroxyquinoline42)
where the donor and acceptor are too far apart for proton tunneling to be effective it is believed that the
reaction is promoted by hydrogenꢀbonding solvents.43,44 Here we provide new insight into how longꢀ
range proton transfer might be effected by proximal alkyl groups. A more detailed quantum chemical
study should provide deeper insight into the subtleties of this mechanism.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It might be recalled that our initial interest in these molecular systems derives from the need to
develop sensitive analytical tests for the detection of formaldehyde present in aqueous solution. In this
respect, we have found that fluorescent analogues can be formed under mild conditions provided the C1
and C5 sites are without alkyl substituents. Future work will build on this realization and seek to
introduce suitable analytical tests using waterꢀsoluble analogues. It is also important17 to identify
strategies leading to dyes possessing increased levels of photostability.
EXPERIMENTAL SECTION
General methods: All reactions were performed under a dry atmosphere of argon. All raw chemicals
were used as received from commercial sources without further purification. Reaction solvents were
distilled according to standard procedures. Thin layer chromatography (TLC) was performed on silica
gel or aluminium oxide plates coated with fluorescent indicator. Chromatographic purifications were
1
conducted using (40ꢀ63 ꢁm) silica gel. All mixtures of solvents are given as the v/v ratio. H NMR
13
(400.1 MHz) and C NMR (100.5 MHz) spectra were recorded at room temperature (rt) on a Bruker
1
Advance 400 MHz spectrometer, H NMR (300.1 MHz) and 13C NMR (75.5 MHz) spectra were
20
ACS Paragon Plus Environment

Page 21 of 32
The Journal of Physical Chemistry
recorded at rt on a Bruker Advance 300 MHz spectrometer, 1H NMR (200.1 MHz) and 13C NMR (50.5
MHz) spectra were recorded at rt on a Bruker Advance 200 MHz spectrometer using perdeuterated
1
2
3
4
5
6
11
solvents as internal standards. The B (128 MHz) NMR spectra were recorded at rt with borosilicate
7
8
9
glass as reference. FTꢀIR spectra were recorded using a spectrometer equipped with an ATR “diamond”
apparatus. Absorption spectra were recorded using a dualꢀbeam grating spectrophotometer with a
Shimadzu UVꢀ3600 spectrophotometer. All fluorescence spectra were recorded with a fully corrected
Horiba Fluoromaxꢀ4 spectrofluorimeter. The fluorescence quantum yields were determined for optically
dilute solutions, having an absorbance less than 0.08 at the excitation wavelength, by comparison to
standard compounds. Quinine sulfate45 was used as reference (Φ= 0.55 in 1N H2SO4, λex = 366 nm) for
dyes emitting below 480 nm and Rhodamine 6G45 (Φ= 0.88 in ethanol, λex = 488 nm) was used for the
other dyes. Luminescence lifetimes were measured on a spectrofluorimeter equipped with an R928
photomultiplier and pulsed laser diode connected to a GwInstect delay generator. The excitation source
was a laser diode (λ = 310 nm). The instrument response function was determined by using a lightꢀ
scattering solution of LUDOX in water. Solidꢀstate fluorescence and excitation spectra were recorded in
KBr pellets of the desired compound dispersed at 10ꢀ6ꢀ10ꢀ7 M concentration for emission and excitation
spectra measurements using an Fꢀ3029 integrating sphere.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents: Ammonium acetate, acetylacetone, dimedone, ethyl acetoacetone, ethyl benzoylacetate,
ethyl chloroformate, ethyl propiolate, 4ꢀethynyltoluene, paraꢀformaldehyde, propane sultone and
propiolic acid were purchased from commercial sources. The preparations of compounds 1,46 ethyl 3ꢀ
tolylpropiolate,47 2,48 3,49 4,50 and 651 were based by the cited references.
Synthesis of 1: To a stirred solution of ethyl propiolate (0.30 mL, 3.10 mmol) in a mixture of
ethanol/H2O solution (5 mL, 1:1) were added paraꢀformaldehyde (120 mg, 3.98 mmol) and ammonium
acetate (307 mg, 3.98 mmol). The resulting solution was heated at 90°C for 1h. After cooling the
solution at room temperature, the reaction mixture was extracted with CH2Cl2. The organic layer was
washed twice with water, dried over MgSO4 and concentrated under reduced pressure. The crude
product was purified by column chromatography on silica gel, eluting with a mixture of CH2Cl2/ethyl
21
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 22 of 32
acetate (95:5) to afford compound 1 as a yellow powder. The solid was recrystallized from
CH2Cl2/cyclohexane and washed with cyclohexane and pentane. Finally, it was dried under vacuum
1
2
3
4
5
6
1
(120 mg, 17%). H NMR (200 MHz, CDCl3) δ (ppm): 7.08 (2H, d, J = 4.9 Hz, CH), 5.78 (1H, s, NH),
7
8
4.18 (4H, q, J = 7.3 Hz, CH2CH3), 3.25 (2H, s, CH2), 1.27 (6H, t, J = 7.3 Hz, CH3CH2).
