The exploration of hydrogen bonding properties of 2,6- and 3,5-diethynylpyridine by IR spectroscopy

Article abstract of DOI:10.1016/j.saa.2014.10.107

Hydrogen bonding properties of 2,6- and 3,5-diethynylpyridine were analyzed by exploring of their interactions with trimethylphosphate, as hydrogen bond acceptor, or phenol, as hydrogen bond donor, in tetrachloroethene C₂Cl₄. The employment of IR spectroscopy enabled unravelling of their interaction pattern as well as the determination of their association constants (K_c) and standard reaction enthalpies (^ΔrHθ). The association of diethynylpyridines with trimethylphosphate in stoichiometry 1:1 is established through CH?O hydrogen bond, accompanied by the secondary interaction between CC moiety and CH₃ group of trimethylphosphate. In the complexes with phenol, along with the expected OH?N interaction, CC?HO interaction is revealed. In contrast to 2,6-diethynylpyridine where the spatial arrangement of hydrogen bond accepting groups enables the simultaneous involvement of phenol OH group in both OH?N and OH?CC hydrogen bond, in the complex between phenol and 3,5-diethynylpyridine this is not possible. It is postulated that cooperativity effects, arisen from the certain type of resonance-assisted hydrogen bonds, contribute the stability gain of the latter. Associations of diethynylpyridines with trimethylphosphate are characterized as weak (K_c ≈0.8-0.9 mol^-1 dm³; -^ΔrHθ ≈5-8 kJ mol^-1), while their complexes with phenol as medium strong (K_c ≈5 mol^-1 dm³; -^ΔrHθ ≈15-35 kJ mol^-1). Experimental findings on the studied complexes are supported with the calculations conducted at B3LYP/6-311++G(d,p) level of theory in the gas phase. Two conformers of diethynylpyridine?trimethylphosphate dimers are formed via CH?O interaction, whereas dimers between phenol and diethynylpyridines are established through OH?N interaction.

Full text of DOI:10.1016/j.saa.2014.10.107

Accepted Manuscript
The exploration of hydrogen bonding properties of 2,6- and 3,5-diethynylpyri-
dine by IR spectroscopy
Danijela Vojta, Goran Kovačević, Mario Vazdar
PII:
S1386-1425(14)01604-7
http://dx.doi.org/10.1016/j.saa.2014.10.107
SAA 12922
DOI:
Reference:
To appear in:
Spectrochimica Acta Part A: Molecular and Biomo-
lecular Spectroscopy
Received Date:
Revised Date:
Accepted Date:
15 July 2014
17 October 2014
23 October 2014
Please cite this article as: D. Vojta, G. Kovačević, M. Vazdar, The exploration of hydrogen bonding properties of
2
,6- and 3,5-diethynylpyridine by IR spectroscopy, Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy (2014), doi: http://dx.doi.org/10.1016/j.saa.2014.10.107
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The exploration of hydrogen bonding properties of 2,6- and 3,5-
diethynylpyridine by IR spectroscopy
a*
b
a
Danijela Vojta , Goran Kovačević , Mario Vazdar
a
Division of Organic Chemistry and Biochemistry, Ruđer Bošković Institute, Bijenička 54,
1
b
0001 Zagreb
Division of Materials Physics, Ruđer Bošković Institute, Bijenička 54, 10001 Zagreb
*
Corresponding author. Tel.:+385 1 457 1220; Fax: +385 1 168 0195; E-mail: dvojta@irb.hr
Danijela Vojta)
(
Abstract
Hydrogen bonding properties of 2,6- and 3,5-diethynylpyridine were analyzed by exploring of
their interactions with trimethylphosphate, as hydrogen bond acceptor, or phenol, as hydrogen
2 4
bond donor, in tetrachloroethene C Cl . The employment of IR spectroscopy enabled
unravelling of their interaction pattern as well as the determination of their association
ꢀ
constants (K
c
) and standard reaction enthalpies ( ∆ H ). The association of diethynylpyridines
r
with trimethylphosphate in stoichiometry 1:1 is established through ≡C-H···O hydrogen
bond, accompanied by the secondary interaction between C≡C moiety and CH
group of
3
trimethylphosphate. In the complexes with phenol, along with the expected OH···N
interaction, C≡C···HO interaction is revealed. In contrast to 2,6-diethynylpyridine where the
spatial arrangement of hydrogen bond accepting groups enables the simultaneous involvement
of phenol OH group in both OH···N and OH···C≡C hydrogen bond, in the complex between
phenol and 3,5-diethynylpyridine this is not possible. It is postulated that cooperativity
effects, arisen from the certain type of resonance-assisted hydrogen bonds, contribute the
stability gain of the latter. Associations of diethynylpyridines with trimethylphosphate are
-1
3
ꢀ
-1
characterized as weak (K
≈0.8-0.9 mol dm ; −∆ H ≈5-8 kJ mol ), while their complexes
r
c
-1
3
ꢀ
-1
with phenol as medium strong (K
findings on the studied complexes are supported with the calculations conducted at B3LYP/6-
11++G(d,p) level of theory in the gas phase. Two conformers of
c
≈5 mol dm ; −∆ H ≈15-35 kJ mol ). Experimental
r
3
diethynylpyridine···trimethylphosphate dimers are formed via ≡C-H···O interaction, whereas
dimers between phenol and diethynylpyridines are established through OH···N interaction.
1

Keywords: diethynylpyridine, hydrogen bonding, IR spectroscopy, thermodynamic
parameters, cooperativity
1
. Introduction
When discussing the bonding capability of multifunctional molecules with interaction
centers being weak hydrogen bond (HB) donors and acceptors, ethynylbenzene (EB) can be
presented as one of the most comprehensively explored systems [1]. The possibility of
employing both ethynyl and phenyl moieties, either as HB accepting or HB donating centers,
has been found to be dependent on the another protagonist in a HB complex, which in turn
have resulted with the uncertainty in the prediction of EB interaction pattern [2-4].
The attempts to clarify the EB bonding hierarchy have been further continued by
undertaking the studies on substituted EBs in which one of its H-atoms have been replaced by
the HB accepting centers such as fluorine (F) atoms [5-6] or cyano (CN) groups [7]. In
contrast to expectations that strong HB acceptors may take a primacy in the bonding pattern,
in the former example they are not even directly involved in HB interplay, while in the latter
they can be, at best, regarded as equal with the phenyl moiety.
The certain predictability in the bonding hierarchy in multifunctional molecules with weak
HB donating and accepting sites has been found in 2 and 3-ethynylpyridine (2- and 3-EP) [8].
By employing IR spectroscopy, coupled with the calculations, Vojta et al. [8] have analyzed
and, in thermodynamic terms, characterized HBs formed by 2- and 3-EP and EB with
trimethylphosphate (TMP) and phenol (PH) in C
2
Cl , respectively. Expectedly, the complexes
4
between EPs or EB with TMP have been found to be weak and directed by C≡C-H···O
interaction, while when PH is their partner in HB complex, the association is of medium
strength and governed by OH···N interaction. In structural terms, however, some exceptional
phenomena addressed to the participation of C≡C moiety in HB formation, have appeared -
when in interaction with TMP (via CH
group), C≡C bond becomes longer (the stretching of
3
associated C≡C oscillator (C≡C···) shifts to the lower frequencies when compared with the
stretching of unassociated C≡C oscillator (C≡C)), while when in interaction with PH (via OH
group), it shortens (C≡C··· stretching shifts to the higher frequencies, respectively). The latter
was observed only in PH···2-EP complex since simultaneous OH···N and OH···C≡C
interactions have been enabled. In 3-EP the additional interaction was absent and explained
due to two reasons: i) sterical limitation does not allow simultaneous arrangements of OH···N
and OH···C≡C HB so OH group certainly prefers HB with N-atom over C≡C group, and ii)
2

the molar absorption coefficient (ε ) of C≡C stretching is rather low so even if the bond is
formed, it is too weak to be observed in these experimental conditions. However, the
participation of C≡C moiety in making of HB complex, presumably as secondary interaction
established due to favorable geometrical arrangement in which N-atom takes hierarchical
primacy in HB formation, resulted with the following idea: C≡C group, through cooperative
actions, may be promoted to the interaction site competitive to the N-atom when exposed to
the HB donor such as OH group.
The understanding of the cooperativity [9] in self-organizing structures [10], as well as the
quantitative assessment of cooperativity related effects [11], is usually based on well-defined
systems mostly capable for the simultaneous formation of several strong or at least medium
strong HB like in trialkylbenzene-1,3,5-tricarboxamide [12]. To the best of our knowledge,
the most recent attempt to quantify the contributions of weak intermolecular interactions in
certain complex requires a development of a platform based on strong HB donors and HB
acceptors such as bisurea [13]; then, by combining theoretical and experimental methods,
-1
subtle differences as weak as 1 kJ mol can be registered. However, in systems able only to
form weak or medium strong HBs in solutions such investigations are scarce and, in general,
weak HBs in solutions are rather poorly explored and characterized. As the interactions as
2
weak as C≡C···HCH and C≡C···OH have been observed in solution at room temperature [8],
one may expect that the interactions, formed by 2- and 3-EP, might be amplified if additional
substituent on ethynylpyridines is introduced. With this respect, it seems interesting to
investigate the influence of additional ethynyl moiety on 2- and 3-EP, especially when
ethynyl groups are placed symmetrically regarding N-atom.
Thus, in this paper, HB capability of 2,6-diethynylpyridine (2,6-DEP) and 3,5-
ethynylpyridines (3,5-DEP) was explored (Fig. 1.). Their associations with TMP as HB
acceptor and PH as HB donor was characterized by using high dilution method [14] and
ꢀ
ꢀ
expressed in terms of K , ∆ H and ∆ S both experimentally and computationally. Spectral
c
r
r
features associated with engagement of C≡C moieties of DEPs in HB complexes, at least at
qualitative level, were used for discussing the hierarchy between HB centers in terms of their
cooperative action.
3

