Synthesis and characterisation of substituted diphenylamines - Charge-transfer, donor-acceptor systems localised at water-oil interfaces

Article abstract of DOI:10.1039/b818110a

A range of charge-transfer, donor-acceptor systems have been synthesised and subjected to in-depth crystallographic, spectroscopic and interfacial studies. The pyridine- or phenyl-substituted (N,N-diphenyl)amino derivatives contain an identical donor unit

Full text of DOI:10.1039/b818110a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Synthesis and characterisation of substituted
diphenylamines—charge-transfer, donor–acceptor systems localised
at water–oil interfacesw
K. Kowalski,ab N. J. Long,*a M. K. Kuimova,a A. A. Kornyshev,a A. G. Taylora
and A. J. P. Whitea
Received (in Durham, UK) 15th October 2008, Accepted 20th November 2008
First published as an Advance Article on the web 15th December 2008
DOI: 10.1039/b818110a
A range of charge-transfer, donor–acceptor systems have been synthesised and subjected to
in-depth crystallographic, spectroscopic and interfacial studies. The pyridine- or phenyl-
substituted (N,N-diphenyl)amino derivatives contain an identical donor unit, NR3, and acceptor
species with varying acceptor strength, with three compounds (3, 6 and 9) being characterised by
X-ray structural analysis. Two of the compounds, the methyl 6 and octyl 8 derivatives, show
interfacial localisation via fluorescence confocal microscopy and illustrate their potential as
light-driven optical machines.
charge-transfer, donor–acceptor properties, which could
localise at ITIES and be suitable to act as the light-driven
molecular machines.
Introduction
The chemistry and physics of liquid–liquid interfaces is an
increasingly important area of interdisciplinary research.
Donor–acceptor systems with large dipole moments have
already found successful applications in non-linear optics,9
Many fundamental processes in nature and technology
solar energy conversion10 and molecular electronics. The
donor and acceptor moieties can be chosen such that
take place at liquid interfaces e.g. uptake and reactions of
pollutants at the surface of water droplets and ice particles in
the atmosphere, phase transfer catalysis and ion-, electron-
the highest lying molecular orbital (HOMO) is localised on
and proton-transfer reactions at liquid–liquid and liquid–
membrane interfaces.1–3 Such interfaces are also ideal
(LUMO) on the acceptor part. Thus a transition from HOMO
the donor part and the lowest unoccupied molecular orbital
candidates for the assembly of functional supramolecular
to LUMO following excitation results in the formation of
architectures, with potential supply of building blocks
the intramolecular charge-transfer (ICT) excited state. The
dissolved in either of the liquid phases. One very promising
particular strength of this approach is that the energy of the
system is the interface of two immiscible electrolytic solutions
ICT transitions can be finely tuned by chemical modifications
(ITIES), a solution of organic ions in oil and inorganic ions in
which affect the redox potential of either the donor or
water that has a large free energy of transfer between the two
acceptor parts.
liquids.4 Following an application of an external electric field,
For an efficient utilisation of the excitation energy it
two ‘back-to-back’ electrical double layers are formed on each
is important to have the maximum overlap between the
side of the liquid–liquid interface. Thus created voltage drops,
up to 0.5 V AÀ1,5 can offer an additional degree of control of
For example, for the solar energy conversion the overlap with
absorption spectrum of the molecule and the light source.
supramolecular assembly and its functions. Current interest in
the solar spectrum, ca. 400–500 nm maximum, is required. For
ITIES stems from the fact that it could be used as a medium
non-linear optical applications, such as second harmonic
for phase transfer catalysis,6 an environment for artificial
generation (SHG), the spectral overlap with the output of
the commercially available pulsed lasers is desirable. Widely
photosynthesis and solar energy conversion.7
Since ITIES are accessible for self-assembly processes from
commercially available Ti:sapphire femtosecond lasers have
the emission maximum of ca. 800 nm, indicating that for
both phases, they constitute a very promising environment for
the self-assembly of interfacial supramolecular nano-scale
objects with amphiphilic compartments.8 Our goal is to
In the course of this project a theory was developed11 that
design and synthesise a series of molecules with promising
resonance enhancement of SHG the absorption at ca.
400 nm might be beneficial.
describes the operation of such machines, termed ‘optical
teaspoons’, that encompasses the modes of conformational
a Department of Chemistry, Imperial College London, South
Kensington, London, UK SW7 2AZ
motion of the molecule depending on different initial orienta-
b Department of Organic Chemistry, Faculty of Chemistry, University
tion and charge distribution, as well as various applied
voltages. That treatment describes the anticipated optical
response to the application of an external electric field, on
both steady state and time resolved properties of the interfacial
chromophore (electrochromic effects). In parallel we carried
´
´
´
´
of Łodz, Narutowicza 68, 90-136 Łodz, Poland
w Electronic supplementary information (ESI) available: Details
of crystallographic refinements, thermal ellipsoid plots and CIF
files. CCDC reference numbers 696198–696200. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/b818110a
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
598 | New J. Chem., 2009, 33, 598–606

View Article Online
out the synthesis of the candidate molecules, which could be
exploited as ‘optical teaspoons’, thus in this manuscript we
report the synthesis of a series of compounds, 1–10, all of
which contain an identical donor moiety, NR3, and acceptor
moieties with varying acceptor strengths. The primary goal of
the present publication was to ascertain that our target
molecules (i) localise at the water–oil interface, suitable for
the formation of ITIES, (ii) possess desirable charge-transfer
properties and (iii) show the spectral overlap with the tuning
curve of the Ti:sapphire femtosecond laser, to enable future
SHG studies.
Synthesis and spectroscopic characterisation
Fig. 1 The molecular structure of 3.
