Highly Soluble p-Terphenyl and Fluorene Derivatives as Efficient Dopants in Plastic Scintillators for Sensitive Nuclear Material Detection

Article abstract of DOI:10.1002/chem.201700877

Plastic scintillators are commonly used as first-line detectors for special nuclear materials. Current state-of-the-art plastic scintillators based on poly(vinyltoluene) (PVT) matrices containing high loadings (>15.0 wt %) of 2,5-diphenyloxazole (PPO) off

Full text of DOI:10.1002/chem.201700877

A Journal of
Accepted Article
Title: Highly Soluble p-Terphenyl and Fluorene Derivatives as Efficient
Dopants in Plastic Scintillators for Sensitive Nuclear Material
Detection
Authors: Alan Sellinger, Henok A Yemam, Adam Mahl, Uwe Greife,
Jonathan Tinkham, and Joshua Koubek
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201700877
Link to VoR: http://dx.doi.org/10.1002/chem.201700877
Supported by

10.1002/chem.201700877
Chemistry - A European Journal
Highly Soluble p-Terphenyl and Fluorene Derivatives as Efficient Dopants in Plastic
Scintillators for Sensitive Nuclear Material Detection
Henok A.Yemam,1 Adam Mahl,2 Jonathan Tinkham,1 Joshua T. Koubek,1 Uwe Greife,2 Alan Sellinger 1,3*
1. Department of Chemistry Colorado School of Mines, Golden, CO 80401, USA
2. Department of Physics, Colorado School of Mines, Golden, CO 80401, USA
3. Materials Science Program, Colorado School of Mines, Golden, CO 80401, USA
*Corresponding author: aselli@mines.edu
Abstract
Plastic scintillators are commonly used as first-line detectors for special nuclear materials. Current state-
of-the-art plastic scintillators based on poly(vinyltoluene) (PVT) matrices containing high loadings (>15.0
wt%) of 2,5-diphenyloxazole (PPO) offer neutron signal discrimination in gamma radiation background
(termed pulse shape discrimination, PSD), however they suffer from poor mechanical properties. In this
work, a series of p-terphenyl and fluorene derivatives were synthesized and used as dopants in PVT based
plastic scintillators as possible alternatives to PPO to address the mechanical property issue and to study
the PSD mechanism. The derivatives were synthesized from low cost starting materials in high yields
using simple chemistry. The photophysical and thermal properties were investigated for their influence on
radiation sensitivity/detection performance, and mechanical stability. A direct correlation was found
between the melting point of the dopants and the subsequent mechanical properties of the PVT based
plastic scintillators. Select fluorene derivatives produced scintillator samples whose mechanical
properties exceeded those of the commercial PPO based scintillators while producing acceptable PSD
capabilities. The physical properties of the synthesized dopants were also investigated to examine their
effect on the samples. Planar derivatives of fluorene were found to be highly soluble in PVT matrices
with little to no aggregation induced effects.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Introduction
Special Nuclear Materials (SNM), such as 233/235U and 239Pu, are radioactive isotopes classified as specific
nuclear materials that could potentially be used in atomic bombs or related ‘dirty’ explosives (Title I of
the Atomic Energy Act of 1954). The potential danger of these isotopes in significant quantities makes
them strictly regulated and controlled substances throughout the world. Therefore, it would be beneficial
if the domestic and international borders, ports of entry, and airports around the world were equipped with
radiation detectors to deter possible illicit trafficking of SNM. The current sparsely deployed detectors
use a combination of thermalizing plastic scintillators and 3He gas proportional chambers for
identification of SNM.[1] The widespread application of these combined detector systems has become
3
prohibitive due to the increasing cost of He gas. As a result, new plastic scintillators are being
investigated as a first-line of detectors for ionizing radiation (i.e. gamma and neutron) due to their low-
cost, mass reproducibility, ease of handling and installation. Despite these qualities, the current base
formulations are not inherently capable of differentiating the signals of neutron and gamma radiation.