Fig. 1. 1) 2,6-DEP; 2) 3,5-DEP; 3) TMP; 4) PH.
2
. Calculations
Geometries of monomers 1-4 (Fig. 1), were optimized at B3LYP/6-311++G(d,p) level of
theory. Geometries of dimers DEPs···TMP and PH···DEPs (Fig. 2) were treated at the same
level of theory. HB systems require flexible [15] basis set in order to properly describe its
thermodynamic properties. Triple-zeta basis set 6-311G++(d,p) with diffuse functions on both
heavy and hydrogen atoms were chosen. On all optimized structures, harmonic frequencies
were calculated on the same level of theory as the geometry optimizations. NBO charges were
also calculated on optimized geometries [16]. Association energies were calculated as a
difference between energies of monomers and energies of a dimer. Correction for the basis set
superposition error (BSSE) was accounted for with the Boys-Bernardi counterpoise scheme
[
17]. ∆rH was calculated according to equation (1):
BSSE
trv
trv
trv
∆rH = ∆E (AB) + ZPE(AB) – ZPE(A) – ZPE(B) + H (AB) – H (A) – H (B)
(1)
BSSE
trv
where ∆E
is the association energy, corrected for BSSE, ZPE is zero point energy and H
is translational, rotational and vibrational contribution to the enthalpy at temperature T. All
harmonic ZPE are scaled with scaling factor 0.9877 [18] before using them in the calculation
of ∆ H. All calculations have been performed in Gamess program package [19-20].
r
3
. Experimental
3
.1 Chemicals
2,6-diethynylpyridine (2,6-DEP) and 3,5-diethynylpyridine (3,5-DEP) are white powders
prepared by Sonogashira rection [21]. The procedure on the compounds preparation and
purification can be found in Supporting Materials.
2 4
Tetrachloroethene (C Cl ; b. p. = 122 °C, spectroscopic grade) and trimethylphosphate
(
TMP; b. p. = 192-194 °C, 99+ % purity) are transparent liquids purchased from Acros
Organics and Sigma Aldrich, respectively, and were used as received. Phenol (PH) is white
crystalline substance obtained from Kemika (m. p. = 40-41 °C, pro analysis) and was
additionally purified by recrystalization. Densities of the studied liquids were determined by a
densitometer DMA 5000-Anton Paar.
4

3
.2 Preparation of solutions
High dilution method, used for determination of relevant thermodynamic parameters of
HB systems, is comprehensive described in the literature (for instance, see ref. [22]); more
details are also given in Supporting Materials. Here, the details on the preparation of
particular set of solution are presented.
Acidity of DEPs. Three sets of the solutions were prepared:
i) Solutions of DEPs in C
2
Cl
4
in certain concentration range (c
0
(2,6-DEP): 0.00082 mol
-
3
-3
-3
-3
dm - 0.07763 mol dm ; c
0
(3,5-DEP): 0.00066 mol dm -0.09708 mol dm ) in order to
determine molar absorption coefficient (ε ) of ≡C-H stretching at the band maximum in both
cases.
ii) a set composed of binary solutions of TMP in C
2
Cl
(TMP)=0.09816 mol dm - 1.02256 mol dm , used for the preparation of ternary solutions
set, which is DEP(s) dissolved in TMP in C Cl . In the latter, the concentration of DEPs
across ternary solutions set were held almost constant (c
sets were used for determination of association constant K
iii) a set composed of one binary solution of DEPs in C
4
in concentration range
-3
-3
c
0
2
4
-3
0
(DEPs)≈0.025 mol dm ). These two
of DEPs···TMP complex.
Cl (c (2,6-DEP)=0.01894 mol
(3,5-DEP)=0.02037 mol dm ), used for determination of temperature dependence of
ε of ≡C-H stretching band, and three ternary solutions made from DEPs and TMP in C
Cl
In 2,6-DEP···TMP complex, the concentrations were c (2,6-DEP)=0.02369 mol dm ,
c
2
4
0
-
3
-3
dm ; c
0
2
4
.
-
3
0
-3
-3
-3
-3
0
.02808 mol dm and 0.02549 mol dm while c
0
(TMP)=0.18211 mol dm , 0.60178 mol dm
-3
and 0.92479 mol dm , respectively. In 3,5-DEP···TMP complex, the concentrations were
-
3
-3
c
c
0
0
(3,5-DEP)= 0.02677 mol dm-3, 0.02248 mol dm and 0.024 mol dm while
-
3
-3
-3
(TMP)=0.1971 mol dm , 0.54035 mol dm and 0.92479 mol dm , respectively. The latter
three sets were used for determination of temperature dependence of K
c
.
Basicity of DEPs. Two sets of the solutions were prepared:
i) a set composed of binary solutions of DEPs in C
2
Cl
4
in concentration range c
0
(2,6-
-3
-3
-3
DEP)=0.0091 mol dm – 0.03984 mol dm (4 solutions); c
0
(3,5-DEP)=0.02127 mol dm –
-
3
0
.05164 mol dm (4 solutions), used for the preparation of ternary solutions set, which is PH
dissolved in DEPs in C Cl . In the latter, the concentration of PH across ternary solutions set
were held almost constant (c
determination of K of PH···DEPs complexes.
2
4
-3
0
(DEPs)≈0.005 mol dm ). These two sets were used for
c
5

ii) a set composed of one ternary solution made from PH and DEPs in C
2
Cl . In PH···2,6-
4
-3
DEP complex, the concentrations were c
0
(PH)=0.00713 mol dm and c
0
(2,6-DEP)=0.0104
-3
-3
mol dm . In PH···3,5-DEP complex, the concentrations were c
0
(PH)=0.00471 mol dm and
c
0
(3,5-DEP)= 0.02037. The latter set was used for determination of temperature dependence
of K
c
.
The concentration of each compound was determined gravimetrically, i. e. by weighting
of each compound in particular mixture up to five decimal points.
3
.3 Measurements
FTIR spectra were recorded on an ABB Bomem MB102 spectrometer equipped with CsI
optics and DTGS detector. IR spectra of solutions at room temperature (t = 26 ± 1 °C) were
recorded in sealed cells equipped with CaF windows with effective pathlengths determined
by employing Bertie's procedure using either dichloromethane (CH Cl ) or benzene (C ) as
a secondary standard [23] (2,6-DEP in C Cl : d=4812.53 µm, 1000.73 µm, 518.19 µm; 2,6-
DEP+TMP in C Cl : d=985.14 µm; PH+2,6-DEP in C Cl : d=988.83 µm; 3,5-DEP in C Cl
d=4816.47 µm, 999.94 µm, 226.33 µm; 3,5-DEP+TMP in C Cl : d=985.14 µm; PH+3,5-DEP
in C Cl : d=987.06 µm, 985.72 µm). Each sample was recorded three times; between
sequential measurements, the cell was emptied, dried with N , washed twice with the studied
solution and the data were recollected.
2
2
2
6
H
6
2
4
2
4
2
4
2
4
:
2
4
2
4
2
Temperature dependent IR spectra were recorded by using Specac 3000 Series high
stability temperature controller with heating jacket for the regulation of the heating rate. IR
2
spectra of solutions were recorded in a sealed cell with CaF windows and d=984.18 µm; the
-1
heating rate was 2 °C min and the sample was held at each temperature for 3 min. IR spectra
of solids were recorded in KBr pellets (about 2 mg of DEPs and 120 mg of KBr) with heating
-
1
rate of 1 °C min and stabilization time of 5 min. In solution samples the spectra were
collected every 10 °C between 25 °C and 115 °C, while in measurements of the pellets the
spectra were collected every 5 °C.
-1
All the spectra were obtained with 10 scans and with nominal resolution of 2 cm . 10
scans were considered to be sufficient for obtaining very smooth band features in spectral
range where ≡C-H moieties of DEPs absorb. Regarding OH stretching band of PH, the
absorbance of OH stretching band has to be determined with certain accuracy and reliability,
while, at the same time, the accumulation of water from the atmosphere which absorbs in the
same spectral range and, consequently, brings up the unreliability in the absorbance
6

