Design, synthesis and noncentrosymmetric solid state organization of three novel pyridylphosphonic acids

Article abstract of DOI:10.1039/c1ce05140g

Three pyridylphosphonic acids, 3-pyridylphosphonic acid 1, 5-(dihydroxyphosphoryl)nicotinic acid 2, and 3,5-pyridinediyldiphosphonic acid 3, were synthesized and structurally investigated by solid state FT-IR and single crystal X-ray diffraction methods. All compounds appear in zwitterionic forms in the solid state with a proton transferred from the phosphonic group toward the pyridine N-atom. Strong hydrogen-bond interactions O-H...O and N-H...O organize the molecules of the compounds into polar three-dimensional networks. The crystal structures are additionally stabilized via weaker C-H...O hydrogen bonds and π...π or CO...π interactions. Powder second harmonic generation and solution NMR spectra were measured as well. The NMR spectra revealed that the double bonds of the functional groups (PO and CO), although differently oriented versus the N _py atom, are coplanar with the aromatic rings in all stable, low energy conformers of compounds 1-3 in solution. In crystals, however, the PO bonds are tilted toward the ring by 10.77(7)° in 1, 70.42(8)° in 2 and by 43.97(7)° and 6.56(8)° in 3. All three compounds exhibit moderate powder SHG efficiencies compared to that of urea.

Full text of DOI:10.1039/c1ce05140g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
CrystEngComm
Cite this: CrystEngComm, 2011, 13, 3474
www.rsc.org/crystengcomm
PAPER
Design, synthesis and noncentrosymmetric solid state organization of three
novel pyridylphosphonic acids†
a
Jerzy Zon, Veneta Videnova-Adrabinska, a Jan Janczak,b Magdalena Wilk,a Anna Samoc,c Roman Gancarza
*
*
and Marek Samocac
Received 27th January 2011, Accepted 24th February 2011
DOI: 10.1039/c1ce05140g
Three pyridylphosphonic acids, 3-pyridylphosphonic acid 1, 5-(dihydroxyphosphoryl)nicotinic acid 2,
and 3,5-pyridinediyldiphosphonic acid 3, were synthesized and structurally investigated by solid state
FT-IR and single crystal X-ray diffraction methods. All compounds appear in zwitterionic forms in the
solid state with a proton transferred from the phosphonic group toward the pyridine N-atom. Strong
hydrogen-bond interactions O–H/O and N–H/O organize the molecules of the compounds into
polar three-dimensional networks. The crystal structures are additionally stabilized via weaker C–H/
O hydrogen bonds and p/p or C]O/p interactions. Powder second harmonic generation and
solution NMR spectra were measured as well. The NMR spectra revealed that the double bonds of the
functional groups (P]O and C]O), although differently oriented versus the Npy atom, are coplanar
with the aromatic rings in all stable, low energy conformers of compounds 1–3 in solution. In crystals,
however, the P]O bonds are tilted toward the ring by 10.77(7)ꢀ in 1, 70.42(8)ꢀ in 2 and by 43.97(7)ꢀ and
6.56(8)ꢀ in 3. All three compounds exhibit moderate powder SHG efficiencies compared to that of urea.
However, the properties of the metal–organic framework
Introduction
(MOF), and therefore the physicochemical and optical charac-
The rapid development of metallosupramolecular chemistry is
teristics of the material, are determined by the network connec-
derived from the enhanced need for novel functional materials.
tivity, which requires a selection and preparation of the modulus
To a great extent it is oriented toward synthesis and character-
(building unit), as well as a selection and structural modification
ization of hybrid coordination compounds, since the unique
of the organic linker by changing the topology and the juxta-
characteristics of the organic and inorganic components in them
position of the functional groups. Very recently,
are combined in a complementary fashion, which leads to
pyridylphosphonic acids were applied as ligands. Search in the
unusual solid state structures with novel properties. A variety of
Cambridge Structural Database (CSD, Version 5.31)19 has
crystalline architectures with different dimensionalities, pore
retrieved thirty five structures of 2-pyridylphosphonates and
sizes and geometries, and versatile properties have been obtained
seven of 4-pyridylphosphonates with d- and f-block metal ions.
by changing the compositional range, and varying the metal and
Three metal complexes (CdII, CoII and SnIV) with 3-pyr-
the coordination sites, which gives an access to a vast area of
idylphosphonic acid are known, but the crystal structure of the
complex multifunctional materials useful for selective adsorption
acid itself has not been determined yet. Pyridylphosphonic acids
and separation,1–5 gas storage,6–8 ion exchange9–14 and catal-
may appear also as important agents in design of Bio-MOFs,20
ysis.15–18 Prototypes of this class of solids are the metal–
since many phosphonates containing a pyridine fragment display
carboxylates and phosphonates, which have been extensively
biological activity and are used as agrochemicals or drugs.
investigated for a long time.
Therefore, synthesis and characterization of novel phosphonic
acids with nitrogen containing aromatic systems and the phos-
phonic/ate group(s) attached as a ring substituent are highly
recent review21 summarizes different synthetic
direct phosphonylation of azaheterocyclic
desired.
A
aDepartment of Chemistry, Wrocław University of Technology, 27
Wybrzez_e Wyspianꢀskiego Str., 50-370 Wrocław, Poland. E-mail: Veneta.
Videnova-Adrabinska@pwr.wroc.pl
methods for
aromatics.
a
bInstitute of Low Temperature and Structural Research, Polish Academy of
Science, 2 Okoꢀlna Str., P.O. Box 1410, 50-950 Wrocław, Poland
cLaser Physics Centre, Research School of Physics and Engineering,
Australian National University, Canberra, ACT 0200, Australia
† Electronic supplementary information (ESI) available. CCDC
reference numbers 810542–810544. For ESI and crystallographic data
in CIF or other electronic format see DOI: 10.1039/c1ce05140g
In this paper we describe the synthesis and discuss the solid
state organization of three pyridylphosphonic acids, namely 3-
pyridylphosphonic acid 1, 5-(dihydroxyphosphoryl)nicotinic
acid 2 and 3,5-pyridinediyldiphosphonic acid 3. Two of them
(2 and 3) are novel compounds, previously not reported in the
literature. All three compounds are phosphonic analogous of the
3474 | CrystEngComm, 2011, 13, 3474–3484
This journal is ª The Royal Society of Chemistry 2011

View Article Online
nicotinic 4 and dinicotinic 5 acids, which have a broad use as
ligands. In solid state the molecules of 4 and 5 self-recognize via
O–H/N interactions forming helices which are further arranged
into polar monolayers.22,23 However, the layers are packed into
centrosymmetric crystal structures by electrostatic and offset
face-to-face (OFF) interactions. Our aim was to design organic
ligands predisposed for: (a) hydrogen-bond controlled
self-recognition and self-assembly into acentric structures; (b)
polytopic coordination in order to form multidimensional
coordination polymers. Therefore, we postulated that the
replacement of the planar carboxylic group(s) with the tetrahe-
dral phosphonic group(s) will lead to non-centrosymmetric
structures. There were some arguments for this presumption. The
way the molecules of 1–3 self-recognize should not be very
different from that in 4 and 5 and the meta-position of the
functional groups versus the Npy atom may support helical
arrangements via N–H/O and/or O–H/O interactions.
