Synthesis and characterisation of luminescent fluorinated organoboron compounds

Article abstract of DOI:10.1039/b700317j

The reaction of 8-hydroxyquinoline (HQ) with B(C₆F ₅)₃ leads to the formation of the zwitterionic compound (C₆F₅)₃BQH (1), involving a proton migration from O to N. Compound 1 can be converted thermally to (C₆F ₅)₂BQ (2), which can also be prepared from (C ₆F₅)₂BCl and HQ. The reaction of HQ with (C₆F₅)B(OC₆F₅)₂ generates initially (C₆F₅)(OC₆F₅)BQ (3), which easily hydrolyses to give the diboron compound ((C₆F ₅)BQ)₂O (4). Compounds 1, 2 and 4 have been fully characterised, including X-ray analysis. The spectroscopic properties of these compounds, including photoluminescence (PL) have been investigated and compared with the non-fluorinated luminescent boron compound (C₆H ₅)₂BQ and also with AlQ₃. The changes in luminescent behaviour upon fluorination of these boron quinolinate compounds have been rationalised using computational studies. The Royal Society of Chemistry.

Full text of DOI:10.1039/b700317j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis and characterisation of luminescent fluorinated organoboron
compounds†
Juri Ugolotti, Sondra Hellstrom, George J. P. Britovsek,* Tim S. Jones,* Patricia Hunt and Andrew J. P. White
Received 9th January 2007, Accepted 8th February 2007
First published as an Advance Article on the web 28th February 2007
DOI: 10.1039/b700317j
The reaction of 8-hydroxyquinoline (HQ) with B(C6F5)3 leads to the formation of the zwitterionic
compound (C6F5)3BQH (1), involving a proton migration from O to N. Compound 1 can be converted
thermally to (C6F5)2BQ (2), which can also be prepared from (C6F5)2BCl and HQ. The reaction of HQ
with (C6F5)B(OC6F5)2 generates initially (C6F5)(OC6F5)BQ (3), which easily hydrolyses to give the
diboron compound ((C6F5)BQ)2O (4). Compounds 1, 2 and 4 have been fully characterised, including
X-ray analysis. The spectroscopic properties of these compounds, including photoluminescence (PL)
have been investigated and compared with the non-fluorinated luminescent boron compound
(C6H5)2BQ and also with AlQ3. The changes in luminescent behaviour upon fluorination of these boron
quinolinate compounds have been rationalised using computational studies.
An alternative method for tuning the emission wavelength and
Introduction
the quantum yield might be to vary the R substituents at the
Luminescent organic and organometallic compounds are cur-
boron centre. Thus far, the R groups have been generally phenyl
or electronically similar groups such as naphthyl,5 thienyl or
benzothienyl.6 Changing the electron-withdrawing or electron-
donating properties of the R substituents will affect the Lewis
acidity of the boron centre, which in turn may affect the boron–
quinoline interaction and thereby the HOMO–LUMO levels. In
order to explore this possibility, we decided to prepare a series of
pentafluorophenyl substituted boron quinolyl compounds of the
general formula (C6F5)2−n(OC6F5)nBQ where n = 0, 1 or 2. Com-
pared to a phenyl group, the electron-withdrawing C6F5 groups
will increase the Lewis acidity of the boron centre and thereby
affect the energy of the HOMO–LUMO levels, which are located
on the quinoline moiety of the molecule. Pentafluorophenyl groups
may also offer an advantage in terms of stability, cf . B(C6F5)3 is
considerably less air and moisture sensitive compared to B(C6H5)3
due to the increased steric protection of the Lewis acidic boron
centre.13 This, however, does not apply to compounds containing
BOC6F5 groups, which are much more susceptible to hydrolysis.14,15
Fluorination may also increase the volatility of the precursors,
which may prove beneficial in device formation. A few studies
on the effect of fluorination have been reported but these were
all on fluorination of the fluorescent heterocyclic part of the
molecule.9,16,17 We report here the synthesis and characterisation
of a series of novel fluorinated quinolyl boron compounds and
an experimental and computational assessment of the effects of
fluorination on their luminescent properties.
rently of great interest due to their potential application in organic
light emitting devices (OLEDs).1 Since the original report by
Tang and VanSlyke2 on the electroluminescent properties of AlQ3
(HQ = 8-hydroxyquinoline, see Fig. 1), many derivatives have
been prepared with emission wavelengths now covering most of
the visible region of the spectrum.3,4 More recently, luminescent
organoboron compounds have received increased attention due
to their better stability compared to aluminium compounds. Ex-
amples include mononuclear four-coordinate boron compounds
of the type R2B(N,N)4,9,10 and R2B(N,O),5–8 where N,N and N,O
are bidentate heterocyclic ligands with one neutral N and one
negative N or O donor, respectively, for example Ph2BQ (Fig. 1). In
addition, some luminescent oxygen-bridged diboron compounds
have been reported recently.11,12
Fig. 1 Examples of luminescent aluminium and boron compounds.