This property is critical as sources of gamma radiation are significantly more common when compared to
neutron radiation; the latter being a key indicator of SNM. Being able to differentiate the two signals will
help identify SNM from naturally occurring radioactive materials (NORM) or medical isotopes such as
230Th and 57Co.[2]
The working principle of plastic scintillators is the absorption of incoming radiation by an inexpensive
polymer matrix [e.g. poly(vinyltoluene) (PVT)] followed by the subsequent cascade of energy transfers to
a primary fluorescent dopant and an optional wavelength shifter. Then finally the emitted photons are
collected by a photodetector such as a photomultiplier tube (PMT) or photodiode.[3] A pulse is generated
that is proportional to the type and energy of the incident radiation. In recent years, research groups have
begun investigating and describing methods to enhance the sensitivity of plastic scintillators for
distinguishing fast neutron signals from background radiation. A simple, reproducible method has
emerged by introducing an increased amount of highly soluble, fluorescent primary dopant (over-doping)
into a standard plastic scintillator formulation to induce interactions that allow for pulse shape
discrimination (PSD) analysis.[4] PSD analysis makes use of the slight difference in decay signatures
between signals induced by different types of incident radiation. Gamma radiation, converted to electrons
in the plastic, inherently deposits less energy over the same distance relative to neutron radiation that
produces elastic scattering (also referred to as proton recoils).[5] As a result, neutrons scattering through
the plastic will result in higher atomic excitation density distributed in singlet and triplet excited states.[6]
The singlet state electrons decay promptly via fluorescence whereas the triplet state, that is populated at a
higher ratio through the protons, decays via phosphorescence: a quantum-mechanically forbidden
transition that results in different emission wavelength with a much longer decay signature. However,
another interaction can occur when excited triplet states are allowed to interact with each other. These
triplet states can combine into a higher triplet state, or annihilate (referred to as triplet-triplet annihilation
(TTA)).[7] The resulting singlet states from this process then decay via fluorescence, but on a longer time
scale (delayed fluorescence) due to the nature of the interaction. The signals of the collected wave forms
can be differentiated by how much of the collected light was generated by delayed fluorescence, thus
discriminating neutron from gamma signals.
At lower primary dopant concentrations (<5.0 wt%), the signals from neutron and gamma radiation
appear identical (due to the lack of any significant delayed fluorescence) and hence identifying the
incident radiation is not possible without employing more complex and expensive inorganic solid state
detectors (such as NaI, HPGe, CsI, BaF2).[8] However, when samples were produced that contained highly
soluble primary dopants that were admixed at >10.0 wt% concentration, the resulting spectra were found
to allow for meaningful PSD analysis. Currently, the accepted theory for PSD capabilities at higher
primary dopants concentrations is that the migration and annihilation of first excited triplet states
increases, that leads to increased delayed fluorescence (Scheme 1).[9] To our knowledge, only PPO, a
diphenylanthracene (DPA) derivative, and m-terphenyl have been reported in literature to show PSD
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
capabilities when doped at high concentrations in common polymer matrices.[10] Furthermore, PPO is the
only primary dopant capable of PSD used in a commercial plastic scintillator product.
Scheme 1: Collision of first triplet excited states and the products of their annihilation [T1= first triplet excited state,
S1=First singlet excited state, S0= Singlet ground state, T5= Higher Triplet excited state]
At higher concentrations of PPO, the mechanical properties of the final plastics soften, making it
challenging to process into final form/shape, as well as transportation and field deployment. There is no
definitive explanation as to why over-doping PPO leads to enhanced PSD capabilities compared to other
dopants such as m-terphenyl or the DPA derivative in plastic scintillators. The main reason for this is due
to the lack of highly soluble fluorescent dopants in order to systematically study the parameters that affect
the PSD efficiency. If efficient TTA is the driving mechanism, then over-doping other highly emissive
primary dopants should introduce similar PSD capabilities as well.
In this work, alternative primary dopants were synthesized that could be admixed into common plastic
matrices at high concentrations to: 1. find alternatives to PPO that lead to PSD with enhanced mechanical
properties; and 2. help identify the PSD mechanism. Therefore, we have prepared highly luminescent
alkylated p-terphenyl and fluorene derivatives with the intent of identifying the main parameters that
affect the onset and efficiency of PSD. To systematically study these parameters, we took two
approaches: First, the effects of attaching alkyl/solubilizing groups onto p-terphenyl were explored in
relation to the melting point, solubility limit in PVT, and PSD performance of the final plastic. Second,
fluorene derivatives with varied photoabsorption and photoemission were compared to PPO based on
their general scintillation characteristics in PVT and PSD performance.