However, the additional hydrogen-bond donor and acceptor
sites and the non-planarity of the phosphonic group may allow
for hydrogen-bond controlled interactions in three dimensions.
Scheme 1 Synthetic route to compounds 1–3.
1257s, 1200m, 1174m, 1111m, 1052s, 1023s, 972m, 764m, 703m,
657m, 570m cmꢁ1
.
Tetraethyl 3,5-pyridinediyldiphosphonate 300(A). 2.015 g, 43%,
Rf ¼ 0.06 (silica gel on PET foils with fluorescent indicator 254
nm, ethyl acetate, UV lamp for visualization). 31P{1H} NMR
(121 MHz, CDCl3): d 12.8 (s). IR (neat, film): nmax 2985m,
1579m, 1562m, 1411m, 1394m, 1260s, 1201m, 1124m, 1264m,
Experimental section
1098m, 1043s, 1023s, 972s, 853m, 795m, 711m, 565s cmꢁ1
An alternative synthesis (B) of the esters 200 and 300 is described
.
Syntheses
Synthesis of diethyl pyridylphosphonates 100, 200(A) and 300(A)
(Scheme 1) by general procedure24. The corresponding bromo-
derivative of pyridine (20.0 mmol 3-bromopyridine 10, methyl 5-
bromonicotinate25 20 or 10.0 mmol 3,5-dibromopyridine 30),
diethyl phosphite (24.0 mmol, 3.1 mL), triethylamine (24.0
mmol, 3.3 mL), toluene (10 mL) and tetrakis-
(triphenylphosphine)palladium (0.80 mmol, 0.9244 g) were
heated at 85 ꢀC while continuously stirred under argon atmo-
sphere for 12 h. A precipitate of triethylamine hydrobromide,
which was formed during the reaction, was filtered under reduced
pressure and the by-product was washed with toluene (5 mL).
The combined toluene filtrates were evaporated under reduced
pressure to dryness. The residue was subjected to column chro-
matography on silica gel using ethyl acetate or CHCl3–ethyl
acetate as an eluent to obtain pure esters as oils. The eluates were
monitored by TLC (Rf data are given in the experimental part
below) and 1H, as well as, 31P NMR in CDCl3. The structures of
final products were additionally confirmed by 1H, 13C{1H} NMR
in CDCl3 (esters 100, 200(A) and 300(A)), D2O (acids 1 and 3), D2O/
DCl (acid 2). The 31P{1H} NMR data are given in the experi-
mental section below.
in the ESI†.
General procedure for synthesis of pyridylphosphonic acids 1–3
(Scheme 1). Pyridylphosphonic acids 1–3 were obtained from
their esters 100–300 (10 mmol) by refluxing for 12 h with a mixture
of concentrated hydrochloric acid (10 mL) and water (10 mL).
After that the reaction mixture was evaporated to dryness under
water aspirator pressure, then water (5 mL) was added and the
mixture was evaporated again to dryness. The residue was
treated with methanol (5 mL), evaporated under reduced pres-
sure to obtain a solid crude product which was crystallized from
water–ethanol or water (as indicated in the Experimental section)
to give pure pyridylphosphonic acids 1–3.
3-Pyridylphosphonic acid 1. 1.0340 g, 65%, mp 249–253 ꢀC
(H2O–EtOH). 31P{1H} NMR (121 MHz, D2O): d 2.1 (s).
Elemental analysis calc. for C5H6NO3P: C, 37.75; H, 3.80; N,
8.81; P, 19.47%. Found: C, 37.44; H, 3.53; N, 8.68; P, 19.38%.
5-(Dihydroxyphosphoryl)nicotinic acid 2. 1.1374 g, 56%, mp
305–310 ꢀC (H2O). 31P{1H} NMR (121 MHz, D2O–DCl): d 14.2
(s). Elemental analysis calc. for C6H6NO5P: C, 35.48; H, 2.98; N,
6.90; P, 15.25%. Found: C, 35.38; H, 2.82; N, 6.74; P, 15.18%.
Diethyl 3-pyridylphosphonate 100. 3.486 g, 81%, Rf ¼ 0.17 (silica
gel on PET foils with fluorescent indicator 254 nm, ethyl acetate,
UV lamp for visualization). 31P{1H} NMR (121 MHz, CDCl3):
d 14.5 (s). IR (neat, film): nmax 2985m, 1582m, 1408m, 1255s,
3,5-Pyridinediyldiphosphonic acid 3. 1.1714 g, 49%, mp 285–
292 ꢀC (H2O). 31P{1H} NMR (121 MHz, D2O): d 13.9 (s).
Elemental analysis calc. for C5H7NO6P2: C, 25.12; H, 2.95; N,
5.86; P, 25.91%. Found: C, 25.30; H, 2.82; N, 5.82; P, 25.83%.
1145m, 1053s, 1023s, 971s, 793m, 711m, 569m cmꢁ1
.
Methyl 5-(diethoxyphosphoryl)nicotinate 200(A). 2.186 g, 40%,
Rf ¼ 0.08 (silica gel on PET foils with fluorescent indicator 254
nm, CHCl3 : ethyl acetate ¼ 5 : 1 v/v, UV lamp for visualiza-
tion). 31P{1H} NMR (121 MHz, CDCl3): d 24.4 (s). IR (neat,
film): nmax 2986m, 1733s, 1591m, 1444m, 1415m, 1318m, 1292s,
X-Ray single crystal analysis
Suitable single crystals of 1, 2 and 3 were used for data collection
on a four-circle k-geometry KUMA KM4 diffractometer
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 3474–3484 | 3475

View Article Online
ꢀ
ꢁ
Table 2 Selected bond lengths (A) and bond angles ( ) for compounds
1–3
equipped with a two-dimensional area CCD detector. The
ꢁ
graphite monochromatized Mo-Ka radiation (l ¼ 0.71073 A)
and the u-scan technique (Du ¼ 1ꢀ) were used for data collection.
