Low dose detection of γ radiation via solvent assisted fluorescence quenching

Article abstract of DOI:10.1021/ja500262n

Development of low cost, easy-to-use chemical sensor systems for low dose detection of radiation remains highly desired for medical radiation therapy and nuclear security monitoring. We report herein on a new fluorescence sensor molecule, 4,4'-di(1H-phenanthro[9,10-d]imidazol-2-yl)biphenyl (DPI-BP), which can be dissolved into halogenated solvents (e.g., CHCl₃, CH ₂Cl₂) to enable instant detection of radiation down to the 0.01 Gy level. The sensing mechanism is primarily based on radiation induced fluorescence quenching of DPI-BP. Pristine DPI-BP is strongly fluorescent in halogenated solvents. When exposed to radiation, the halogenated solvents decompose into various radicals, including hydrogen and chlorine, which then combine to produce hydrochloric acid (HCl). This strong acid interacts with the imidazole group of DPI-BP to convert it into a DPI-BP/HCl adduct. The DPI-BP/HCl adduct possesses a more planar configuration than DPI-BP, enhancing the π-π stacking and thus molecular aggregation. The strong molecular fluorescence of DPI-BP gets quenched upon aggregation, due to the π-π stacking interaction (forming forbidden low-energy excitonic transition). Interestingly the quenched fluorescence can be recovered simply by adding base (e.g., NaOH) into the solution to dissociate the DPI-BP/HCl adduct. Such sensing mechanism was supported by systematic investigations based on HCl titration and dynamic light scattering measurements. To further confirm that the aggregation caused fluorescence quenching, a half size analogue of DPI-BP, 2-phenyl-1H-phenanthro[9,10-d]imidazole (PI-Ph), was synthesized and investigated in comparison with the observations of DPI-BP. PI-Ph shares the same imidazole conjugation structure with DPI-BP and is expected to bind the same way with HCl. However, PI-Ph did not show fluorescence quenching upon binding with HCl likely due to the smaller π-conjugation structure, which can hardly enforce the π-π stacking assembly. Combining the low detection limit, fast and reversible fluorescence quenching response, and low cost of halogenated solvent composites, the sensor system presented may lead to the development of new, simple chemical dosimetry for low dose detection of γ radiation.

Full text of DOI:10.1021/ja500262n

Article
pubs.acs.org/JACS
Low Dose Detection of γ Radiation via Solvent Assisted Fluorescence
Quenching
Ji-Min Han,† Miao Xu,† Brian Wang,‡ Na Wu,† Xiaomei Yang,† Haori Yang,§ Bill J. Salter,∥
and Ling Zang*,†
†Department of Materials Science and Engineering, University of Utah, 36 South Wasatch Drive, Salt Lake City, Utah 84112, United
States
§Department of Nuclear Engineering & Radiation Health Physics, Oregon State University, 3451 SW Jefferson Way, Radiation Center
E108, Corvallis, Oregon 97331, United States
‡Department of Radiation Oncology, University of Louisville, 529 South Jackson Street, Louisville, Kentucky 40202, United States
∥Huntsman Cancer Institute, Department of Radiation Oncology, University of Utah, 1950 Circle of Hope, Salt Lake City, Utah
84112, United States
S
* Supporting Information
ABSTRACT: Development of low cost, easy-to-use chemical
sensor systems for low dose detection of γ radiation remains
highly desired for medical radiation therapy and nuclear
security monitoring. We report herein on a new fluorescence
sensor molecule, 4,4′-di(1H-phenanthro[9,10-d]imidazol-2-yl)-
biphenyl (DPI-BP), which can be dissolved into halogenated
solvents (e.g., CHCl3, CH2Cl2) to enable instant detection of γ
radiation down to the 0.01 Gy level. The sensing mechanism is
primarily based on radiation induced fluorescence quenching of
DPI-BP. Pristine DPI-BP is strongly fluorescent in halogenated solvents. When exposed to γ radiation, the halogenated solvents
decompose into various radicals, including hydrogen and chlorine, which then combine to produce hydrochloric acid (HCl). This
strong acid interacts with the imidazole group of DPI-BP to convert it into a DPI-BP/HCl adduct. The DPI-BP/HCl adduct
possesses a more planar configuration than DPI-BP, enhancing the π−π stacking and thus molecular aggregation. The strong
molecular fluorescence of DPI-BP gets quenched upon aggregation, due to the π−π stacking interaction (forming forbidden low-
energy excitonic transition). Interestingly the quenched fluorescence can be recovered simply by adding base (e.g., NaOH) into
the solution to dissociate the DPI-BP/HCl adduct. Such sensing mechanism was supported by systematic investigations based on
HCl titration and dynamic light scattering measurements. To further confirm that the aggregation caused fluorescence
quenching, a half size analogue of DPI-BP, 2-phenyl-1H-phenanthro[9,10-d]imidazole (PI-Ph), was synthesized and investigated
in comparison with the observations of DPI-BP. PI-Ph shares the same imidazole conjugation structure with DPI-BP and is
expected to bind the same way with HCl. However, PI-Ph did not show fluorescence quenching upon binding with HCl likely
due to the smaller π-conjugation structure, which can hardly enforce the π−π stacking assembly. Combining the low detection
limit, fast and reversible fluorescence quenching response, and low cost of halogenated solvent composites, the sensor system
presented may lead to the development of new, simple chemical dosimetry for low dose detection of γ radiation.