measurements (Fig. S3 in Supporting Materials) should be minimized. Thus, 10 scans in the
systems related with PH is the result of a certain compromise.
3
.4 Spectral analysis
As the band maximum, i. e. the band absorbance, provides the most sensitive and the most
reliable response of HB donors (DEPs or PH) in the presence of HB acceptor (TMP or DEPs)
at given concentration, this spectroscopic parameter was aimed to be determined in all
mixtures sets. (More details on tuning optimal absorbance value at the band maximum is
given in Supporting Materials).
In order to determine the ε of ≡C-H stretching in binary (DEPs in C
Cl
2 4
) mixtures and,
-
c
2 4
consequently, K from ternary (DEPs+TMP in C Cl ) mixtures, spectral range 3360-3260 cm
1
was analyzed. After the baseline correction, the absorbance of ≡C-H stretching band of both
-1
DEPs was determined from the band maximum (3308 cm ) and used in calculation of ε from
Lambert-Beer's law and K
In PH+DEPs in C Cl
c
.
2
4
mixtures, K
c
was determined by analyzing O-H stretching band of
-1
-1
PH in spectral range 3680-3540 cm (maximum at 3610 cm ) after the baseline correction.
Additionally, C≡C stretching in IR spectra of both 2,6-DEP in C
2
Cl
4
and, respectively,
-1
3
,5-DEP in C
2
Cl
4
binary mixtures were analyzed (2155-2080 cm ). Only value ε at the band
-1
-1
maximum is estimated; for 2,6-DEP at 2125 cm and for 3,5-DEP at 2117 cm .
Mentioned quantities were measured both at room temperature and in temperature
dependence.
4
. Results and discussion
4
. 1 Calculations: structures and vibrational spectra of monomers and dimers
HB complexes of monomers 1 and 2 (Fig. 1) with 3 are formed via hydrogen bonds
between DEP ethynyl group and oxygen atoms in TMP (structures 5 and 6). In the case of
dimers 7 and 8, HBs are formed between hydroxyl group of 4 and DEP nitrogen atom (Fig.
2
). Thus, for discussing the structures of HB complexes, the most important spectral
parameters are the positions of the stretching of ≡C-H, C≡C and OH moieties, both free and
associated ones. The former originate from monomers (1, 2 and 4), while the latter from
species involved in HB complexes. Both are presented in Table 1. The main difference in
designation of the stretching of free and associated moieties is presented using the mark ···
that refers to the associated moieties. As far as monomers 2,6-DEP and 3,5-DEP are
7

concerned (structures 1 and 2 in Fig. 1), their complete vibrational analysis (gas phase
calculations, IR spectra of their solutions in C Cl , IR spectra of solid and melt 2,6-DEP and
,5-DEP) is given in Table S1 in Supporting Materials.
The structural analysis of DEPs···TMP and PH···DEPs dimers is elaborated in more
2
4
3
detail. The initial geometry for the search was taken from the already determined
conformations of dimers of TMP and 2-EP and 3-EP [8]. The most stable conformers of the
dimers 5 and 6 are shown in Fig. 2. We should note that since different conformers of TMP
exist depending on the orientation of methyl groups, we used the most stable one for the
further analysis (other conformers are located as well and are higher in energy by ca. 2-3 kJ
-1
mol ). The most significant interaction between monomers in the dimers 5 and 6 is the HB
P=O···H-C≡ which makes a significant change in vibrational frequencies of ethynyl groups.
The shift in vibrational frequencies makes this conformer detectable in vibrational spectra
(
Table 1). Aside P=O···H-C≡ HB, a secondary interaction between methyl groups of TMP
and one C≡C moiety of particular DEP is predicted. This is manifested by the low-frequency
-1
shift of the symmetric stretching of C≡C oscillators for ∆=-17 cm in 2,6-DEP···TMP (dimer
-1
5
) association and, respectively, ∆=-16 cm in 3,5-DEP···TMP (dimer 6) association. The
analogous phenomenon was previously observed in [8], where it was hypothesized that
ethynyl group of EPs, when in interaction with HB acceptor, becomes longer. The same
conclusion can be drawn in the systems studied here.
The optimized structures of the dimers PH···DEP complexes (7 and 8) areexpectedly
defined by OH···N HB (Fig. 2). Simultaneously with OH···N HB, in PH···2,6-DEP complex
(
dimer 7), an additional OH···π interaction can be theoretically formed (π is for C≡C groups).
According to the calculations conducted on this level of theory, OH group of PH is not shared
between N-atom and C≡C moieties of 2,6-DEP. Since this is in contrast with results obtained
for the complex between PH and 2-EP [8], the explanation is proposed as follows: HB
donating and HB accepting sites of monomers, although close enough, are not nearly collinear
and directional [24] due to unfavorable steric interactions imposed by two ethynyl groups. In
PH···3,5-DEP complex formation (dimer 8), we performed additional conformational analysis
in order to examine the theoretical possibility for OH···π interaction as an interaction
-1
competitive to OH···N HB. We found that this dimer is less stable than 8 by ca. 21 kJ mol
and therefore considered as not probable.
8

Fig. 2. Structures 5-8, optimized at B3LYP/6-311++G(d,p) level of theory. The most
important interactions between monomers are shown as dashed bonds and the distance
between atoms in Ångström units are shown next to it.
The analysis of NBO charges of the structure of dimers 5-8 and their illustration is given
in Supporting Materials (Figs. S6-S9).
The existence of trimers DEPs···2TMP, although possible in theory, is suspected to be
experimentally improbable. The theoretical predictions of their structures and thermodynamic
parameters, along with experimental findings that exclude their existence in studied solutions,
are presented in the Supporting Materials.
9

-
1
Table 1. Calculated fundamental vibrational modes (in cm ) for ≡C−H and C≡C stretching in DEPs monomers and DEPs dimers, as well as O-H
vibration of free PH and PH in complexes with DEPs. All frequencies are scaled with scaling factor 0.9679 [18]. Intensities are reported in
-1
parenthesis. Calculated intensities are in km mol units.
a
νꢁ
system
≡C-H
≡C-H···
C≡C
C≡C···
OH
OH···
calc
exp
calc
exp
-
calc
-
exp
calc
exp
-
calc
-
exp
-
calc
-
exp
-
2
125 (w)
3
308 (s)
3363 (125.50)
3364 (50.12)
2142 (1.23)
2141 (2.81)
1
2115(w)
102 (vw)
-
3
289 (w)
2
3
308 (s)
289 (w)
-
3363 (168.40)
3364 (38.33)
-
2117 (w)
2138 (1.18)
2136 (4.52)
-
2
4
-
-
-
-
-
-
-
-
-
-
-
-
-
-
3
2105 (vw)
-
3610
3710 (61.38)
-
2
122 (w)
b
b
b
3
308 (s)
3211-
3208
3237 (586.75)
2125 (84.74)
5
6
7
-
-
-
-
2108 (w)
-
-
-
-
2101 (w)
-
-
-
-
b
b
3
3
3
289 (w)
3365 (87.61)
2140 (0.52)
2
101 (vw)
2117 (w)
2103 (vw)
b
b
3
308 (s)
3209-
3206
3238 (688.72)
3364 (103.31)
2122 (42.38)
2110 (w)
-
-
-
-
-
b
289 (w)
2135 (7.06)
2
122 (w)
2129 (w)
2118 (w)
2108 (vw)
3
308 (s)
3308 (s)
3289(w)
3362 (131.65)
3362 (21.00)
2145 (14.71)
2146 (3.47)
c
c
2108 (w)
3610
3610
3306
3204
3402 (1134.73)
3336 (1932.51)
289 (w)
2
101 (vw)
3
308 (s)
3308 (s)
3289(w)
3360 (166.79)
3361 (12.72)
2117 (w)
2140 (0.03)
2141 (0.27)
8
2120 (vw)
3
289 (w)
2103 (vw)
c
a
-1; b
cm structures 5 and 6 (≡C-H···O and C≡C··· HCH
); structures 7 and 8 (N···HO).
2
1
0

4
. 2 IR spectra of binary mixtures at room temperature
Acidity. The most useful spectral feature for probing HB acidity of 2,6-DEP and 3,5-DEP
in C
2
Cl
4
solution is the symmetric and antisymmetric stretching of ≡C-H groups (Fig. 3a and
3
b). In IR spectra of both 2,6-DEP and 3,5-DEP solutions the corresponding normal modes
-1
-1
produce the bands at 3289 cm (weak) 3308 cm (strong), the former being assigned as
symmetric and the latter as antisymmetric stretching. In addition, there is a shoulder at high
-
1
-1
frequency side of 3308 cm band in IR spectrum of 2,6-DEP (3329 cm , very weak) which
-1
remains unassigned at the moment. Molar absorption coefficient of the band at 3308 cm of
2
,6-DEP and, respectively, 3,5-DEP dissolved in C Cl is obtained from Lambert-Beer's law
2
4
-
1
3
-1
-1
3
-1
and is ε =518 ± 4 mol dm cm , and, respectively, ε =575 ± 2 mol dm cm .
Basicity. When exposed to certain HB donor like phenol (PH), 2,6-DEP and 3,5-DEP can
serve as HB acceptors primary via N-atom on pyridine ring. In characterizing their basicity in
-1
termodynamical terms, OH signal of PH (3610 cm ) is used as a probe. Its molar absorption
-1
3
-1
coefficient in C
2
Cl is determined elsewhere [8] and is ε =235 ± 1 mol dm cm . Aside N-
4
atom, DEPs possess two C≡C groups that can act as weak HB acceptors. The antisymmetric
and symmetric stretching of C≡C groups produces, in general, rather weak and complex
-
1
-1
feature in IR spectrum of dissolved 2,6-DEP: 2102 cm (very weak), 2115 cm (weak,
-1
3
-1
-1
-1
ε =29.90 ± 0.01 mol dm cm ), 2125 cm (weak) and 2135 cm (very weak) bands (Fig.
c). In IR spectrum of 3,5-DEP solution the envelope attributed to the stretchings of C≡C
groups is somewhat simpler since it is composed from seemingly two bands; the one at 2105
3
-1
-1
cm (very weak) attributed to the antisymmetric stretching, and the other at 2117 cm (weak,
-
1
3
-1
ε =1.77 ± 0.04 mol dm cm ), attributed to the symmetric stretching (Fig. 3d). In both of the
-
1
compounds there is one very weak band at 2158 cm which remains unassigned. Aside to the
appearance of multiband structure in 2,6-DEP solution, the C≡C stretching bands are, in both
cases, rather weak, as indicated from ε values given in parenthesis. Thus, these signals will
be used only as diagnostic tools for assessment whether the C≡C moieties also participate in
HB formation and C≡C will be free from any quantitative analysis.
Since thermodynamic parameters are to be determined only from the maximum of the
-1
band at 3308 cm , we will not discuss eventual mechanical and / or electrical anharmonicites
that appear in IR spectra of studied compounds. This implies that thorough interpretation of
all the signals observed in the spectral range where only ≡C-H and C≡C stretching, symmetric
and antisymmetric is expected, is not the aim of this paper.
4
.3 Temperature dependent IR spectra of binary mixtures
1
1

