A mechanistic investigation of mechanochromic luminescent organoboron materials

Article abstract of DOI:10.1039/c2jm32809g

Mechanochromic luminescence (ML) refers to the luminescence color and/or intensity change of solid-state materials induced by mechanical perturbations. For organic molecular solids, this phenomenon is related to the specific packing modes and orientations of individual fluorophores, which could give rise to different excited-state interactions. The molecular solids of difluoroboron dibenzoylmethane (BF₂dbm) derivatives were previously found to exhibit reversible ML at room temperature and are promising as self-healing optical materials. In this report, we aim to shed some light on the mechanism of BF₂dbm ML by trying to understand the excited-state interactions among solid-state BF₂dbm molecules and elucidate how these interactions change upon mechanical stimulation. We first investigated the optical properties of monomeric, dimeric, and polymeric BF₂dbm derivatives in optically dilute solutions and demonstrated unambiguously that BF₂dbm moieties have a propensity to form H-aggregates. Next, we studied the physical properties of these boron complexes in the solid state including their crystal structures, fluorescence emissions, and mechanochromic luminescence. By correlating solution data with the solid-state characterization results, it was concluded that two coupled processes, force-induced emissive H-aggregate formation and energy transfer to the emissive H-aggregates, are responsible for the observed BF₂dbm ML in the solid state. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2jm32809g

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 17332
www.rsc.org/materials
PAPER
A mechanistic investigation of mechanochromic luminescent organoboron
materials†
*
Xingxing Sun, Xuepeng Zhang, Xinyang Li, Shiyong Liu and Guoqing Zhang
*
Received 4th May 2012, Accepted 3rd July 2012
DOI: 10.1039/c2jm32809g
Mechanochromic luminescence (ML) refers to the luminescence color and/or intensity change of solid-
state materials induced by mechanical perturbations. For organic molecular solids, this phenomenon is
related to the specific packing modes and orientations of individual fluorophores, which could give rise
to different excited-state interactions. The molecular solids of difluoroboron dibenzoylmethane
(BF2dbm) derivatives were previously found to exhibit reversible ML at room temperature and are
promising as self-healing optical materials. In this report, we aim to shed some light on the mechanism
of BF2dbm ML by trying to understand the excited-state interactions among solid-state BF2dbm
molecules and elucidate how these interactions change upon mechanical stimulation. We first
investigated the optical properties of monomeric, dimeric, and polymeric BF2dbm derivatives in
optically dilute solutions and demonstrated unambiguously that BF2dbm moieties have a propensity to
form H-aggregates. Next, we studied the physical properties of these boron complexes in the solid state
including their crystal structures, fluorescence emissions, and mechanochromic luminescence. By
correlating solution data with the solid-state characterization results, it was concluded that two coupled
processes, force-induced emissive H-aggregate formation and energy transfer to the emissive H-
aggregates, are responsible for the observed BF2dbm ML in the solid state.
generate a drastic alteration in the emission color and/or inten-
1. Introduction
sity, which is one of the known causes for piezochromic or
mechanochromic luminescence (ML).6 Other mechanisms of
mechanochromism may involve breakage of covalent bonds such
as the force-activated spiropyran to merocyanine reaction.7
Studies on ML not only concern apparent applications such as
monitoring material deformation and stress,8 but also help in the
understanding of excited-state interactions which are important
for optoelectronics and solar energy research.9 In the past few
years, many ML molecules have been reported based on chem-
ically distinct core structures.10 The advantage of molecular ML
materials is that their response typically does not involve the
breaking of chemical bonds and is therefore reversible, by
thermal or solvent treatment in many cases. This allows for
repeated application of force and makes them useful fluorescent
responsive materials. For example, one of the early studies based
on gold(I) thiouracilate complexes, reported by Eisenberg,10a that
mechanically induced an intense emission could be attributed to
the change in aurophilic interactions. Also based on Au
complexes, Ito and coworkers10d reported that the fluorescence
emission of (C6F5Au)2(m-1,4-diisocyanobenzene) complexes
turns from blue to yellow upon grinding and the ML process can
be reversed by solvent treatment. Pure organic ML materials
were previously reported by Sagara and Kato,10k who synthe-
sized liquid crystals with polyaromatic hydrocarbon cores such
as pyrene or anthracene. The emission intensity and color of
these solid state molecular ensembles can be conveniently tuned
For organic molecular materials, the organization of the building
blocks through non-covalent bonds determines the macroscopic
properties of such systems.1,2 A classical example is the formation
of J- and H-aggregates, where blue- and red-shifts in absorption
spectra are caused by a mere variation in the slippage angle of the
same stacked molecules.3 When external stimuli are introduced,
it is possible to switch the molecular organization from one form
to another.4 As reported by Yi and co-workers,5 the hydropho-
bicity of a naphthylimide–cholesterol coating may be altered by
ultrasonic irradiation, resulting from the so called T-gel to S-gel
phase transition. This apparent water repellency change occurs
due to the adoption of a new preferential bonding mode at the
molecular level. As the most fundamental power source,
mechanical force may also be employed to harness new proper-
ties from molecular assemblies.4 For example, mechanically
changing the molecular organization of fluorescent species can
CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer
Science and Engineering, University of Science and Technology of China,
96 Jinzhai Road, Hefei, 230026, China. E-mail: gzhang@ustc.edu.cn;
sliu@ustc.edu.cn; Fax: +86 551 3606607; Tel: +86 551 3606607
† Electronic supplementary information (ESI) available: Experimental
details, characterization methods, and all relevant characterization data
are described. CCDC reference numbers 880808–880810. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/c2jm32809g
17332 | J. Mater. Chem., 2012, 22, 17332–17339
This journal is ª The Royal Society of Chemistry 2012

View Online
by mechanical force and heating to achieve dual-colored or even
tri-colored luminescence.
ordered crystals tend to become amorphous based on previous
studies.10,12,13 Shown below, BF2dbm complexes 1–3, D1, and P1
are the monomeric, dimeric, and polymeric derivatives chosen as
the model system in the present study.