detectors. However, ion chambers are vulnerable to low
sensitivity,2 and semiconductor detectors suffer from energy
and angular dependence.3 The measured data by film-based
detectors are time-consuming to read out. The single-crystal-
based scintillation detectors are usually too costly to scale up.4
To overcome these challenges, various organic chemical sensors
(e.g., those based on molecules or polymers) have been
developed and studied for gamma detection. Molecular and
polymeric sensors provide many advantages over the current
detector systems, including ease of processing, low cost, high
adaptability for size miniaturization, and high conformability for
INTRODUCTION
■
Low dose detection of γ radiation remains essential in medical
radiation treatment of cancer and for nuclear relevant security.1
For example, the use of γ rays to treat cancerous tumors
requires the precise calibration and delivery of γ radiation
intensity (dose) to the tumor, for which the source intensity
must be controlled down to the resolution of 0.01 Gy.
Meanwhile, greater awareness of the threat of nuclear and
radiological terrorism has been raised as a result of the growing
violence in terrorist attacks. It becomes more critical now than
ever to develop facile and low cost dosimeters for instant
detection of γ radiation.
Current popular radiation dosimeters include ion chambers,
semiconductor detectors, radiographic films, and scintillation
Received: January 9, 2014
Published: March 10, 2014
© 2014 American Chemical Society
5090
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096

Journal of the American Chemical Society
Article
fabrication into various shapes.5 While many types of chemical
sensors have been developed,6 the best sensor system available
so far can only detect γ radiation down to a level of about 10
Gy,6g,h still several orders of magnitude less sensitive than what
is required for practical use in personnel dose monitoring or
environmental radiation measurement. There is a large
technical gap between laboratory research of chemical sensor
systems and the practical application of such systems. We
report herein on a new molecular sensor system that can
significantly improve the detection limit by 3 orders of
magnitude, down to 0.01 Gy.
fluorophores, can only detect protons in polar solvents (e.g.,
water, DMSO),10 whereas in nonpolar solvents HCl can hardly
be ionized into free protons. To overcome this technical
challenge, we have herein developed a new fluorescence sensor
molecule, DPI-BP, which can dissolve into halogenated solvents
to enable instant sensing of HCl generated under γ radiation
(Figure 1). The sensing mechanism is primarily based on the
addition reaction between HCl and the imidazole moieties of
DPI-BP. The HCl adduct of DPI-BP intends to aggregate,
causing quenching of the molecular fluorescence of DPI-BP due
to the intermolecular π−π stacking interaction.
The design of DPI-BP was based on 3-fold consideration.
First, the imidazole moiety has been widely employed in
synthesis of organic semiconducting materials,11 because of its
ease for structure modification,12 strong chemical durability13
and high fluorescence efficiency of the π-conjugated system.14
Indeed, DPI-BP was synthesized in this work with a high yield
of 90%, and its fluorescence quantum yield (Φ) was
determined to be as high as 0.81 (see Supporting Information).
Second, as common alkaloid heterocycles, the π-expanded
derivatives of imidazole have high affinity to acids;15 thus they
form a stable adduct with HCl instantly in nonpolar solvents.
Lastly, the decreased solubility of the DPI-BP/HCl adduct, in
conjunction with the strong intermolecular π−π stacking
interaction, leads to aggregation of DPI-BP molecules, which
in turn results in effective fluorescence quenching of the
molecules.16 To confirm the aggregation induced fluorescence
quenching, we have synthesized a reference compound, PI-Ph,
a half sized analogue of DPI-BP (Scheme 1), and investigated it
for γ radiation sensing following the same procedures as those
tested for DPI-BP. As expected, due to the smaller size (and
thus the higher solubility) of PI-Ph, no aggregation was
observed for this compound when interacting with HCl.
Combining the low detection limit and expedient sensing
response (via rapid radical reactions), the reported sensor
system of DPI-BP will help open new opportunities for
developing low-cost, easy-to-use devices for low dose detection
of γ radiation.