Acidity. In the temperature interval 25-115 °C the most intensive band originated due to
-1
≡
C-H stretching (Fig. 3a and 3b), the one at 3308 cm (25 °C) in both 2,6-DEP and 3,5-DEP
solutions, decrease in intensity and shift to the higher wavenumbers; at 115 °C it absorbs at
-1
-1
-
3
311 cm in 2,6-DEP solution, while in 3,5-DEP solution at 3310 cm . The band at 3289 cm
1
(25 °C) in IR spectra of both of the compounds moves to the higher wavenumbers with the
-
1
-1
temperature increase (3292 cm at 115 °C), although, in contrast to the 3308 cm band, it
gains intensity in temperature range 25-115 °C. In 2,6-DEP solution the additional band at
-
1
3
329 cm diminishes immediately when heated above 25 °C. Temperature dependence of
-
1
molar absorption coefficient at 3308 cm is given by the equation
o
-1
-1
3
-1
-1
3
-1 o -1
o
o
ε
(
t / C, 3308 cm
)
=
(
532 ±1mol dm cm
while corresponding
572 ± 2 mol dm cm
)
−
(
1.56 ± 0.02 mol dm cm
C
)
⋅
(
t / C − 25 C
is
t / C − 25 C
)
for
2,6-DEP,
expression
for
3,5-DEP
o
-1
-1
3
-1
-1
3
-1 o -1
o
o
ε
(
t / C, 3308 cm
)
=
(
)
−
(
1.63± 0.02 mol dm cm
C
)
⋅
(
)
In the above equations, t stands for the temperature expressed in °C (if it were expressed in
Kelvins, then T would stand).
The signals attributed to the C≡C moieties in both compounds (Fig. 3c and 3d) in general
decrease upon heating, although certain changes in band shapes and in relative intensities can
-
1
be observed. In 2,6-DEP solution spectra the band at 2102 cm observed at 25 °C vanishes
-1
-1
-1
when heated, while the bands at 2115 cm , 2125 cm and 2135 cm at 25 °C shift to 2112
-1
-1
-1
cm , 2123 cm and 2134 cm at 115 °C. The bands also became broader and, consequently,
the splitting between them is not so pronounced at 115 °C as at 25 °C. In 3,5-DEP in C Cl
2
4
-1
-1
solution at 25 °C, two bands at 2105 cm and at 2117 cm appear; the former, being weaker
than the latter, generates one broad asymmetric band at 115 °C with the asymmetry
-1
pronounced on low-frequency side and the center estimated at 2108 cm . In both of the
-
1
-1
compounds the band at 2158 cm (25 °C) shifts to 2157 cm (115 °C).
1
2

Fig. 3. Temperature study (25-115 °C) of DEPs in C
2
Cl
4
2 4
, after C Cl subtraction: a) ≡C-H
-1
-3
stretching (3360-3260 cm ) of 2,6-DEP (c
0
=0.01894 mol dm ); b) ≡C-H stretching (3360-
-1
-3
-1
3
260 cm ) of 3,5-DEP (c
=0.02037 mol dm ); c) C≡C stretching (2145-2080 cm ) of 2,6-
0
-3
-1
DEP (c
0
=0.01894 mol dm ); d) C≡C stretching (2145-2080 cm ) of 3,5-DEP (c
=0.02037
0
-3
mol dm ).
Basicity. Temperature dependence of O-H stretching band of PH in C
2
Cl , the purpose of
4
which is to exerts HB accepting abilities of 2,6-DEP and 3,5-DEP, is described in our
previous paper [8] and can be presented with the equation
o
-1
-1
3
-1
-1
3
-1 o -1
o
o
ε
(
t / C, 3610 cm
)
=
(
271±16 mol dm cm
)
−
(
1.25± 0.32 mol dm cm
C
)
⋅
(
t / C − 25 C
)
4
.4 IR spectra of ternary mixtures at room temperature
Acidity. Upon mixing DEPs and TMP, the first and obvious evidence that HB is formed it
the appearance of new band at low-frequency side of ≡C-H stretching bands, with intensity
and position being dependent on the concentration of TMP (Fig. 4a and 4b). In concentration
-
3
-3
range 0.09816 mol dm ≤ c
0
(TMP) ≤ 1.02256 mol dm the band attributed to the stretching
-1
-1
of ≡C-H···O oscillator shifts from 3211 cm to 3208 cm in 2,6-DEP+TMP in C
2
4
Cl mixture,
-
1
-1
while from 3209 cm to 3206 cm in 3,5-DEP+TMP in C
2 4
Cl mixture. The positions of
-
1
-
-
symmetric and antisymmetric ≡C-H stretching bands remain unchanged (3289 cm , 3308 cm
1
1
-1
-1
and 3329 cm in 2,6-DEP+TMP in C
in 3,5-DEP+TMP in C
2
Cl mixture and, respectively, 3289 cm and 3308 cm
4
2
Cl4 mixture), while their intensities are, as expected, reduced with an
increase in TMP's concentration. The presence of TMP in both of the compounds induced the
1
3

changes in C≡C stretching region as well (Fig. 4c and 4d), but it has to emphasized that they
2 4 2 4
are more distinguished when TMP in C Cl mixture is subtracted from DEPs+TMP in C Cl
mixtures.
2 4
First, we shall discuss 2,6-DEP+TMP in C Cl mixtures. Gradual increase in TMP's
-1
concentration causes two phenomena: i) one additional band at about 2101 cm appears and
gains intensity as TMP is getting more concentrated. Consequently, very weak band at 2102
-1
-1
-1
cm merges with the newly formed band; ii) The band at 2115 cm shifts to 2108 cm , the
-1
-1
-1
band at 2125 cm shifts to 2122 cm , while the band at 2135 cm remains on the same
-
1
position. In 3,5-DEP+TMP in C
2 4
Cl mixtures new band appears at about 2110 cm and, as
the concentration of TMP increases, it rapidly gains intensity so that it exceeds by far the
-1
-1
bands at 2105 cm and 2117 cm , where the former exhibits slight shift to the lower
-
1
wavenumbers (2103 cm ). Due to the position of the newly formed band and the fact that it is
much greater than other two bands, it is hard to estimate whether symmetric stretching of
-1
C≡C moieties (2117 cm ) shifts or remains at the same position. Overall, the low-frequency
shift of C≡C stretching when involved in HB complex indicates the shortening of C≡C when
DEPs interact with TMP as HB donors [8].
Fig. 4. TMP's concentration dependence (at 26±1°C) of: a) ≡C-H and ≡C-H··· stretching of
-
1
2
,6-DEP in a 2,6-DEP+TMP in C
stretching of 3,5-DEP in a 3,5-DEP+TMP in C
C≡C··· stretching of 2,6-DEP in a 2,6-DEP+TMP in C
and C≡C··· stretching of 3,5-DEP in a 3,5-DEP+TMP in C
2
Cl
mixture (3360-3160 cm ); b) ≡C-H and ≡C-H···
4
-1
2
Cl
mixture (3360-3160 cm ); c) C≡C and
4
-1
2
4
Cl mixture (2145-2080 cm ); d) C≡C
-
1
2
4
Cl mixture (2145-2080 cm ). IR
1
4