9
Synthesis of ethyl 3-tolylpropiolate: To a stirred solution of pꢀtolylacetylene (0.65 mL, 5.12 mmol)
in freshly distilled THF (5 mL) at ꢀ78°C was added dropꢀwise 1.6M nꢀBuLi in pentane (2.88 mL, 4.61
mmol). The solution was maintained at room temperature during 1h before being returned to ꢀ78°C.
Ethyl chloroformate (0.49 mL, 5.12 mmol) was added dropꢀwise. The resulting solution was stirred for
8h at room temperature before being extracted with CH2Cl2. The organic layer was washed twice with
water, dried with MgSO4 and concentrated under reduced pressure. The crude material was purified by
column chromatography on silicaꢀgel, eluting with a mixture of AcOEt/petroleum ether (10:90) to give
the desired compound as a yellow oil which was dried under reduced pressure (940 mg, 97%). 1H NMR
(200MHz, CDCl3) δ (ppm): 7.48 (2H, d, J = 8.0 Hz, CH arom), 7.18 (2H, d, J = 8.0 Hz , CH arom), 4.29
(2H, q, J = 6.9 Hz, CH2CH3), 2.38 (3H, s, CH3), 1.35 (3H, t, J = 6.9 Hz , CH3CH2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of 2: To a stirred solution of ethyl 3ꢀtolylpropiolate (200 mg, 1.06 mmol) in ethanol (2 mL)
were added paraꢀformaldehyde (42 mg, 1.38 mmol) and ammonium acetate (106 mg, 1.38 mmol). The
resulting solution was heated at 85°C for 5 days. After cooling to room temperature, the reaction
mixture was extracted with AcOEt. The organic layer was washed twice with water, dried over MgSO4
and concentrated under reduced pressure. The crude product was purified by column chromatography on
silica gel, eluting with a mixture of AcOEt/petroleum ether (70:30) to afford compound 2 as a yellow
powder. The solid was recrystallized from AcOEt/pentane and washed with pentane. Finally, it was
dried under vacuum (70 mg, 12%). 1H NMR (300 MHz, CDCl3) δ (ppm): 7.16 (8H, dd, 3J = 12.0 Hz, 4J
= 3.6 Hz, CH arom), 5.45 (1H, s, NH), 3.95 (4H, q, J = 7.2 Hz, CH2CH3), 3.58 (2H, s, CH2), 2.33 (6H, s,
CH3), 0.98 (6H, t, J = 7.2 Hz, CH3CH2). 13C NMR (75 MHz, CDCl3) δ (ppm): 167.44, 146.94, 139.14,
134.04, 129.11, 127.90, 99.96, 59.83, 34.27, 26.01, 22.48, 21.48, 14.20, 14.09. ESIꢀMS 405.1 (100,
22
ACS Paragon Plus Environment

Page 23 of 32
The Journal of Physical Chemistry
[M]) Elemental analysis for C25H27O4: Calc: C, 74.05; H, 6.71; N, 3.45. Found: C, 73.88; H, 6.49; N,
3.17.