Organoboron compounds11 are also of particular interest as
fluorescent materials. With potential commercial relevance, a
boron-sunscreen complex difluoroboron avobenzone (BF2AVB)
was demonstrated to possess reversible ML with a dramatic
emission color change from green-blue to yellow under shear
force.12 The original green-blue fluorescence can be restored by
simple thermal treatment. It was found later that many of the
difluoroboron dibenzoylmethane (BF2dbm) derivatives can
exhibit ML in the solid state.13 However, no conclusive mecha-
nisms were provided given the difficulty in interpreting solid-
state spectra, which are highly prone to self-absorption and light
scattering.14 In this article, we attempted to conduct a thorough
investigation of the BF2dbm ML mechanism: when the BF2dbm
crystals/solids are subjected to mechanical force, what and how
relevant processes are involved in changing the emission color?
To answer this question, one must first understand the excited-
state interactions among BF2dbm moieties in the solid state.
Experimentally, a common strategy to circumvent self-absorp-
tion and scattering from solid-state samples would be to spatially
confine two or more fluorophores and study the photophysical
properties of these confinements in dilute solution. Using cova-
lently linked dimers is a popular strategy since it is simple to work
with.15 However, dimers do not fully reflect the complex envi-
ronment of the solid state samples, especially after mechanical
force application. Polymers, on the other hand, can serve as a
bridge from dimers to solid-state systems.16 Polymers can be
molecularly dissolved in solution and unambiguous optical
spectra can be obtained. They can also be fabricated into
nanoparticles, which are solid-state in nature but can be largely
immune to self-absorption and light scattering as well.17 In this
study we shall employ both solution and nanoparticle dispersion
of BF2dbm-containing polymers to help identify the excited-state
interactions among BF2dbm dyes in the solid state.
2. Results
2.1 Synthesis
The syntheses of all three monomeric BF2dbm derivatives are
similar, as shown in Scheme 1, by using methoxy- and allyloxy-
substituted acetophenones and methyl benzoates. The ketone and
ester pairs are refluxed in dry THF for ꢀ48 h to yield the
corresponding diketones, which are then chelated with boron
trilfluoride in CH2Cl2 at room temperature to afford bright green-
yellow crystals 1–3 after recrystallization from CH2Cl2–n-hexane.
The purity and composition of the products were verified with 1H
NMR and HRMS, respectively (ESI†). To prepare the linked
dimer D1 and polymer P1, polycondensation reaction was carried
out in toluene at ꢀ105 ꢁC under N2. Hydrosilylations of the boron
dye and 1,1,3,3-tetramethyldisiloxane (TMDS) at 2 : 1 and stoi-
chiometric ratios for D1 and P1, respectively, were catalyzed by
H2PtCl4 (0.05%). D1 was purified by silica gel chromatography
and was verified by 1H NMR and HRMS. For P1, the resulting
gummy solid resulting from toluene evaporation was re-dissolved
in CH2Cl2 for methanol precipitation to remove the monomer and
catalyst. The final polymer precipitate was obtained as a yellow
1
powder. P1 was first examined with H NMR spectroscopy in
CDCl3, and the characteristic vinyl proton resonance peaks
usually found at around 6.05 and 5.40 had completely dis-
appeared, instead the methylene proton signal at 0.62 (CH2–Si)
was observed. No peak at ꢀ17 ppm (enol proton of the diketone)
was seen, confirming that the methanol precipitation process did
not cleave the boron complex to form a diketone, as BF2bdk
complexes are conditionally sensitive towards protic solvents.18
The polymer molecular weight was determined by GPC (Mn ¼
11 900 Da, PDI ¼ 3.80), showing a broad distribution typical of
polycondensation/hydrosilylation reactions.
To study the dynamic process of ML, one must also elucidate
the excited-state interactions before and after mechanical stim-
ulation. Crystal structures of BF2dbm monomers can help
predict excited-state information prior to force application. (To
avoid confusion with polymer chemistry, the term ‘‘monomer’’ in
this study refers to the complex with a single BF2dbm moiety.) By
correlating information obtained from BF2dbm randomly
orientated polymeric aggregates (e.g., nanoparticle dispersion),
post-mechanical-force interactions can be inferred since initially
2.2 Optical characterization in solution state
In dilute CH2Cl2, complexes 1–3 exhibit almost identical
absorption spectra (lmax ¼ 412 nm, Fig. 1) characteristic of
Fig. 1 Absorption and normalized emission spectra of the boron
complexes, indicated by the arrows in the order of 1–3, dimer D1, and
polymer P1 in CH2Cl2. Inset: photo showing the deep blue and white
fluorescence emission of 1 (left) and P1 (right) in dilute CH2Cl2 under UV
excitation (lex ¼ 365 nm).
Scheme 1
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17332–17339 | 17333

View Online
strong p–p* transitions (3max ꢀ 8 ꢂ 104 Mꢃ1 cmꢃ1). For the
BF2dbm dimer D1, the absorption maximum remains the same
as those of 1–3, varying only in the relative intensity of a shoulder
at ꢀ390 nm. The BF2dbm-containing polymer P1 however,
exhibits a blue-shifted absorption maximum at 391 nm in addi-
tion to the 412 nm peak. The steady-state emission spectra for all
the boron complexes were also collected in CH2Cl2 (Fig. 1). As
expected, 1–3 have almost indistinguishable spectra mostly in the
blue region (lem ¼ 437 nm, FF ¼ 0.93, 0.90, 0.95 for 1–3). For D1
and P1 however, despite their unaltered emission maxima at 437
nm, an additional lower energy shoulder was observed at around
550 nm. As a result, unlike the deep blue emission from monomer
1, the dilute solution of polymer P1 in CH2Cl2 exhibits a white-
colored photoluminescence (PL) to the naked eye under UV
irradiation (Fig. 1 inset). The excitation spectra monitored at 550
nm for D1 and P1 are also distinctly different from that moni-
tored at 440 nm (Fig. S1 and S2†): for both D1 and P1, the blue-
shifted absorption at 391 nm results in emission at 550 nm but a
red-shifted absorption at 412 nm corresponds to fluorescence
emission at a much shorter wavelength (437 nm).