spectra of a) and b) are obtained after C
subtraction.
2
Cl
4
subtraction, while c) and d) after TMP+C
2
Cl
4
According to the calculations, both free ≡C-H and associated ≡C-H···O stretching is
shifted to the higher wavenumbers when compared to experimental data (Table 1). Although
-1
the magnitude of change (∆≈-30 cm ) is much smaller than the observed change in
-1
experiment (∆≈-100 cm ), .the observed shift in wavenumbers corresponds to formation of
dimers 5 and 6. The difference in C≡C and C≡C···stretching bands, although smaller in
-1
-1
predicted (∆≈-16 cm ) than experimentally obtained spectra (∆≈-25 cm ), also corresponds
to the formation of?dimers 5 and 6.
Basicity. The differences between PH+2,6-DEP in C
the sum of PH in C Cl and 2,6-DEP in C Cl mixtures, although being rather subtle, are still
enough to serve as conclusive evidence on the HB formation between PH and 2,6-DEP (Fig.
a); there is the rise of one really broad and shallow band with the maximum that almost
2 4
Cl mixtures, when compared with
2
4
2
4
5
-
1
coincides with the band maximum of ≡C-H stretching bands (~ 3306 cm ) and the
appearance of the bands attributed to the stretchings of C≡C··· oscillators shifted with respect
-
1
-1
to the bands of monomers 2,6-DEP (from 2102 to 2108 cm , from 2115 to 2118 cm , from
-
1
-1
2
125 to 2129 cm and from 2135 to 2140 cm ) (Fig. 5c), while their relative intensities
remain the same. In both of the phenomena a concentration dependence of intensities is
observed. The shift in frequencies of C≡C··· vibrations can be attributed to the charge
redistribution caused by the interaction between C≡C and phenyl groups [25]. In PH+3,5-DEP
in C
2
Cl mixtures (Fig. 5b), the maximum at the OH···N stretching band is estimated to be at
4
-1
-1
about 3204 cm so, according to the shift of almost 100 cm wavenumbers in lower spectral
range, one can assume that PH and 3,5-DEP form stronger HB that PH and 2,6-DEP.
Optimized structures (Fig. 2) indeed show shorter HB in the dimer 8 than in the dimer 7.
The lack of any other interactions between monomers in 8 enables ideal relative orientation
between PH and 3,5-DEP. This may be expected when taking into account the steric
hindrance of N-atom in 2,6-DEP.. Furthermore, simultaneous engagement of PH's OH group
in OH···N and OH···C≡C HBs is not expected due to spatial limitations, i.e. due to too large
distance between N-atom and ≡C-H groups in the positions 3 and 5. However, according to
the IR spectra in Fig. 5d, with the rise in concentration of 3,5-DEP, one very weak and broad
-
1
feature emerges at about 2120 cm . This phenomenon implies that ≡C-H group, regarding its
weakness in HB accepting capability in comparison with N-atom of pyridine ring, probably
form C≡C···OH complexes but not as secondary interaction. The number of such HB
1
5

complexes is certainly by far below the number of OH···N complexes, but they still live long
enough to be captured by IR spectroscopy and in certain amount. Unfortunately, due to
-
exceptionally low ε value of even C≡C stretching band profile in the maximum (ε (2117 cm
1
-1
3
-1
)
= 1.77 ± 0.04 mol dm cm ), it is not very likely to expect that at given experimental
conditions (the room temperature and the medium) the feature assigned as C≡C··· stretching
band could be better resolved. In both cases, however, the observed high-frequency shift of
C≡C··· oscillator goes along with the assumption that in PH···DEPs complexes the respective
moiety lengthens when DEPs play role of HB acceptors [8, 25].
Fig. 5. PH's concentration dependence (at 26±1°C) of: a) OH and OH··· stretching of 2,6-
-1
DEP in a PH+2,6-DEP in C
,5-DEP in a PH+3,5-DEP in C
of 2,6-DEP in a PH+2,6-DEP in C
stretching of 3,5-DEP in a PH+3,5-DEP in C
and b) are obtained after C Cl subtraction, while c) and d) after DEPs+C
2
Cl mixture (3680-2965 cm ); b) OH and OH··· stretching of
4
-1
3
2
Cl
mixture (3680-2965 cm ); c) C≡C and C≡C··· stretching
4
-1
2
4
Cl mixture (2145-2085 cm ); d) C≡C and C≡C···
-
1
2
Cl
4
mixture (2145-2085 cm ). IR spectra of a)
Cl subtraction.
2
4
2
4
4
.5 Temperature dependent IR spectra of ternary mixtures
-
3
Acidity. When mixture of 2,6-DEP+TMP in C
2
Cl
(TMP)= 0.92479 mol dm ) is heated from 25 °C to 115 °C, the most important spectral
changes are related to the stretching of ≡C-H oscillator, both free and associated (Fig. 6a). At
4
(c
0
(2,6-DEP)=0.02549 mol dm ,
-
3
c
0
-1
2
5 °C, the band at 3208 cm , attributed to the stretching of ≡C-H··· oscillator, shifts to 3216
-1
cm and continuously decrease intensity when heated up to 115 °C. At the same time, the
-1
-1
-1
-1
bands at 3289 cm and 3308 cm shifts to, respectively, 3292 cm and 3310 cm . Aside the
1
6

shift of the band center, another phenomenon that should be discussed is the gain of their
intensities when heated up to 85 °C; above that temperature the former keeps increasing,
-
1
while the latter starts to decrease. The band at 3329 cm , as when TMP is absent, vanishes
above 25 °C.
-3
In a 3,5-DEP+TMP in C
2
Cl
4
mixture (c
0
(3,5-DEP)=0.024 mol dm , c
0
(TMP)= 0.92479
-3
-1
-1
mol dm ), ≡C-H··· oscillator shifts from 3206 cm (25 °C) to 3214 cm (115 °C)
simultaneously with an intensity reduction (Fig. 6b). The shift of ≡C-H oscillators, from 3289
-1
-1
-1
-1
cm and 3308 cm at 25 °C to 3292 cm to 3310 cm at 115 °C, is accompanied by
continuous increase of the former band and the redistribution of intensity of the latter. As in
2 4
the case of 2,6-DEP+TMP in C Cl mixture, up to certain temperature (here 65 °C) there is
gain in intensity, while at temperatures above that it decreases.
Regarding C≡C stretching bands in 2,6-DEP+TMP in C
2
Cl
4
mixture, the most
distinguished change is associated with the decrease of C≡C··· stretching band(s) (Fig. 6c)
-1
-1
-1
and their shift from 2101 cm (25 °C) to 2099 cm (115 °C). The bands at 2108 cm , 2122
-1
-1
cm and 2135 cm absorb at the same wavenumbers in the whole temperature interval and
gradually decrease in intensity.
3
,5-DEP+TMP in C
2
Cl
mixture the bands attributed to the stretchings of C≡C and C≡C···
4
oscillators form rather asymmetric band (Fig. 6d), with asymmetry being changed from 25 °C
to 115 °C. One might argue that, at analytical concentrations of 3,5-DEP and TMP, two band
-1
-1
could be distinguished at 25 °C, the one at 2105 cm and another at 2110 cm . When the
temperature reaches 115 °C, one rather broad band with slight asymmetry on its low-
-1
frequency side appears at 2109 cm .
1
7

Fig. 6. Temperature study (25-115 °C) of DEPs+TMP in C
2
Cl
4
mixture: a) ≡C-H and ≡C-H···
-1
stretching (3360-3160 cm ) of 2,6-DEP; b) ≡C-H stretching and ≡C-H··· stretching (3360-
-1
-1
3
160 cm ) of 3,5-DEP ; c) C≡C stretching (2145-2080 cm ) of 2,6-DEP; d) C≡C stretching
-1
(
2145-2080 cm ) of 3,5-DEP. IR spectra of a) and b) are obtained after C
while c) and d) after TMP+C Cl subtraction. Initial concentrations in a) and c) were c
DEP)=0.02549 mol dm and c
2
Cl
4
subtraction,
(2,6-
(3,5-
2
4
0
-3
-3
0
(TMP)=0.92479 mol dm , while in b) and d) c
0
-
3
-3
DEP)=0.024 mol dm and c
0
(TMP)=0.92479 mol dm .
Basicity. Since 2,6-DEP and 3,5-DEP are not very soluble in C
higher analytical concentrations so that temperature dependence of K
explored, they will be described only qualitatively. In both cases, when PH+DEPs in C
mixtures are heated in the temperature interval 25-115 °C (Fig. 7a and 7b), the broad feature
2
Cl
can be more thorough
Cl
4
and, hence, prevent
c
2
4
-1
centered at about 3308 cm , vanishes at the temperatures above 65 °C, as well as the
envelope attributed to the stretching of C≡C··· oscillators (Fig. 7c). In 3,5-DEP in C
Cl
mixtures (Fig. 7d) this band is very weak and disappears immediately when heated above 25
C.
2
4
°
Fig. 7. Temperature study (25-115 °C) of PH+DEPs in C
stretching of PH in a PH+2,6-DEP in C Cl mixture (3680-2965 cm ); ); b OH and OH···
stretching of OH in a PH+3,5-DEP in C Cl
stretching of 2,6-DEP in a PH+2,6-DEP in C
C≡C··· stretching of 3,5-DEP in a PH+3,5-DEP in C
spectra of a) and b) are obtained after C Cl subtraction, while c) and d) after DEPs+C
subtraction. Initial concentrations in a) and c) were c
2 4
Cl mixture: a) OH and OH···
-
1
2
4
-1
2
4
mixture (3680-2965 cm ); c) C≡C and C≡C···
-1
2
4
Cl mixture (2145-2080 cm ); d) C≡C and
-1
2
Cl
4
mixture (2145-2080 cm ). IR
Cl
0
(PH)=0.00713 mol dm and c (2,6-
2
4
2
4
-3
0
1
8