Various derivatives of these boron-based emitters have been
prepared, most of which are variations on the N,O or N,N part
of the molecule, whereby substituents on the heterocycle affect the
HOMO–LUMO levels and thereby the colour of the emission.9
Results and discussion
Synthesis of organoboron compounds
The reaction of 8-hydroxyquinoline (HQ) with BPh3 has been
reported to proceed smoothly at room temperature within
several hours to give Ph2BQ and benzene.5 In contrast, we found
that the reaction of B(C6F5)3 with HQ at room temperature
results in the formation of the zwitterionic adduct (C6F5)3BQH
Department of Chemistry, Imperial College London, Exhibition Road,
London, SW7 2AY, UK. E-mail: g.britovsek@imperial.ac.uk; t.jones@
imperial.ac.uk; Tel: +44-(0)20-75945863; +44-(0)20-75945794
† Electronic supplementary information (ESI) available: Molecular struc-
tures of compounds 1, 2 and 4 (Fig. S1–S6). See DOI: 10.1039/b700317j
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1425–1432 | 1425
©

View Article Online
(1), whereby the oxygen is coordinated to the boron centre and a
proton shift from oxygen to nitrogen has occurred (Scheme 1).
Interestingly, a similar proton shift has been observed in the
reaction between B(C6F5)3 and a-naphthol, whereby a proton
migrates from oxygen to the para-carbon centre upon coordination
of boron to oxygen.18 The adduct 1 can be converted to the
bis(pentafluorophenyl)borinic acid quinolyl ester (C6F5)2BQ (2)
by refluxing 1 in benzene, concomitant with the loss of pentafluo-
robenzene. Alternatively, 2 can be prepared by reacting (C6F5)2BCl
with 8-hydroxyquinoline at room temperature in CH2Cl2 for 2 h
(Scheme 1).
This compound is much more moisture sensitive compared to 2
due to the ease of hydrolysis of the remaining pentafluorophenoxy
moiety. Indeed, dissolution of compound 3 in normal grade
hexane resulted in the formation of the dimeric product
((C6F5)BQ)2O (4), together with the loss of pentafluorophenol.
The synthesis and electroluminescent properties of a similar non-
fluorinated analogue ((C6H5)BQꢀ)2O were recently reported (Qꢀ =
2-methyl-8-hydroxyquinoline).11
In analogy to the formation of compound 3, the reaction
of B(OC6F5)3 with 8-hydroxyquinoline was expected to produce
the boric acid ester (F5C6O)2BQ (5). However, this product
could never be isolated with acceptable purity as all attempts
resulted in the formation of by-products, probably the dimeric
compound ((F5C6O)BQ)2O and further degradation products due
to hydrolysis of the BOC6F5 moieties.
All compounds 1–4 are bright yellow solids, which show
fluorescence in solution and the solid state. They have been
1
characterised by H, 19F and 11B NMR spectroscopy, MS, CHN
analysis as well as solution UV-vis and PL spectroscopy. The solid
state structures of compounds 1, 2 and 4 have been determined
by X-ray crystallography. The energies of the HOMO and LUMO
levels for compounds 1, 2 and 4 and the corresponding excitation
energies have been determined computationally and compared
with experimental values.
Solid state structures
The X-ray analysis of crystals of compound 1 revealed the presence
of two independent molecules (I and II) in the asymmetric unit;
molecule I is shown in Fig. 2, and molecule II in Fig. S2 in the ESI.†
The conformations of the two molecules are distinctly dissimilar;
while the OB(C6F5)3 portions are very comparable (having an
˚
Scheme 1
r.m.s. fit of ca. 0.12 A), the quinoline rings adopt noticeably
different orientations (see Fig. S4†), the torsion angle about the
C–O bond being ca. +30 and −5◦ in molecules I and II respectively.
The 8-quinolinol ring has adopted a monodentate O-coordination
mode to the boron centre rather than a bidentate N,O mode, which
The preparation of (pentafluorophenyl)boronic acid penta-
fluorophenyl quinolyl ester (C6F5)(OC6F5)BQ (3) was carried out
by reacting (C6F5)B(OC6F5)2 with 8-hydroxyquinoline (Scheme 2).
Scheme 2
1426 | Dalton Trans., 2007, 1425–1432
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
an extended chain of molecules along the crystallographic b-axis
direction.