When monitored at around the emission maximum (550 nm), the
excitation spectrum resembles the absorption of P1 in CH2Cl2
with a slightly more blue-shifted maximum at 384 nm (Fig. 2).
2.4 Crystal structures of monomers 1–3
To seek structure–property relationships of BF2dbm dyes in the
solid state, single crystal XRD measurements were performed.
During a period of six months, we were unable to obtain single-
crystals for D1, probably due to the dynamic nature of the
flexible linker and the increased freedom of motion. For
complexes 1–3, upon crystallization, minor difference in the
carbon number or saturation of the alkyl chain leads to
conspicuous variations in crystal packing. Although 3 is a known
compound, crystal structures were solved for all three complexes
given this class of molecule has been found to possess multiple
polymorphs depending on the crystallization conditions.12 We
did not identify more than one polymorph in this experiment
when single crystals were obtained by n-hexane diffusion into
CH2Cl2 at room temperature.
Fluorescence lifetimes of these boron complexes in CH2Cl2
were measured as well. 1–3 have almost the same single-expo-
nential decay of ꢀ1.68 ns. Although the emission band at 437 nm
of D1 and P1 can almost overlap with those of 1–3, the fluo-
rescence decay profiles of D1 and P1 at 440 nm are quite
different. D1 has a pseudo single-exponential decay of 1.34 ns but
P1 is fitted to a triple-exponential decay with a pre-exponential
averaged value of 1.29 ns. When monitored at 550 nm, the life-
time of D1 and P1 (39 and 40 ns) is much longer.
Fig. 3 shows the unit cell packing of each of the three
complexes 1–3. Consistent with previous studies, the BF2dbm
backbones are rather rigid and mostly planar.12,19 All three
crystals are monoclinic with space groups of C2/c, Pc, and C2/c
for 1–3, respectively. The BF2dbm complexes all adapt to an
anti-parallel packing manner (i.e., the BF2 moiety points to the
opposite direction compared to the BF2 group on the nearest
neighbor) and significant p–p stacking is noticed for 1 and 3. For
3, the diketone ring lays above the phenyl ring of its nearest
ꢀ
neighbor at a distance of 3.54 A.
Weak interactions including ArH/F, OCH/F, CH]CH/
O, CH/HC, C]C/O are identified in these molecular systems
2.3 Optical characterization of the P1 nanoparticle dispersion
We also investigated the photophysics of BF2dbm dye aggregates
in the form of nanoparticles so that optical distortions remain
non-problematic while its solid-state properties are retained. As a
result, the investigation of the nanoparticles links the solution
characterization results to those in the solid state. Polymer P1 in
acetone (BF2dbm concentration 1 ꢂ 10ꢃ3 M) was fabricated into
polymeric nanoparticles with an intensity-average hydrodynamic
diameter of 54.4 nm and a size polydispersity of 0.088 as deter-
mined by DLS, (Fig. S3†).18 The nanoparticle dispersion has a
greenish yellow fluorescence emission under UV light and
exhibits a Gaussian-shaped emission spectrum (lex ¼ 545 nm).
(Fig. S4†). Specifically, ArH/F interactions occur in
a
symmetric fashion involving both fluorine atoms on the BF2
moiety and two hydrogen atoms from the 2 and 20 positions on
the phenyl ring in complexes 1 and 3. The two fluorine hydrogen
ꢀ
bonds distances are 2.628 and 2.664 A for 1 and 3, respectively.
These fluorine hydrogen bonds alternate between adjacent
molecules and propagate along one dimension. For 2 however,
one boron molecule directly interacts with eight neighboring
molecules (versus two in 1 and 3) through multiple weak inter-
actions (O/CH]CH, CH/HC, C]C/O) in addition to
ArH/F hydrogen bonding, which is the only form of interaction
in the case of 1 and 3. In 2, only one fluorine is involved and the
Fig. 2 Normalized excitation and emission spectra of P1 nanoparticle
dispersion in water; inset: photo of the nanoparticle dispersion under UV
light (lex ¼ 385 nm, excitation monitored at 550 nm).
Fig. 3 Illustration of unit cell packing for boron dye single crystals 1–3.
Hydrogen atoms are omitted for clarity.
17334 | J. Mater. Chem., 2012, 22, 17332–17339
This journal is ª The Royal Society of Chemistry 2012

View Online
dual hydrogen bonding (H/F/H) consists of one hydrogen
atom from the allyloxy group (ArOCH/F) and the other from
the phenyl ring (ArH/F) (Fig. S4†). They are 2.470 and 2.572 A
away from the fluorine atom, respectively. This combination of
and their corresponding spectra in small regions could be
obtained simultaneously, minimizing sample heterogeneity
problems. In this section, to avoid problems associated with
instrumentation discrepancy, all fluorescence spectra were
recorded with the array spectrometer.