-3
-3
DEP)=0.0104 mol dm , while in b) and d) c
0
(PH)=0.00471 mol dm and c (3,5-
0
-
3
DEP)=0.02037 mol dm .
. 6 Determination of thermodynamical parameters
When discussing theoretically determined thermodynamic parameters, one has to consider
4
that all calculations were done on isolated molecules, without the influence of solvent
molecules. These conditions are more appropriate for isolated molecules in vacuum and some
thermodynamic quantities should be different if determined for reactions in vacuum. For
example, entropy of reactions in great deal depends on the rotational and translational degrees
of freedom of reactants and products and interaction with solvent molecules can limit these
degrees of freedom. Nevertheless, it is informative to report calculated thermodynamic
parameters, since in this case weak interactions are not influenced by solvent molecules.
As the dimer 5 have two intermolecular interactions, whereas the dimer 6 has only one,
respectively, the lower stabilization energy and enthalpy in 5 is observed when compared with
6
. Reaction enthalpy for association of dimer 8 (Table 2) is lower than the enthalpy of
association of structurally similar dimer 8, due to the stronger hydrogen bond between its
constituents (Figure 2).
Calculated reaction entropy (in vacuum) (Table 2) is much more negative than the
experimental result. The large difference can be explained with the interaction of solute
molecules with the molecules of solvent. That interaction makes translational and rotational
contributions to entropy (that are larger in monomers) lower since the interaction with the
solvent restricts these degrees of freedom.
Table 2. Energies (BSSE values), enthalpies and entropies at 298.15 K for HB complexes of5-
8
in the gas phase. Values are calculated at the B3LYP6-311++G(d, p) level of theory.
a
a
a
b
system ∆E
∆
r
H ∆
r
G
r
∆ S
5
6
7
8
-17.4 -7.9 25.1 -110.5
-16.9 -6.5 24.5 -104.2
-27.6 -16.4 19.6 -121.0
-29.4 -17.6 15.7 -111.8
1
9

a
-1
b
-1
-1
In kJ mol ; In J K mol .
Assuming the 1:1 stoichiometry in both DEPs···TMP and PH···DEPs complexes,
association constants K were determined from the relationship
c
Af
c0 (A) −
d ⋅ε
c (B) − c (A) +
K =
(2)
c
Af

A 
f
⋅


0
0

d ⋅ε
d ⋅ε

In expression (2), Af designates the band absorbance of unassociated HB donor (A)
estimated in ternary mixtures (A+B in C Cl ). In particular, when DEPs+TMP in C Cl are to
2
4
2
4
-
1
be analyzed, the absorbance of ≡C-H stretching at 3308 cm (for both DEPs) is to be
measured in dependence of the concentration of DEPs (A) and TMP (B). When PH+DEPs in
-
1
C
2
Cl
4
mixtures are concerned, the absorbance of OH stretching (3610 cm ) is measured as the
function of concentration of PH (A) and DEPs (B). ε is molar absorption coefficient of ≡C-H
-1
-1
stretching of DEPs (3308 cm ) and, respectively, OH stretching of PH (3610 cm ) bands
-1
3
-1
determined from binary mixtures measurements (ε=518 ± 4 mol dm cm for 2,6-DEP,
-1
3
-1
-1
3
-1
ε =575 ± 2 mol dm cm for 3,5-DEP and ε =235 ± 1 mol dm cm for PH in, respectively,
Cl ). c (A) and c (B) are analytical concentrations of HB donor (DEPs or PH) and,
C
2
4
0
0
respectively, HB acceptor (TMP or DEPs). The procedure for determination of the absorbance
of the analyzed bands is described in detail in Section 3.4, while more details on the
relationship between Lambert-Beer’s law and K are given in Supporting Materials and in ref.
c
[
23].
Experimentally obtained values for K
solution set ii) in Section 3.2) and temperature dependent measurement (solution set iii) in
c
, from both room temperature measurements
(
ꢀ
ꢀ
ꢀ
ꢀ
Section 3.2), employing the relations ∆ G = −RT ln K and ∆ G = ∆ H −T∆ S , are
r
c
r
r
r
presented in Table 3.
Table 3. Thermodynamic quantities of DEPs···TMP and PH···TMP complexes obtained
from IR spectra at 26±1 °C.
a, d
a, e
ꢀ
b, d
ꢀ
b, e
ꢀ
c, e
ꢀ
b, e
system
K
c
K
c
−
∆ G
−∆ H
−∆rS
−∆ G
r
r
r
5
0.81±0.09 -0.5±0.3
8.6±0.8
28±4
-0.2±0.9 0.9±.0.4
2
0

a, d
ꢀ
b, d
ꢀ
b, e
ꢀ
c, e
ꢀ
b, e
a, e
system
K
c
K
c
−
∆ G
−∆ H
−∆rS
−∆ G
r
r
r
6
7
0.89±0.07 -0.3±0.2
5.7±0.3
36±5
17.6±0.9
103±15
27±2
0.4±0.4
5±7
1.2±0.2
7±20
5±3
5±2
4±1
4±1
8
15.3±0.8
7±1
17±9
a
-1
3
b
-1
c
-1
-1
d
In mol dm at 26±1 °C; In kJ mol ; In J K mol ; from concentration dependent
e
measurements at room temperature; from temperature dependent measurements.
Let us begin with the interpretation of thermodynamic parameters for DEPs+TMP in
C
2
Cl
4
mixtures. According to K values obtained from the measurements at room temperature,
c
-
1
3
DEPs ···TMP complexes are weak (K
c
(2,6-DEP···TMP)=0.81±0.09 mol dm , K
DEP···TMP)= 0.89±0.09 mol dm ) and can be considered as equal. Moreover, they are
almost equal to the corresponding values of monosubstituted EPs [8] (K (2-
(3-EP···TMP)=0.74±0.05 mol dm ), implying that the
c
(3,5-
-
1
3
c
-1
3
-1
3
EP···TMP)=0.82±0.06 mol dm , K
c
additional ethynyl moiety on ethynylpyridines does not influence its HB donating properties.
ꢀ
The differences between the K
c
and, consequently, ∆ G values measured at room
r
temperature and those obtained from temperature dependent measurements in DEPs···TMP
complexes are explained as follows: i) when applying high dilution method in studying HB
ꢀ
-1
systems, the uncertainty in the ∆ H values of HB can be up to ± 2.1 kJ mol [26]. Although
r
this finding does not present significant obstacle when medium strong or strong HBs are to be
ꢀ
-1
characterized ( −∆ H ≈ 17-170 kJ mol ), in the analysis of weak HBs, as are in the systems
r
ꢀ
-1
ꢀ
-1
studied here ( ∆ H = -8.6±0.8 kJ mol for 2,6-DEP···TMP and ∆ H = -5.7±0.3 kJ mol for
r
r
3
,5-DEP···TMP), this can be rather serious drawback since it counts 20-30 % of the
ꢀ
ꢀ
magnitude of measured ∆ H ; ii) ∆ G values for complex 6, obtained from room
r
r
temperature measurements and from single solution in temperature dependent regime,
-1
3
although of opposite signs, in both cases are, expectedly (for K ~ 1 mol dm ), around zero
c
ꢀ
d
-1
ꢀ
e
-1
d
e
(
∆ G
= +0.3 kJ mol and ∆ G
= -0.4 kJ mol . The superscripts and designate
r
r
different measurements sets, as given in Table 3.). The magnitude of their range
ꢀ
ꢀ
d
ꢀ
e
-1
(
∆ ∆ G ≡∆ G - ∆ G = 0.7 kJ mol ) is still smaller than the average thermal energy in
r r r
2
1

-
1
solution which counts up to 2 kJ mol . In turn, in both cases, K
c
determined from temperature
obtained from concentration
dependent measurements is considered as less reliable than K
c
study at room temperature.
As far as PH···DEPs complexes are concerned, certain predictions regarding K
c
’s orders
of magnitude can be made. For HB complexes of PH and pyridines substituted with electron
-1
3
withdrawing groups such as F, Cl or CN in CCl
27-30]. As CCl is found to be very similar to C
values are expected to be found in this interval as well. For obtaining as accurate as possible
values of these orders of magnitude, it is of utmost importance that HB acceptors (DEPs)
4
, K
c
values span the range 5-15 mol dm
[
4
2
Cl
4
[31], in the systems studied here, K
c
K
c
-3
-3
are of concentrations 0.01 mol dm - 0.1 mol dm , if not even higher [32]. Unfortunately, due
to the low solubility of DEPs in C
2
Cl
4
, it was impossible to make their solutions with
-
3
-3
concentrations higher than 0.04 mol dm for 2,6-DEP and 0.05 mol dm for 3,5-EP. Thus, K
c
2 4
values, determined only from 4 solutions of PH+2,6-DEP in C Cl mixtures and, respectively,
-
1
3
PH+3,5-in C Cl mixtures will be reported and they are K
2
4
c
(PH···2,6-DEP)= 5±3 mol dm
-1
3
ꢀ
ꢀ
and K
c
(PH···3,5-DEP)= 5±2 mol dm . Their ∆ H values are, respectively, ∆ H (PH···2,6-
r
r
-1
ꢀ
-1
DEP)= -36±5 kJ mol and ∆ H (PH···3,5-DEP)= -15.3±0.8 kJ mol . Additional obstacle in
r
obtaining more precise values of thermodynamic parameters is the analytical concentration of
PH in ternary mixtures. As PH undergoes self-association at concentrations higher that 0.1
-3
-3
mol dm [33], although some authors reported even as low as 0.003 mol dm [34], the most
-3
often used analytical concentration of PH in studies concerned HB was about 0.005 mol dm
356] and this was also the case in this research. However, when collecting IR spectra of PH,
[
either in binary or ternary mixtures, the OH signal is at the limit of recommended interval of
absorbance values (see Supporting Materials, Fig. S2) and, thus, of poor reproducibility. In
turn, great uncertainty is generated in these measurements; this is reflected in the values of
thermodynamic parameters for complexes 7 (PH···2,6-DEP) and 8 (PH···3,5-DEP) and,
respectively, their uncertainties. Additional issue that naturally arises in investigation of HBs
produced by OH oscillators is their overlap with OH stretching of water which is always
present in measurements, regardless to the fact that background spectrum was collected before
collecting the IR spectrum of each sample.
However, one other interesting phenomenon rises related to the involvement of C≡C
moieties of DEPs in HBs. C≡C moieties, according to their IR spectra, certainly participate in
PH···2,6-DEP complex. This is not surprised since, sterically, if PH's OH group associates
with N-atom, simultaneous interaction with C≡C groups is straightforward. In PH···3,5-DEP
2
2