The 960 images, covering over 99% of the Ewald sphere, were
performed for six different runs. One image was used as a stan-
dard after every 40 images for monitoring the stability of the
crystals, as well as for monitoring the data collection. No
correction on the relative intensity variation was necessary. Data
collection and reduction along with absorption correction were
performed using CrysAlis software package.27 The minimum and
maximum transmission factors are 0.952 and 0.975 for 1, 0.922
and 0.985 for 2 and 0.966 and 0.997 for 3. The structures, solved
by direct methods using SHELXS-97, gave the positions of
almost all non-hydrogen atoms. The remaining atoms were
located from subsequent difference Fourier syntheses. The
structures were refined using SHELXL-97 with the anisotropic
thermal displacement parameters.28 The hydrogen atoms of the
aromatic ring were located from the difference Fourier maps, but
in the final refinement their positions were constrained to thermal
parameters and distances. Details of the data collection para-
meters, crystallographic data and final agreement parameters are
collected in Table 1. Selected geometrical parameters are listed in
Table 2.
1
C3–P3
1.8215(14)
1.4954(12)
1.5031(10)
1.5633(11)
1.3325(18)
1.3373(17)
C3–P3–O31
C3–P3–O32
C3–P3–O33
O31–P3–O32
O31–P3–O33
O32–P3–O33
C2–N1–C6
107.81(6)
107.77(7)
103.83(6)
117.68(6)
111.61(6)
107.22(5)
121.59(12)
ꢁ13.35(13)
P3–O31
P3–O32
P3–O33
N1–C6
N1–C2
O31–P3–C3–C2
2
C3–C31
C31–O31
C31–O32
C5–P5
P5–O51
P5–O52
P5–O53
N1–C6
N1–C2
1.510(3)
1.198(2)
1.303(2)
1.8159(16)
1.4948(14)
1.5165(13)
1.5658(17)
1.341(2)
C3–C31–O31
C3–C31–O32
O31–C31–O32
C5–P5–O51
C5–P5–O52
C5–P5–O53
O51–P5–O52
O51–P5–O53
O52–P5–O53
C2–N1–C6
O31–C31–C3–C2
O51–P5–C5–C6
122.64(17)
112.30(15)
125.06(19)
108.66(8)
105.72(7)
104.71(9)
114.83(8)
114.10(8)
108.01(8)
122.87(16)
ꢁ1.4(3)
1.337(3)
95.24(17)
3
C3–P3
P3–O31
P3–O32
P3–O33
C5–P5
P5–O51
P5–O52
P5–O53
N1–C6
N1–C2
1.8013(17)
1.4880(12)
1.5212(12)
1.5375(14)
1.8062(16)
1.4916(13)
1.5229(13)
1.5456(13)
1.331(3)
C3–P3–O31
C3–P3–O32
C3–P3–O33
O31–P3–O32
O31–P3–O33
O32–P3–O33
C5–P5–O51
C5–P5–O52
C5–P5–O53
O51–P5–O52
O51–P5–O53
O52–P5–O53
C2–N1–C6
108.39(7)
105.62(7)
105.59(7)
116.85(7)
112.38(7)
107.25(7)
108.66(7)
105.72(7)
106.76(7)
115.63(7)
110.28(7)
109.34(7)
123.36(16)
ꢁ132.38(14)
ꢁ173.78(13)
Powder second harmonic measurements and their computations
Powdered samples of the investigated compounds and of urea,
which was used as the standard, were mounted between glass
microscope slides. The powders were ungraded, but with similar
grain sizes for all samples.
1.341(2)
Table 1 Crystallographic data and structure refinement parameters for
compounds 1–3
O31–P3–C3–C2
O51–P5–C5–C6
Compounds
1
2
3
The laser system used for the investigation was previously
described.29,30 Briefly, a beam from a SESAM mode-locked Nd-
YLF oscillator is amplified in a flashlamp-pumped Nd-YLF
amplifier resulting in approx. 6 ps pulses at 1054 nm with
the repetition rate of 20 Hz and the energies in mJ range. The
amplifier beam is attenuated and used without focussing. The
measurements were performed in transmission geometry,
the second harmonic being detected, after filtering off the
fundamental, with a photomultiplier whose output was moni-
tored with a digital oscilloscope. The relative powder second
harmonic efficiencies (Table 4) were calculated as an average
from the ratios of the reading obtained for the investigated
sample and that of urea at several different laser pulse energies
(in the range ꢂ10 mJ per pulse).
Formula
Molecular weight 159.08
C5H6NO3P
C6H6NO5P
203.09
Orthorhombic Orthorhombic Orthorhombic
C5H7NO6P2
239.06
Crystal system
Space group
Pna21 (No. 33) Fdd2 (No. 43) P212121 (No. 19)
ꢁ
a/A
7.5460(15)
10.948(2)
7.8040(16)
644.7(2)
4
1.639
1.630
0.365
19.948(3)
28.534(4)
5.304(1)
3019.0(8)
16
7.0495(14)
10.490(2)
10.813(2)
799.6(3)
4
ꢁ
b/A
ꢁ
c/A
ꢁ
V/A
3
Z
Dcalcd/g cmꢁ3
Dobs./g cmꢁ3
m/mmꢁ1
1.787
1.790
0.352
1.986
1.980
0.548
Crystal size/mm3
0.26 ꢃ 0.24 ꢃ 0.27 ꢃ 0.16 ꢃ 0.26 ꢃ 0.21 ꢃ
0.22 0.14 0.16
295 295 295
0.71073 0.71073 0.71073
8140/1643/1466 9629/1680/1525 10503/2077/1814
T/K
ꢁ
l/A
Total/unique/
observed
Reflections (Rint
R [F2 > 2s(F2)]a
wR [F2 all refls]
Flack parameter
S
Second-order nonlinear optical properties of the investigated
molecules were also evaluated using the AM1 method31,32 as
implemented in the MOPAC-7 quantum chemical program. The
second-order hyperpolarizabilities were computed for AM1-
optimized molecular geometries taken in two forms: that of
a neutral molecule and that of a zwitterion built by transferring
one of the protons from the phosphonic group to the pyridine
nitrogen atom (Table 5).
)
(0.018)
0.0209
0.0531
0.12(7)
0.955
(0.026)
0.0249
0.0592
ꢁ0.02(8)
1.001
(0.024)
0.0220
0.0548
0.04(8)
1.002
ꢁ3
ꢁ
Drmax, Drmin/e A +0.162, ꢁ0.174 +0.248, ꢁ0.192 +0.235, ꢁ0.256
a
w ¼ 1/[s2(Fo2) + (0.0380P)2] for 1, w ¼ 1/[s2(Fo2) + (0.0367P)2] for 2 and
w ¼ 1/[s2(Fo2) + (0.0372P)2] for 3, where P ¼ (Fo2 + 2Fc )/3.