In the synthesis of the candidate molecules, the compounds
were designed to possess clear hydrophobic and hydrophilic
functionalities to enable interfacial localisation. Pyridine or
phenyl-substituted (N,N-diphenyl)amino derivatives were
focussed on due to their ease of synthesis, rod-like structures,
low cost, photochemical and SHG properties. For example,
2-(N-methyl-N-isopropylamino)-5-cyanopyridine exhibits so-
called twisted internal charge-transfer (TICT),12 molecules of
(N,N-di-p-tolylamino-p-styryl)benzene can self-organise on
surfaces to form microdots,13 tetraaryl amines exhibit SHG
properties14 and oligomeric and polymeric materials based on
triphenylamine15–17 can be electroluminescent and electron
conductors. The synthetic strategy for compounds 1–5
was based on palladium-catalysed Suzuki coupling and is
illustrated in Scheme 1.
structure of 3 is shown in Fig. 1 and reveals a series of twists
along the N–Ar–Ar–NO2 axis with torsion angles of ca. +22,
+24 and À14 about the N(1)–A, A–B and B–NO2 bonds,
respectively, suggesting less than perfect alignment (and thus
conjugation) of the p systems. Adjacent molecules stack to
form a centrosymmetric ‘‘dimer pair’’ held together by
N(p)Á Á Áp and p–p contacts (interactions a and b in Fig. 2 here,
and Fig. S2, ESIw). The N(p)Á Á Áp interaction involves ring A in
one molecule and the NO2 group in a Ci-related counterpart,
and has an NÁ Á Ácentroid separation of ca. 3.72 A [though the
O(28) atom is closer to ring A centroid at ca. 3.49 A]. The p–p
interaction b between ring B and its centrosymmetrically
related counterpart has centroidÁ Á Ácentroid and mean inter-
planar separations of ca. 3.73 and 3.60 A, respectively, and the
two rings are perfectly parallel due to them being Ci-related.
Adjacent ‘‘dimer pairs’’ are held together by a C–HÁ Á Áp
interaction between the C(23) proton on ring D in one dimer,
and the ‘‘unused’’ face of ring B in another dimer (interaction c
in Fig. 2 and S2, ESIw), resulting in an extended 2D sheet of
dimers arranged in a parquet-like fashion (Fig. 2). The
C–HÁ Á Áp interaction c has HÁ Á Áp 2.90 A, C–HÁ Á Áp 1221, with
the HÁ Á Áp vector being inclined by ca. 791 to the aromatic ring
plane; the centroidÁ Á Ácentroid and HÁ Á Áp vectors of inter-
actions b and c, respectively, subtend an angle of ca. 1571 at
the ring B centroid.
Compounds 1–5 were isolated in high yields (ca. 80%) and
easily purified via column chromatography. The structures and
purity of 1–5 were confirmed by spectroscopic analysis
(1H, 13C, MS, IR) and CHN microanalysis. For example,
1
the H NMR of 3 shows characteristic doublets for the nitro-
substituted phenyl ring protons at 8.26 ppm and 7.69 ppm.
A signal for the second disubstituted phenyl ring protons is
present at 7.50 ppm. The remaining protons of the mono-
substituted and disubstituted phenyl rings give two multiplets
at 7.06–7.16 and 7.27–7.32 ppm, respectively. The high-
resolution mass spectrum (HRMS, EI, 70 eV) of 3 shows a
mass peak at 366.1368 m/z with the calculated value for
C24H18N2O2 being 366.1368. Slow diffusion of n-hexane to
a saturated CH2Cl2 solution of 3 afforded red-orange
monocrystals suitable for X-ray analysis. The molecular
Fig. 2 Part of one of the 2D parquet-like sheets of Ci-related pairs of
molecules present in the structure of 3.
Scheme 1 Formation of compounds 1–5.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 598–606 | 599

View Article Online
Scheme 2 Formation of compounds 6–10.
Compounds 1–5 were then derivatised to include enough
amphiphilic character to localise them onto a water–organic
interface. This was executed via addition of carbon chains to
the structure, protonation or oxidation of the pyridine nitro-
gen atom to form compounds 6–10. For example, reaction of 1
with methyl iodide at room temperature afforded the N-methyl
iodide salt 6 in good yields, as did the similar reaction of 2 to
form 7 (Scheme 2). 1 was reacted with 1-iodooctane overnight
at 70 1C to yield the long chain derivative 8 and the hydro-
chloride salt 9 was obtained via direct acidification of 1 via
concentrated HCl addition. The N-oxide 10 was obtained by
oxidation of 1 with 3-chloroperoxo benzoic acid (MCPBA).
The structures of compounds 6, 7, 8 and 10 were confirmed
by spectroscopic analysis (1H, 13C, MS, IR) and microanalysis.
Structural X-ray analysis was obtained for compounds 6
and 9, though good microanalysis of 9 could not be obtained
due to the relatively high amount of water present in the
crystal lattice as confirmed by the X-ray analysis. The 1H
NMR spectrum of 6 shows 2H doublets for the pyridine ring
at 9.07 ppm and 8.08 ppm, and 2H doublets for the C6H4 ring
protons at 7.65 ppm and 7.03 ppm. A multiplet for the ortho-
protons of the C6H5 ring is observed at 7.34–7.31 ppm and for
the meta-, para-C6H5 ring protons it is at 7.18–7.13 ppm. The
methyl group gives rise to a signal at 4.53 ppm. In the 13C
NMR spectrum, the diagnostic signal for the methyl group
carbon is at 47.94 ppm. The mass spectrum (FAB+) of 6
shows a mass peak for the cationic fragment at 337 m/z
confirming the molecular ion. Slow diffusion of n-pentane into
a saturated CH2Cl2 solution of 6 afforded deep-orange mono-
crystals suitable for X-ray analysis and the molecular structure
of 6 is shown in Fig. 3. The structure was found to be highly
disordered, with two 50% occupancy orientations (I and II)
for the whole of the cation being identified (in fact, the iodine
anion is the only ordered atom present)—see the ESIw for
more details. Orientation I is shown in Fig. 3, orientation II in
Fig. S4 (ESIw), and an overlay of both in Fig. S6 (ESIw). As a
result of this extensive disorder neither the individual mole-
cular or extended structures can be discussed in detail, but
some broad features are worthy of mention. As was seen in the
structure of 3, here for both orientations the N–Ar–Ar axis is
twisted, and the packing is dominated by a stack of molecules
in an alternating –I–II–I–II– fashion (Fig. 4, see the ESIw
for why the stack has to be alternating). The
centroidÁ Á Ácentroid separations for the p–p interactions
that form this stack are ca. 3.8 A.