Scheme 1 shows the molecular structure of the sensor, 4,4′-
di(1H-phenanthro[9,10-d]imidazol-2-yl)biphenyl (DPI-BP),
Scheme 1. (a) Generation of HCl from CHCl3 under γ
Radiation, (b) Formation of Imidazole/HCl Adduct in
Nonpolar Solvents, and (c) Molecular Structures of the
Sensor Molecule, DPI-BP, and Its Half Size Analogue, PI-Ph
and the sensing mechanism based on γ radiation initiated
radical reactions. The half size analogue of the sensor molecule,
2-phenyl-1H-phenanthro[9,10-d]imidazole (PI-Ph), was also
synthesized and used as a reference to study the sensing
mechanism. Under γ radiation, the halogenated solvents, such
as chloroform (CHCl3) and dichloromethane (CH2Cl2),
decompose into free radicals as shown in Scheme 1a.7 These
radicals are unstable, undergoing fast recombination to form
new compounds including hydrochloric acid (HCl). The
presence of water (humidity) and oxygen under ambient
condition does not affect significantly the production of HCl,
mainly due to their slow reaction with the radicals.8 This
inspired us to develop a sensing pathway that can detect γ
radiation simply by detecting HCl thus produced. However, in
situ detection of HCl in nonpolar solvents still remains a great
challenge.9 Current pH sensors, particularly those based on
MATERIALS AND METHODS
■
PI-Ph was synthesized following the literature methods.12a DPI-BP was
synthesized as follows: Phenanthrene-9,10-quinone (1.0 g, 4.8 mmol),
ammonium bicarbonate (5.0 g, 63 mmol, excess), and biphenyl-4,4′-
dicarbaldehyde (0.5 g, 2.4 mmol) were added into 150 mL of EtOH in
a 250 mL round flask; the suspension thus obtained was refluxed for 6
h. During the refluxing, bright yellow precipitate was eventually
Figure 1. (a) Schematic drawing of the fluorescence sensing mechanism of DPI-BP. Protonation interaction with radiation generated HCl causes
molecular aggregation of DPI-BP, which in turn results in fluorescence quenching due to π−π stacking. (b) Fluorescence photograph of a DPI-BP
solution in CHCl3 taken before (left) and after (right) 3.0 Gy of γ radiation, indicating complete fluorescence quenching.
5091
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096

Journal of the American Chemical Society
Article
Figure 2. (a) Absorption and (b) fluorescence spectra of a CHCl3 solution of DPI-BP (5 × 10−6 mol/L) recorded under increasing doses of γ
radiation up to 5.0 Gy. (c) A plot showing the relative decrease in absorption (measured at 378 nm) of the same DPI-BP solution shown in panel a
as a function of the dosage of γ radiation; the data points in low dosage range (0−2.0 Gy) are fitted in linear relationship (R2 = 0.988). (d) A plot
showing the relative fluorescence quenching (measured at 451 nm) of the same DPI-BP solution shown in panel b as a function of the dosage of γ
radiation; the data points in low dosage range (0−2.0 Gy) are fitted in linear relationship (R2 = 0.980).
formed. After cooling to room temperature, the mixture was filtered,
and the residue was collected. The crude product was then washed
with deionized water (2 × 25 mL), ethanol (2 × 25 mL), and cold
CHCl3 (2 × 25 mL), and finally dried under vacuum. The final
product of DPI-BP was obtained as a yellow solid (1.27 g, 90% yield).
1H NMR (DMSO-d6, 300 MHz, ppm): δ= 13.66 (2 H, s), 8.90 (4 H,
d, J = 8 Hz), 8.65 (4 H, d, J = 8 Hz), 8.5 (4 H, d, J = 8 Hz), 8.11 (4 H,
d, J = 8 Hz), 8.79 (4 H, t, J = 8 Hz), 8.68 (4 H, t, J = 8 Hz). 13C NMR
(DMSO-d6, 75 MHz, ppm): δ = 149.6, 140.6, 130.5, 128.5, 128.0,
127.9, 127.6, 126.2, 124.8, 122.9. LRMS (ESI): m/z calcd 586.2; found
587.2 [M + H]+.
Halogenated methanes are commonly used for radical generation
under radiation, with the efficiency in the order of CH2I2 > CHBr3 >
CH2Br2 > CHCl3 > CH2Cl2.17 Considering the fact that DPI-BP is not
fluorescent in iodinated and brominated solvents due to the enhanced
intersystem crossing (ISC) by the so-called “heavy-atom effect”,18 we
chose CHCl3 and CH2Cl2 as the two major solvents to study the
sensor reactions. Though halogenated ethanes and other larger alkanes
may function similarly regarding radical generation, these solvents
were not employed for testing mainly because of their decreased
efficiency in production of hydrogen halides. CHCl3 and CH2Cl2 were
purchased from Sigma-Aldrich. Before being used for experiments,
these solvents were further purified to remove the excessive acids and
stabilizers. Briefly, CHCl3 (200 mL) was washed with deionized water
(200 mL) to remove EtOH and HCl, followed by drying with CaCl2.