2
3476 | CrystEngComm, 2011, 13, 3474–3484
This journal is ª The Royal Society of Chemistry 2011

View Article Online
ꢀ
ꢁ
Table 3 Geometry of the hydrogen bonds for compounds 1–3 (A,
)
D–H/A
D–H
H/A
D/A
D–H/A
1
1. O33–H33/O31i
2. N1–H1/O32ii
1.006(16)
1.049(14)
1.522(16)
1.519(15)
2.5181(15)
2.5559(16)
170.1(15)
169.0(15)
3
Symmetry codes: (i) ꢁ½ + x, 3/2 ꢁ y, z; (ii) /2 ꢁ x, ½ + y, ꢁ½ + z
2
1. O53–H53/O51i
2. O32–H32/O52ii
3. N1–H1/O52iii
0.87(3)
0.87(3)
1.00(3)
1.72(3)
1.72(3)
1.64(3)
2.5195(19)
2.589(2)
2.632(2)
153(3)
175(2)
175(2)
Symmetry codes: (i) ꢁx, ½ ꢁ y, ꢁ½ + z; (ii) ½ ꢁ x, ½ ꢁ y, 1 + z; (iii) ¼ ꢁ x, ꢁ¼ + y, ꢁ¼ + z
3
1. O32–H32/O52i
2. O53–H53/O31ii
3. O33–H33/O51iii
4. N1–H1/O51ii
5. N1–H1/O53iv
6. N1–H1/O52v
7. C2–H2/O32
1.199(14)
0.986(14)
0.918(16)
0.88(2)
0.88(2)
0.88(2)
0.9300
0.9300
0.9300
0.9300
1.286(15)
1.543(16)
1.634(17)
2.34(2)
2.4711(17)
2.5248(14)
2.5458(18)
2.999(2)
2.959(2)
3.050(2)
2.936(2)
3.158(2)
3.348(2)
3.382(2)
168.1(17)
173.5(14)
171.5(19)
131.8(16)
112.7(15)
125.3(16)
106.00
149.00
166.00
145.00
2.50(2)
2.454(19)
2.5400
2.3200
2.4400
2.5800
8. C6–H6/O31ii
9. C4–H4/O52iii
10. C2–H2/O33vi
Symmetry codes: (i) 1 ꢁ x, ½ + y, ½ ꢁ z; (ii) ½ ꢁ x, ꢁ y, ꢁ½ + z; (iii) ꢁx, ½ + y, ½ ꢁ z; (iv) 1 + x, y, z; (v) ½ + x, ꢁ½ ꢁ y, ꢁ z; (vi) ½ + x, ½ ꢁ y, ꢁz
Table 4 Relative powder SHG efficiencies obtained at 1054 nm with 6 ps
pulses
Infrared spectroscopy
The FT-IR spectra of the compounds were measured in KBr
pellets and nujol mulls (4000–400 cmꢁ1) using a Bruker IFS-88
spectrometer with a resolution of 2 cmꢁ1. In addition the spectra
of the partly deuterated 3-pyridylphosphonic acid were
measured. Dideutero-3-pyridylphosphonic acid was obtained by
dissolving the compound 1 (0.020 g) in deuterium oxide (2 mL),
followed by evaporation to dryness under reduced pressure. This
procedure was repeated one more time and the residual crystal-
line compound was air dried. The frequencies of all observed
bands and their tentative assignments are collected in Table S1
(ESI†).
Compound
1
2
3
Urea
1
Relative powder
SHG efficiency
0.37 ꢄ 0.06
0.58 ꢄ 0.04
0.13 ꢄ 0.03
Table 5 Computed vector parts of the second-order hyper-
polarizabilities for compounds 1–3
Molecule
form
Hyperpolarizability
component
Value at
0 eV (at.u.)a
Value at
0.5 eV (at.u.)
NMR spectroscopy in solution
The 13C, 1H{31P}, 1H, 31P{1H}, 13C{1H} NMR spectra were
recorded at room temperature on a Bruker Avance DRX300
instrument, operating at 300.13 MHz (1H), 121.50 (31P) and 75.46
MHz (13C). Proton and phosphorus decoupling was achieved by
1 zwitterion
bx,SHG
by,SHG
bz,SHG
bvect,SHG
bx,SHG
by,SHG
bz,SHG
bvect,SHG
bx,SHG
by,SHG
bz,SHG
bvect,SHG
bx,SHG
by,SHG
bz,SHG
bvect,SHG
bx,SHG
by,SHG
bz,SHG
bvect,SHG
bx,SHG
by,SHG
bz,SHG
bvect,SHG
322.25
48.54
59.93
331.35
28.74
336.73
47.45
62.83
345.81
28.06
129.78
ꢁ269.36
300.30
255.43
184.07
ꢁ81.37
325.19
ꢁ55.18
ꢁ107.88
11.40
1 neutral
power gated decoupling using Waltz 16 sequence. Reference for
13C and H NMR was solvent-chloroform lines (singlet dH
117.38
ꢁ244.26
272.52
241.70
178.58
ꢁ77.17
310.27
ꢁ53.08
ꢁ100.81
9.22
114.30
ꢁ224.64
327.52
ꢁ46.87
399.92
ꢁ60.76
ꢁ19.99
ꢁ10.16
64.77
1
¼
7.26 ppm, triplet dC ¼ 77.00 ppm). NMR spectra were analyzed
using MestreC program.33
2 zwitterion
2 neutral
Theoretical NMR data calculations
A
conformational search was performed using molecular
mechanic (MM+) and semiempirical methods (AM1). The found
conformers were optimized by the density functional theory
(DFT).34 The Becke’s three parameter functional35 with the
Vosko et al.36 local correlation part and the Lee et al.37 non-local
part, abbreviated as B3LYP, has been applied and the calcula-
tions were performed using the standard 6-31G(d,p) basis set.38
The effect of the solvent on the calculated structures was taken
into account through the polarized continuum model for chlo-
roform. The ideal gas, rigid rotor, and harmonic oscillator
approximations39 were applied for studying the vibrational
121.71
ꢁ239.78
340.56
ꢁ50.79
419.59
ꢁ65.39
ꢁ22.49
ꢁ6.45
3 zwitterion
3 neutral
69.45
a
The factor to convert atomic units to esu units is 8.6571 ꢃ 10ꢁ33
.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 3474–3484 | 3477

View Article Online
frequencies and the thermodynamic properties of the
compounds. All structures were in their energetical minimum,
which was confirmed by the frequency calculations. The NMR
parameters were calculated using the coupled perturbed density
functional theory (CP-DFT) method with B3LYP functional and
including the diamagnetic spin-orbit, paramagnetic spin-orbit,
Fermi-contact and spin-dipolar terms. This functional is believed
to give results that best fit the experimental data. A IGLOII base
was applied to the calculations. For the calculation of scalar
couplings were used the same solvent models as those employed
for the geometry optimization. The computations by DFT, AM1
and MM+ methods were carried out using the Gaussian 03 suite
of codes program.40
O–H/Npy/Npy–H/O interactions. However, the crystal
packing of these compounds takes place with a generation of an
inversion center, since there are no specific interactions between
the layers. Our design strategy was based on the postulate that by
replacing the planar carboxylic group for the three-dimensional
phosphonic group, we may promote an acentric three-dimen-
sional arrangement due to hydrogen bonds formed between the
monolayers. Indeed, all three compounds display non-centro-
symmetric crystal structures. Compounds 1 and 3, which are
phosphonic analogous of nicotinic and dinicotinic acids, crys-
tallise in Pna21 and P212121 space groups, respectively.