1
2
3
4
5
6
7
8
Synthesis of 3: To a stirred solution of ethyl acetoacetate (9.7 mL, 77 mmol) in ethanol (100 mL) was
added paraꢀformaldehyde (1.15 g, 38 mmol) and ammonium acetate (4.45g, 58 mmol). The resulting
solution was heated at 80°C for 15 min. A pale yellow precipitate formed after cooling to room
temperature. The mixture was diluted with water and the yellow suspension was isolated by filtration,
washed firstly with water and then with cold ethanol. A recrystallization of the filtrate was made from
hot ethanol and the solution was allowed to cool at 0°C. The bright yellow precipitate was removed by
filtration, washed with cold ethanol and dried under vacuum (15.4 g, 79%). 1H NMR (200MHz, CDCl3)
δ (ppm): 5.18 (1H, s, NH), 4.16 (4H, q, J = 7.3 Hz, CH2CH3), 3.26 (2H, s, CH2), 2.18 (6H, s, CH3), 1.28
(6H, t, J = 6.9 Hz, CH3CH2).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of 4: To a stirred solution of acetylacetone (10.3 mL, 100 mmol) in ethanol (100 mL) was
added paraꢀformaldehyde (1.50 g, 50 mmol) and ammonium acetate (3.50g, 50 mmol). The resulting
solution was heated at 45°C for 1h. After cooling, a pale yellow precipitate formed. This was isolated by
filtration and washed with water. The resulting pale yellow powder was recrystallized twice from hot
1
ethanol. Finally, it was dried under vacuum (16.7g, 86%). H NMR (400MHz, CDCl3) δ (ppm): 5.42
(1H, s, NH), 3.42 (2H, s, CH2), 2.22 (6H, s, CH3CO), 2.17 (6H, s, CH3).
Synthesis of 5: To a stirred solution of 3ꢀphenylꢀ1ꢀethyl acetoacetate (0.27 mL, 1.56 mmol) in a
mixture of ethanol/H2O solution (5 mL, 1:1) were added paraꢀformaldehyde (61 mg, 2.03 mmol) and
ammonium acetate (157 mg, 2.03 mmol). The resulting solution was heated at 90°C for 1h. After
cooling to room temperature, the reaction mixture was extracted with CH2Cl2. The organic layer was
washed twice with water, dried over MgSO4 and concentrated under reduced pressure. The crude
product was purified by column chromatography on silica gel, eluting with a mixture of
AcOEt/petroleum ether (70:30) to give compound 5 as a yellow powder. The solid was recrystallized
1
from AcOEt/pentane and washed with pentane before being dried under vacuum (70 mg, 12%). H
23
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 24 of 32
NMR (200MHz, CDCl3) δ (ppm): 7.35 (10H, m, CH arom), 5.47 (1H, s, NH), 4.20 (4H, q, J = 7.0 Hz,
CH2CH3), 3.53 (2H, s, CH2), 0.96 (6H, t, J = 7.3 Hz, CH3CH2).
1
2
3
4
5
6
7
8
Synthesis of 6: To a stirred solution of dimedone (200 mg, 1.43 mmol) in a mixture of ethanol/H2O
solution (5 mL, 1:1) were added paraꢀformaldehyde (56 mg, 0.71 mmol) and ammonium acetate (21
mg, 0.71 mmol). The resulting solution was heated at 90°C for 4h. After cooling to room temperature, a
yellow precipitate formed, which was isolated by filtration, washed successively with cold ethanol,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
diethyl ether and pentane. It was dried under vacuum (120 mg, 36%). H NMR (400 MHz, CDCl3) δ
(ppm): 5.16 (1H, s, NH), 3.12 (2H, s, CH2), 2.27 (4H, s, CH2CO), 2.19 (4H, s, CH2), 1.09 (12H, s,
CH3).
Computational methods: A wide variety of quantum chemical calculations were performed based on
prior work which concluded that semiꢀempirical AMI SCF approaches could give satisfactory estimates
of the barrier heights for proton transfer in substituted salicylaldehyde anils.36 Semiꢀempirical methods,
using PM652ꢀRHF(CI)ꢀCOSMO53 as implemented by AMPAC,54 were used to obtain first estimates for
the lowestꢀenergy conformations at both groundꢀ and excitedꢀstate levels. Convergence was judged in
terms of the steepest descent and conjugate gradients methods. Calculation of second derivatives was
made by the NewtonꢀRaphson method. Subsequent refinement of the geometries was made using DFT
calculations, by way of Gaussianꢀ09.55 The hybrid Beckeꢀ3ꢀLeeꢀYangꢀParr56ꢀ59 (B3LYP) exchange
correlation functional was applied. Geometries were fully optimised at the B3LYP level using the 6ꢀ
311G** basis set. The absence of imaginary frequencies was confirmed for all original structures.
Dipole moments were taken from the latter outputs. The same procedure was used to determine
geometries and singleꢀpoint energy levels for the corresponding enols and for intermediates akin to
Species A. Related studies, starting with AMPAC and moving to Gaussianꢀ09, addressed the
corresponding dipolar intermediates associated with proton transfer from the C4 position to an adjacent
carbonyl oxygen atom. In all cases, the solute was embedded in a reservoir of water molecules. The
DFT/B3LYP/6ꢀ311G** optimised structures were used for subsequent electronicꢀstructure analysis of
the HOMO/LUMO isoꢀsurfaces.