The solid state structure of 2 (Fig. 3), where the 8-quinolinol
ring coordinates in a bidentate N,O fashion, is very similar to the
structure of the unfluorinated bis(phenyl) analogue (C6H5)2BQ,
which has two independent molecules in the solid state.19 Despite
the rather high estimated standard deviations (esds) for both
independent molecules in (C6H5)2BQ, it is possible to see the
differences in the boron coordination geometry caused by having
pentafluorophenyl rather than phenyl rings. The B–O bond in 2
˚
˚
[1.494(2) A] is shorter [1.56(1) and 1.54(1) A], whilst the B–C
˚
bonds in 2 [1.624(3) and 1.629(3) A] are longer than those in
˚
(C6H5)2BQ [in the range 1.58(1)–1.59(1) A]. For the B–N bonds,
the large esds in (C6H5)2BQ make the apparent lengthening in 2
not significant (Table 2). The most notable feature of the angles at
the boron centre is an enlargement of ca. 3◦ in 2 for the O–B–N
Fig. 2 The molecular structure of one (I) of the two independent
molecules present in the crystals of (C6F5)3BQH (1).
appears to be the first such example for a 8-quinolinol ring linked
to a boron atom. The geometry at the boron centre is distorted
tetrahedral with angles in the range 103.2(2)–115.6(2)◦ and
103.8(2)–113.9(2)◦ for molecules I and II respectively. In both
cases the B–O bond length is significantly shorter than the B–C
bonds (Table 1). The N–H proton in each independent molecule
is involved in intermolecular N–H · · · F hydrogen bonding to a
symmetry-related counterpart. For molecule I the proton links
to an ortho fluorine [F(15)] on the C(10) pentafluorophenyl ring
˚
˚
in a Ci-related neighbour [N · · · F 2.929(3) A, H · · · F 2.27 A,
N–H · · · F 130◦], and vice versa to give centrosymmetric dimer
pairs. For molecule II, by contrast, the proton donates to a meta
fluorine [F(14ꢀ)] on the C(10ꢀ) ring in a glide relat◦ed counterpart
˚
˚
[N · · · F 2.938(3) A, H · · · F 2.33 A, N–H · · · F 124 ] giving rise to
Fig. 3 The molecular structure of (C6F5)2BQ 2.
◦
˚
Table 1 Selected bond lengths (A) and angles ( ) for 1
Molecule I
Molecule II
Molecule I
Molecule II
B–O(1)
B–C(16)
O(1)–B–C(10)
O(1)–B–C(22)
C(10)–B–C(22)
1.508(4)
1.664(4)
107.8(2)
112.9(3)
115.6(2)
1.491(4)
1.666(4)
107.7(2)
113.8(2)
113.9(2)
B–C(10)
B–C(22)
O(1)–B–C(16)
C(10)–B–C(16)
C(16)–B–C(22)
1.654(5)
1.656(5)
103.2(2)
112.7(3)
104.0(2)
1.659(5)
1.655(4)
104.0(2)
113.3(2)
103.8(2)
◦
19
˚
Table 2 Selected bond lengths (A) and angles ( ) for (C6F5)2BQ (2) and the two independent molecules in its bis(phenyl) analogue (C6H5)2BQ
2
(C6H5)2BQ molecule I
(C6H5)2BQ molecule II
B–N(1)
B–O(9)
B–C(11)
B–C(17)
N(1)–B–O(9)
N(1)–B–C(11)
N(1)–B–C(17)
O(9)–B–C(11)
O(9)–B–C(17)
C(11)–B–C(17)
1.637(2)
1.494(2)
1.624(3)
1.629(3)
99.43(13)
111.35(14)
107.64(14)
108.26(14)
111.90(15)
116.89(15)
1.61(1)
1.56(1)
1.59(1)
1.58(1)
96.6(7)
111.1(8)
111.1(8)
109.2(8)
109.0(8)
117.7(9)
1.61(1)
1.54(1)
1.59(1)
1.58(1)
96.7(8)
109.5(9)
109.7(8)
110.8(8)
110.2(9)
117.9(9)
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1425–1432 | 1427
©

View Article Online
bite angle, a consequence of the contracted B–O bond length.
B(2) respectively]. The five-membered chelate rings have planar
˚
The five-membered chelate ring in 2 adopts a slight envelope
conformations [C2NOB coplanar to within ca. 0.01 and 0.02 A for
˚
conformation, the boron lying ca. 0.14 A out of the C2NO plane
the B(1)- and B(2)-containing rings respectively] with O–B–N bite
angles of 98.57(18) and 98.10(18)◦ at B(1) and B(2) respectively.
The two 8-quinolinol rings are inclined to each other by ca. 16◦,
and are oriented such that there is very little overlap between
them; considering the view in Fig. 4, the O(1)/N(1) ring system is
positioned out of the plane of the paper (towards the viewer) whilst
the O(2)/N(2) ring system is positioned into this same plane (away
from the viewer). The closest analogue for complex 4 is (l2-oxo)-
bis((2-methyl-8-quinlinolato)phenylborane ((C6H5)BQꢀ)2O where
Qꢀ has a methyl group in the 2-position of each 8-quinolinol ring,
and phenyl instead of pentafluorophenyl groups.11 The parameters
for this compound along with those for complex 4 are given
in Table 3. Unfortunately, the differences are not large enough
compared to the esds to be considered significant.
˚
(which is coplanar to better than 0.01 A) in the direction of C(17).
The crystal structure of 4 shows the oxygen-bridged complex to
have molecular C2 symmetry about an axis that passes through
the bridging oxygen and bisects the B–O–B angle (Fig. 4).