After refluxing with P2O5, the CHCl3 was collected by distillation.
CH2Cl2 (200 mL) was then refluxed with CaH2 and collected by
distillation. The purified CHCl3 and CH2Cl2 thus obtained were sealed
and stored in the dark to avoid photochemical decomposition.
γ-Irradiation experiments were performed with a 6 MV photon
beam on a Varian/BrainLab Novalis Classic (Varian Medical Systems,
Palo Alto, CA; BrainLAB AG, Feldkirchen, Germany) linear
accelerator (LINAC) at room temperature. The radiation output
was calibrated by an ionization chamber to generate 0.01 Gy/MU
(monitor unit) at maximum dose depth of 1.4 cm in water with a
source to surface distance (SSD) of 100 cm. The ionization chamber
used has a calibration that is traceable to an accredited dosimetry
calibration laboratory. The radiation beam from the LINAC machine
head was angled to the downward direction, wherein the sensor
samples were placed underneath at a height that was calculated to yield
a specific dose of radiation. Slabs of solid water were placed atop the
samples to provide dose building up. Before exposure to γ radiation,
the CHCl3 solution of DPI-BP was colorless, with strong blue
fluorescence (Figure 1b). Upon irradiation of 3.0 Gy, the fluorescence
of DPI-BP was completely quenched, implying instant sensor response
for γ radiation.
5092
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096

Journal of the American Chemical Society
Article
Figure 3. (a) A plot showing the relative decrease in absorption (measured at 378 nm) of a CHCl3 solution of DPI-BP (5 × 10−6 mol/L) as a
function of the concentration of HCl added; the data points in low concentration range ((0−15) × 10−6 mol/L) are fitted in linear relationship (R2 =
0.987). (b) A plot showing the relative fluorescence quenching (measured at 451 nm) of the same DPI-BP solution shown in panel a as a function of
the concentration of HCl added; the data points in low concentration range ((0−15) × 10−6 mol/L) are fitted in linear relationship (R2 = 0.991). (c)
Result of dynamic light scatting measured over a CHCl3 solution of DPI-BP (5 × 10−6 mol/L) before and after 3.0 Gy of γ radiation. (d)
Comparison of the fluorescence quenching efficiency of DPI-BP (5 × 10−6 mol/L) dissolved in CHCl3 and CH2Cl2 upon 3.0 Gy of γ radiation; also
compared are the results of the same concentration of DPI-BP dissolved in several other nonhalogen solvents, though upon 100 Gy of γ radiation.
consistent with the new absorption band measured at 428 nm.
More strikingly, the optimized (energy minimized) structure of
DPI-BP/HCl adduct showed approximately coplanar config-
uration, that is, only ca. 11° between the two BPI planes
(Figure S2, Supporting Information). In contrast, the pristine
DPI-BP demonstrates a typical twisted configuration between
the two BPI planes, with a torsional angle as large as 47°
(Figure S1, Supporting Information). The coplanar geometry of
DPI-BP/HCl adduct is highly conducive to the cofacial
intermolecular π−π stacking.5d The calculation result supports
the hypothesis that binding with HCl enhances the molecular
aggregation of DPI-BP, which is also consistent with the
increased light scattering observed at longer wavelength (Figure
2a). As to be demonstrated below, the dynamic light scattering
measurement also confirmed the formation of aggregation of
DPI-BP upon binding with HCl.
RESULTS AND DISCUSSION
■
The halogenated solvent mediated γ radiation sensor system
was investigated in detail using both UV−vis absorption and
fluorescence spectral measurements. As shown in Figure 2a, the
main absorption peak of DPI-BP centered at 378 nm decreased
gradually with increasing dose of γ radiation. Meanwhile, a new
absorption band emerged around 428 nm, corresponding to the
formation of DPI-BP/HCl adduct. Along with this new
absorption band, the absorption baseline at the longer
wavelength was also increased, indicating the formation of
molecular aggregation that caused light scattering. The
electronic structure of DPI-BP before and after binding with
HCl was calculated by using time dependent density functional
theory (TD-DFT) (see Supporting Information). For pristine
DPI-BP, the energy band gap between the highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular
orbital (LUMO) levels is calculated as 3.40 eV, which matches
the absorption peak measured at 378 nm. The band gap of the
DPI-BP/HCl adduct is calculated to be 3.10 eV, which is also
Parallel with the absorption measurement, the fluorescence
spectra of the same CHCl3 solution of DPI-BP was also
measured as shown in Figure 2b. Before exposure to γ
5093
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096

Journal of the American Chemical Society
Article
radiation, the DPI-BP solution demonstrated strong blue
emission with maximal wavelength centered at 451 nm. Upon
γ irradiation, the emission intensity was gradually quenched
with increase in radiation dosage. Remarkably, after only 0.1 Gy
of γ radiation exposure, the emission of DPI-BP decreased
about 5.5%. Such significant fluorescence quenching implies
great potential of the DPI-BP sensor for low dose detection.