Results and Discussion
a. Synthesis and physical characteristics of PTP derivatives
PTP has been used as a primary fluorescent dopant together with 1,4-bis(5-phenyloxazol-2-yl) benzene
(POPOP) as a wavelength shifter in basic plastic scintillator formulations by several groups including
commercial producers of scintillators, eg. Saint Gobain Crystals.[11] Its high fluorescence quantum yield
(93%), fast decay lifetime (1.05 ns) and high photospectral overlap with PVT and POPOP makes it an
efficient dopant.[12] However, due to its limited solubility in current plastic scintillator formulations, it
has never been investigated for PSD application in plastics outside of the early attempts by Birks.[13]
Birks had used 2.5 wt% p-terphenyl in polystyrene as a proof of concept to show how fast neutron signals
could be separated from gamma radiation, though exhibiting only weak PSD.[13] We here attempt to
increase the solubility of PTP to further study this promising family of fluorescent dopants.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Scheme 2: Synthesis of para-alkylated p-terphenyl derivatives
Scheme 3: Synthesis of meta-alkylated p-terphenyl derivatives
The solubility of PTP in PVT was increased by changing the position and length of alkyl substituents on
PTP. As expected, this minimally affected the absorption and emission properties (see Figures 2 & 3 and
SI 15-16), and yielded a more useful form of the dopant for possible PSD properties in plastic scintillators
via over-doping. Finkelstein and modified Miyaura borylation reactions were used to synthesize
precursors from inexpensive starting materials, whereas Suzuki coupling of the precursors afforded the
final products with good to excellent reaction yields (Schemes 2-3, Figure 1).[14] Synthetic procedures for
1
precursors of PTP derivatives, and their corresponding H and 13C NMR can be found in the
Supplementary Information. The choice of the alkylated substituents was made based on the price of
starting materials, as well as the ease of scale up and purification schemes.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Figure 1: PTP derivatives and their reaction yields. [ditBuPTP is not shown on Schemes 2/3 as a matter of
convenience.]
The relative photoluminescent quantum yields (PLQY) of these compounds are also very similar (>90%),
as shown in Table 1, and these deviations are within the systematic error of the relative quantum yield
measurement.[15] The thermal properties of PTP derivatives were studied using differential scanning
calorimetry (DSC) and results shown in Figures S65-S73. The Tm vary greatly from one another and as
expected, the m-alkylated PTP derivatives showed much lower values compared to their p-alkylated
counterparts. Interestingly, some of these derivatives appear to have multiple thermal transitions (Table
1), suggesting liquid crystalline (LC) phases (that has been previously reported for terphenyl
molecules)[16], but we found no correlation between the existence of the LC-phases and the position of the
solubilizing group. In addition, the existence and position of the LC-phase did not noticeably affect
scintillation efficiency of the dopants. Results for PPO is shown for comparison.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Table 1: p-terphenyl derivatives and their physical properties.
Primary dopants Tm
LC-phase
UV-max
(nm)
304
281
277
277
285
278
280
274
275
275
274
PL-max
(nm)
367
341
344
345
353
344
349
342
345
346
346
Relative PLQY
(oC)
(oC)
(%)
100
93
93
91
93
98
-
90
97
95
100
PPO
72
N/A
N/A
N/A
126-180
N/A
121-174
245-255
N/A
N/A
57-62
N/A
PTP
213
187
180
179
174
255
94
136
64
11
tBuPTP
nBuPTP
ehPTP
hepPTP
ditBuPTP
mtBuPTP
mtBu2PTP
mopPTP
mehPTP
Melting Point (Tm) was determined by TA Instruments Q2000 Differential Scanning Calorimetry; UV-VIS was run in spectral grade
cyclohexane with Beckman Coulter DU 800 Spectrometer; Photoluminescence (PL) was run with Fluorolog Spectrofluorimeter FL1057 excited
at 290 nm. PL Quantum Yield (PLQY) was run with PTP as the reference dye and excitation wavelength at 270 nm.[15]
Figure 3: Photoemission of m-alkylated PTP
Figure 2: Photoabsorption of p-alkylated PTP
a. Plastic scintillator sample preparation
An alumina column plug was used to remove the inhibitor from 4-vinyltoluene monomer (TCI America).