Results and discussion
Syntheses
Ethyl and isopropyl esters of pyridylphosphonic acids 100–300 were
obtained from the corresponding derivatives of 3-bromopyridine
10–30 and diethyl (Hirao et al.,24 procedure A, or diisopropyl
phosphite (Montchamp et al.,26 procedure B) in the presence of
amine using two palladium catalyzed cross-coupling reactions
(Scheme 1). Both used methods differ in the type and loading of
catalyst. Yield of desired products varied from high to low. The
ethyl and isopropyl esters 100–300 were chromatographically
purified and spectroscopically characterized. The pure esters
were hydrolyzed with 6 N hydrochloric acid at reflux to give
corresponding pyridylphosphonic acids 1–3 with good yields.
Molecular and crystal structures
All compounds in solid state appear in zwitterionic form (NH+
and PO3Hꢁ) with a proton transferred from the phosphonic
(–PO3H2) group toward the nitrogen atom of the pyridine ring
(Fig. 1). The proton transfer is manifested in the P–O bond
lengths of the phosphonic group as well as in C–N–C angle of the
pyridine ring. The deprotonated P–O bond in all compounds is
intermediate between the P–OH bond and the formal double
bond P]O (see Table 2). It is close in length to the double P]O
bond in crystals 1 and 2. In crystal 3 the intermediate (P x O32
and P x O52) bonds become significantly longer and the
protonated (P–O33 and P–O53) bonds become slightly shorter,
which reflects the proton dynamic from both phosphonic groups
toward the Npy atom, as well as, between the phosphonic groups
of neighbouring molecules. The protonation of the nitrogen
atom is reflected also in the enlargement of the internal C–N–C
angle (121.59(12)ꢀ in 1, 122.87(16)ꢀ in 2 and 123.36(16)ꢀ in 3)
versus an average C–N–C angle of 117.3(2)ꢀ found in CSD for
pyridine derivatives.19 This change is consistent with the valence-
shell electron pair repulsion model (VSEPR),41 according to
which the bonded pair of electrons affords a smaller region than
ꢁ
the lone-pair. The C–O bond distances of 1.198(2) A and 1.303(2)
ꢁ
A in 2 are typical for C]O and C–OH bonds of the non-disso-
ciated carboxylic group, which proves that also in this compound
the proton of the ring nitrogen atom Npy is transferred from the
phosphonic group in 5-position.
The meta-position of the carboxylic group(s) versus the pyri-
dine nitrogen atom in nicotinic and dinicotinic acids22,23
determines the molecular arrangement into polar layers via
Fig. 1 The asymmetric units with the atom-labelling schemes for
compounds 1–3. The displacement ellipsoids are given at the 50% prob-
ability level.
3478 | CrystEngComm, 2011, 13, 3474–3484
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Moreover, compound 3, containing two phosphonic groups,
symmetrically arranged versus the pyridine N-atom, displays
a chiral crystal structure with helical arrangement of the mole-
cules in all three crystallographic directions. Note that P212121,
though belonging to the Sohncke family, is not a chiral space
group. On the other hand, compound 2 can be considered as
a derivative of the nicotinic acid with an additional functional
group in 5-position or as a derivative of the dinicotinic acid in
which one of the carboxylic groups is replaced with a phosphonic
group. The total asymmetry of the zwitterion and the chemical
and geometrical bias of the functional groups prevent recogni-
tion events governed by screw operators enforcing diamond
relations in solid state and the compound crystallizes in Fdd2
space group.
Supramolecular networks
Numerous hydrogen bond interactions, established between the
functional groups of the molecules, as well as the nitrogen site of
the pyridine ring, are the main driving forces responsible for the
molecular self-recognition and self-association leading to well
defined crystal structures 1–3 (see Table 3).
There are two strong hydrogen bonds O33–H33/O31 of
ꢁ
ꢁ
2.5181(15) A and N1–H1/O32 of 2.5559(16) A in crystal 1. The
first of them is formed between the phosphonate groups of
adjacent glide related zwitterions and serves to arrange them into
chains along the a-crystallographic axis (Fig. 2a). The second
hydrogen bond, established between the protonated N-atom and
a phosphonate oxygen site, cross-links n-glide chains in [011] and
ꢂ
[011] directions in order to form 3D hydrogen-bonded network
shown in Fig. 2b. The aromatic rings are arranged into stacks
ꢁ
ꢁ
along the a-axis with 3.8297(11) A and 3.8298(11) A distances
between their gravity centres (Fig. 2c).
The unit cell of crystal 2 contains 16 symmetry related zwit-
terions. Three different hydrogen bonds are used to link them
into 3D hydrogen-bonded network. The shortest bond O53–
ꢁ
H53/O51 (2.5195(19) A) is formed between the phosphonate
groups of rotation related zwitterionic units and extends them
into a chain along the c-crystallographic axis. The carboxylic
groups, arranged from both sides along the chain, approach the
phosphonate groups of two neighbouring chains via O32–H32/
ꢁ
O52 bond (of length 2.589(2) A), linking them into a formal
monolayer parallel to (010) crystallographic plane (Fig. 3a). The
aromatic rings are inclined versus the main plane of the mono-
layer and amassed into two different stacks along c-axis. There
are no p/p interactions between the rings along the stacks. The
hydrogen atom of the pyridine nitrogen sites and the carbonyl
oxygen site of the carboxylic groups are arranged outward from
both sides of the layers (Fig. 3b). However, the carbonyl O31
atom is inaccessible for hydrogen-bond formation since the
geometrical and directional requirements of the interaction are
incompatible with the symmetry constraints of the crystal
network. Instead, the H1(N) atoms of one layer approach the
phosphonate O52 atoms of two adjacent layers (N1–H1/O52),
completing thus the 3D hydrogen-bonded network. On the other
Fig. 2 (a and b) The interconnection between the [100] chains in [011]
ꢂ
and [011] directions to form two and three-dimensional hydrogen bonded
network in 1. (c) The crystal packing and the hydrogen-bonded network
viewed along the chains. The O–H/O (assigned as 1) bonds are shown in
red colour and the N–H/O (assigned as 2) bonds in blue colour. The
hydrogen bond numbering is consistent with that in Table 3.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 3474–3484 | 3479

View Article Online
hand, the carbonyl O31 atoms project the p system of the rings in
the next glide-related layer establishing interlayer C]O/p
planes in compound 2. The hydrogen-bonded extensions of the
zwitterions are helical in all three crystallographic directions and
the structure of the measured crystal is right-handed with a Flack
parameter of 0.04(8).42 Nonetheless, the space group P212121 is
not a chiral group, since the even-fold screw operations do not
produce enantiomorphic pairs.43 There are four strong proton
donors in this system (compared to two in 1 and three in 2),
which form three O–H/O and three N–H/O hydrogen bonds
(listed in Table 3). Additionally, six C–H/O hydrogen bonds
stabilize the crystal structure. The protonation of the pyridyl Npy
atom prompts a marked migration of the phosphonic H32 atom
toward the monodeprotonated phosphonate group of an adja-
cent zwitterion. The implemented hydrogen bond O32–H32/
ꢁ
interaction (the C]O/Cg and O/Cg distances are 3.940(2) A
ꢁ
and 3.1542(18) A, respectively and the angle C]O/Cg is
123.19(13)ꢀ) (Fig. 3c).