Continuous exposure to radiation up to 3.0 Gy resulted in
complete fluorescence quenching of the DPI-BP solution.
Figure 2c,d presents the change of absorption and fluorescence
intensity (data obtained from Figure 2a,b, respectively) as a
function of the γ radiation dose. Both plots show a linear
relationship in the low dose range (0 to 2.0 Gy), which
provides quite wide dynamic range for quantitative analysis of
the detection limit. If we define a decrease in spectral intensity
three times higher than the standard deviation as the detectable
signal, the detection limit for the DPI-BP sensor system shown
in Figure 2 is projected to be ca. 0.01 Gy (by using the
absorption data) or 0.007 Gy (by using fluorescence data) (see
Supporting Information). The detection limit thus projected is
about 3 orders of magnitude better than those previously
reported for the other organic or polymeric fluorescence
sensors.6g,h Moreover, the DPI-BP sensor can be completely
recovered after exposure to γ radiation simply by adding strong
base, such as NaOH or Et3N (Figure S9, Supporting
Information), which deprotonates the HCl from the DPI-BP/
HCl adduct, thus dissolving the aggregate back to molecular
solution of free base DPI-BP. Both absorption and fluorescence
spectra were measured for this recovered DPI-BP solution and
showed no difference compared with the original solution
before γ radiation. This observation also helps confirm the
critical role of HCl in the sensing process of DPI-BP as outlined
in Scheme 1 and Figure 1. The reversibility, together with the
high sensitivity and fast response, thus obtained for the DPI-BP
solution makes it an ideal sensor system for expedient detection
of γ radiation.
To further prove that the above observed fluorescence
quenching of DPI-BP was induced by the radiation generated
HCl acid, a series of titration experiments were conducted
within the same DPI-BP solution in CHCl3. As expected, upon
addition of HCl, both the absorption and fluorescence spectra
of DPI-BP (Figure S3, Supporting Information) underwent
almost identical spectral change as obtained under γ radiation
(Figure 2a,b). Moreover, the fluorescence quenching was
observed instantly upon titration with HCl, indicating the
expedient acid−base interaction between HCl and the
imidazole moiety of DPI-BP. We also examined the same
titration process by replacing DPI-BP with a conventional
fluorescent pH sensor based on rhodamine. Since the
rhodamine sensor only reacts with free protons and HCl can
hardly be ionized into free protons in hydrophobic solvent like
CHCl3, the presence of HCl in CHCl3 solution (10−5 mol/L)
did not turn on the fluorescence of rhodamine molecules (5 ×
10−6 mol/L) (Figure S4, Supporting Information). In contrast,
when the same experiment was repeated in 1:1 (volume)
water/ethanol solvent (where free protons are available),
significant fluorescence turn-on of rhodamine was observed.
When the concentration of HCl in CHCl3 solution was
increased to 10−4 mol/L (i.e., 20 times excess of the
concentration of rhodamine), the fluorescence could still just
barely be activated, and the increase in emission intensity took
over 500 min to reach equilibrium, indicating the slow
protonation process of rhodamine sensor in hydrophobic
solvent (Figure S5, Supporting Information). Even at
equilibrium, the emission intensity observed was only about
one-third of the intensity observed in the hydrophilic water/
ethanol solvent, wherein the concentration of HCl was only
two times that of the rhodamine sensor. Moreover, with the
assumption that the fluorescence quenching of DPI-BP
observed under γ radiation is solely due to the interaction
with HCl, the titration curves shown in Figure 3a,b can be used
to estimate the concentration of HCl formed under different
doses of γ radiation. For instance, for 1 Gy of delivered γ
radiation dose, the fluorescence of DPI-BP was quenched by
32%, which corresponds to 6.0 × 10−6 mol/L of HCl (as shown
in Figure 3b). This concentration is in good agreement with the
theoretically projected value,7b 8.0 × 10−6 mol/L of HCl
produced under 1.0 Gy of delivered γ radiation. In addition to
HCl, other molecules may also be formed under the γ radiation
through recombination of the radicals produced (Scheme 1).