The two boron centres have very similar geometries (Table 3),
and it is notable that in each case the B–O bonds to the
˚
bridging oxygen O(3) [1.406(4) and 1.397(3) A for B(1) and B(2)
˚
respectively] are ca. 0.10 A shorter than those to the oxygens of the
˚
chelating 8-quinolinol rings [1.507(3) and 1.518(3) A for B(1) and
UV-vis and photoluminescence (PL) spectroscopy
The UV-vis spectra for compounds 2 and 4 have been measured
in CH2Cl2 solution and the kmax and emax values are collected in
Table 4, together with those determined for AlQ3 and
(C6H5)2BQ. For comparison, we have also included values re-
ported by others for AlQ3 and (C6H5)2BQ. It can be seen that
compared to the non-fluorinated compound (C6H5)2BQ, only a
slight blue shift of the absorption maximum kmax is observed for
compound 2, which could be due to the increased Lewis acidity
of the boron centre. In the case of the oxo-bridged compound 4,
which displays a similar kmax as (C6H5)2BQ, the Lewis acidity of
the boron centres is likely to be reduced due to the donating ability
of the bridging oxygen atom. Overall, considering the variation in
kmax values, fluorination seems to have only a small effect on the
position and intensity of the absorption maximum in the UV-vis
spectra.
Fig. 4 The molecular structure of ((C6F5)BQ)2O 4.
The PL spectra were determined in CH2Cl2 solution and are
shown in Fig. 5. Again, there appears to be very little difference
in the peak emission wavelength between the non-fluorinated
and fluorinated boron compounds. However, the quantum yield
appears to be affected by fluorination. The relative quantum
yield of (C6H5)2BQ was found to be 2.5 times better than AlQ3,
which correlates well with previous reports.7 For the fluorinated
compound 2, the relative quantum yield was found to be 1.8 times
that of AlQ3.
◦
˚
Table 3 Selected bond lengths (A) and angles ( ) for 4 and its closely
related phenyl analogue ((C6H5)BQꢀ)2O11a
4
((C6H5)BQꢀ)2O
1.522(5), 1.526(5)
1.670(5), 1.672(5)
1.612(6), 1.609(6)
1.397(5), 1.392(5)
97.3(3), 96.8(3)
111.5(3), 110.8(3)
110.9(3), 110.7(3)
109.1(3), 109.6(3)
112.6(3), 113.2(3)
114.3(3), 114.6(3)
129.8(3)
B(1)–O(1)
B(2)–O(2)
B(1)–N(1)
B(2)–N(2)
B(1)–C(10)
B(2)–C(25)
B(1)–O(3)
B(2)–O(3)
O(1)–B(1)–N(1)
O(2)–B(2)–N(2)
O(1)–B(1)–C(10)
O(2)–B(2)–C(25)
N(1)–B(1)–O(3)
N(2)–B(2)–O(3)
N(1)–B(1)–C(10)
N(2)–B(2)–C(25)
O(3)–B(1)–C(10)
O(3)–B(2)–C(25)
O(1)–B(1)–O(3)
O(2)–B(2)–O(3)
B(1)–O(3)–B(2)
1.507(3)
1.518(3)
1.648(4)
1.652(4)
1.641(4)
1.633(4)
1.406(4)
1.397(3)
98.57(18)
98.10(18)
112.5(2)
112.2(2)
112.1(2)
111.2(2)
108.2(2)
108.9(2)
111.2(2)
111.7(2)
113.6(2)
113.9(2)
128.2(2)
a Due to the molecular C2 symmetry possessed by both species the bond
lengths and angles have been grouped together on the basis of this pseudo
symmetry.
Fig. 5 Photoluminescense (PL) spectra for compounds Ph2BQ, 2 and 4
in CH2Cl2.
1428 | Dalton Trans., 2007, 1425–1432
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 4 Physical properties of compounds (C6F5)2BQ 2 and ((C6F5)BQ)2O 4
NMRa
UV-visb
PLb
Compound
AlQ3
11B (ppm)
kmax/nm
emax/M−1 cm−1
kmax/nm
Reference
387
390
387c
384c
383b
387
395
396
400c
395
389
395
319c
10000
516
514
515c
517c
524b
530
505
504
497c
515
515
508
496c
8
20
17
21
22
6600
3200
This work
8
7
21
This work
This work
This work
11
(C6H5)2BQ
11.1
7.0
8.5
8440
6600
(C6F5)2BQ 2
((C6F5)BQ)2O 4
((C6H5)BQꢀ)2O
10.35
a Measured in CDCl3. b Measured in CH2Cl2. c Measured in CHCl3.
Computational studies
in compound 2. In compound 4, the stabilisation is reduced be-
cause each boron centre is bonded to only one pentafluorophenyl
group, but it is still noticeably larger compared to (C6H5)2BQ.
In luminescent compounds containing 8-quinolinolate groups,
the HOMO and LUMO are localised on the 8-quinolinolate group,
primarily on the phenoxy and pyridyl part, respectively.7,8,23 This
is also the case for 2 and 4. The delocalised HOMO and LUMO
orbitals for 4 are shown in Fig. 7.