Fig. 3 The molecular structure of one (6-I) of the two 50% occu-
pancy orientations present in the asymmetric unit of 6.
Fig. 4 Part of one of the alternating stacks of molecules present in the
asymmetric unit of 6 (orientation I has been drawn with dark bonds,
and II with open bonds).
1
The H NMR spectrum of 7 shows signals for the pyridine
protons at 9.32 ppm (1H singlet), 9.03 ppm (1H doublet),
8.47 ppm (1H doublet), 8.03 ppm (1H doublet of doublets).
A doublet for the C6H4 ring protons is observed at 7.68 ppm
and all the other aromatic proton signals give multiplets
at 7.33–7.25 ppm (ortho-C6H5 protons) and 7.15–7.07 ppm
(meta-, para-C6H5 and C6H4 protons). A singlet for the methyl
group is observed at 4.74 ppm. The mass spectrum (FAB+)
of 7 shows a mass peak of the cation fragment at 337 m/z
1
confirming the presence of the molecular ion. The H NMR
spectrum of 8 shows doublets for the pyridine ring at
9.11 ppm and 8.10 ppm, and doublets for C6H4 ring protons
at 7.66 ppm and 7.05 ppm. Multiplets for the ortho-protons
of the C6H5 ring are observed at 7.37–7.33 ppm and for
the meta-, para-C6H5 ring protons at 7.18–7.15 ppm. The
signal at 4.75 ppm can be assigned to the two protons of
the methylene group attached to the pyridine nitrogen, the
remaining protons of the carbon chain give multiplets at
2.04–1.96 ppm and 1.41–1.16 ppm, and there is a three proton
triplet for terminal methyl group protons at 0.84 ppm. The
mass spectrum (FAB+) of 8 shows a mass peak of the cation
fragment at 435 m/z.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
600 | New J. Chem., 2009, 33, 598–606

View Article Online
Fig. 5 The molecular structure of 9. The geometries of the N–HÁ Á ÁCl hydrogen bonds [NÁ Á ÁCl] A, [HÁ Á ÁCl] A and [N–HÁ Á ÁCl]1 are (a) 3.041(3),
2.15, 170, and (b) 3.040(2), 2.14, 173. The hydrogen bonds c and d have OÁ Á ÁCl separations of 3.249(4) and 3.261(3) A, respectively. The C–HÁ Á Áp
interaction e has HÁ Á Áp 2.90 A, C–HÁ Á Áp 1451, with the HÁ Á Áp vector being inclined by ca. 811 to the aromatic ring plane.
The 1H NMR spectrum of 10 shows a doublet for C5H4NO
ring protons at 8.20 ppm, whilst all the other signals are
grouped into three multiplets at 7.46–7.43 ppm (C5H4NO
and C6H4 protons), 7.31–7.27 (ortho-C6H5 protons) and
7.14–7.07 (meta-, para-C6H5 and C6H4 protons). The
high-resolution mass spectrum (HRMS, EI, 70 eV) of 10
shows a mass peak at 338.1417 m/z with the calculated value
of C23H18N2O being 338.1419. Attempts to obtain a good
C–HÁ Á Áp angle of ca. 1451, and with the HÁ Á Áp vector being
inclined by ca. 811 to the ring plane. There is also a much
weaker C–HÁ Á Áp interaction between the C(25) proton and
ring E (not shown in the picture) with HÁ Á Áp 3.27 A, C–HÁ Á Áp
1151, and the HÁ Á Áp vector inclined by ca. 601 to the ring plane.
Both of the chloride anions sit above the centroids of
pyridyl rings. As can be seen from Fig. 5, Cl(2) approaches
the centroid of ring B with a ClÁ Á Ácentroid separation of
ca. 3.95 A (ClÁ Á Ácentroid vector inclined by ca. 641 to the ring
plane), whilst Cl(1) is only ca. 3.98 A from the centroid of ring
F of a lattice translated cation (ClÁ Á Ácentroid vector inclined
by ca. 591 to the ring plane).
There are numerous other intermolecular interactions in this
structure that serve to link the unit shown in Fig. 5 to form a
complex 3D array. The centroid of ring A is approached by the
C(49) proton of a lattice translated unit with HÁ Á Áp 3.01 A,
C–HÁ Á Áp 1351, and the HÁ Á Áp vector inclined by ca. 631 to the
ring plane. Ring B is involved in a p–p stacking interaction
with ring E of a likewise lattice translated unit with
centroidÁ Á Ácentroid and mean interplanar separations of
ca. 3.91 and 3.48 A, respectively, the two rings being inclined
by ca. 61 (the EÁ Á ÁB centroidÁ Á Ácentroid vector and the
BÁ Á ÁCl(2) vector subtend an angle of ca. 1771 at the ring B
centroid). Ring C is approached by the C(46) proton from a
Ci-related unit with HÁ Á Áp 3.05 A, C–HÁ Á Áp 1361, and the
HÁ Á Áp vector inclined by ca. 841 to the ring plane. The
‘‘unused’’ face of ring D is involved in a second C–HÁ Á Áp
interaction, here with the C(45) proton from a lattice trans-
lated unit with HÁ Á Áp 2.92 A, C–HÁ Á Áp 1411, and the HÁ Á Áp
vector inclined by ca. 781 to the ring plane; the two inter-
actions subtend an angle of ca. 1601 at the ring D centroid.