The radical initiator azobisisobutyronitrile (AIBN) (Aldrich) was recrystallized twice from methanol and
stored in a freezer. A stock solution of 0.01 wt% AIBN in 4-vinyltoluene was prepared and stored in a
freezer. Glass jars from VWR were used as molds to make 4.7 cm diameter by 1.1-1.3 cm thickness
samples (approximately 20.0 g). The sample compositions are X wt% primary dopant, 0.1 wt% POPOP
and (99.9-X) wt% 4-vinyltoluene unless otherwise noted. For samples above 10 wt% primary dopant, the
glass vial was silanized with dichlorodimethylsilane (Sigma-Aldrich) in accordance with a literature
procedure.[17] After the dopant solutions were prepared in 4-vinyltoluene, argon gas was gently bubbled in
the solutions for 20 minutes to displace oxygen. The glass jars were then capped and put in an argon filled
o
vacuum oven under slight vacuum (Fisher Scientific isotemp 280A) and left to cure for 4 days at 80 C
and 24hr at 90 oC. Once the samples were cured and slowly air cooled to room temperature, the glass jars
were broken and the plastic discs were collected (Figure 4 and Figure 11).
6
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
b. Solubility test and radiation characterization
The solubility limit of PTP in PVT was determined to be 2.0-3.0 wt% in agreement with literature.[3a] The
p-alkylated derivatives (Figure 1) solubility in PVT were determined to be between 5.0-7.0 wt% with the
exception of tBuPTP (Figure 1) and ditBuPTP (Figure 1). The higher melting points of PTP, tBuPTP
and ditBuPTP when compared to the rest of the p-alkylated derivatives correlate to the lower solubility
limit. ditBuPTP had the highest melting point and lowest solubility in PVT (<1.0 wt%). The relationship
between high melting points of the dopants and low solubility in PVT were directly correlated and
provided motivation to synthesize lower melting point PTP derivatives (to increase the solubility within
PVT), through the synthesis of m-alkylated PTP derivatives (Table 1, Scheme 3).
m-alkylated PTP derivatives had much higher solubility limits; up to 20.0 wt% in PVT. The increase in
solubility correlates with the lower melting points of the derivatives. The high solubility allowed for
testing of PSD capabilities via over-doping. Note, mehPTP (Figure 1) has a melting point lower than
room temperature and thus was not explored further for over-doping due to extreme softening (gel-like)
of the final plastic at concentrations of 10.0 wt%.
Figure 4: Plastic scintillators doped with the new terphenyl derivatives. [Each small square is 1cm x 1cm]
To test their radiation response, each sample was polished (Figure 4), coupled to a PMT (Hamamatsu
H2431-50) and then exposed to various radiation sources. First, a 137Cs gamma source was used to
measure the gamma response. The dominant photoelectric interaction in plastic scintillators is via the
Compton effect. Therefore, the collected spectra are analyzed for the Compton edge feature, the location
7
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
of which can be used to quantify the light yield (LY) and generate an energy calibration unique to each
sample. Light yield is a general property of plastic scintillators that describe their scintillation or energy
transfer efficiency. As in previous publications, light yields are presented by comparing with signals from
commercial scintillators such as Saint Gobain Crystals (BC-408) and Eljen Technology. In this study, the
light outputs of the samples (in Table 2) are compared to a similarly sized sample of BC 408 acquired
from Saint Gobain Crystals. The 137Cs response spectra of all the samples are shown in the Supplementary
Information.
In comparison of light yield, p-alkylated PTP derivatives are similar to PTP at 1.0 wt%. Note that the
systematic error of the light yield (LY) measurements of this experiment is estimated to be +5.0% (Table
2). The variation in LY outside of the experimental error is believed to be due to the morphological
changes within the plastic as the solubilizing chain of the derivatives increases, that in turn decreases the
π−π stacking of the derivatives.[18] This leads to a decrease in non-radiative energy transfer from the PVT
matrix to the primary dopant. Similar trends were observed with 1.0 wt% m-alkylated PTP derivatives,
with mopPTP (Figure 1) showing the lowest light yield as shown in Table 2. The steric hindrance affects
the π−π stacking of the derivatives and decreases the homogeneity of the solution. This lowers the
likelihood of Förster Resonance Energy Transfer (FRET) between PVT to dopants, resulting in lower
scintillation efficiency or light output.