In crystal 3 the proton transfer takes place from one of the
phosphonic groups, meta-positioned versus the pyridine nitrogen
atom. The molecular symmetry is only slightly reduced, which is
the probable reason for appearing of screw axes instead of glide
ꢁ
O52 is very strong (2.4711(17) A) and links the zwitterionic units
into a helical chain along the b-crystallographic axis. The folding
of the b-helix is stabilized by p/p interactions established
between the p electronic systems of the double bond P]O31 and
the aromatic ring of the next molecule. (P]O31/Cg ¼
ꢁ
ꢁ
3.8888(11) A, O31/Cg ¼ 3.4234(15) A, :P]O31/Cg ¼
96.70(5)ꢀ) (see Fig. 4b). A moderate strong phosphonic–phos-
ꢁ
phonate interaction (O33–H33/O51 of 2.5458(18) A) connects
the a-translated b-helixes to form undulate molecular mono-
layers parallel to the (001) crystallographic plane (Fig. 4d). The
implemented two-dimensional network is a grid with large R44(26)
ring motifs enclosed between six phosphonic groups belonging to
four molecules. Another O–H/O bond, donated from the
phosphonate H53 hydrogen atom toward the phosphonic O31
oxygen site, executes helical arrangement along the c-crystallo-
graphic axis and serves to associate the monolayers (Fig. 4c and
e). On the other hand the pyridine H1(N1) atom serves as triple
hydrogen-bond donor in order to form hydrogen bonds with all
three oxygen atoms of the phosphonate group (Fig. 4f). One of
them (N1–H1/O52) executes the helical arrangement along a-
axis (21(X) operation), while the other (N1–H1/O53) is holding
the translation related molecules along the a-helix, preventing
thus its unfolding (Fig. 4a). The third bond N1–H1/O51 is
firming the c-helix. Four C–H/O bonds stabilize the three-
ꢁ
dimensional network. The C2–H2/O32 bond (of 2.936(2) A) is
intramolecular, formed between the pyridine ring C2 atom
(ortho-positioned versus Npy) and the phosphonic group, and
serves to fix the configuration of the zwitterion. C6–H6/O31 is
established along the c-helix and stabilizes it. The remaining two
bonds (C2–H2/O33 and C4–H4/O52) are formed between the
helices.
Nonlinear optical properties
The powder second harmonic generation (SHG) technique
provides only a very general indication on the merit of a second-
order nonlinear optical (NLO) material, but the results obtained
attest to some potential of phosphonic acids for forming inter-
esting crystal structures that, especially with introduction of
stronger NLO chromophores, may lead to useful material
properties. Table 4 shows the powder SHG results obtained for
the three investigated compounds against urea. Clearly,
compound 3 has an efficiency lower than the other two
compounds.
Fig. 3 The molecular organization in 2. (a) A view of the hydrogen-
bonded monolayer. (b) A side view of the monolayer. (c) The 3D
hydrogen-bonded network viewed along the c-crystallographic axis. The
O–H/O and N–H/O bonds are shown in red and blue colours. The
interlayer electron pair/p interactions are shown in gray colour.
The hydrogen bond numbering is consistent with that in Table 3.
Table 5 shows results obtained by the AM1 method for the
three investigated molecules, each of them modelled as
3480 | CrystEngComm, 2011, 13, 3474–3484
This journal is ª The Royal Society of Chemistry 2011

View Article Online
a zwitterion and as a neutral molecule. The values listed in the
table are the components and the value (length) of the vector part
of the hyperpolarizability b(ꢁ2u;u,u) computed for the funda-
mental frequency corresponding to photon energy of 0 eV and
0.5 eV. The vector parts of the hyperpolarizabilities of the
molecules, computed by the AM1 method, are on the order of 102
atomic units (ꢂ10ꢁ30 esu), which is in agreement with these
molecules exhibiting only moderate strength nonlinear effects. It
should be noted, however, that while the vector part of the
hyperpolarizability is a convenient quantity characterizing the
nonlinearity of a polar molecule, the actual performance of
a material containing such molecules will depend on particulars
of the employed nonlinear process and on the orientation of
molecules in the material.
Infrared spectra
The IR spectra of the compounds 1–3 are presented in Fig. 5. The
frequencies of the observed bands together with their tentative
assignments are given in Table S1 in the ESI†. Generally one can
state that the measured spectra reflect the structural peculiarities
both at a molecular and at a supramolecular level. The different
chemical groups in positions 1, 3 and 5 on the benzene ring of the
compounds give rise to different band frequencies and intensities
of the ring stretching vibrations C–H (3300–3000 cmꢁ1) and C–C
(1630–1550 cmꢁ1) and especially of the ring deformation vibra-
tions in the region 730–650 cmꢁ1. The nC–P band is observed at
940–955 cmꢁ1. The bands at 1708 cmꢁ1 (nC]O), 1316 cmꢁ1 (nC–
OH) and 1266 cmꢁ1 (nC–C), observed in the spectrum of
compound 2, are connected with the vibrations of the carboxylic
group.
The strong and broad absorption observed in the whole region
3500–1650 cmꢁ1 is connected with the stretching vibrations nO–H
and nN–H. It provides evidence for the strong hydrogen bond
interactions (P)O–H/O(P), (C)O–H/O(P), that organize the
molecules of these compounds in solid state. The spectrum of the
partly deuterated analogue of 1 displays a marked red shift and
narrowing of the absorption and the observed frequency shift
nOH/nOD roughly corresponds to the ratio of the mass factors.