ꢀ
interactions causes the allyl group to bend towards the contact-
ꢀ
ing fluorine atom and increases the B–F bond by 0.033 A
Under a fluorescence microscope (lex ¼ 405 nm), needle-sha-
ped, sheet-like, and thick needle crystalline species were observed
for 1–3, respectively (Fig. S5†). The dimer D1 appeared as an
irregular polycrystalline solid. Among complexes 1–3, the single
crystals of complex 2 exhibit the most narrow and blue-shifted
fluorescence emission (lmax ¼ 500 nm, sF ¼ 1.54 ns, Fig. 4)
compared to 1 (lmax ¼ 505 nm, sF ¼ 4.35 ns) and 3 (lmax ¼ 509
nm, sF ¼ 2.87 ns), whose spectra are red-shifted and broadened
in the lower energy region. The polymer solids have a broad and
structureless spectrum at 554 nm (sF ¼ 24.2 ns) while that of the
dimer solids (lmax ¼ 511 nm, sF ¼ 8.92 ns) is an intermediate
between the monomers and the polymer.
compared to the other B–F bond (that does not participate) and
ꢀ
by 0.025 A compared to the B–F bond in 1 (that forms a mono
ArH/F contact). Detailed crystal data are also provided (ESI,
Tables S1–S21†).
2.5 Photophysical properties in the solid state
Given that most BF2dbm derivatives exhibit reversible ML,12,13
namely that their emission spectra spontaneously recover over
time after mechanical perturbation, it is problematic to record
ML spectra on a FluoroMax spectrometer since collecting a
decent emission spectrum typically takes 3–5 min over the visible
range while many BF2dbm dyes have ML recovery timeframes
shorter than a few minutes.13 To solve this problem, an optical
fiber attached to a linear CCD-array spectrometer with 25 ms
integration time equipped with an built-in LED excitation
module (lex ¼ 385 nm) was mounted to the sample holder of a
fluorescence microscope. In this setup, fluorescence micrographs
2.6 Mechanochromic luminescence
Like many BF2dbm derivatives, ML was observed in 1–3 and D1
(Fig. 4 photos). When the crystals of 1–3 and solid D1 were
sandwiched between microscope glass slides and covers, 2 and 3
left bright yellow streaks on both slides and cover slips upon the
very gentle sliding of the thin covers (ꢀ500 mg in weight),
demonstrating high sensitivity towards mechanical force. On the
other hand, 1 and D1 did not exhibit ML until the solids were
pressed very firmly and sheared in between the two pieces of
glass. Although the roles of the substituents and substrates on
ML are not within the scope of this study, clearly the substituent
effects, which are practically non-existent in dilute solution, are
once again manifested in the solid state. The polymer P1 did not
show any visual fluorescence color or intensity change after being
ground against the substrate with substantial shear force.
The photos, fluorescence micrographs, emission spectra and
fluorescence lifetimes (collected and averaged over ꢀ2 min) were
recorded for the solid smears. After hard pressing, the crystals
were mostly amorphous under the microscope (Fig. S5†). Their
emission spectra were instantly recorded by the array spec-
trometer. Unlike the relatively narrow fluorescence emission
found in the crystalline state, the smeared solids all exhibit
similar Gaussian-shaped spectra (Fig. 4). The fluorescence life-
times for the smears (1–3 and D1) range from 7–10 ns with
fluctuations due to the spontaneous morphological changes of
Fig. 4 Photos of 1–3, D1, and P1 solids on microscope slides with glass
covers under UV light. Left: the solids were gently moved by the glass
covers; right: the solids were repeatedly pressed hard. The plots at the
bottom show the emission spectra of each solid before (left) and after
hard pressing (right).
Fig. 5 (a) Reversible ML of 3 loaded on a piece of weighing paper:
writing ‘‘Exciton Migration’’ with a wood stick and erasing it by heating
at 120 ꢁC for 30 s; (b) quickly dimmed ML of 1 after shear force and
brightness restoration by repeated smearing with a cotton swab on the
areas sheared ꢀ15 s ago. (lex ¼ 365 nm).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17332–17339 | 17335

View Online
the solid samples.12,13 It was interesting to see that irrespective of
the initial emissions, all the samples ended up with similar spectra
after being ground via mechanical force, suggesting similar
amorphous states for all molecules.
Consistent with previous studies,12,13 the ML of 1–3 and D1 is
reversible, namely that the yellow ML spontaneously reverts to
the blue-green. This process takes longer ꢁat room temperature
but can be substantially expedited at 120 C, which is still well
below the melting temperatures of these solids (>185 ꢁC). The
reversibility was demonstrated with 3 on a piece of weighing
paper as shown in Fig. 5a. Again, details of the substituent and
substrate effects are not discussed in the current study. A
previously unnoticed interesting feature of BF2dbm ML is that
the freshly sheared sites tend to become dimmer very quickly
(extended time leads to color change) but the brightness can be
reinstated by repeated shear force (Fig. 5b). In this process
however, the emission color is not altered, only the intensity is.
3. Discussion
3.1 Excited state interactions among BF2dbm(OR)2
Fig. 6 (a) Absorption spectra of D1 in a mixture of CH2Cl2 and
n-hexane with different CH2Cl2 content percentages: 100%, 80%, 50%,
20%, and 10% at a fixed concentration of 5.6 ꢂ 10ꢃ6 M. Inset: schematic
representation of increased transition probability at higher energy level of
the BF2dbm H-dimer with ‘‘face-to-face’’ parallel transition dipoles. The
strength of the transition is denoted with ‘‘f’’ (oscillator strength21). Black
arrows denote the transition dipole moments. Solid and dotted lines
denote allowed and forbidden excited states.21 (b) Normalized steady-
state emission spectra of D1 in a mixture of CH2Cl2 and n-hexane with
different CH2Cl2 content percentages: 100%, 80%, 50%, 20%, and 10% at
a fixed concentration of 5.6 ꢂ 10ꢃ6 M (lex ¼ 385 nm). Inset: corre-
sponding photos of the five samples under UV light and schematic
presentation of monomer and H-dimer fluorescence (FL).