Next, a 244Cm/13C source was used to generate a mixed neutron and gamma radiation field. The individual
samples response was energy calibrated using the 137Cs spectrum, and then analyzed for any indication of
n/γ PSD. The PSD analysis was performed using the charge integration method. By integrating the
digitized waveforms of the response for the total charge in each given pulse (total fluorescence), and
comparing it to the charge in the tail of that pulse (delayed fluorescence), a ratio is calculated (delayed
energy fraction) which is different due to the slight difference in decay signal between incident gammas
and neutrons. The efficiency of PSD is then quantified by a figure-of-merit (FoM) measuring the
separation of the two types of signals at a given energy. PSD performance is measured by a dimensionless
Figure-of-Merit (FoM) metric which is calculated as detailed in Equation 1. These values are obtained by
examining the signals within an energy range of the neutron and gamma lines. A y-projection is
performed and a double Gaussian fit is assigned to the peaks. The centroids and FWHM are extracted and
a FoM is obtained.
FoM = !"#$%&'(#!!"#$%&'()
!"#$%!!"#$%
Equation 1: FoM Calculation
p-alkylated PTP derivatives did not have high enough solubility in PVT (<7.0 wt%) to generate sufficient
TTA, resulting in a low FoM (Table 2). Thus, only the initial onset of PSD was demonstrated close to the
solubility limit. The drop in the light yield from 1.0 to 5.0 wt% is expected due to concentration
fluorescence quenching (Table 2). The results of PPO are shown as the last entry in Table 2 for
comparison.
Table 2: Radiation response of plastic scintillator with soluble derivatives of PTP.
Primary
Dopants
PTP
t-buPTP
n-buPTP
ehPTP
hepPTP
mtBuPTP
1.0 wt%
LY (%)
97.1
85.4
87.5
89.6
77.1
86.0
5.0 wt%
LY (%) FOM
10.0 wt%
15.0 wt%
20.0 wt%
LY (%)
FOM LY (%) FOM LY (%) FOM
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
71.0
69.0
65.0
55.0
94.0
0.58
0.67
0.52
0.56
0.79
-
-
-
-
90.0
0.81
89.0
0.72
8
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
mtBu2PTP
mopPTP
PPO
79.0
63.3
90.0
68.0
62.0
80.0
0.57
0.64
0.65
88.0
76.0
87.0
0.79
0.71
1.10
67.0
67.0
88.0
0.81
0.77
1.21
-
-
64.0
89.0
0.87
1.61
The increased solubility of the m-alkylated PTP derivatives allowed for more significant PSD analysis.
mtBuPTP showed promising PSD at 5.0 wt% with a FoM value higher than the corresponding PPO at
5%, but the FoM stopped increasing with higher concentrations, showing even a decrease at 15.0 wt%.
This was counterintuitive to our expectation as higher concentrations of primary dopants should lead to
efficient TTA and hence better PSD. However, the same trend was also observed with mtBu2PTP in
which there was not a significant increase in FoM as the concentration increased from 10.0 to 15.0 wt%.
Due to this trend, samples with higher concentrations were not pursued. The reason for the leveling PSD
performance with increasing dopant concentration could be attributed to morphological effects allowing
the energy transfer from triplet excited states to be quenched because of vibrational dissipation.[19] The
effect of morphology on triplet excited state migration and density within a non-conjugated system such
as PVT needs to be investigated further. In addition, the light yield of the dopants with respect to
concentration shows no discernable pattern, which is currently being investigated.
Figure 5 and 6 show the n/γ PSD analysis of 5.0 wt% and 20.0 wt% mopPTP respectively. The increase
in FoM as the amount of mopPTP increases is thought to be due to enhanced triplet state migration that
leads to higher concentration of TTA and subsequent delayed fluorescence.[4] The 5.0 wt% sample
showed onset of discrimination at higher energies where the discrimination in the 20.0 wt% sample
becomes better resolved at lower energies. PSD spectra for 10.0 wt% and 15.0 wt% mopPTP samples can
be found in the Supplementary Information.
The mechanical properties were evaluated by using a Shore-D durometer (GxPro model# 560-10D). For
this measurement, six equidistance points on the flat side of a sample were measured and averaged. Only
m-alkylated PTP samples were measured due to insufficient data points for the p-alkylated derivatives to
make a meaningful comparison. The graph (hardness as a function of % wt dopant concentration can be
found in Figure S64) shows hardness dropping off with increasing dopant concentration. This could be a
result of fluorescent dopants retarding polymerization that creates softer plastic scintillators.[20] mopPTP
has the most drastic drop off correlating with the lower melting point. Addition of cross-linkers such as
1,4-divinyl benzene to curb this problem was not attempted but will be investigated in future work.