These new molecules may include tetrachloromethane,
pentachloroethane, 1,1,2,2-tetrachloroethane, and tetrachloroe-
thene. To exclude the possibility that the observed fluorescence
quenching above was due to these byproducts, we also
performed the fluorescence titration experiments of the same
DPI-BP solution in CHCl3 (as used in Figure 3) by adding the
four molecules. The results showed that none of these
molecules caused significant fluorescence quenching of the
sensor even when the concentration added was increased up to
5 × 10−4 mol/L (Figure S8, Supporting Information). This is in
sharp contrast to the titration result with HCl, for which close
to 100% fluorescence quenching was achieved even when the
concentration of HCl was as low as 1.5 × 10−5 mol/L. Clearly
the observed fluorescence quenching of DPI-BP is due to the
binding reaction with HCl.
To confirm the molecular aggregation of DPI-BP upon
interaction with HCl generated under γ radiation, dynamic light
scatting (DLS) measurements were performed over the CHCl3
solution of DPI-BP (5 × 10−6 mol/L) before and after exposure
to 3.0 Gy of γ radiation (Figure 3c). The hydrodynamic radius
(Rh) of DPI-BP before the γ irradiation has a narrow
distribution below 1 nm, indicating that DPI-BP was
molecularly dissolved in the CHCl3 solution. In contrast, larger
aggregates with wider range of size (averaged around 100 nm)
were formed in the solution after γ irradiation, consistent with
the discussion above that the DPI-BP/HCl adduct is more
favored for molecular assembly through π−π stacking.
To further explore the essential role of halogenated solvent
in the radiation induced fluorescence quenching of DPI-BP, we
also performed the same experiments in another halogenated
solvent, CH2Cl2, and several other nonhalogenated solvents,
such as THF, DMSO, toluene, and DMF (Figure 3d). As
expected, the CH2Cl2 solution demonstrated almost the same
quenching efficiency as CHCl3, indicating the same radical
reaction mechanism as shown in Scheme 1. However, no
detectable fluorescence quenching was observed for all the
nonhalogenated solvents, even under much higher dose of γ
radiation, 100 Gy. This observation further supports the
hypothesis that the sensing of DPI-BP relies on the radiation
induced generation of HCl.
The design of the sensor molecule DPI-BP is also well
justified regarding the expanded π-conjugation that is
conducive to increasing the π−π stacking interaction, thus
leading to molecular aggregation. For comparison, we designed
and synthesized a smaller analogue, PI-Ph, which possesses the
same conjugation of DPI-BP but only half size and is suited for
5094
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096

Journal of the American Chemical Society
Article
Figure 4. (a) A plot showing the absorption change of PI-Ph (5 × 10−6 mol/L CHCl3 solution) recorded at 356 nm as a function of the
concentration of HCl added; data points are fitted with the bimolecular reaction equilibrium (R2 = 0.995). (b) DLS measurement over a CHCl3
solution of PI-Ph (5 × 10−6 mol/L) before and after 5.0 Gy of γ radiation.
being used as a reference compound to study the sensing
mechanism. When examined under the same experimental
conditions (for both γ radiation and HCl titration in CHCl3
solution), PI-Ph underwent a very similar spectral shift,
indicating the acid−base binding with HCl in both cases
(Figures S6 and S7, Supporting Information). However, no
fluorescence quenching was observed. The spectral change
caused by HCl titration shows well-defined isosbestic points in
both absorption and fluorescence spectra, indicating the
stoichiometric transformation of PI-Ph to the PI-Ph/HCl
adduct. The bathochromic shift of fluorescence observed for
the PI-Ph/HCl adduct is consistent with our theoretical
calculation and previous observation on similar imidazole
based fluorophores.19 The lack of fluorescence quenching is
consistent with the smaller conjugation structure of PI-Ph, for
which the π−π stacking interaction is not as strong as that of
DPI-BP to afford molecular aggregation. This was further
confirmed by the DLS experiment performed on the CHCl3
solution of PI-Ph before and after 5.0 Gy of gamma irradiation
(Figure 4b), wherein both the hydrodynamic radii Rh were
smaller than 1 nm, indicating no molecular aggregation. To
quantify the strong interaction between imidazole moiety and
HCl in CHCl3, we fit the titration data with bimolecular
reaction equilibrium (Figure 4a), from which the binding
constant of PI-Ph/HCl was estimated to be 1.92 × 105 L/mol
(see Supporting Information). This high value of binding
constant implies effective formation of the PI-Ph/HCl adduct
in diluted CHCl3 solution. Considering the same imidazole
based structure of PI-Ph and DPI-BP, we speculate that the
interaction of DPI-BP with HCl would share similar binding
strength, though it undergoes multistep equilibrium because it
contains two imidazole moieties. Combination of all the
observations above clearly supports the hypothesis that the γ
radiation induced fluorescence quenching of DPI-BP is mainly
due to the binding interaction with HCl, which in turn causes
molecular aggregation through π−π stacking assembly.