complex, on the contrary, simultaneous engagement of OH group of PH with both N-atom
and C≡C moieties is disabled due to spatial arrangement. Interestingly, the bands attributed to
the C≡C stretching of 3,5-DEP, when PH is added and its concentration is gradually
increased, and shifted to higher wavenumbers. This observation indicates that C≡C groups
participate in HB and, in contrast to our calculations, likely compete for OH group with N-
atom of 3,5-DEP. This is certainly different situation than in PH···3-EP complex, in which
this phenomenon has not been observed. Although the absence of this phenomenon in the
2

dµ 
dQ
latter may be explained by small value of
of C≡C stretching, it is also possible that




the appearance of this phenomenon in PH···3,5-DEP complex is a result of the cooperativity
of two ethynyl moieties. In effect, it can be considered that C≡C groups, together with OH
group, form some kind of resonance-assisted HB (RAHB), as proposed by Gilli et al. [36].
Although they based this model on strong and medium strong HBs, due to electron flow
through triple bond system, mediated by phenyl (pyridine) ring, it may be that the strength of
mutual interaction by far overcomes the solitary C≡C-H···OH interaction (Fig. 8).
Fig. 8. Possible resonance-assisted HB in a) PH···3,5-DEP complex in contrast to b) PH···3-
EP complex.
Due to the weakness of the observed band it is not possible to quantify the magnitude of
the proposed effect and, in turn, characterize the cooperativity in numeric terms. However, it
is very likely that this phenomenon may be more thorough investigated and characterized by
using other spectroscopic techniques , such as IR UV double resonance spectroscopy or at
least matrix isolated IR, that are more efficient in both detection and subsequent
2
3

characterization of weak HBs. The employment of explicit solvation computational models,
along with the more sensitive spectroscopic techniques, is also highly recommended.
Conclusions
Hydrogen bonding properties of 2,6- and 3,5-DEP were explored by IR spectroscopy and
were further supported by computational approach. High dilution method was used for
ꢀ
determination of thermodynamic parameters, K
c
and ∆ H , that characterize the associations
r
2 4
of DEPs with TMP and, respectively, DEPs with PH molecules in C Cl . In the former case it
is assumed that the association stoichiometry is 1:1 and the corresponding parameters, K and
c
ꢀ
∆r H , for 2,6-DEP···TMP and, respectively, 3,5-DEP···TMP HB systems are 0.81±0.09
-1
3
-1
-1
3
-1
mol dm and -8.6±0.8 kJ mol and, respectively, 0.89±0.09 mol dm and -5.7±0.3 kJ mol .
ꢀ
-1
3
In PH···DEPs HB complexes, with 1:1 stoichiometry, K
c
and ∆ H are 5±3 mol dm and -
r
-1
-1
3
-1
3
6±5 kJ mol for PH···2,6-DEP, while 5±2 mol dm and -15.3±0.8 kJ mol for PH···3,5-
DEP. In all associations, regardless to the HB donating or HB accepting role of DEPs, C≡C
moiety was found to participate in the HB complex formation. This was shown to be of
exceptional importance in PH···3,5-DEP complexes, were simultaneous involvement of OH
group in both OH···N and OH···C≡C association is disabled due to spatial arrangement.
Therefore, it is postulated that cooperativity between two ethynyl groups covalently linked to
a resonance system, plays an important role in augmentation of HB accepting strength of C≡C
moieties of 3,5-DEP.
The calculations were conducted at the B3LYP/6-311++G(d,p) level of theory in the gas
phase and are supporting experimental data. The most stable conformers in gas phase are
characterized by HB formed via ≡C-H···O interactions, whereas dimers between phenol and
diethynylpyridines are formed only through OH···N HB interaction. Both experimental and
calculated values excluded the existence of trimers, i. e. DEPs···2TMP hydrogen bonded
complexes.
Acknowledgements
This work was supported by a grant No. 0982904-2927 from Ministry of Science, Education
and Sport of the Republic of Croatia. D. V. thanks to J. Alerić and T. Parlić-Risović from
Croatian Metrology Institute for measuring the densities of trimethylphosphate and
2
4

tetrachloroethene. M. V. acknowledges Croatian Academy of Sciences and Arts for financial
support.
Supporting Materials
Supporting Materials associated with this paper can be found on the online version at
http://.
References:
[
1] S. Maity, M. Guin, C. Singh, G. Naresh Patwari, Phenylacetylene: A Hydrogen Bonding
Chameleon, ChemPhysChem 12 (2011) 26-46.
2] P. Chandra Singh, B. Bandyopadhyay, G. Naresh Patwari, Structure of the
[
Phenylacetylene-Water Complex as revealed by Infrared-Ultraviolet Double Resonance
Spectroscopy, J. Phys. Chem. A 112 (2008) 3360-3363.
[
3] R. Sedlak, P. Hobza, G. Naresh Patwari, Hydrogen-Bonded Complexes of
Phenylacetylene with Water, Methanol, Ammonia, and Methylamine. The Origin of Methyl
Group-Induced Hydrogen Bond Switching, J. Phys. Chem. A, 113 (2009) 6620-6625.
[
4] P. Chandra Singh, G. Naresh Patwari, IR-UV Double Resonance Spectroscopic
Investigation of Phenylacetylene-Alcohol Complexes. Alkyl Group Induced Hydrogen Bond
Switching, J. Phys. Chem. A 112 (2008) 5121-5125.
[
5] S. Maity, G. Naresh Patwari, Hydrogen Bonding to Multifunctional Molecules:
Spectroscopic and ab Initio Investigation of Water Complexes with Fluorophenylacetylenes,
J. Phys. Chem. A 113 (2009) 1760-1769.
[
6] S. Maity, D. K. Maity, G. Naresh Patwari, Interaction of Alcohols with 2-Fluoro- and 4-
Fluorophenylacetylenes: Infrared-Optical Double Resonance Spectroscopic and
Computational Investigation, J. Phys. Chem. A 115 (2011) 11229-112237.
[
7] S. Maity, G. Naresh Patwari, Hydrogen Bonding to Multifunctional Molecules:
Spectroscopic and ab Initio Investigation of 4-Ethynylbenzonitrile-(Water)1-3 Complexes, J.
Phys. Chem. A 114 (2010) 8337-8344.
[
8] D. Vojta, I. Matanović, G. Kovačević, G. Baranović, The study of secondary effects in
vibrational and hydrogen bonding properties of 2- and 3-ethynylpyridine and ethynylbenzene
by IR spectroscopy, Spectrochim. Acta A 132 (2014) 215-224.
[
9] C. A. Hunter, H. L. Anderson, What is Cooperativity?, Angew. Chem. Int. Ed. 48 (2009)
488-7499.
7
2
5

[
10] T. F. A. de Greef, M. M. J. Smulders, M. Wolffs, A. P. H. J. Schenning, R. P. Sijebesma,
E. W. Meijer, Supramolecular Polymerization, Chem. Rev. 109 (2009) 5687-5754.
11] A. Camara-Campos, D. Musumeci, C. A. Hunter, S. Turega, Chemical Double Mutant
Cycles for the Quantification of Cooperativity in H-Bonded Complexes, J. Am. Chem. Soc.
31 (2009) 18518-18524.
12] I. A. W. Filot, A. R. A. Palmans, P. A. J. Hilbers, R. A. v. Santen, E. A. Pidko, T. F. A.
[
1
[
de Greef, Understanding Cooperativity in Hydrogen-Bond-Induced Supramolecular
Polymerization: A Density Functional Study, J. Phys. Chem. B 114 (2010) 13667-13674.
[
13] M. Roman, C. Cannizzo, T. Pinault, B. Isare, B. Andrioletti, P. v. d. Schoot, L.
Boutieller, Supramolecular Balance: Using Cooperativity to Amplify Weak Interactions, J.
Am. Chem. Soc. 132 (2010) 16818-16824.
[
14] E. M. Arnett, L. Joris, E. Mitchell, T. S. S: R. Murty, T. M. Gories, P. v. R. Schlayer,
Studies of Hydrogen-Bonded Complex Formation III. Thermodynamics of Complexing by
Infrared Spectroscopy and Calorimetry, J. Am. Chem. Soc. 92 (1970) 2365-2377.
[
15] S.-i- Kawahara, T. Uchimaru, Basis set effect on hydrogen bond stabilization energy
estimation of the Watson-Crick type nuclei acid base pairs using medium-size basis sets:
single point MP2 evaluations at the HF optimized structures, Phys. Chem. Chem. Phys. 2
(
[
2000) 2869-2872.
16] A. E. Reed, R. B. Weinstock, F. Weinhold, Natural Population Analysis, J. Chem. Phys.
8
3 (1985) 735-746.
17] S. F. Boys, F. Bernardi, The calculations of small molecular interactions by the
differences of separate total energies. Some procedures with reduced errors, Mol. Phys. 19
[
(
[
1970) 553-566.
18] M. P. Andersson, P. Uvdal, New scale factors for harmonic vibrational frequencies using
the B3LYP density functional method with the triple-zeta basis set 6-311+G(d,p), J. Phys.
Chem. A 109 (2005) 2937-2941.
[
19] M.W. Schmidt, K. K. Baldridge, J. A. Boatz, S.T. Elbert, M. S. Gordon, J. J. Jensen, S.
Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L.Windus, M. Dupuis, J. A. Montgomery, J.
Comput. Chem. 14 (1993) 1347-1363.
[
(
20] M. S. Gordon, M. W. Schmidt in: C. E. Dykstra, G. Frenking, K. S. Kim,G. E. Scuseria
Eds),, Theory and Applications of Computational Chemistry, the first forty years" , Elsevier,
Amsterdam, 2005, pp. 1167-1189.
21] R. Chinchilla, C. Nájero, The Sonogashira reaction: A Booming Methodology in
Synthetic Organic Chemistry, Chem. Rev. 107 (2007) 874-922.
[
2
6