9
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Figure 5: Onset of Pulse Shape Discrimination (PSD) of 5.0 wt% mopPTP and 0.1 wt% POPOP in PVT
Figure 6: Pulse Shape Discrimination (PSD) of 20 wt% mopPTP and 0.1 wt% POPOP in PVT
c. Fluorene derivatives
Fluorene derivatives have been frequently used in Organic Light Emitting Diode (OLED) research due to
their ease of chemical manipulation, efficient and predictable photophysical properties, thermal stability,
10
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
efficient light output, and relatively low cost.[21] We tested five fluorene derivatives that could potentially
be used as emissive dopants to better understand the scintillation process as related to PPO. Dimethyl
fluorene (Me2F) and PPO were directly purchased from Sigma-Aldrich. Me2F was tested both as
purchased and then after further purification. 9,9-dimethyl-2-phenylfluorene (PhF), 9,9-dimethyl-2,7-
diphenylfluorene (PFP), 9,9-dimethyl-2-styrylfluorene (SF) and 9,9-dimethyl-2,7-di((E)-styryl)fluorene
(SFS) were synthesized from the respective 2-bromo-9,9-dimethylfluorene and 2,7-dibromo-9,9-
1
dimethylfluorene precursors via Suzuki and Heck coupling. The synthesis procedure, as well as H and
13C NMR spectra can be found in the Supplementary Information. The fluorene derivatives (Figure 7)
were designed to evaluate the effect of both blue- and red-shifting of photo-absorption and emission
spectra as compared to PPO as shown in Figure 8 and 9. Scintillation samples for this study were
prepared and tested in the same manner as the PTP derivative based samples.
Figure 7: Fluorene derivatives for use as dopants in plastic scintillators in this study.
Figure 8: Absorption of fluorene derivatives.
Figure 9: Emission of fluorene derivatives.
Melting points and optical properties of the fluorene derivatives and PPO are outlined in Table 3. The
range in values of these properties will be key in determining the parameter(s) that affects their
scintillation efficiency in PVT. PPO is used as a control to compare FoM values for PSD capabilities and
thermomechanical properties of the final plastic scintillator. BC-408 from Saint Gobain Crystals was
again used for light yield comparison.
11
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Table 3: Photophysical properties and melting points of fluorene derivatives and PPO.
Primary dopants Tm
UV-max
(nm)
304
264
289
342
338
371
PL-max
(nm)
367
304, 317
330, 346
390
387
411, 436
PLQY
(%)
100
45
64
89
(oC)
PPO
Me2F
PhF
PFP
SF
72
96
86
175
113
125
72
100
SFS
p-terphenyl was used as a reference dye for PLQY measurement of Me2F and PhF, excitation wavelength at 270 nm whereas
PPO was used as a reference dye for PFP and SF PLQY measurement, excitation wavelength at 320nm. 9,10-diphenylanthracene
was used as a reference dye for SFS, excitation wavelength at 350nm. Cyclohexane was used as a solvent for all UV/VIS and PL
measurements. Melting point was measured using the same DSC procedure as used for the PTP.
The photoemission of Me2F (Figure 9) is blue-shifted from PPO by 60 nm so the overlap with the
absorption of POPOP is decreased relative to PPO, and the expected light output should decrease as well
due to less energy transfer. As predicted, 1.0 wt% Me2F shows lower light output compared to the same
PPO sample (Table 4). The purified 1.0 wt% Me2F showed a drastic increase in light output, nearly 1.5-
fold compared to as received Me2F. Therefore, the purity of materials (>99% by GC) is a key parameter
for the optimization of energy transfer (light output) as impurities may quench fluorescence. The
photoemission of PhF is also blue-shifted from PPO, but only by 22 nm and a 1.0 wt% PhF showed
similar light output compared to PPO. Similar light output was observed for 1.0 wt% PFP as well. The
photoemission from SF is red-shifted from PPO by 20 nm and a 1.0 wt% SF sample has comparable light
output to a corresponding PPO control. This indicates that the optimization of the overlap between
primary dopant emission and secondary absorption is not a dominant factor for efficient scintillation in
agreement with literature.[22] SFS absorbs and emits within the same region as POPOP so as expected,
showed a high light output as a stand-alone dopant (Table 4). We have used SFS as a wavelength shifter
and showed equivalent performance as DPA and POPOP, two of the most commonly used wavelength
shifters. Currently, there is no correlation observed between light output and measured PLQY of the
fluorene based dopants.
Table 4: Radiation responses of fluorene derivatives plastic scintillators and PPO.