Both faces of ring G act as acceptors of C–H hydrogen bonds,
from the C(53) and C(21) protons of lattice translated and
Ci-related units, respectively, C(53)–HÁ Á ÁG has HÁ Á Áp 2.91 A,
1
quality H NMR spectrum of 9 failed due to our assumption
that in solution 9 is in equilibrium with nonprotonated 1, and
such fast proton exchange limited clear assignation of the
signals. However by slow diffusion of n-pentane into a
saturated CH2Cl2 solution of 9, monocrystals suitable for
X-ray analysis were obtained. The molecular structure
of 9 is shown in Fig. 5 and revealed the presence of two
crystallographically independent cation–anion pairs in the
asymmetric unit, along with one water molecule and a highly
disordered 50% occupancy dichloromethane molecule
(the latter has been omitted from Fig. 5). Both independent
cations show twists along their N–Ar–Ar axis with torsion
angles of ca. +17 and À28 about the N(1)–A and A–B bonds,
respectively, and ca. À27 and À11 about the N(31)–E and E–F
bonds, respectively. Each chloride anion is linked to one of the
cations by an N–HÁ Á ÁCl hydrogen bond (interactions a and b
in Fig. 5), and to each other via the cocrystallised water
molecule (interactions c and d). The N–HÁ Á ÁCl hydrogen
bonds have [NÁ Á ÁCl] and [HÁ Á ÁCl] separations (A), and
[N–HÁ Á ÁCl] angles (1) of 3.041(3), 2.15, 170, and 3.040(2),
2.14, 173 for a and b, respectively. The OÁ Á ÁCl hydrogen bonds
c and d have heteroatom separations of 3.249(4) and 3.261(3) A,
respectively (the water protons could not be located—see
the ESIw for more details). The two cations are further linked
by a C–HÁ Á Áp hydrogen bond between the C(45) proton and
ring D (interaction e) with an HÁ Á Áp distance of ca. 2.90 A, a
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 598–606 | 601

View Article Online
C–HÁ Á Áp 1501, HÁ Á Áp vector inclined by ca. 761 to the ring
plane; C(21)–HÁ Á ÁG has HÁ Á Áp 3.21 A, C–HÁ Á Áp 1551, HÁ Á Áp
vector inclined by ca. 541 to the ring plane; the two inter-
actions subtend an angle of ca. 1311 at the ring G centroid.
A similar pattern is seen for ring H, with approaches from the
C(15) and C(24) protons of different lattice translated
units—C(15)–HÁ Á ÁH has HÁ Á Áp 2.83 A, C–HÁ Á Áp 1471, HÁ Á Áp
vector inclined by ca. 821 to the ring plane; C(24)–HÁ Á ÁH has
HÁ Á Áp 2.90 A, C–HÁ Á Áp 1511, HÁ Á Áp vector inclined by ca. 731
to the ring plane; the two interactions subtend an angle of
ca. 1561 at the ring H centroid.
Physical characterisation and interfacial studies
Compounds 1–10 contain an identical donor moiety, NR3,
and acceptor moieties with varying acceptor strength. These
compounds are analogous to donor–acceptor systems9 in
which the lowest excited state has been previously identified
as ICT and these molecules have been shown by Hyper
Rayleigh scattering with 800 or 1300 nm femtosecond pulses
to possess high static first hyperpolarisabilities corresponding
to high non-linear optical properties.9
Compounds 1–5 incorporate weak acceptor groups
(high lying LUMOs), resulting in the high energy of the ICT
transition, centred at ca. 350 nm. Compound 3 has additional
strong bands associated with p–p* NO2 group transitions
at ca. 400 nm. Compounds 6–9 incorporate a stronger
acceptor—the pyridinium group—which should result in the
lower lying ICT states. Consistent with this, we observe the
marked red shift of the lowest lying absorption maxima of 6, 8
and 9 to ca. 420 nm (Table 1) indicating the significant
lowering of the excited state energy. Surprisingly, compound
7 is characterised by an absorption maximum of 350 nm,
similar to 1, even though it has acceptor moiety similar to 6.
This could result from a different charge distribution pattern
in 6 and 7.
Fig. 6 (a) Absorption spectra of 1, 7 and 8 and (b) emission spectra of
1, 6 and 8 in methanol.
Following excitation into the lowest absorption band,
emission is observed for 1, 2, 4–9 in several organic solvents
and in water. The emission maximum is significantly red-
shifted compared to the absorption maximum (the large
Stokes shift), confirming the ICT nature of the lowest excited
state. In addition the emission maxima of 6, 8, 9 are shifted to
lower energy compared to 1, consistent with lower lying
LUMO (Fig. 6b). Absorption and emission maxima of 6, 8,
9 change as a function of solvent, e.g. see Fig. 7 for 6. In
conclusion, in the present series of compounds the absorption
spectra of 6, 8, 9 represent the best match for the solar
Table 1 Absorption and emission properties of 1, 6, 8, 9 in different
solvents
lmax/nm (abs)/(emis)
1
6
8
9
Solvent
Methanol
CH2Cl2
Toluene
345/455
350/460
345/460
425/645
456/640
427/470a
423/640
460/640
425/466a
425/645
460/645
425/—b
a
The small Stokes shifts and the presence of a well defined vibrational
structure in the emission spectra of 6 and 8 in toluene suggest that the
b
emitting state is not ICT. Not measured.
Fig. 7 (a) Absorption and (b) emission spectra of 6 in different
solvents.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
602 | New J. Chem., 2009, 33, 598–606

View Article Online
spectrum and for the output of the ultrafast Ti:sapphire lasers.
In addition, there is the possibility to tune absorption and
emission energy of these compounds with the environmental
parameters.
Our target applications such as molecular electronics and
SHG generation can benefit from selective localisation of a
probe on the interface. Due to the preferential orientation of
the probe, a much higher degree of order in the interfacial
region could be achieved, resulting in a large cooperative
dipole moment. The preferential localisation on the interface
of the condensed/air phases is normally achieved by standard
methods such as coating of the solid–air interface or Langmuir
Blodgett film formation on the liquid–air interface. However
these methods require deposition techniques to be used.
An alternative solution is the spontaneous formation of the
self-organised monolayers of the probe on the interface of
two immiscible liquids. Compounds 6–9 consist of a large
hydrophobic moiety with a single positive charge localised on
one end. This charge renders compounds 6–9 water soluble
and could assist in self-assembly on the liquid–liquid interface.
Here we have directly tested the possibility of spontaneous
formation of monolayers of 6, 8 and 9 on water–oil interfaces
using three complementary observations: (i) of surface tension
in the interfacial region, (ii) of the bulk absorption change
upon formation of the oil–water interface and (iii) of inter-
facial fluorophore localisation using confocal fluorescence
microscopy.
The photographs of unmodified water–CH2Cl2 and
water–toluene interfaces and interfaces in the presence of 6
and 8 (Fig. 8) clearly show the reduction in surface tension
between the two immiscible liquids in the presence of 6 and 8.