24
ACS Paragon Plus Environment

Page 25 of 32
The Journal of Physical Chemistry
Rotation of the carbonyl function was carried out using semiꢀempirical methods (PM6ꢀRHFꢀCOSMO)
used with the computed geometry for the ground state in a water bath. The relevant dihedral angle was
varied systematically, the energy minimised, and a singleꢀpoint calculation performed using AMPAC. A
limited number of molecular dynamics simulation runs were performed using YASARA in order to
verify the free rotation of these carbonyl groups.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Transition states were searched within Gaussianꢀ09 using the TS keyword. All transition states were
verified by the existence of a single imaginary frequency mode. In all cases, intrinsic reaction coordinate
(IRC) calculations were performed to confirm that the transition state connects keto and species A and
enol and species A. Activation barriers were taken as the difference between the zero point vibrational
energy corrected electron energies for transition state and the stable conformation. These IRC
calculations worked well for 1 and 4 but were somewhat less successful for 3. In contrast, all three
compounds could be handled well within AMPAC, although the derived barriers were roughly 15ꢀ20
kJ/mol higher than those found by DFT. For compound 3, related IRC calculations were made with
GAMESS.60,61 A set of 30 points along a Linear Least Motion (LLM) path connecting the keto and
species A was calculated and single point energies at the B3LYP/6ꢀ311G** level were calculated at
these points to provide an upper bound for the protonꢀtransfer reaction. Another set of 30 LLM points
was calculated for the species A to enol conversion and single point energies were calculated again at
the B3LYP/6ꢀ311G** level.
ACKNOWLEDGMENT
We thank ECPMꢀStrasbourg, CNRS and Newcastle University for financial support. This work was
supported by OSEO through the project MINI FORMALAIR (A1106007) and ANR through the project
CAPFEIN (ANRꢀ11ꢀECOTꢀ0013).
SUPPORTING INFORMATION: Provided are absorption, emission and excitation spectra in
solution and in the solid state for all compounds. This material is available free of charge via the internet
at http://pubs.acs.org.
25
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 26 of 32
REFERENCES
1
2
3
4
(1) Allouch, A.; Le Calvé, S.; Serra, C. A. Portable, Miniature, Fast and High Sensitive RealꢀTime
5
Analyzers: BTEX Detection. Sensors Actuators B: Chem., 2013, 182, 446ꢀ452.
6
7
8
9
(2) Marchand, C.; Bulliot, B. ; Le Calvé, S. ; Mirabel, P. Aldehyde Measurements in Indoor
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Environments in Strasbourg. Atm. Environ., 2006, 40, 1336–1345.
(3) IARC, 2004. Overall Evaluation of Carcinogenicity to Humans, Formaldehyde [50ꢀ00ꢀ0] The
IARC Monographs Series. International Agency for Research on Cancer 88.
(4) Toda, K.; Yoshioka, K.ꢀI.; Mori, K.; Hirata, S. Portable System for NearꢀRealꢀTime
Measurement of Gaseous Formaldehyde by Means of Parallel Scrubber StoppedꢀFlow Absorptiometry.
Anal. Chim. Acta, 2005, 531, 41ꢀ49.
(5) Li, J.; DasGupta, P.K.; Genfa, Z.; Hutterli, M. A. Measurement of Atmospheric Formaldehyde
with a Diffusion Scrubber and LightꢀEmitting Diode LiquidꢀCore Waveguide Based Fluorometry. Field
Anal. Chem. and Technol., 2001, 5, 2ꢀ11.
(6) Li, J.; DasGupta, P. K.; Luke W. Measurement of Gaseous and Aqueous Trace Formaldehyde –
Revisiting the Pentanedione Reaction and Field Applications. Anal. Chim. Acta, 2005, 531, 51ꢀ68.
(7) Le Calvé, S.; Zheng, W.; Ponche, J. L.; Bernhardt, P. in 'Device and Method for Determining the
Concentration of a Compound in an Aqueous or Gaseous Phase', WO 2010/142908 A1, 16,12,2010.
(8) Chung, P.ꢀR.; Tzeng, C.ꢀT.; Ke, M.ꢀT.; Lee, C.ꢀY. Formaldehyde Gas Sensors: A Review.
Sensors, 2013, 13, 4468ꢀ4484.