In order to explain the results from the luminescence studies,
a computational study was carried out on the compounds
(C6H5)2BQ, 2 and 4, both in the gas phase and in dichloromethane
solution. The effects of solvation have been mimicked by placing
the solute within a cavity (large or small) in a polarisable medium
that responds to the local charge distribution of the solute (see
Experimental). The calculations have shown that fluorination of
the phenyl rings at the boron centre has the effect of decreasing,
i.e. stabilising, both the HOMO and LUMO energy levels of these
compounds, as depicted in Fig. 6. The largest stabilisation occurs
We have also computed the three lowest electronic excitation
energies (vertical) for the singlet states of each compound. TD-
DFT calculations were performed on geometries optimised at the
B3LYP/6–31G(d) level in the gas phase and with dichloromethane
as a solvent. In order to compare our results with previous
studies,23 we have also computed the TD-DFT excitation ener-
gies for the HF optimised geometry of (C6H5)2BQ. Absorption
wavelengths of 406 and 404 nm (see Table 5) were obtained for
(C6H5)2BQ and 2, respectively, using the small cavity solvation
model, which are in reasonable agreement with the experimental
values of 395 and 389 nm (Table 4) and are also comparable
with those reported previously for (C6H5)2BQ.23 The excitation
wavelengths for the HF structure of (C6H5)2BQ are ca. 20 nm
lower than those computed using the DFT-B3LYP structure,
consistent with the findings of Teng et al.23 Compound 4 could
not be optimised within the small cavity model and we therefore
investigated the large cavity model, resulting in a rather large
excitation wavelength of 411 nm. It was noticed that the use of
the large cavity model increases the first excitation wavelength by
an average of 16 nm for (C6H5)2BQ and 2. A correction of −16 nm
applied to the large cavity result (411 nm) produces an estimate
of 395 nm for the first singlet excitation wavelength of 4, which
matches exactly the experimentally determined value.
Fig. 6 Difference in energy between the HOMO and LUMO levels of
Ph2BQ, 2 and 4. The dashed lines represent MO energy levels for the small
cavity solvent model and the solid lines the MO energy levels for the large
cavity solvent model.
Table 5 Excitation energies (nm) and oscillator strength in brackets for the three singlet states of (C6H5)2BQ, (C6F5)2BQ (2) and ((C6F5)BQ)2O (4)
(C6H5)2BQ HFa
(C6H5)2BQ B3LYPa
(C6F5)2BQ (2) B3LYPa
((C6F5)BQ)2O (4) B3LYPb
S 1
S 2
S 3
385 (0.08)
316 (0.008)
307 (0.008)
406 (0.09)
334 (0.005)
323 (0.01)
404 (0.07)
321 (0.0003)
316 (0.0005)
411 (0.06)
402 (0.03)
390 (0.03)
a A dichloromethane polarisable continuum model was used. b A large cavity was employed.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1425–1432 | 1429
©

View Article Online
Fig. 7 (a) HOMO and (b) LUMO orbitals of ((C6F5)BQ)2O (4). Light areas have positive phase and dark areas negative phase, depicted for an isovalue
of 0.02.
In conclusion, the experimental and computational studies
on the spectroscopic and luminescent properties of the boron
quinolinate compounds (C6H5)2BQ, 2 and 4 have shown that
fluorination of the phenyl rings results in a stabilisation of
both the HOMO and LUMO levels and therefore the effect
on the absorption and emission maxima in the UV-vis and PL
spectra, respectively, is only minimal. However, the difference
in stability and volatility between fluorinated and unfluorinated
luminescent boron compounds may have an effect on their solid
state properties and their performance in OLED devices, which
will be the subject of future investigations. The good agreement
between the computational and experimental results has shown
that modest B3LYP/6–31G(d) calculations (which include basic
solvation effects) can be used to offer valuable insights into factors
affecting absorbance and emittance, and may potentially be used
as a predictive tool for similar compounds.
the relevant sample at the excitation wavelength of 396 nm. For
each compound, data is taken for six optical densities all less than
0.1 and the fraction I/OD is calculated as the slope of the best-fit
line through these points.
Computational details
All calculations were performed using the Gaussian 03 program
package.24 Structures were optimised at the HF and DFT levels,
and confirmed as minima by frequency analysis. We have employed
the B3LYP hybrid functional25,26 which has recently been shown to
perform reasonably well for these types of compounds.9,17,23,27 Both
6–31G(d) and 6–31G(d,p) basis sets have been tested.28,29 Very
similar results were obtained with both sets and we have therefore
used the smaller 6–31G(d) basis set because of the faster computa-
tional times. Excitation energies have been calculated for the fully
optimised DFT-B3LYP (or HF) structures using time-dependent
(TD) DFT.30–32 The effects of a solvent, dichloromethane, were
included in an approximate way by employing the polarisable
conductor-like screening model implemented in Gaussian (C–
PCM).33–35 For the “small cavity model” the cavity surface is
determined using spheres placed around each heavy atom of the
solute (hydrogen atoms are enclosed in the sphere of the atom to
which they are bonded). However, optimisation of ((C6F5)BQ)2O
(4) in the solvent proved problematic because of the proximity
of the two nearly-parallel 8-quinolinolate groups, hence a “large
cavity model” was employed where the cavity volume is increased
by adding the radius of the solvent to the radii of the solute atoms.