The interfacial tension of oil vs. water is 43018 as indicated
by the curvature of the unmodified interface. Surfactants
localised at the interface reduce the surface tension, causing
the reduction in contact angle. A large reduction in contact
angle is observed for 6 and 8, indicating that these molecules
are acting as surfactants and are preferentially localised on the
water–oil interface.
Fig. 9 Absorption study of interfacial concentration of 9 (a), 6 (b)
and 8 (c). The change in absorption spectrum of 9 in CH2Cl2 upon
addition of the aqueous layer is consistent with deprotonation. For 6
and 8 given the extinction coefficient of 1.8 Â 104 MÀ1 cmÀ1 in CH2Cl2
and 1.4 Â 104 MÀ1 cmÀ1 in water, we estimate from absorption change
of ca. 0.05 OD that ca. 1e15 molecules cmÀ2 are localised at the
interface.
To estimate the interfacial concentration of 6, 8 and 9 we
have used the change in absorption of either the aqueous or
the oil layer containing the known concentration of these
probes upon addition of the second layer. For example, the
initial absorption of 6 in CH2Cl2 stock solution is ca. 0.05
(e = 1.8 Â 104 MÀ1 cmÀ1 in CH2Cl2 and 1.4 Â 104 MÀ1 cmÀ1
in water), which is equivalent to the concentration of 2.8 mM
(in 0.5 ml). Upon addition of the second aqueous layer the
resulting absorption of both CH2Cl2 and water layers is
close to zero (Fig. 9b and c). The calculation shows that
{(2.8 Â 10À6 M) Â (5 Â 10À4 l) Â (6.03 Â 1023 molÀ1)}/
1 cm2 = 8 Â 1014 molecules cmÀ2 are localised on the
liquid–liquid interface. For all the systems studied we obtain
the interfacial concentration in the order 1 Â 1015 molecules cmÀ2
.
Contrary to the observations for 6 and 8, the surface tension
of the water–oil interface does not change in the presence of 9.
In addition, the absorption spectroscopic study shows that
upon addition of aqueous layer to the CH2Cl2 stock solution
containing 9, the absorption spectrum changes (Fig. 9a).
The change from 420 nm absorption maximum to 350 nm
maximum is consistent with deprotonation of 9 yielding
compound 1 in the CH2Cl2 phase and protons in aqueous
Fig. 8 Photographs of unmodified and modified interfaces.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 598–606 | 603

View Article Online
as the reference solvents. Mass spectra were recorded using
positive FAB methods for the methyl iodide salts and EI
for the rest of the complexes, on a micromass Autospec Q
spectrometer. Infrared spectra were recorded as KBr disks on
a Perkin Elmer Spectrum RX FTIR spectrometer. Micro-
analyses were carried out by Mr S. Boyer at SACS (Scientific
Analysis and Consultancy Services) at the University of North
London. All other chemicals were purchased from the Aldrich
Chemical Co.
Fluorescence imaging was performed using a confocal laser
scanning microscope (Leica TCS SP2), coupled to a CW
argon-ion laser (488 nm). The fluorescence emission of 6 and
8 from the thin layer of solution placed between the glass
coverslips was spectrally dispersed using a prism and detected
using a PM tube. Water immersion 63Â objective (NA = 1.2)
was used in all measurements.
Fig. 10 Confocal fluorescence images of water–oil interfaces
modified with 6 (water–CH2Cl2) and 8 (water–toluene).
phase, see Fig. 9a. Thus 9 is not sufficiently stable in the
presence of water and is not suitable for formation of
self-organised monolayers at aqueous–oil interface.
Additional confirmation of the interfacial localisation of 6
and 8 comes from the confocal fluorescence microscopy study
(Fig. 10). The images of the interfacial region between water
and oil phases obtained following 488 nm excitation clearly
show intense fluorescence originating from the interfacial
layer. The intensity of the fluorescence from the interfacial
region is considerably higher than that observed from either of
the bulk phases.
Synthesis of 1
4-Iodopyridine (205 mg, 1.0 mmol), 4-(diphenylamino)phenyl
boronic acid (318 mg, 1.1 mmol) and dimethoxyethane (DME)
(12 ml) were placed in a Schlenk tube under an atmosphere of
nitrogen. Sodium carbonate (315 mg, 3 mmol) was dissolved in
a minimum volume of water. The catalyst, Pd(PPh3)4 (46 mg,
0.022 equiv.), was added to the DME solution of the reactants,
followed immediately by addition of the sodium carbonate
solution. After stirring at room temperature for 1 h, the
temperature was increased to 90 1C for 16 h. The solvent
was removed under reduced pressure, the resultant oily residue
dissolved in ethyl acetate–chloroform (2 : 1) and subjected to
column chromatography on SiO2. The second major, yellow
fraction was collected. Evaporation of solvent gave the
required product 1 as a colourless solid (240 mg, 75% isolated
yield).
Conclusions and future work
In summary, we have synthesised a range of molecules with
desirable charge-transfer properties that can localise on the
ITIES and bear capability of light induced intramolecular
electron transfer. We have performed a series of absorption
measurements for a number of immiscible solvent combina-
tions to determine which molecules and solvent systems would
best fit their application in future light-driven molecular
machines, such as ‘interfacial optical teaspoons’.11
These studies yielded two candidates that show interfacial
localisation in water–toluene and water–dichloroethane
mixtures—the methyl 6 and octyl 8 derivatives and on-going
work will establish the orientations of our molecules relative to
the interface plane. Work is in progress to take our findings
here—at the pure ‘solvent–solvent’ interface—and localise
similar molecules at ITIES, control their orientation by
electric field, and study different electrochromic effects, be it
in optical adsorption or SHG signal with applied potential, in
order to create the first example of a light-driven molecular
machine acting at ITIES.