solvents (e.g., CHCl3, CH2Cl2) to enable instant detection of γ
radiation down to the 0.01 Gy level. The sensing mechanism is
primarily based on fluorescence quenching of DPI-BP upon
binding to HCl, forming DPI-BP/HCl adduct, which enhances
the π−π stacking interaction and thus the molecular
aggregation in the halogenated solvents. HCl was produced
from the γ radiation induced decomposition of CHCl3 or
CH2Cl2 and the subsequent radical recombination. Such a
sensing mechanism was extensively investigated and proven by
comparative experiments between DPI-BP and its half size
analogue, PI-Ph, under both γ radiation and HCl titration. The
uniqueness of the reported sensor system lies in the design of
DPI-BP, which undergoes molecular aggregation upon binding
with HCl. This feature, in combination with the high reactivity
of halogenated solvents toward γ radiation, enables the
development of a solvent assisted fluorescence quenching
sensor for low dose radiation detection. Given the high
sensitivity, expedient response, and low cost, the DPI-BP sensor
composite may find broad applications in nuclear relevant
security monitoring and dosage calibration and measurement in
the medical γ ray treatment of cancer.
ASSOCIATED CONTENT
■
S
* Supporting Information
Materials, general instrumentations, and sensor testing in
solutions. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
LZANG@eng.utah.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
CONCLUSION
■
This work was supported by DHS (Grant 2009-ST-108-
LR0005). We thank Dr. Marc Porter for allowing us to use the
DLS facility in his laboratory.
In conclusion, we have developed a fluorescence sensor
molecule, DPI-BP, which can be dissolved in halogenated
5095
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096

Journal of the American Chemical Society
Article
Valley, CA, 1991. (c) Malkin, J. Photophysical and Photochemical
Properties of Aromatic Compounds; CRC Press: Boca Raton, FL, 1992;.
(d) Birks, J. B. Photophysics of Aromatic Molecules; Wiley: New York,
1970; Vol. 71.
(17) Bronze-Uhle, E. S.; Borin, J. F.; Batagin-Neto, A.; Alves, M. C.
O.; Graeff, C. F. O. Mater. Chem. Phys. 2012, 132, 846.
(18) Valeur, B. Molecular Fluorescence: Principles and Applications;
John Wiley & Sons: New York, 2013.
REFERENCES
■
(1) (a) Knoll, G. F. Radiation Detection and Measurement; Wiley: New
York, 2010. (b) Herz, R. H. The Photographic Action of Ionizing
Radiations in Dosimetry and Medical, Industrial, Neutron, Auto- and
Microradiography; Wiley-Interscience: New York, 1969.
(2) Graham, S. C.; Friend, R. H.; Fung, S.; Moratti, S. C. Synth. Met.
1997, 84, 903.
(3) Webster, J. G. The Measurement, Instrumentation, and Sensors:
Handbook; CRC Press: Boca Raton, FL, 1999.
(19) Kim, H. J.; Heo, C. H.; Kim, H. M. J. Am. Chem. Soc. 2013, 135,
17969.
(4) Storm, L.; Israel, H. I. At. Data Nucl. Data Tables 1970, 7, 565.
(5) (a) James, T. D.; Shinkai, S. Top. Curr. Chem. 2002, 218, 159.
(b) Janata, J.; Josowicz, M. Nat. Mater. 2003, 2, 19. (c) Grate, J. W.
Chem. Rev. 2008, 108, 726. (d) Zang, L.; Che, Y.; Moore, J. S. Acc.
Chem. Res. 2008, 41, 1596.
(6) (a) Kudoh, H.; Sasuga, T.; Seguchi, T.; Katsumura, Y. Polymer
1996, 37, 2903. (b) Silva, E. A. B.; Borin, J. F.; Nicolucci, P.; Graeff, C.
F. O.; Netto, T. G.; Bianchi, R. F. Appl. Phys. Lett. 2005, 86,
No. 131902. (c) Zhao, Y. S.; Zhong, H.; Pei, Q. Phys. Chem. Chem.
Phys. 2008, 10, 1848. (d) Zhong, H.; Zhao, Y.; Li, Y.; Pei, Q.
Nanotechnology 2008, 19, No. 505503. (e) Lobez, J. M.; Swager, T. M.
Angew. Chem., Int. Ed. 2010, 49, 95. (f) Liu, Z.; Xue, W.; Cai, Z.;
Zhang, G.; Zhang, D. J. Mater. Chem. 2011, 21, 14487. (g) Bronze-
Uhle, E. S.; Batagin-Neto, A.; Lavarda, F. C.; Graeff, C. F. O. J. Appl.
Phys. 2011, 110, No. 073510. (h) Singh, A.; Chen, K.; Adelstein, S. J.;
Kassis, A. I. Radiat. Res. 2007, 168, 233.