The HOMO and LUMO values for (C6H5)2BQ and (C6F5)2BQ (2)
are therefore reported as an energy band in Fig. 6 rather than an
energy level, because they are calculated with both solvent models.
A single value is reported for ((C6F5)BQ)2O (4), because only the
large cavity model would converge.
Experimental
General
All moisture-sensitive compounds were manipulated using stan-
dard vacuum line, Schlenk or cannula techniques, or in a conven-
tional nitrogen-filled glove box. 1H, 19F and 11B NMR spectra were
recorded on a Bruker AC-250 or a Jeol JNM-EX270 spectrometer;
chemical shifts for 1H and 13C NMR are referenced to the residual
protio impurity and to the 13C NMR signal of the deuterated
solvent. 19F and 11B chemical shifts are reported relative to CFCl3
and BF3·OEt2, respectively. Elemental analyses were performed by
the Science Technical Support Unit at the London Metropolitan
University. Absorption spectra were measured using a Perkin
Elmer UV-Vis Lambda-2 spectrometer. PL spectra were measured
using a Jobin-Yvon Fluorolog-3 with an integration time of 1.0 s
and monochromator slits set at 1 nm. All solution state spectra
were obtained from 10−5 M solutions in dry CH2Cl2; measurements
were taken in right-angle mode. All spectra were taken in 5 nm
intervals and corrected against dark counts and detector factors.
Relative PL efficiency is calculated using the formula
Solvents and reagents
Toluene and pentane were dried by passing through a column,
filled with commercially available Q-5 reagent (13 wt% CuO on
alumina) and activated alumina (pellets, 3 mm). Dichloromethane
and acetonitrile were dried over CaH2 and distilled under nitrogen.
The synthesis of B(C6F5)3,13 (C6F5)2BCl,36 (C6F5)B(OC6F5)2,14
Q1−3
QAlQ3
ODAlQ3 I1−3
IAlQ3 OD1−3
=
where Q is the PL efficiency, I is the integral of the corresponding
PL spectrum from 400–750 nm and OD is the optical density of
37
B(OC6F5)3 and (C6H5)2BQ5 have been reported previously. All
1430 | Dalton Trans., 2007, 1425–1432
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 6 Crystallographic data for compounds 1, 2 and 4a
Data
Chemical formula C27H7BF15NO C21H6BF10NO C30H12B2F10N2O3
other chemicals and NMR solvents were obtained commercially
and used as received.
1
2
4
Quinolinium tris(pentafluorophenyl)borate (C6F5)3B(OC9H6NH)
(1). B(C6F5)3 (100 mg, 0.19 mmol) and 8-hydroxyquinoline
(28 mg, 0.19 mmol) were placed together into a Schlenk flask
and dissolved in 40 mL of CH2Cl2. The colour of the solution
was bright yellow. The solution was stirred at room temperature
for 1 h, after which the solvent was removed under reduced
pressure, leaving the product as a yellow solid. Crystals suitable for
single crystal X-ray diffraction were obtained in hexane at room
temperature.
Solvent
FW
0.5CH2Cl2
699.61
—
—
489.08
−100
660.04
−100
T/◦C
−100
¯
P21/c (no. 14) P1 (no. 2)
Space group
C2/c (no. 15)
26.9700(11)
10.5284(4)
37.9995(14)
—
˚
a/A
15.3409(13)
7.8053(6)
15.4746(14)
—
96.164(7)
—
1842.2(3)
4
1.763
8.0303(8)
12.2898(12)
13.6929(15)
86.956(8)
89.875(9)
85.670(8)
1345.6(2)
2
1.629
1.54248
1.342
0.050
0.099
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
104.723(3)
—
1H NMR (ppm, CDCl3, 250 MHz): 8.95 (d, 3J = 8.2 Hz, 1 H),
3
˚
8.83 (t, 3J = 6.4 Hz, 1 H), 7.96 (t, 3J = 5.4 Hz, 1 H), 7.66 (t, 3J =
V/A
10435.7(7)
Z
16b
3
3
8.3 Hz, 1 H), 7.48 (d, J = 8.3 Hz, 1 H), 7.04 (d, J = 8.2 Hz,
1 H). 19F (ppm, CDCl3, 235 MHz): −135.0 (o-F), −159.7 (p-F),
−165.1 (m-F). 11B NMR (ppm, CDCl3, 86 MHz): −3.1. Elemental
analysis (calc. for (C6F5)3B(OC9H6NH): C = 49.35, H = 1.07, N =
2.13%): C = 49.33, H = 1.03, N = 2.17%.