Primary
1.0
5.0 wt%
10.0 wt%
15.0 wt%
20.0 wt%
25.0 wt%
Dopants wt%
LY
LY
FOM
LY
FOM
LY
FOM
LY
FOM
LY
FOM
(%)
(%)
52.0
85.0
83.0
89.0
76.0
-
(%)
37.0
70.0
90.0
-
61.0
-
87.0
(%)
42.0
64.0
66.0
-
63.0
-
88.0
(%)
43.0
61.0
70.0
-
55.0
-
89.0
(%)
-
-
-
-
64.0
-
91.0
Me2F
Me2F
PhF
53.0
77.0
96.0
92.0
88.0
82.0
90.0
0.41
0.53
0.62
0.69
0.43
-
0.49
-
0.97
-
0.63
-
1.10
-
-
-
-
-
-
-
-
0.94
-
0.72
-
1.16
-
0.73
-
PFP
SF
0.78
-
1.73
SFS†
PPO
80.0
0.65
1.21
1.61
* as received Me2F was purified by passing through a silica column with 10% activated carbon. † Sample was made without a
POPOP wavelength shifter.
Three out of the five tested fluorene dopants can be overdoped in PVT to a significant degree (Table 4).
Despite increased purity, the PSD for all over-doped Me2F was negligible (Table 4). It shows that not all
soluble fluorescent dopants could produce PSD when exposed to a mixed radiation environment. PhF also
has a high solubility limit in PVT at over 20.0 wt% with a comparable light output relative to
corresponding PPO samples. Purified crystals of PhF were crushed and used to make samples that
showed a 1.16 FoM for 20% PhF, to our knowledge the highest FoM reported for a non-PPO derivative
(Table 4, Figure 10).[23] Radiation response data for PhF based samples can be found in the
Supplementary Information. PFP precipitated out at 10%, and was thus not studied further for PSD. This
12
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
suggests that the inverse relationship between melting points of dopants and solubility in PVT generally
holds true for fluorene derivatives as well as PTP derivatives. SF showed >25.0 wt% solubility in PVT
with good light output and mechanical properties. However, the increase in FoM from 5.0 to 25.0 wt% SF
was unexpectedly minimal. All PSD data for SF can be found in the Supplementary Information. The
reason for the low PSD at high SF concentrations is proposed to be attributed to a small percentage of cis-
isomer present in the synthesized SF as shown in GC/MS data (Figure S14). The first excited triplet state
for cis-SF (1.93eV) was found to be lower than that of trans-SF (2.21eV) in our computational modeling
results (Supplementary Information). Hence, it could be acting as a trap site to the first excited triplet
states of trans-SF in agreement with what others have observed in a mixed crystal scintillator system.[24]
In addition to the negligible increase in FoM, samples with a higher concentration of SF developed a
yellow color during polymerization that is attributed to aggregation effects at higher concentrations
(Figure 11). We are exploring the synthesis of pure trans-SF to test this theory.
Figure 10: Pulse Shape Discrimination (PSD) of 20 wt% PhF and 0.1 wt% POPOP in PVT
Selected fluorene derivatives showed similar FoM values (PhF) as compared to the corresponding PPO
samples at 10 wt% loadings, and also displayed similar light output (SF) and stronger thermomechanical
material properties. Hardness measurements were done using the Shore-D durometer following the same
procedures as with the PTP derivatives. As shown in Figure 12, as the concentration of PPO is increased,
the hardness of the final plastic scintillator decreases dramatically. The same was true for purchased Me2F
and purified Me2F*. PhF and SF both displayed a significantly lower drop off in hardness as concentration
increased.
13
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Figure 11: Plastic scintillators doped with fluorene derivatives and PPO. [Each small square is 1cm x 1cm]
Generally, fluorene derivatives and PPO had higher solubility in PVT than PTP derivatives regardless of
melting point or photophysical properties. For instance, SF had a higher melting point than mtBuPTP but
it also had higher solubility in PVT (~25.0 vs ~20.0 wt%). The same is true for SF and PPO vs mopPTP.
To investigate this, the molecular geometry and angle measurements were modeled with Density
Functional Theory (DFT) based code using the PBE generalized gradient approximations and TZ2P basis
set in an all electron calculation. These calculations were made using the Slater type orbitals of
Amsterdam Density Functional 2016.104 (ADF2016).[25] The models (Figures S48-S63) showed that
dopants with the highest solubility (SF and PPO) were completely planar whereas the less soluble PTP
derivatives had one of the phenyl rings out of plane from the other two phenyl rings by 142.0°. Going
forward we believe that computation can play an important role in understanding how dopants work so
that we can predict and design more efficient dopants.