1H NMR (400 MHz) d (CDCl3): 8.62 (d, 2H, J = 6.0 Hz,
C5H4N), 7.52 (d, 2H, J = 11.0 Hz, C6H4), 7.47 (d, 2H,
J
= 6.0 Hz, C5H4N), 7.32–7.27(m, 4H, ortho-C6H5),
7.16–7.06 (m, 8H, meta-, para-C6H5, C6H4); 13C NMR
(100 MHz) 149.17, 148.92, 147.68, 147.16, 129.37, 127.58,
124.93, 123.57, 122.80, 120.80; MS (EI 70 eV) m/z = 322
[M+], 244 [M+ À C5H4N], 78 [C5H4N]; HRMS: m/z =
322.1463 (calc. for C23H18N2: 322.1469); anal. calc. for
C23H18N2: C, 85.68; H, 5.63; N, 8.69. Found C, 85.59; H,
5.68; N, 8.71%.
Synthesis of 2
4-(Diphenylamino)phenyl boronic acid (318 mg, 1.1 mmol),
3-iodo-pyridine (205 mg, 1.0 mmol) and DME (12 ml) were
placed in a Schlenk tube under an atmosphere of nitrogen.
Pd(PPh3)4 (46 mg) was added to the solution, along with
sodium carbonate (315 mg, 3 mmol) dissolved in a minimum
volume of water. After stirring at room temperature for 1 h the
mixture was heated for 16 h at 70 1C. The solvent was
evaporated under reduced pressure and the crude product
was subjected to column chromatography on silica gel (eluent:
n-hexane–CHCl3 1 : 1). The first main fraction contained
the pure product, which after evaporation gave red-orange
crystals of 2 (yield 293 mg, 80%).
Experimental section
General procedures
All preparations were carried out using standard Schlenk
techniques. All solvents were distilled over standard drying
agents under nitrogen directly before use, and all reactions
were carried out under an atmosphere of nitrogen.
All NMR spectra were recorded using a Delta upgrade on
JEOL EX270 400 and 500 MHz Bruker spectrometers.
Chemical shifts are reported in d (ppm) using CDCl3 (1H d
7.26 ppm), acetone-d6 (1H d 2.05 ppm), D2O (1H d 4.75 ppm)
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
604 | New J. Chem., 2009, 33, 598–606

View Article Online
1H NMR (400 MHz) d (CDCl3): 8.84 (d, 1H, J = 1.6 Hz,
C5H4N), 8.54 (dd, 1H, J = 4.8 Hz, 1.6 Hz, C5H4N), 7.84
(d of t, 1H, J = 8.4 Hz, 1.6 Hz, C5H4N), 7.45 (d, 2H, J = 8.3 Hz,
C6H4), 7.33 (dt, 1H, J = 8.4 Hz, 4.8 Hz, C5H4N), 7.30–7.26
(m, 4H, ortho-C6H5), 7.17–7.03 (m, 8H, meta-, para-C6H5,
C6H4); 13C NMR (100 MHz) 148.00, 147.92, 147.43, 136.13,
133.68, 131.25, 129.34, 127.74, 124.68, 123.55, 123.48, 123.27;
MS (EI 70 eV) m/z = 322[M+]; anal. calc. for C23H18N2: C,
85.68; H, 5.63; N, 8.69. Found C, 85.61; H, 5.57; N, 8.59%.
CH2Cl2–n-pentane giving yellow needles of 6 (132 mg,
92% yield).
1H NMR (400 MHz) d (CDCl3) 9.07 (d, 2H, J = 6.0 Hz,
C5H4N+), 8.08 (d, 2H, J = 6.0 Hz, C5H4N+), 7.65 (d, 2H,
J = 8.4 Hz, C6H4), 7.34–7.31 (m, 4H, ortho-C6H5), 7.18–7.13
(m, 6H, meta-, para-C6H5), 7.03 (d, 2H, J = 8.4 Hz, C6H4),
4.53 (s, 3H, CH3); 13C NMR (100 MHz) 154.85, 145.75,
144.79, 129.76, 129.02, 126.19, 125.29, 123.87, 122.66,
120.27, 47.94; MS FAB (+ve, nba) m/z = 337 [M+]; anal.
calc. for C24H21N2I: C, 62.08; H, 4.56; N, 6.03. Found C,
61.98; H, 4.51; N, 5.94%.
Synthesis of 3
A similar procedure to the synthesis of 2 was followed except
that 1-iodo-4-nitrobenzene (249.0 mg, 1.0 mmol) was utilised.
3 was isolated as red-orange crystals (yield 293 mg, 80%).
1H NMR (400 MHz) d (CDCl3) 8.26 (d, 2H, J = 8.8 Hz,
C6H4NO2), 7.69 (d, 2H, J = 8.8 Hz, C6H4NO2), 7.50 (d, 2H,
J = 8.4 Hz, C6H4), 7.32–7.27 (m, 4H, ortho-C6H5), 7.16–7.06
(m, 8H, meta-, para-C6H5, C6H4); 13C NMR (100 MHz)
148.86, 147.18, 147.04, 146.49, 131.56, 129.45, 128.05,
126.85, 125.05, 124.17, 123.70, 122.82; MS (EI 70 eV) m/z =
366 [M+], 320 [M+ À NO2]; HRMS: m/z = 366.1368 (calc.
for C24H18N2O2: 366.1368); anal. calc. for C24H18N2O2: C,
78.67; H, 4.95; N, 7.65. Found C, 78.60; H, 4.89; N, 7.61%.
Synthesis of 7
1 (100 mg, 0.30 mmol) was dissolved in methyl iodide (7 ml).
The solution was stirred for 30 min at room temperature and
the product precipitated as a yellow solid. After filtration and
washing with Et2O, the crude product was recrystallised from
CH2Cl2–n-pentane giving yellow needles of 7 (120 mg, 83%
yield).
1H NMR (400 MHz) d (CDCl3) 9.32 (s, 1H, C5H4N), 9.03
(d, 1H, J = 5.6 Hz, C5H4N), 8.47 (d, 1H, J = 8.42 Hz,
C5H4N), 8.03 (dd, 1H, J = 5.6 Hz, 4.0 Hz, C5H4N), 7.68
(d, 2H, J = 8.9 Hz, C6H4), 7.33–7.25 (m, 4H, ortho-C6H5),
7.15–7.07 (m, 8H, meta-, para-C6H5, C6H4), 4.74 (s, 3H, CH3);
MS FAB (+ve, nba) m/z = 337 [M+]; anal. calc. for
C24H21N2I: C, 62.08; H, 4.56; N, 6.03. Found C, 61.99; H,
4.47; N, 5.93%
Synthesis of 4
A similar procedure to the synthesis of 2 was followed except
that 4-iodobenzaldehyde (232.0 mg, 1.0 mmol) was utilised. 4
was isolated as red-orange crystals (yield 297 mg, 82%).