qcalcd/g cm−3
1.781
1.54248
2.561
0.051
0.130
˚
k/A
1.54248
1.605
0.035
l/mm−1
c
R1
d
wR2
0.086
a Oxford Diffraction Xcalibur PX Ultra diffractometer, Cu Ka radiation,
refinement based on F2. b There are two crystallographically independent
Bis(pentafluorophenyl)borinic acid quinolyl ester (C6F5)2-
B(OC9H6N) (2). (C6F5)2BCl (150 mg, 0.394 mmol) and 8-
hydroxyquinoline (57 mg, 0.394 mmol) were placed into a Schlenk
flask and dissolved in 50 mL of CH2Cl2. The solution, which
was bright yellow, was stirred under nitrogen for 2 h, after which
the solvent was removed by high vacuum, leaving a yellow solid
(yield = 91%). Crystals suitable for single crystal X-ray diffraction
molecules in the asymmetric unit. c R1 = RꢁFo|–|Fcꢁ/R|Fo|. d wR2
=
{R[w(Fo2–Fc2)2]/R[w(Fo2)2]}1/2; w−1 = r2(Fo2) + (aP)2 + bP.
CDCl3, 235 MHz): −135.2 (o-F), −156.1 (p-F), −164.5 (m-F). 11
B
NMR (ppm, in CDCl3, 86 MHz): 8.5.
1
were obtained in hexane at room temperature. H NMR (ppm,
CDCl3, 250 MHz): 8.85 (d, 3J = 5.3 Hz, 1 H), 8.56 (d, 3J = 8.2 Hz,
1 H), 7.78-7.67 (m), 7.39 (d, 3J = 8.2 Hz, 1 H), 7.21 (d, 3J = 7.6 Hz,
1 H). 19F (ppm, CDCl3, 235 MHz): −135.4 (o-F), −156.0 (p-F),
−163.3 (m-F). 11B NMR (ppm, CDCl3, 86 MHz): 7.0. MS-EI+
(m/z): 489 ([M]+ 322 ([M–C6F5]+). Elemental analysis (calc. for
(C6F5)2B(8-OC9H6N): C = 51.57, H = 1.24, N = 2.86%): C =
51.53, H = 1.20, N = 2.79%.
Crystallographic details
Table 6 provides a summary of the crystallographic data for
compounds 1, 2 and 4.
CCDC reference numbers 619418, 632949 and 632950.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b700317j
Pentafluorophenylboronic acid quinolyl pentafluorophenyl ester
(C6F5)B(OC9H6N)(OC6F5) (3). (C6F5)B(OC6F5)2 (300 mg,
0.55 mmol) was placed in a Schlenk flask together with 8-
hydroxyquinoline (80 mg, 0.55 mmol). After the addition of 50 mL
of CH2Cl2 to the flask, a bright yellow solution was produced.
After 1 h stirring, the solvent was removed by reduced pressure,
leaving a bright yellow powder. The product was dissolved in
Acknowledgements
We are grateful to the EPSRC for financial support (GR/
R92042/01).
References
◦
hot hexane and left at 4 C overnight. The day after, the yellow
1 Organic Light Emitting Devices, ed. K. Mu¨llen and U. Scherf, Wiley-
VCH, Weinheim, 2005.
2 C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett., 1987, 51, 913.
3 C. H. Chen and J. Shi, Coord. Chem. Rev., 1998, 171, 161.
4 S. Wang, Coord. Chem. Rev., 2001, 215, 79.
5 Q. Wu, M. Esteghamatian, N.-X. Hu, Z. Popovic, G. Enright, Y. Tao,
M. D’Iorio and S. Wang, Chem. Mater., 2000, 12, 79.
6 Y. Qin, C. Pagba, P. Piotrowiak and F. Ja¨kle, J. Am. Chem. Soc., 2004,
126, 7015.
7 Y. Cui, Q.-D. Liu, D.-R. Bai, W.-L. Jia, Y. Tao and S. Wang, Inorg.
Chem., 2005, 44, 601.
8 S. Anderson, M. S. Weaver and A. J. Hudson, Synth. Met., 2000, 111–
112, 459.
9 Q.-D. Liu, M. S. Mudadu, R. Thummel, Y. Tao and S. Wang, Adv.
Funct. Mater., 2005, 15, 143.
10 Q.-D. Liu, M. S. Mudadu, H. Schmider, R. Thummel, Y. Tao and S.
Wang, Organometallics, 2002, 21, 4743.
11 H. J. Son, H. Jang, B.-J. Jung, D.-H. Kim, C. H. Shin, K. Y. Hwang,
H.-K. Shim and Y. Do, Synth. Met., 2003, 137, 1001.
12 Q. Wu, M. Esteghamatian, N.-X. Hu, Z. Popovic, G. Enright, S. R.
Breeze and S. Wang, Angew. Chem., Int. Ed., 1999, 38, 985.