The positions of the gravity centre of the nOH absorption in
the spectra of the compounds correlate well with the lengths of
the hydrogen bonds in them. On the other hand, the differenti-
ated structure of the nOH absorption, issued from strong reso-
nance and coupling effects of the stretching vibrations with
overtones and combinations of internal vibrations, reflects the
structural relationships between the molecules.44,45 So, the four
maxima (at 2650 cmꢁ1 and 2450 cmꢁ1 (band A), 2250 cmꢁ1 (band
B) and 1735 cmꢁ1 (band C)) and two minima (at 2550 cmꢁ1 and
2400 cmꢁ1), observed in the spectrum of 1, are a result from
packing patterns viewed along the c-axis. For transparency reasons the
O–H/O (shown in red colour) and the N–H/O (shown in blue colour)
hydrogen bonds are visualized in two separate pictures. The H1(Npy
)
atoms are removed from the picture (e) and the H atoms of the phos-
phonate groups are removed from picture (f). The (C)H-atoms of the
benzene ring are removed from all pictures. The C–H/O bond is shown
in green and the p/Cg interaction in the b-helix is shown in gray colours.
The hydrogen bond numbering is consistent with that in Table 3.
Fig. 4 (a–c) The 1D helical arrangement of the molecules in 3: the a-
helix, the b-helix and the c-helix. (d) The (001) monolayer. (e and f) The
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 3474–3484 | 3481

View Article Online
Fig. 5 The IR spectra of compounds 1–3. The spectrum of the partly deuterated compound 1 is shown in red colour.
couplings with overtone and combination of vibrations of the
phosphonate group. For example, the minimum at 2550 cmꢁ1 is
caused by the overtone of nP]O vibration (1379 cmꢁ1), which is
clearly observed as a weak bond in the deuterated spectrum. The
spectrum of compound 2 displays maxima at 2400 cmꢁ1 (A), 2007
cmꢁ1 (B) and 1650 cmꢁ1 (C) with some additional structure,
which reflects both the existence of two different hydrogen bonds
ꢁ
ꢁ
(P)O–H/O(P) (2.5195 A) and (C)O–H/O(P) (2.589 A) and
coupling effects with overtones of the two different functional
groups. The two bands observed at 2660 cmꢁ1 (A) and 2280 cmꢁ1
(B) in the spectrum of compound 3 arise from the two medium
strong hydrogen bonds (O53–H53/O31 and O33–H33/O51).
3482 | CrystEngComm, 2011, 13, 3474–3484
This journal is ª The Royal Society of Chemistry 2011

View Article Online
The broad absorption (band C), appearing in the region below
1600 cmꢁ1 of 3, is due to the very short O32–H32–O52 hydrogen
ꢁ
bond of 2.4711 A, which has a great extent of a covalent char-
direction. The crystal structure of 3,5-pyridinediyldiphosphonic
acid 3 is chiral, with a helical propagation of the molecules in all
three crystallographic directions. The pronounced differences in
molecular conformation and torsion angles observed in solid
state versus those in solution result from the proton transfer
acter.46
The spectra in the fingerprint region (1400–900 cmꢁ1) are very
complicated and cannot be understood solely on the ground of
intramolecular bonding considerations. For example, the two
P–O bonds in 1 and 2, to which no H atom is attached, are very
close in length while the third bond P–OH is significantly longer.
toward the Npy atom and the strong (P)O–H/O(P) and Npy
–
H/O(P) interactions between the zwitterions in the crystals. The
lack of inversion center in the solid state structures allows for
SHG and makes the compounds 1–3 interesting building blocks
for nonlinear optical materials. On the other hand, the studied
compounds are very attractive as polytopic ligands for design of
coordination polymers. Synthesis and study of variable hybrid
compounds with 3-pyridylphosphonic linkers are now in prog-
ress in our laboratory.
Therefore, the three strong bands observed at 1275 cmꢁ1
,
1192 cmꢁ1 and 1125 cmꢁ1 in 1 and at 1265 cmꢁ1, 1175 cmꢁ1 and
1125 cmꢁ1 in 2 are assigned to the naPO2 and nsPO2, while the two
strong bands at 1205 cmꢁ1 and 1133 cmꢁ1 in the spectrum of 3 are
connected with the two independent stretching vibrations nP]O
of both phosphonic groups. On the other hand, the whole
functional groups (–PO3H2 and –CO2H) participate in the
hydrogen bond interactions and reflect to a greater or smaller
extent the proton motions in O–H/O bridges. In particular, the
stretching vibrations of the long P–O(H) bonds are coupled with
the deformation vibration dOH and gOH and split into several
bands observed at 1060 cmꢁ1, 1037 cmꢁ1, and 986 cmꢁ1 in 1, at
1041 cmꢁ1, 1010 cmꢁ1, 978 cmꢁ1, and 915 cmꢁ1 in 2 and at
1026 cmꢁ1, 1000 cmꢁ1, and 958 cmꢁ1 in 3.
Acknowledgements
The authors thankfully acknowledge the Grant supported from
the Chemistry Department, Wroclaw University of Technology.
The authors also wish to thank P. Garczarek and P. Pa1etko for
their help at the initial step of the synthesis.
Notes and references
The above discussion clearly shows that the differences
observed in the spectra reflect not only the atom arrangement
and the bonding relationships inside the molecules of the
compounds 1–3, but also the structural peculiarities issued from
the different connectivity patterns between the molecules in the
crystals. In particular, they reflect the variable hydrogen-bond
interactions responsible for the supramolecular organization of
the compounds.
1 J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38,
1477.
2 Y.-S. Bae, O. K. Farha, A. M. Spokoyny, C. A. Mirkin, J. T. Hupp
and R. Q. Snurr, Chem. Commun., 2008, 4135.
3 J. An, S. J. Geib and N. L. Rosi, J. Am. Chem. Soc., 2010, 132, 38.
4 Y.-Y. Bae, O. K. Farha, J. T. Hupp and R. Q. Snurr, J. Mater. Chem.,
2009, 19, 2131.
5 D. Britt, H. Furukawa, B. Wang, T. G. Glover and O. M. Yaghi,
Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 20637.
ꢃ
6 L. J. Murray, M. Dinca and J. R. Long, Chem. Soc. Rev., 2009, 38,
1294.
7 Y. H. Hu and L. Zhang, Adv. Mater., 2010, 22, E117.
8 G. Tian, G. S. Zhu, X. Y. Yang, Q. R. Fang, M. Xue, J. Y. Sun,
Y. Wei and S. L. Qiu, Chem. Commun., 2005, 1396.
9 X. Shi, G. S. Zhu, S. L. Qiu, K. L. Huang, J. H. Yu and R. R. Xu,
Angew. Chem., Int. Ed., 2004, 43, 6482.