From characterization in dilute CH2Cl2, it is clearly evident that
when BF2dbm moieties are spatially constrained (D1 and P1),
intermolecular interactions give rise to a new emission band
(ꢀ550 nm) that is absent when the dye molecules are discrete
(1–3). There are two common causes for a red-shifted fluores-
cence band for proximal dyes of the same species: excited-state or
ground-state dimers. Excited-state dimers are also known as
excimers,20 which are only attractive to each other in the excited
state but are non-interacting or repulsive in the ground state.
Thus excimers absorb light as discrete molecules and emit light as
associated ones. Ground-state dimers however, can be excited by
light as one delocalized entity which differs in transition energy
to the monomer and results in new absorption bands.21
BF2dbm moieties in the ‘‘face-to-face’’ configuration (Fig. 6a
inset). Meanwhile, the fluorescence emission contributed by the
emissive H-dimer (ꢀ550 nm, sF ¼ 39 ns) is boosted with
diminished solvent polarity (Fig. 6b). The red-shifted and long-
lived fluorescence is also typical of an H-dimer due to cancelled
transition dipoles in a lower energy state (Fig. 6b inset). The
absorption spectra of 1 in decreasing polarity solvents were
measured in addition to a control experiment (Fig. S6†).
Although the absorption maximum of 1 shows a slight blue-shift
(412 to 403 nm), neither the shape nor the intensity of the
absorption shows a perceivable change, indicating that the blue-
shift is merely a result of the solvatochromic effect. The mild
solvatochromic effect was observed in the fluorescence emission
of 1 too with no additional band for the various solvent polarities
(Fig. S6†).
From absorption data obtained for 1–3, D1 and P1 in solution
and nanoparticle dispersion of P1, we are quite convinced that
BF2dbm dyes have a strong tendency to form ground-state
dimers for the reason mentioned above. A blue-shifted absorp-
tion band at 391 nm is responsible for the new fluorescence band
at ꢀ550 nm for D1 and P1 (Fig. S1 and S2†). This hypsochromic
dimer, or H-dimer (H-aggregate) is formed when two (or more)
dye molecules are arranged in ‘‘face-to-face’’ configurations.21
Although the absorption evidence is not clear enough for D1, it is
possible that the intramolecular H-dimer22 and the unassociated
counterpart are at equilibrium in solution.23 Presumably the
enthalpy-driven ground-state dimer prefers less polar solvents,
due to the large dipoles24 and multiple non-covalent interactions
of BF2dbm solute molecules (inferred from crystal structures,
Fig. S4†). It is expected to see increased dimer probability
(population) when the solvent polarity is decreased. Fig. 6a
shows the D1 absorption (5.6 ꢂ 10ꢃ6 M) change in response to
decreased solvent polarity, where a CH2Cl2–n-hexane mixture at
various volume ratios is used as the solvent. When the content of
polar CH2Cl2 decreases, not only does the absorption show a
blue shift, the intensity of the absorption maximum also
increases by more than 50% in pure CH2Cl2 compared to 9 : 1
n-hexane–CH2Cl2. The increase in the new absorption band is
presumably due to aligned transition dipole moments of the two
For the polymer P1, higher-ordered H-aggregates are possible
since from Fig. 1, P1 exhibits a more dramatic blue-shifted
absorption and a greater contribution from low energy emission
than does D1. The alternating polar, rigid fluorophores and non-
polar, flexible linkers could allow for intrapolymer H-aggregate
formation.25,26 The equally long-lived lifetime of P1 (40 ns) is also
characteristic of H-aggregate emission. Although impurities,
defects, excimers, and exciplexes which all render red-shifted
emissions are not uncommon in conjugated fluorescent
17336 | J. Mater. Chem., 2012, 22, 17332–17339
This journal is ª The Royal Society of Chemistry 2012

View Online
polymers,27 P1 is essentially a site-isolated system and is thus far
less complicated to study than conjugated systems. Therefore,
these solution experiments unambiguously demonstrated the
propensity of H-aggregate, formation by BF2dbm dyes.
shown). It was noted that neither at room temperature nor at
77 K did the smeared region exhibit maximal ML effect; it was a
temperature region in between. The simple experiments have
convincingly demonstrated the temperature-dependent energy
transfer processes within the solid organoboron materials.
Previous studies also confirm that H-aggregate-like structure
of BF2dbm derivatives have the lowest emissive states: the only
known BF2dbm derivative that does not show ML is a single-
methoxy substituted (Fig. 8). The dye, BF2dbmOMe, was used to
monitor bulk polymerization of lactide because of its robust
crystalline emission at ꢀ550 nm (sF ¼ 41 ns),33 which coincides
with the spectra of the sheared crystals of 1–3, D1 and the NP
and solids of polymer P1. Unsurprisingly, the solved crystal
structure19 of BF2dbmOMe reveals a fully overlapped ‘‘face-to-
face’’ H-aggregate-like packing mode which is absent in any other
BF2dbm derivatives studied, including the current crystal struc-
tures for 1–3. As a result, the energy levels cannot be brought
down further by mechanical stimulation. Nonetheless, a more
symmetric spectrum was noted after grinding, which either could
be caused by reduced self-absorption or increased chaos in the
molecular orientations caused by mechanical force.
The excitation spectrum of the NPs in water reveals a further
blue-shifted maximum relative to that of the monomers 1–3
(Fig. 4, from 411 nm to 384 nm, or 1770 cmꢃ1), which indicates
that the same type, but possibly stronger, H-aggregate associa-
tion occurs in more firmly packed BF2dbm dye clusters. Thus the
NPs fluorescence emission (ꢀ550 nm, sF ¼ 24 ns) that closely
matches the lower energy part of D1 and P1 (ꢀ550 nm) in
solution is most likely to be caused by emissive H-aggregates as
well. The shorter fluorescence lifetime of the NP (compared to 40
ns of the H-aggregates in CH2Cl2) is possibly caused by increased
quenching events in the solid state. Since the polymer particle
(highly spatially confined and yet with rotationally flexible
linkers) is a close mimic of amorphous BF2dbm solids, we have a
good reason to believe the red-shifted emission (ꢀ550 nm)
induced by mechanical force in 1–3 is also caused by the forma-
tion of H-aggregates. As for the tendency for dimer or aggregate
formation in the solid state of BF2dbm derivatives, in previous
and our current investigations, both single-crystal XRD12,19 and
STM28 studies reveal that p–p stacking and ArH/F hydrogen
bonding could both contribute significantly.