13 A. G. Massey and A. J. Park, J. Organomet. Chem., 1964, 2, 245.
precipitate was filtered and washed with hexane, resulting in 97 mg
1
of product 3 (yield = 56%). H NMR (ppm, CDCl3, 250 MHz):
3
3
8.93 (d, J = 5.1 Hz, 1 H), 8.62 (d, J = 8.3 Hz, 1 H), 7.86-7.81
3
3
(m), 7.68 (t, J = 7.7 Hz, 1 H), 7.40 (d, J = 8.3 Hz, 1 H), 7.16
(d, 3J = 7.7 Hz, 1 H). 19F NMR (ppm, CDCl3, 235 MHz): −134.3
(o-F, C6F5), −155.2 (p-F, C6F5), −157.2 (o-F, OC6F5), −163.3
(m-F, C6F5), −165.3 (m-F, OC6F5), −166.8 (p-F, OC6F5). 11B
NMR (ppm, CDCl3, 86 MHz): 8.4. Elemental analysis (calc. for
(C6F5)(OC6F5)B(8-OC9H6N): C = 49.94, H = 1.20, N = 2.77%):
C = 49.84, H = 1.12, N = 2.66%.
Pentafluorophenyl(8-quinolinolinate)borinic anhydride ((C6F5)-
BQ)2O (4). Attempts to grow crystals suitable for single crystal
X-ray diffraction of compound 3 from hexane at 4 ◦C resulted in
the formation of compound 4, probably due to small amounts of
water in the solvent. Crystals of compound 4 were analysed by
X-ray crystallography and NMR spectroscopy. 19F NMR (ppm,
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1425–1432 | 1431
©

View Article Online
14 G. J. P. Britovsek, J. Ugolotti and A. J. P. White, Organometallics, 2005,
24, 1685.
15 D. C. Bradley, I. S. Harding, A. D. Keefe, M. Motevalli and D. H.
Zheng, J. Chem. Soc., Dalton Trans., 1996, 3931.
16 M. Matsumura and T. Akai, Jpn. J. Appl. Phys., 1996, 35, 5357.
17 Y.-W. Shi, M.-M. Shi, J.-C. Huang, H.-Z. Chen, M. Wang, X. D. Liu,
Y.-G. Ma, H. Xu and B. Yang, Chem. Commun., 2006, 1941.
18 D. Vagedes, R. Froehlich and G. Erker, Angew. Chem., Int. Ed., 1999,
38, 3362.
19 H. Ho¨pfl, V. Barba, G. Vargas, N. Farfan, R. Santillan and D. Castillo,
Chem. Heterocycl. Compd., 1999, 35, 912.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03, Revision
C.01, Gaussian, Inc., Wallingford, CT, 2004.
25 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
26 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785.
27 H.-Y. Chen, Y. Chi, C. S. Liu, J.-K. Yu, Y.-M. Cheng, K.-S. Chen, P.-T.
Chou, S.-M. Peng, G.-S. Lee, A. J. Carty, S.-J. Yeh and C.-T. Chen, Adv.
Funct. Mater., 2006, 15, 567.
20 T. A. Hopkins, K. Meerholz, S. Shaheen, M. L. Anderson, A. Schmidt,
B. Kippelen, A. B. Padias, H. K. Hall, Jr., N. Peyghambarian and N. R.
Armstrong, Chem. Mater., 1996, 8, 344.
28 J. S. Binkley, J. A. Pople and W. J. Hehre, J. Am. Chem. Soc., 1980, 102,
939.
21 X.-Y. Wang and M. Weck, Macromolecules, 2005, 38, 7219.
22 M. Brinkmann, G. Gadret, M. Muccini, C. Taliani, N. Masciocchi and
A. Sironi, J. Am. Chem. Soc., 2000, 122, 5147.
23 Y. L. Teng, Y. H. Kan, Z. M. Su, Y. Liao, L. K. Yan, Y. J. Yang and
R. S. Wang, Int. J. Quantum Chem., 2005, 103, 775.
29 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213.
30 R. Bauernschmitt and R. Ahlrichs, Chem. Phys. Lett., 1996, 256,
454.
31 M. E. Casida, C. Jamorski, K. C. Casida and D. R. Salahub, J. Chem.
Phys., 1998, 108, 4439.
24 M. J. Frisch, G. W. T., H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
32 R. E. Stratmann, G. E. Scuseria and M. J. Frisch, J. Chem. Phys., 1998,
109, 8218.
33 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995.
34 M. Cossi, N. Rega, G. Scalmani and V. Barone, J. Comput. Chem.,
2003, 24, 669.
35 S. Miertus, E. Scrocco and J. Tomasi, Chem. Phys., 1981, 55, 117.
36 D. J. Parks, W. E. Piers and G. P. A. Yap, Organometallics, 1998, 17,
5492.
37 D. Naumann, H. Butler and R. Gnann, Z. Anorg. Allg. Chem., 1992,
618, 74.
1432 | Dalton Trans., 2007, 1425–1432
This journal is
The Royal Society of Chemistry 2007
©