1H NMR (270 MHz) d (CDCl3) 10.02 (s, 1H, CHO), 7.92
(d, 2H, J = 8.1 Hz, C7H5O), 7.72 (d, 2H, J = 8.1 Hz, C7H5O),
7.51 (d, 2H, J = 8.6 Hz, C6H4), 7.32–7.25 (m, 4H, ortho-
C6H5), 7.15–7.04 (m, 8H, meta-, para-C6H5, C6H4); IR
(KBr, cmÀ1): 1691; MS (EI 70 eV) m/z = 349 [M+]; HRMS
m/z = 349.1464 (calc. for C25H19NO: 349.1466); anal. calc. for
C25H19NO: C, 85.93; H, 5.48; N, 4.01. Found C, 85.87; H,
5.64; N, 3.95%.
Synthesis of 8
A mixture of 1 (100 mg, 0.30 mmol) and 1-iodooctane (7 ml)
was heated for 16 h at 70 1C. The product precipitated as a
yellow solid, and after filtration under reduced pressure and
washing with Et2O, it was recrystallised from CH2Cl2–
n-pentane giving yellow needles of 8 (140 mg, 80% yield).
1H NMR (400 MHz) d (CDCl3) 9.11 (d, 2H, J = 7.2 Hz,
C5H4N+), 8.10 (d, 2H, J = 7.2 Hz, C5H4N+), 7.66 (d, 2H,
J = 8.8 Hz, C6H4), 7.37–7.33 (m, 4H, ortho-C6H5), 7.18–7.15
(m, 6H, meta-, para-C6H5), 7.05 (d, 2H, J = 8.8 Hz, C6H4),
4.75 (t, 2H, J = 7.6 Hz, N+CH2(CH2)6CH3), 2.04–1.96
(m, N+CH2(CH2)6CH3), 1.41–1.16 (m, N+CH2(CH2)6CH3),
0.84 (t, 3H, J = 6.8 Hz, N+CH2(CH2)6CH3); 13C NMR
(100 MHz) 154.98, 145.75, 143.94, 129.81, 129.04, 126.27,
125.39, 123.82, 122.76, 120. 24, 60.63, 31.62, 28.98, 28.96,
16.02, 22.53, 14.01; MS FAB (+ve, nba) m/z = 435 [M+], 322
[M+ À C8H17]; anal. calc. for C31H35N2I: C, 66.19; H, 6.27; N,
4.98. Found C, 66.15; H, 6.26; N, 4.92%.
Synthesis of 5
A similar procedure to the synthesis of 2 was followed except
that 4-iodo-acetophenone (246.0 mg, 1.0 mmol) was utilised. 5
was isolated as red-orange crystals (yield 308 mg, 85%).
1H NMR (400 MHz) d (CDCl3) 8.01 (d, 2H, J = 8.0 Hz,
C8H7O), 7.66 (d, 2H, J = 8.0 Hz, C8H7O), 7.52 (d, 2H,
J
= 8.4 Hz, C6H4), 7.31–7.26 (m, 4H, ortho-C6H5),
7.16–7.05 (m, 8H, meta-, para-C6H5, C6H4), 2.63 (s, 3H,
CH3); 13C NMR (100 MHz) 197.56, 148.13, 147.36, 145.13,
135.28, 133.07, 129.33, 128.91, 127.86, 126.42, 124.73, 123.31,
123.23, 26.55; IR (KBr, cmÀ1): 1678; MS (EI 70 eV) m/z = 363
Synthesis of 9
To finely powdered 1 (200 mg, 0.60 mmol) concentrated HCl
solution (40 ml) was added. The resulting suspension was
sonicated and then stirred at room temperature for 30 min.
Upon stirring colourless 1 was slowly dissolved to form a
yellow solution. Removal of the solvent in vacuo effected
precipitation of a yellow solid which was filtered off and dried
under vacuum overnight. The crude product was recrystallised
from CH2Cl2–n-pentane giving yellow needles of 9 (345 mg);
anal. calc. for C23H19N2ClÁ2CH2Cl2ÁH2O: C, 54.92; H, 4.61;
N, 5.12. Found C, 55.12; H, 4.91; N, 5.13%.
[M+]; HRMS m/z
= 363.1617 (calc. for C26H21NO:
363.1623); anal. calc. for C26H21NO: C, 85.92; H, 5.82; N,
3.85. Found C, 85.93; H, 5.86; N, 3.79%.
Synthesis of 6
1 (100 mg, 0.30 mmol) was dissolved in methyl iodide (7 ml).
The solution was stirred for 30 min at room temperature and
the product precipitated as a yellow solid. After filtration and
washing with Et2O, the crude product was recrystallised from
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 598–606 | 605

View Article Online
Synthesis of 10
F2 refinement, R1 = 0.062, wR2 = 0.159, 3741 independent
observed absorption-corrected reflections [|Fo| 4 4s(|Fo|),
2ymax = 1421], 525 parameters. CCDC 696200.
To a solution of 1 (100 mg, 0.30 mmol) in CH2Cl2 (2 ml),
3-chloroperoxo benzoic acid (105 mg, 0.60 mmol) was added
at 0 1C. The resultant mixture was stirred at room temperature
for 1 h and then the reaction was quenched by addition of
aqueous NaHCO3 solution. The organic layer was separated
and evaporated to dryness, and the crude product was
subjected to column chromatography on silica (ethyl acetate–
chloroform 80 : 8). The product was eluted as the final yellow
fraction and after evaporation of solvent afforded an oily
yellow solid 10.
Acknowledgements
MKK is thankful to the EPSRC Life Sciences Interface
programme for a personal fellowship. We would like to thank
Dr Klaus Suhling for the use of the confocal fluorescence
microscope, and the EPSRC for funding this work via the
‘Adventurous Chemistry’ programme.