To confirm that the yellow emission from BF2dbmOMe
mostly comes from stable H-aggregates and the effect of exciton
migration is minor in this case, smeared crystals of BF2dbmOMe
were bathed in liquid N2 under UV light. Unlike 1–3 the smeared
crystals of which show blue-shifted emission at 77 K,
BF2dbmOMe turns orange with a second-long, red ‘‘after-glow’’
after the UV excitation had ceased (Fig. 8 inset). This is a clear
indication of strong phosphorescence, or triplet-state emission
from the H-aggregates. As has been explained in the insets of
Fig. 6a and b, H-aggregates have a forbidden (long-lived) lowest
singlet state (sF ¼ 40 ns), which significantly enhances the
probability of intersystem crossing to triplet states.21 Therefore,
it is not surprising to see red-shifted emission from BF2dbmOMe
solids and not from 1–3. This is because the H-aggregate pop-
ulation generated by mechanical force for 1–3 could be small and
does not result in detectable phosphorescence.
3.2 Energy transfer in the solid state
Since the orientations of BF2dbm moieties in polymer solids and
sheared crystals are more likely to be random, with the possibility
to form site-isolated BF2dbm fluorophores and J-aggregate-like
structures, one obvious question to ask is why only H-aggregates
emission is observed? Or how is it possible to achieve uniform
fluorescence color conversion with crude mechanical stimuli, as
has been observed in the experiments? A hypothesis is that
(unperturbed) ordered crystals and other dye–dye interactions
possess higher excited state energy levels than those of the emissive
H-aggregates, which serve as acceptors (traps) in energy trans-
fer.29,30 Given that some energy transfer events (or exciton
migration31 in this case) can be thermally activated,32 it is expected
that BF2dbm ML can be quite different at lower temperatures.
Indeed when the glass slide covered with smeared solid of 3
was bathed in liquid N2 (Fig. 7), it was clearly visible that the
originally homogeneous yellow emission turned into three
different apparent colors (cyan, green, and yellow) readily
observable to the naked eye. The heavily smeared regions were
more yellow while the less smeared parts changed to cyan or
green emission. Similar observations were noted for 1 and 2 (not
3.3 A mechanistic model for BF2dbm ML in the solid state
Based on the experimental data in the current and previous
studies, we have summarized a mechanistic explanation for the
Fig. 7 Photos showing ML manifested by smearing crystals of 3 on glass
at room temperature (left), bottom part of the smeared region bathed in
liquid N2 (middle), and recovery of ML when the glass slide was retrieved
from liquid N2 to warm up to room temperature. (lex ¼ 365 nm).
Fig. 8 Fluorescence emission spectra of crystals and smeared crystals of
BF2dbmOMe on glass substrates (lex ¼ 385 nm). Inset: chemical struc-
ture of BF2dbmOMe and photo showing smeared BF2dbmOMe crystals
partially bathed in liquid N2 under UV light. (lex ¼ 365 nm).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17332–17339 | 17337

View Online
(1) BF2dbm fluorophores have a propensity to form H-dimers or
H-aggregates in solution and in amorphous solids, such as
polymeric nanoparticles or smeared BF2dbm crystals; (2)
mechanical stimuli can cause the formation of the H-dimers or
H-aggregates of BF2dbm that exhibit a low-lying emissive state
around 550 nm; (3) energy transfer may happen from other
higher excited states to the lowest H-aggregate excited states,
which eventually makes the entire solid emit at ꢀ550 nm. This
investigation can serve as a guideline for tailored material design
in the future.
Fig. 9 Pictorial representation of BF2dbm ML process: the crystals are
ordered and emit green-blue fluorescence (blue molecules); when the
order is disrupted by mechanical force, limited numbers of H-aggregate-
forming sites (orange molecules) can give the entire solid a red-shifted
emission mediated by energy transfer (exciton migration). Green mole-
cules in the Fig. represent any other type of intermolecular interaction
with an excited-state energy level above that of the H-aggregates but
below the crystals.
Acknowledgements
The work was supported by the University of Science and
Technology of China startup funds and the National Natural
Science Foundation of China (J1030412). We also thank Mr
Joshua Vaughn for the photos.
Notes and references
1 G. M. Whitesides, E. E. Simanek, J. P. Mathias, C. T. Seto, D. Chin,
M. Mammen and D. M. Gordon, Acc. Chem. Res., 1995, 28, 37.
2 J. A. A. W. Elemans, A. E. Rowan and R. J. M. Nolte, J. Mater.
Chem., 2003, 13, 2661–2670.
3 J. Seo, S. Kim, S. H. Gihm, C. R. Park and S. Y. Park, J. Mater.
Chem., 2007, 17, 5052–5057.
4 K. Ariga, T. Mori and J. P. Hill, Adv. Mater., 2012, 24, 158–176.
5 J. Wu, T. Yi, T. Shu, M. Yu, Z. Zhou, M. Xu, Y. Zhou, H. Zhang,
J. Han, F. Li and C. Huang, Angew. Chem., Int. Ed., 2008, 47,
1063–1067.
6 (a) The first review in the field: Y. Sagara and T. Kato, Nat. Chem.,
2009, 1, 605–610; (b) L. Maggini and D. Bonifazi, Chem. Soc. Rev.,
2012, 41, 211–241; (c) A thorough and more recent review: Z. Chi,
X. Zhang, B. Xu, X. Zhou, C. Ma, Y. Zhang, S. Liu and J. Xu,
Chem. Soc. Rev., 2012, 41, 3878–3896.