(9) Lowe, D. C.; Schmidt, U.; Ehhalt, D. H.; Frischkom, C. G. B. Determination of Formaldehyde in
Clean Air. Environ. Sci. Technol. 1981, 15, 819ꢀ823.
26
ACS Paragon Plus Environment

Page 27 of 32
The Journal of Physical Chemistry
(10) Levin, J. O.; Andersson, K.; Linndahl, R. Determination of Subꢀppm Levels of Formaldehyde in
Air Using Active or Passive Sampling on 2,4ꢀDinitrophenylhydrazineꢀCoated Glass Fibre Filters and
HLPC. Anal. Chem. 1985, 57, 1032ꢀ1035.
1
2
3
4
5
6
7
8
9
(11). Hopkins, J. R.; Still, T.; AlꢀHaider, S.; Fisher, I. R.; Lewis, A. C.; Seakins, P. W. A Simplified
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Apparatus for Ambient Formaldehyde Detection via GCꢀpHID. Atmos. Environ. 2003, 37, 2557ꢀ2565.
(12) Brackmann, C.; Li, Z.; Rupinski, M.; Docquier, N.; Pengloan, G. Strategies for Formaldehyde
Engines Using a SingleꢀMode Detection in Flames and NdꢀYAG/OPO Laser System. Appl. Spectros.
2005, 59, 763ꢀ768.
(13) Lv, P.; Tang, Z. A.; Yu, J. ; Zhang, F. T.; Wei, G. F.; Huang, Z. X.; Hu, Y. Study on a MicroꢀGas
Sensor With SnO2ꢀNiO Sensitive Film for Indoor Formaldehyde Detection. Sensors Actuators B: Chem.
2008, 132, 74ꢀ80.
(14) Kudo, H.; Suzuki, Y.; Gessei, T.; Takahashi, D.; Arakawa, T.; Mitsubayashi, K. Biochemical Gas
Sensor (BioꢀSniffer) for Ultrahigh Sensitive Gaseous Monitoring. Biosensors Bioelectronics 2010, 26,
854ꢀ858.
(15) Deng, B.; Liu, Y.; Yin, H.; Ning, X.; Lu, H.; Ye, L.; Xu, Q. Determination of UltraꢀTrace
Formaldehyde in Air Using Ammonium Sulfate as Derivatization Reagent and Capillary Electrophoresis
Coupled with OnꢀLine Electrochemiluminescence Detection. Talanta 2012, 91, 128ꢀ133.
(16) Allouch, A.; Guglielmino, M.; Bernhardt, P.; Serra C. A.; Le Calvé, S. Transportable, Fast and
High Sensitive NearꢀRealꢀTime Analyzers: Formaldehyde Detection. Sensors Actuators B: Chem., 2013,
181, 551ꢀ558.
(17) Salthammer, T. Photophysical Properties of 3,5ꢀDiacetylꢀ1,4ꢀdihydrolutidine in Solution –
Application to the Analysis of Formaldehyde. J. Photochem. Photobiol. A. Chem., 1993, 74, 195ꢀ201.
(18) Eisner, U.; Kuthan, J. Chemistry of Dihydropyridines. Chem. Rev., 1972, 72, 1ꢀ42.
27
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 28 of 32
(19) Meyers, A. I.; Stout, D. M. Recent Advances in the Chemistry of Dihydropyridines. Chem. Rev.,
1982, 82, 223ꢀ243.
1
2
3
4
5
6
7
8
(20) Kuthan, J.; Kurfurst, A. Developments in Dihydropyridine Chemistry. Ind. Eng. Chem. Prod.
Res. Dev., 1982, 21, 191ꢀ261.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(21) Rathore, R. S.; Reddy, B. P.; Vijayakumar, V.; Ragavan, R. V.; Narasimhamurthy, T. Hantzsch
1,4ꢀDihydropyridine Esters and Analogs: Candidates for Generating Reproducible OneꢀDimensional
Packing Motifs. Acta Cryst. B, Struct. Sci., 2009, 65, 375ꢀ381.
(22) Lenstra, A. T. H.; Petit, G. H.; Dommisse, R. A.; Alderweireldt, F. C. A Reꢀdetermination of the
Crystal Structure of 4,4,6ꢀTrimethylꢀ1,3,2ꢀdioxathianꢀ2ꢀoxime at 120K. Bull. Soc. Chim. Belg., 1979,
88, 133ꢀ141.