1H NMR (400 MHz) d (CDCl3) 8.20 (d, 2H, J = 7.2 Hz,
C5H4NO), 7.46–7.43 (m, 4H, C5H4NO, C6H4), 7.31–7.27
(m, 4H, ortho-C6H5), 7.14–7.07 (m, 8H, meta-, para-C6H5,
C6H4); 13C NMR (100 MHz) 149.02, 147.03, 139.24, 138.28,
129.49, 128.74, 127.04, 125.13, 123.86, 122.72, 122.69; MS
(EI 70 eV) m/z = 338 [M+], 322 [M+ À O]; HRMS m/z =
338.1417 (calc. for C23H18N2O 338.1419); anal. calc. for
C23H18N2O: C, 81.63; H, 5.36; N, 8.28. Found C, 81.58; H,
5.27; N, 8.19%.
References
1 I. Benjamin, Chem. Rev., 2006, 106, 1212.
2 W. H. Steel and A. W. Walker, Nature, 2003, 424, 296.
3 K. B. Eisenthal and E. A. McArthur, J. Am. Chem. Soc., 2006, 128,
1068.
4 Z. Samec and T. Kakiuchi, Advanced Electrochemistry and
Electrochemical Science, ed. H. Gerischer and C. W. Tobias,
VCH, Weinheim, 1995, p. 297.
5 H. H. Girault and D. H. Schiffrin, Electroanalytical Chemistry, ed.
A. J. Bard, Dekker, New York, 1989, vol. 15, p. 1; H. H. Girault,
Modern Aspects of Electrochemistry, ed. J. O’M. Bockris et al.,
Plenum, New York, 1993, vol. 25, p. 1.
6 C. M. Starks, C. L. Liotia and M. Halpern, Phase Transfer
Catalysis, Chapman and Hall, New York, 1994.
Crystal data for 3: C24H18N2O2, M = 366.40, monoclinic,
P21/c (no. 14), a = 13.3702(3), b = 15.0329(3), c = 9.6952(2) A,
b = 104.763(2)1, V = 1884.34(7) A3, Z = 4, Dc = 1.292 g cmÀ3
,
m(Mo-Ka) = 0.083 mmÀ1, T = 173 K, orange blocky needles,
Oxford Diffraction Xcalibur 3 diffractometer; 6348 indepen-
dent measured reflections, F2 refinement, R1 = 0.045,
7 A. G. Volkov, D. W. Deamer, D. I. Tanelian and V. S. Markin,
Liquid Interfaces in Chemistry and Biology, Wiley, New York,
1998.
wR2
= 0.131, 4504 independent observed absorption-
8 B. Su, J.-P. Abid, D. J. Fermin, H. H. Girault, H. Hoffmannova,
P. Krtil P and Z. Samec, J. Am. Chem. Soc., 2004, 126, 915.
9 B. J. Coe, J. A. Harris, I. Asselberghs, K. Wostyn, K. Clays,
A. Persoons, B. S. Brunschwig, S. J. Coles, T. Gelbricht,
M. E. Light, M. B. Hursthouse and K. Nakatani, Adv. Funct.
Mater., 2003, 13, 347.
10 V. Balzani and F. Scandola, Supramolecular Photochemistry, Ellis
Horwood, New York, 1991, p. 427.
11 A. A. Kornyshev, M. Kuimova, A. M. Kuznetsov, J. Ulstrup and
M. Urbakh, J. Phys.: Condens. Matter, 2007, 19, 375111.
12 J. Dobkowski, J. Wojcik, W. Kozminski, R. Kolos, J. Waluk and
J. Michl, J. Am. Chem. Soc., 2002, 11, 2406.
corrected reflections [|Fo| 4 4s(|Fo|), 2ymax = 651], 253
parameters. CCDC 696198.
Crystal data for 6: (C24H21N2)(I), M = 464.33, monoclinic,
P21/n (no. 14), a = 5.69284(14), b = 40.9581(10), c =
9.1109(2) A, b = 101.802(2)1, V = 2079.46(9) A3, Z = 4,
Dc = 1.483 g cmÀ3, m(Mo-Ka) = 1.550 mmÀ1, T = 173 K,
yellow needles, Oxford Diffraction Xcalibur 3 diffractometer;
4216 independent measured reflections, F2 refinement, R1 =
0.044, wR2 = 0.113, 3113 independent observed absorption-
corrected reflections [|Fo| 4 4s(|Fo|), 2ymax = 541], 384
parameters. CCDC 696199.
13 H. Yangi, N. Matsuoka, M. Kondo, M. Nagawa and
Y. Taniguchi, Langmuir, 2001, 17, 5491.
14 T. Verbiest, D. M. Burland, M. C. Jurich, V. Y. Lee, R. D. Miller
and W. Volksen, Science, 1995, 268, 1604.
Crystal data for 9: (C23H19N2)(Cl)Á0.5H2OÁ0.25CH2Cl2,
ꢀ
M = 389.09, triclinic, P1 (no. 2), a = 9.0670(17), b =
15 M. Yano, Y. Ishida, K. Aoyama, M. Tatsumi, K. Sato, D. Shiomi,
A. Ichimura and T. Takui, Synth. Met., 2003, 137, 1275.
16 S. Tanaka, T. Iso and Y. Doke, Chem. Commun., 1997, 2064.
17 H. Tanaka, S. Tokito, Y. Taga and A. Okada, Chem. Commun.,
1996, 2175.
10.852(2), c = 21.500(4) A, a = 100.431(15), b = 90.436(15),
g = 103.375(16)1, V = 2021.4(7) A3, Z = 4 (2 independent
molecules), Dc = 1.278 g cmÀ3, m(Cu-Ka)=2.367 mmÀ1
T = 173 K, yellow plates, Oxford Diffraction Xcalibur PX
Ultra diffractometer; 7479 independent measured reflections,
,
18 G. Barnes and I. Gentle, Interfacial Science An Introduction,
Oxford University Press Inc., New York, 2005, p. 264.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
606 | New J. Chem., 2009, 33, 598–606