BF2dbm ML phenomenon shown in Fig. 9. Before mechanical
stimulation, the emission color (energy level) of the BF2dbm
crystals is determined by the manner of molecular packing in
crystals (hypothetically, green-blue in the scheme).34 Nonethe-
less, irrespective of the starting emissive energy levels of the
crystals, after shear force is applied, the ordered state is disrupted
into randomness and all possible dye–dye interactions can be
adopted, which could give rise to many non-vanishing excited
states.35 Among the many dye–dye interactions, shear force could
always bring about the possibility of H-aggregate-like associa-
tion (a metastable state), which has the lowest emissive energy
level and serves as the acceptor in efficient solid-state energy
transfer. Therefore exciton migration can happen from other
allowed states (resulted from different dye–dye interactions) to
H-aggregate emissive states.
7 D. A. Davis, A. Hamilton, J. Yang, L. D. Cremar, D. Van Gough,
S. L. Potisek, M. T. Ong, P. V. Braun, T. J. Martinez, S. R. White,
J. S. Moore and N. R. Sottos, Nature, 2009, 459, 68–72.
€
8 C. Lowe and C. Weder, Adv. Mater., 2002, 14, 1625–1629.
9 (a) S. R. Forrest and M. E. Thompson, Chem. Rev., 2007, 107, 923–
925; (b) Y. Kim, J. Bouffard, S. E. Kooi and T. M. Swager, J. Am.
Chem. Soc., 2005, 127, 13726–13731; (c) S. Kim, J. K. Lee,
S. O. Kang, J. Ko, J. H. Yum, S. Fantacci, F. De Angelis, D. Di
The proposed model predicts that for BF2dbm derivatives,
different solids with different starting fluorescence emission
colors will end in the same, lowest emissive state (specifically
ꢀ550 nm) after mechanical stimulation, provided that H-aggre-
gates are not highly unfavorable. That is to say that BF2dbm ML
should be dependent on the Gibbs free energy for H-aggregate
formation in the ground state. If the free energy is negative,
BF2dbm H-aggregates are stable and thus no ML, in terms of
emission color change, should be observed (Fig. 8). This is the
case of BF2dbmOMe which has an H-aggregate-like crystal
structure. If the free energy is positive, ML should be observable
since the initial stable packing mode differs from the metastable
H-aggregates. This is the case for the majority of the crystals and
solids observed.
€
Censo, M. K. Nazeeruddin and M. Gratzel, J. Am. Chem. Soc.,
2006, 128, 16701–16707; (d) V. Bulovic, R. Deshpande,
ꢁ
M. E. Thompson and S. R. Forrest, Chem. Phys. Lett., 1999, 308,
317–322.
10 (a) Y.-A. Lee and R. Eisenberg, J. Am. Chem. Soc., 2003, 125, 7778–
7779; (b) J. W. Chung, Y. You, H. S. Huh, B.-K. An, S.-J. Yoon,
S. H. Kim, S. W. Lee and S. Y. Park, J. Am. Chem. Soc., 2009, 131,
8163–8172; (c) C. Dou, L. Han, S. Zhao, H. Zhang and Y. Wang,
J. Phys. Chem. Lett., 2011, 2, 666–670; (d) H. Ito, T. Saito,
N. Oshima, N. Kitamura, S. Ishizaka, Y. Hinatsu, M. Wakeshima,
M. Kato, K. Tsuge and M. Sawamura, J. Am. Chem. Soc., 2008,
130, 10044–10045; (e) H. Li, X. Zhang, Z. Chi, B. Xu, W. Zhou,
S. Liu, Y. Zhang and J. Xu, Org. Lett., 2011, 13, 556–559; (f)
X. Luo, J. Li, C. Li, L. Heng, Y. Q. Dong, Z. Liu, Z. Bo and
B. Z. Tang, Adv. Mater., 2011, 23, 3261–3265; (g) J. Ni, X. Zhang,
N. Qiu, Y.-H. Wu, L.-Y. Zhang, J. Zhang and Z.-N. Chen, Inorg.
Chem., 2011, 50, 9090–9096; (h) Y. Ooyama and Y. Harima, J.
Mater. Chem., 2011, 21, 8372–8380; (i) A. Pucci and G. Ruggeri, J.
Mater. Chem., 2011, 21, 8282–8291; (j) D. R. T. Roberts and
S. J. Holder, J. Mater. Chem., 2011, 21, 8256–8268; (k) Y. Sagara
and T. Kato, Angew. Chem., 2011, 123, 9294–9298; (l) S.-J. Yoon,
J. W. Chung, J. Gierschner, K. S. Kim, M.-G. Choi, D. Kim and
S. Y. Park, J. Am. Chem. Soc., 2010, 132, 13675–13683; (m)
A. L. Balch, Angew. Chem., Int. Ed., 2009, 48, 2641–2644; (n)
S.-J. Yoon and S. Park, J. Mater. Chem., 2011, 21, 8338–8346; (o)
J. Mei, J. Wang, A. Qin, H. Zhao, W. Yuan, Z. Zhao,
H. H. Y. Sung, C. Deng, S. Zhang, I. D. Williams, J. Z. Sun and
4. Conclusions
In summary, we have synthesized monomeric, dimeric and
polymeric derivatives of difluoroboron dibenzoylmethane
(BF2dbm) complexes to investigate the mechanism of the
mechanochromic luminescence exhibited by most BF2dbm
complexes in the solid state. Using an approach that allows us to
unambiguously obtain excited-state interactions among BF2dbm
fluorophores, we were able to conclude the following statements:
17338 | J. Mater. Chem., 2012, 22, 17332–17339
This journal is ª The Royal Society of Chemistry 2012

View Online
B. Z. Tang, J. Mater. Chem., 2012, 22, 4290–4298; (p) S. Perruchas,
X. F. L. Goff, S. B. Maron, I. Maurin, F. O. Guillen, A. Garcia,
T. Gacoin and J.-P. Boilot, J. Am. Chem. Soc., 2010, 132, 10967–
10969.