[
22] E. M. Arnett, L. Joris, E. Mitchell, T. S. S: R. Murty, T. M. Gories, P. v. R. Schlayer,
Studies of Hydrogen-Bonded Complex Formation III. Thermodynamics of Complexing by
Infrared Spectroscopy and Calorimetry, J. Am. Chem. Soc. 92 (1970) 2365-2377.
[
23] J. E. Bertie, C. D. Keefe, R. N. Jones, Tables of Intensities for the Calibration of Infrared
Spectroscopic Measurement in the Liquid Phase, first ed., Blackwell Science Ltd, Oxford,
995, p. 13.
24] G. R. Desiraju, T. Steiner,The Weak Hydrogen Bond In Structural Chemistry and
Biology, first ed., Oxford University Press, New York, 1999, pp. 40-80.
25] D. Vojta, M. Vazdar, The study of hydrogen bonding and π···π interactions in
phenol···ethynylbenzene complex by IR spectroscopy, Spectrochim. Acta A 132 (2014) 6-14.
26] E. M. Arnett, E. J. Mitchell, T. S. S. R. Murty, „Basicity.“ A Comparison of Hydrogen
Bonding and Proton Transfer to Some Lewis Bases, J. Am. Chem. Soc. 96 (1974) 3875-3889.
27] M. H. Abraham, P. L. Grellier, D. V. Prior, J. J. Morris, P. J. Taylor, Hydrogen Bonding.
1
[
[
[
[
Part 10. A Scale of Solute Hydrogen-bond Basicity using logK Values for Complexation in
Tetrachloromethane, J. Chem. Soc. Perkin Trans. 2 (1990) 521-529.
[
28] L. Joris, P. v. r. Schlayer, π-deficient N-heteroaromatics as proton acceptors in hydrogen-
bonding, Tetrahedron 24 (1968) 5991-6005.
29] H. P. Hopkins, Jr., C. J. Alexander, S. Zakir Ali, A Quantitative Comparison of the
[
relative Hydrogen Bonding Basicities and the Gas-Phase Proton Affinities of Substituted
Pyridines, J. Phys. Chem. 82 (1978) 1268-1272.
[
30] J. Rubin, G. S. Panson, Hydrogen Bonding. II. Phenol Interactions with Substituted
Pyridines, J. Phys. Chem. 69 (1965) 3089-3091.
31] J. Graton, M. Berthelot, C. Laurence, Hydrogen-bond basicity pKHB scale of secondary
amines, J. Chem. Soc., Perkin Trans. 2 (2001) 2130-2135.
32] R. Adámek, Z. Ksandr, Accuracy of the determination of hydrogen bond association
constant by IR spectroscopy, Collect. Czech. Chem. Commun. 33 (1968) 3053-3056.
33] K. B. Whetsel, J. H. Lady, Self-association of phenol in non-polar solvents, Spectrometry
of Fuels, Plenum Press, 1970, 259-279.
34] J. M. Widom, R. J. Philippe, M. E. Hobbs. A study of the association of phenol with
several ketones by infrared, J. Am. Chem. Soc. 79 (1957) 1383-1386.
35] J. Brandmüller, K. Seevogel, Infrarotspektroskopische Untersuchungen zur
Wasserstoffbrükenbindung bei Pyridin-Alkohol-Gemischen, Spectrochim. Acta 20 (1964)
53-465.
[
[
[
[
[
4
2
7

[
36] P. Gilli, L. Pretto, V. Bertolasi, G. Gilli, Predicting Hydrogen-bond Strengths from Acid-
Base Molecular Properties. The pK Slide Rule: Toward the Solution of a Long-Lasting
Problem, Acc. Chem. Res. 42 (2009) 33-44.
Figure captions
Fig. 1. 1) 2,6-DEP; 2) 3,5-DEP; 3) TMP; 4) PH.
Fig. 2. Structures 5-8, optimized at the B3LYP/6-311++G(d,p) level of theory. The most
important interactions between monomers are shown as dashed bonds and the distance
between atoms in Ångström units are shown next to it.
Fig. 3. Temperature study (25-115 °C) of DEPs in C
2
Cl
4
, after C
2
Cl
subtraction: a) ≡C-H
4
-1
-3
stretching (3360-3260 cm ) of 2,6-DEP (c
0
=0.01894 mol dm ); b) ≡C-H stretching (3360-
-1
-3
-1
3
0
260 cm ) of 3,5-DEP (c =0.02037 mol dm ); c) C≡C stretching (2145-2080 cm ) of 2,6-
-3
-1
0 0
DEP (c =0.02037
=0.01894 mol dm ); d) C≡C stretching (2145-2080 cm ) of 3,5-DEP (c
-3
mol dm ).
Fig. 4. TMP's concentration dependence (at 26±1°C) of: a) ≡C-H and ≡C-H··· stretching of
-
1
2
,6-DEP in a 2,6-DEP+TMP in C
stretching of 3,5-DEP in a 3,5-DEP+TMP in C
C≡C··· stretching of 2,6-DEP in a 2,6-DEP+TMP in C
and C≡C··· stretching of 3,5-DEP in a 3,5-DEP+TMP in C
2 4
Cl
mixture (3360-3160 cm ); b) ≡C-H and ≡C-H···
-1
2
Cl
mixture (3360-3160 cm ); c) C≡C and
4
-1
2
Cl
mixture (2145-2080 cm ); d) C≡C
4
-
1
2
Cl
4
mixture (2145-2080 cm ). IR
Cl
4
spectra of a) and b) are obtained after C
subtraction.
2
Cl
4
subtraction, while c) and d) after TMP+C
2
Fig. 5. PH's concentration dependence (at 26±1°C) of: a) OH and OH··· stretching of 2,6-
-1
DEP in a PH+2,6-DEP in C
,5-DEP in a PH+3,5-DEP in C
of 2,6-DEP in a PH+2,6-DEP in C
stretching of 3,5-DEP in a PH+3,5-DEP in C
and b) are obtained after C Cl subtraction, while c) and d) after DEPs+C
2 4
Cl mixture (3680-2965 cm ); b) OH and OH··· stretching of
-1
3
2 4
Cl mixture (3680-2965 cm ); c) C≡C and C≡C··· stretching
-1
Cl
4
mixture (2145-2085 cm ); d) C≡C and C≡C···
2
-
1
2
Cl
4
mixture (2145-2085 cm ). IR spectra of a)
Cl subtraction.
2
4
2
4
Fig. 6. Temperature study (25-115 °C) of DEPs+TMP in C
2
Cl
4
mixture: a) ≡C-H and ≡C-H···
-1
stretching (3360-3160 cm ) of 2,6-DEP; b) ≡C-H stretching and ≡C-H··· stretching (3360-
-1
-1
3
160 cm ) of 3,5-DEP ; c) C≡C stretching (2145-2080 cm ) of 2,6-DEP; d) C≡C stretching
-1
(
2 4
2145-2080 cm ) of 3,5-DEP. IR spectra of a) and b) are obtained after C Cl subtraction,
2
8

while c) and d) after TMP+C
2
Cl
4
subtraction. Initial concentrations in a) and c) were c (2,6-
0
-3
-3
DEP)=0.02549 mol dm and c
0
(TMP)=0.92479 mol dm , while in b) and d) c
0
(3,5-
-
3
-3
DEP)=0.024 mol dm and c
0
(TMP)=0.92479 mol dm .
Fig. 7. Temperature study (25-115 °C) of PH+DEPs in C
stretching of PH in a PH+2,6-DEP in C Cl
stretching of OH in a PH+3,5-DEP in C Cl
stretching of 2,6-DEP in a PH+2,6-DEP in C
C≡C··· stretching of 3,5-DEP in a PH+3,5-DEP in C
spectra of a) and b) are obtained after C Cl subtraction, while c) and d) after DEPs+C
subtraction. Initial concentrations in a) and c) were c
2
Cl
4
mixture: a) OH and OH···
-
1
2
4
mixture (3680-2965 cm ); ); b OH and OH···
-1
2
4
mixture (3680-2965 cm ); c) C≡C and C≡C···
-1
2
4
Cl mixture (2145-2080 cm ); d) C≡C and
-1
2
Cl
4
mixture (2145-2080 cm ). IR
2
4
2
Cl
4
-3
0
(PH)=0.00713 mol dm and c
0
(2,6-
(3,5-
-3
-3
DEP)=0.0104 mol dm , while in b) and d) c
0
(PH)=0.00471 mol dm and c
0
-
3
DEP)=0.02037 mol dm .
Fig. 8. Possible resonance-assisted HB in a) PH···3,5-DEP complex in contrast to b) PH···3-
EP complex
2
9