(7) (a) Ottolenghi, M.; Stein, G. Radiat. Res. 1961, 14, 281.
(b) Werner, H. R.; Firestone, R. F. J. Phys. Chem. 1965, 69, 840.
(8) (a) Kondo, M. M.; Jardim, W. F. Water Res. 1991, 25, 823.
(b) Kormann, C.; Bahnemann, D. W.; Hoffmann, M. R. Environ. Sci.
Technol. 1991, 25, 494. (c) Shirayama, H.; Tohezo, Y.; Taguchi, S.
Water Res. 2001, 35, 1941.
(9) Kim, H. N.; Lee, M. H.; Kim, H. J.; Kim, J. S.; Yoon, J. Chem. Soc.
Rev. 2008, 37, 1465.
(10) (a) Tang, B.; Yu, F.; Li, P.; Tong, L.; Duan, X.; Xie, T.; Wang, X.
J. Am. Chem. Soc. 2009, 131, 3016. (b) Albertazzi, L.; Storti, B.;
Marchetti, L.; Beltram, F. J. Am. Chem. Soc. 2010, 132, 18158.
(c) Myochin, T.; Kiyose, K.; Hanaoka, K.; Kojima, H.; Terai, T.;
Nagano, T. J. Am. Chem. Soc. 2011, 133, 3401. (d) Wu, S.; Li, Z.; Han,
J.; Han, S. Chem. Commun. 2011, 47, 11276. (e) Li, Z.; Wu, S.; Han, J.;
Han, S. Analyst 2011, 136, 3698. (f) Shi, W.; Li, X.; Ma, H. Angew.
Chem., Int. Ed. 2012, 51, 6432. (g) Lee, M. H.; Han, J. H.; Lee, J. H.;
Park, N.; Kumar, R.; Kang, C.; Kim, J. S. Angew. Chem., Int. Ed. 2013,
52, 6206.
(11) (a) Ge, Z.; Hayakawa, T.; Ando, S.; Ueda, M.; Akiike, T.;
Miyamoto, H.; Kajita, T.; Kakimoto, M.-a. Chem. Mater. 2008, 20,
2532. (b) Takizawa, S.-y.; Montes, V. A.; Anzenbacher, P. Chem.
Mater. 2009, 21, 2452. (c) Gong, S.; Chen, Y.; Yang, C.; Zhong, C.;
Qin, J.; Ma, D. Adv. Mater. 2010, 22, 5370. (d) Hung, W.-Y.; Chi, L.-
C.; Chen, W.-J.; Chen, Y.-M.; Chou, S.-H.; Wong, K.-T. J. Mater.
Chem. 2010, 20, 10113. (e) Chen, Y.-M.; Hung, W.-Y.; You, H.-W.;
Chaskar, A.; Ting, H.-C.; Chen, H.-F.; Wong, K.-T.; Liu, Y.-H. J.
Mater. Chem. 2011, 21, 14971. (f) Gong, S.; Chen, Y.; Luo, J.; Yang,
C.; Zhong, C.; Qin, J.; Ma, D. Adv. Funct. Mater. 2011, 21, 1168.
(g) Kulhanek, J.; Bures, F. Beilstein J. Org. Chem. 2012, 8, 25.
(12) (a) Krebs, F. C.; Lindvold, L. R.; Jørgensen, M. Tetrahedron Lett.
2001, 42, 6753. (b) Xie, N.; Chen, Y. J. Photochem. Photobiol., A 2007,
189, 253.
(13) (a) Tamao, K.; Uchida, M.; Izumizawa, T.; Furukawa, K.;
Yamaguchi, S. J. Am. Chem. Soc. 1996, 118, 11974. (b) Kulkarni, A. P.;
Tonzola, C. J.; Babel, A.; Jenekhe, S. A. Chem. Mater. 2004, 16, 4556.
(c) Wang, Z.; Lu, P.; Chen, S.; Gao, Z.; Shen, F.; Zhang, W.; Xu, Y.;
Kwok, H. S.; Ma, Y. J. Mater. Chem. 2011, 21, 5451.
(14) Chou, H. H.; Chen, Y. H.; Hsu, H. P.; Chang, W. H.; Cheng, C.
H. Adv. Mater. 2012, 24, 5867.
(15) Walba, H.; Isensee, R. W. J. Org. Chem. 1961, 26, 2789.
(16) (a) Wise, D. L., Wnek, G. E., Trantolo, D. J., Cooper, T. M.,
Gresser, J. D. Photonic Polymer Systems: Fundamentals: Methods, and
Applications; CRC Press: Boca Raton, FL, 1998. (b) Turro, N. J.
Modern Molecular Photochemistry; University Science Books: Mill
5096
dx.doi.org/10.1021/ja500262n | J. Am. Chem. Soc. 2014, 136, 5090−5096