(23) Goldmann, S.; Born, L.; Kazda, S.; Pittel, B.; Schramm, M. Synthesis, Pharmacological Effects
and Conformation of 4,4ꢀDisubstituted 1,4ꢀDihydropyridines. J. Med. Chem., 1990, 33, 1413ꢀ1418.
(24) Fossheim, R.; Svarteng, K.; Mosad, A.; Romming, C.; Shefter, E.; Triggle, D. Crystal Structures
and Pharmacological Activity of Calcium Channel Antagonists – 2,6ꢀDimethylꢀ3,5ꢀdicarbomethoxyꢀ4ꢀ
(unsubstituted, 3ꢀmethylphenyl, 4ꢀmethylphenyl, 3ꢀnitrophenyl and 2,4ꢀdinitrophenyl)ꢀ1,4ꢀ
Dihydropyridine. J. Med. Chem., 1982, 25, 126ꢀ131.
(25) Klahm, S.; Lüchow, A. Accurate Rotational Barrier Calculations with Diffusion Quantum Monte
Carlo. Chem. Phys. Lett. 2014, 600, 7ꢀ9.
(26) Hickey, A. L.; Rowley, C. N. Benchmarking Quantum Chemical Methods for the Calculation of
Molecular Dipole Moments and Polarizabilities. J. Phys. Chem. 2014, 118, 3678ꢀ3687.
(27) Dommisse, R. A.; LePoivre, J. A.; Alderweireldt, F. C. UVꢀSpectroscopic and IR Spectroscopic
Studies of 4ꢀArylꢀ3,5ꢀdiethoxyꢀcarbonylꢀ2,6ꢀdimethylꢀ1,4ꢀdihydropyridines (Hantzsch Esters). Bull.
Soc. Chim. Belg., 1975, 84, 855ꢀ859.
28
ACS Paragon Plus Environment

Page 29 of 32
The Journal of Physical Chemistry
(28) Kuthan, J.; Palecek, J. On the πꢀElectronic Structure of Hantzsch’s Dihydropyridines. Collect.
Czech. Chem. Commun., 1969, 34, 1339ꢀ1341.
1
2
3
4
5
6
7
8
9
(29) Surendra Babu, N. DFT Studies of Molecular Structure, Equilibrium Constant for KetoꢀEnol
Tautomerism and Geometrical Isomerization (EꢀZ) of 2ꢀAminoꢀ1ꢀphenylpropanꢀ1ꢀone (Cathinone). Adv.
Appl. Sci. Res., 2013, 4, 147ꢀ153.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(30) Barone, V.; Cossi, M. Quantum Calculation of Molecular Energies and Energy Gradients in
Solution by a Conductor Solvent Model. J. Phys. Chem. A 1998, 102, 1995ꢀ2001.
(31) Yanai, T.; Tew, D. P.; Handy, N. C. A New Hybrid ExchangeꢀCorrelation Functional Using the
Coulomb Attenuating Method (CAMꢀB3LYP). Chem. Phys. Lett. 2004, 393, 51ꢀ57.
(32) Huang, S.; Zhang, Q.; Shiota, Y.; Nakagawa, T. ; Kuwabara, K.; Yoshizawa, K.; Adachi, C.
Computational Prediction for Singletꢀ and TripletꢀTransition Energies of ChargeꢀTransfer Compounds.
J. Chem. Theory Comput. 2013, 9, 3872ꢀ3877.
(33) Lu, T. Chen, F. W. Multiwvn: A Multifunctional Wavefunction Analyzer. J. Comput. Chem.
2012, 33, 580ꢀ592.
(34) Chou, P. T. ; Chen, Y. C. ; Yu, W. S. ; Chou, Y. H. ; Wei, C. Y. ; Cheng, Y. M. Excited State
Intramolecular Proton Transfer in 10ꢀHydroxybenzo[h]quinoline. J. Phys. Chem. A, 2001, 105, 1731ꢀ
1740.
(35) Pan, Z.ꢀH.; Zhou, J.ꢀW.; Luo, G.ꢀG. Experimental and Theoretical Study of EnolꢀKeto
Prototropic Tautmerism and Photophysics of AzomethineꢀBODIPY dyads. Phys. Chem. Chem. Phys.
2014, 16, 16290ꢀ16301.
(36) Maran, U.; Karelson, M.; Katritzky, A. R. A Comparative AM1 and ab initio Study of the
Intramolecular Proton Transfer in Tautomeric Organic Compounds. Int. J. Quantum Chem., 1996, 60,
1765ꢀ1773.
29
ACS Paragon Plus Environment