20 A. Pucci, F. Di Cuia, F. Signori and G. Ruggeri, J. Mater. Chem.,
2007, 17, 783–790.
21 M. Kasha, H. R. Rawls and A. El-Bayoumi, Pure Appl. Chem., 1965,
11, 371–392.
22 F. C. Spano, Acc. Chem. Res., 2009, 43, 429–439.
€
11 (a) F. Jakle, Chem. Rev., 2010, 110, 3985–4022; (b) Z. M. Hudson and
€
€
23 U. Rosch, S. Yao, R. Wortmann and F. Wurthner, Angew. Chem.,
S. Wang, Acc. Chem. Res., 2009, 42, 1584–1596; (c) A. Nagai,
J. Miyake, K. Kokado, Y. Nagata and Y. Chujo, J. Am. Chem.
Soc., 2008, 130, 15276–15278; (d) W. L. Jia, M. J. Moran,
Y.-Y. Yuan, Z. H. Lu and S. Wang, J. Mater. Chem., 2005, 15,
3326–3333.
Int. Ed., 2006, 45, 7026–7030.
ꢁ
24 E. Cogne-Laage, J.-F. Allemand, O. Ruel, J.-B. Baudin, V. Croquette,
M. Blanchard-Desce and L. Jullien, Chem.–Eur. J., 2004, 10, 1445–
1455.
25 M. Kasha, Radiat. Res., 1963, 20, 55–71.
26 T.-Q. Nguyen, V. Doan and B. J. Schwartz, J. Chem. Phys., 1999, 110,
4068–4078.
27 S. A. Jenekhe and J. A. Osaheni, Science, 1994, 265, 765–768.
28 (a) X. Zhang, C.-J. Yan, G.-B. Pan, R.-Q. Zhang and L.-J. Wan,
J. Phys. Chem. C, 2007, 111, 13851–13854; (b) D. Rohde,
C.-J. Yan, H.-J. Yan and L.-J. Wan, Angew. Chem., Int. Ed., 2006,
45, 3996–4000.
12 G. Zhang, J. Lu, M. Sabat and C. L. Fraser, J. Am. Chem. Soc., 2010,
132, 2160–2162.
13 (a) G. Zhang, J. P. Singer, S. E. Kooi, R. E. Evans, E. L. Thomas and
C. L. Fraser, J. Mater. Chem., 2011, 21, 8295–8299; (b) N. D. Nguyen,
G. Zhang, J. Lu, A. E. Sherman and C. L. Fraser, J. Mater. Chem.,
2011, 21, 8409–8415; (c) T. Liu, A. D. Chien, J. Lu, G. Zhang and
C. L. Fraser, J. Mater. Chem., 2011, 21, 8401–8408.
ꢀ
14 (a) G. Bourhill, L. O. Palsson, I. D. W. Samuel, I. C. Sage,
I. D. H. Oswald and J. P. Duignan, Chem. Phys. Lett., 2001, 336,
234–241; (b) M. J. Mayoral, P. Ovejero, M. Cano and G. Orellana,
Dalton Trans., 2011, 40, 377–383.
29 H. C. Wolf, in Advances in Atomic and Molecular Physics, ed. D. R.
Bates and E. Immanuel, Academic Press, 1968, vol. 3, pp. 119–
142.
15 (a) C. Curutchet and B. Mennucci, J. Am. Chem. Soc., 2005, 127,
16733–16744; (b) V. Russo, C. Curutchet and B. Mennucci, J. Phys.
Chem. B, 2007, 111, 853–863.
30 (a) C. E. Evans and P. W. Bohn, J. Am. Chem. Soc., 1993, 115, 3306–
3311; (b) C. W. Farahat, T. L. Penner, A. Ulman and D. G. Whitten,
J. Phys. Chem., 1996, 100, 12616–12623.
16 U. H. F. Bunz, V. Enkelmann, L. Kloppenburg, D. Jones,
K. D. Shimizu, J. B. Claridge, H.-C. zur Loye and G. Lieser, Chem.
Mater., 1999, 11, 1416–1424.
31 J.-I. Lee, T. Zyung, R. D. Miller, Y. H. Kim, S. C. Jeoung and
D. Kim, J. Mater. Chem., 2000, 10, 1547–1550.
32 D. Beljonne, G. Pourtois, C. Silva, E. Hennebicq, L. M. Herz,
€
€
R. H. Friend, G. D. Scholes, S. Setayesh, K. Mullen and
17 G. Kuzmanich, S. Simoncelli, M. N. Gard, F. Spanig,
ꢁ
B. L. Henderson, D. M. Guldi and M. A. Garcia-Garibay, J. Am.
Chem. Soc., 2011, 133, 17296–17306.
18 A. Pfister, G. Zhang, J. Zareno, A. F. Horwitz and C. L. Fraser, ACS
Nano, 2008, 2, 1252–1258.
19 A. G. Mirochnik, B. V. Bukvetskii, E. V. Fedorenko and
V. E. Karasev, Russ. Chem. Bull., 2004, 53, 291–296.
J. L. Bredas, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 10982–10987.
33 G. Zhang, S. Xu, A. G. Zestos, R. E. Evans, J. Lu and C. L. Fraser,
ACS Appl. Mater. Interfaces, 2010, 2, 3069–3074.
34 M. Schwoerer and H. C. Wolf, in Organic Molecular Solids, Wiley-
VCH Verlag GmbH, 2008, pp. 125–175.
35 G. D. Scholes and G. Rumbles, Nat. Mater., 2006, 5, 683–696.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17332–17339 | 17339