10 R. Xiong, X. You, B. F. Abrahams, Z. Xue and C. Che, Angew.
Chem., Int. Ed., 2001, 40, 4422.
11 S. L. James, Chem. Soc. Rev., 2003, 32, 276.
12 W. B. Lin, J. Solid State Chem., 2005, 178, 2486.
13 X. Shi, G. S. Zhu, S. L. Qiu, K. L. Huang, J. H. Yu and R. R. Xu,
Angew. Chem., Int. Ed., 2004, 43, 6482.
14 G. Tian, G. S. Zhu, X. Y. Yang, Q. R. Fang, M. Xue, J. Y. Sun,
Y. Wei and S. L. Qiu, Chem. Commun., 2005, 51396.
15 J.-Y. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S.-T. Nguyen and
J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450.
16 L. Ma, C. Abney and W. Lin, Chem. Soc. Rev., 2009, 38, 1248.
17 R.-Q. Zou, H. Sakurai, S. Han, R.-Q. Zhong and Q. Xu, J. Am.
Chem. Soc., 2007, 129, 8402.
ꢀ
18 A. Corma, H. Garcia and F. X. Llabres i Xamena, Chem. Rev., 2010,
110, 4606.
19 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380.
20 A. C. McKinlay, R. E. Morris, P. Horcajada, G. Ferey, R. Grek,
P. Couvreur and C. Serre, Angew. Chem., Int. Ed., 2010, 49, 6260.
21 S. Van der Jeught and C. V. Stevens, Chem. Rev., 2009, 109, 2672.
22 W. B. Wright and G. S. D. King, Acta Crystallogr., 1953, 6, 305;
M. P. Gupta and P. Kumar, Cryst. Struct. Commun., 1975, 4, 365;
A. Kutoglu and C. Scheringer, Acta Crystallogr., Sect. C: Cryst.
Struct. Commun., 1983, 39, 232.
Solution NMR spectra
1
The NMR (1H, H{1H}, 31P, 31P{1H), HHCosy) spectra both of
the studied pyridylphosphonic acids 1–3 and of their esters 100,
200(A) and 300(A) have been measured and analyzed. The spectra
of the esters are composed of several multiplets in their aromatic
region. For the reasons of their complexity and the literature
inconsistency in the assignment of the NMR parameters for
diethyl 3-pyridylphosphonate 100,47 we decided also to perform
calculations at ab initio level. The theoretical and experimental
results for chemical shifts are collected in Table S2† and for the
coupling constants in Table S3 (given in the ESI†). We have
found that all stable low energy conformers have Cs symmetry
and the double bonds P]O and C]O lie in the symmetry plane
of the molecules. However, their bond vectors are differently
orientated and point toward the Npy atom or in the opposite
direction (which is expressed in the Table S3† by ‘up’ or ‘down’
arrows). The torsion angles describing the orientations of the
carboxyl and phosphonate groups in solid state are given in
Table 2. This investigation reveals that the stable conformers in
solution are different to those in the crystals.
23 F. Takusagawa, K. Hirotsu and A. Shimada, Bull. Chem. Soc. Jpn.,
1973, 46, 2292; J. A. Cowan, J. A. K. Howard, G. J. McIntyre,
S. M.-F. Lo and I. D. Williams, Acta Crystallogr., Sect. B: Struct.
Sci., 2005, 61, 724.
24 T. Hirao, T. Masunaga, Y. Ohshiro and T. Agawa, Synthesis, 1981,
56.
Conclusions
One can state that, in agreement with our design considerations,
all three compounds display non-centrosymmetric crystal struc-
tures with helical arrangement of the molecules at least in one
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 3474–3484 | 3483

View Article Online
25 W. J. Thompson and J. Gandino, J. Org. Chem., 1984, 49, 5237.
26 Y. Belabassi, S. Alzghari and J.-L. Montchamp, J. Organomet.
Chem., 2008, 693, 3171.
36 S. H. Vosko, L. Wilk and M. Nusiar, Can. J. Phys., 1980, 58, 1200.
37 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785.
27 CrysAlis CCD and CrysAlis RED, Version 171.32.6, Oxford
Diffraction, Wroc1aw, Poland, 2006.
28 G. M. Sheldrick, in SHELX-97: Program Package for Crystal
38 M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon,
D. J. DeFress and J. A. Pople, J. Chem. Phys., 1982, 77, 3654.
39 N. Davidson, in Statistical Mechanics, McGraw-Hill, New York,
1962.
40 M. J. Frisch, G. W. Trucks, H. B. Schlegel, et al., Gaussian 03,
Revision C.02, Gaussian, Inc., Wallingford, CT, 2004.
41 R. J. Gilespie, Chem. Soc. Rev., 1992, 21, 59.
42 H. D. Flack, Acta Crystallogr., Sect. A: Found. Crystallogr., 1983, 39,
876.
€
Structure Solution and Refinement, University of Gottingen,
€
Gottingen, Germany, 1997.
29 A. Samoc, M. Samoc, B. Luther-Davies, J. Kelly, E. Krausz and
A. Willis, Mol. Cryst. Liq. Cryst., 2004, 415, 179.
30 A. Samoc, M. Samoc, B. Luther-Davies, V. Z. Kolev, R. K. Bagien,
X. Luo and C. Zha, Mol. Cryst. Liq. Cryst., 2006, 446, 123.
31 M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and J. J. P. Stewart, J.
Am. Chem. Soc., 1985, 107, 3902.
43 H. D. Flack, Helv. Chim. Acta, 2003, 86, 905.
44 J. C. Speakman, Struct. Bonding, 1972, 12, 141.
ꢄ
32 H. A. Kurtz, J. J. P. Stewart and K. M. Dieter, J. Comput. Chem.,
1990, 11, 82.
33 http://www.mestrec.com.
34 R. G. Parr and W. Yang, in Density-Functional Theory of Atoms and
Molecules, Oxford University Press, New York, 1994.
35 D. J. Becke, Chem. Phys., 1993, 98, 5648.
45 D. Hadzi and S. Bratos, in Hydrogen Bond, Recent Developments in
Theory and Experiments, ed. P. Schuster, G. Zundel and C.
Sandorfy, North-Holland, Amsterdam, 1976, vol. 2, p. 567.
46 A. Novak, Struct. Bonding, 1974, 18, 177.
47 P. Ayyappan, O. R. Evans, B. M. Foxman, K. A. Wheeler,
T. H. Warren and W. Lin, Inorg. Chem., 2001, 40, 5954.
3484 | CrystEngComm, 2011, 13, 3474–3484
This journal is ª The Royal Society of Chemistry 2011