It’s important to note that both PTP and fluorene-based samples have shown no noticeable physical
change or degradation for over nine months. We would also like to emphasize that these samples were
made via bulk radical polymerization that has its drawbacks with regards to incomplete termination of
propagating radicals and long reaction time.[26]
14
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
90.0
85.0
80.0
75.0
70.0
65.0
60.0
55.0
50.0
PhF
SF
PPO
Me2F
Me2F*
0.0
5.0
10.0
15.0
20.0
25.0
30.0
%wt primary dopant
Figure 12: Hardness of final plastic scintillators as a function of dopant concentration.
Conclusion
The softening of plastics when over-doped with high concentrations of primary dopants is a critical
problem in the development and commercialization of new, next generation plastic scintillators with PSD
properties. It appears that the melting point of primary dopants is a predictive factor that affects the
solubility limit in PVT and the mechanical strength of the resultant plastic. In addition, planar dopants
such as SF and PPO had higher solubility in PVT than the non-planar PTP derivatives. Select m-alkylated
PTP and fluorene derivatives when doped into PVT resulted in encouraging PSD capabilities with
enhanced plastic mechanical properties compared to the analogous samples using PPO as dopants.
We were successful in preparing derivatives with enhanced solubility, and were able to show that over-
doping of fluorescent material does not lead to a significant PSD in all cases. While we successfully
achieved comparable PSD with mopPTP, the similar mtBuPTP showed decreased PSD, implying triplet
exciton migration has a preferred morphological pathway to achieve efficient TTA as a plausible
mechanism for PSD in over-doped samples. We also prepared fluorene derivatives with varied optical
properties and did not find a direct correlation between spectral overlap of dopants, and final scintillating
light output. Over-doped samples also showed PSD, but not to the same degree as PPO. PhF at 20 wt%
showed the highest reported FoM (1.16) for a non-PPO derivative. Furthermore, the hardness of this
sample compared to the corresponding 20 wt% PPO sample was significantly harder (Shore-D of 81 vs
69) resulting in a final plastic that could be easily handled and polished. It is also important to investigate
the effect of isomers on PSD performance as shown by the proposed detrimental effect of cis-SF on trans-
SF.
The exact physics responsible for efficient PSD is still elusive, however we believe that planar molecules
heavily influence how triplet states migrate, collide and annihilate each other. Our future work will focus
on these investigations using computational modeling of excited states as well as the effect of plastic
mechanical properties on PSD performance.
15
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
Acknowledgment
This research was funded through US DHS DNDO grants # DHS-14-DN-077-AR-NC7 and DHS-16-
DNDO-077-001. The authors would like to thank Captain Brett McNichols for his help in the analysis of
proton NMR spectra.
References
[1] G. H. Bertrand, M. Hamel, F. Sguerra, Chem. Eur. J. 2014, 20, 15660-15685.
[2] R. Kouzes, J. Ely, B. Geelhood, R. Hansen, E. Lepel, J. Schweppe, L. Siciliano, R. Warner, Nuc. Sci Symp. Conf.
Rec., IEEE 2003, 1448-1452.
[3] a) J. B. Birks, The Theory and Practice of Scintillation Counting: International Series of Monographs in
Electronics and Instrumentation, Elsevier, 2013, p; b) F. Brooks, Nuc. Inst. Meth. 1979, 162, 477-505.
[4] N. Zaitseva, B. L. Rupert, I. Pawelczak, A. Glenn, H. P. Martinez, L. Carman, M. Faust, N. Cherepy, S. Payne,
Nuc. Instr. Meth. A 2012, 668, 88-93.
[5] G. J. Hine, G. L. Brownell, Rad. Dos., Elsevier, 2013, p.
[6] G. F. Knoll, Radiation Detection and Measurement, John Wiley & Sons, 2010.
[7] V. Jankus, M. Aydemir, F. B. Dias, A. P. Monkman, Adv. Sci. 2015.
[8] A. Perez-Andujar, L. Pibida, Appl. Radiat. Isot. 2004, 60, 41-47.
[9] T. N. Singh-Rachford, F. N. Castellano, Coord. Chem. Rev. 2010, 254, 2560-2573.
[10] a) T. J. Hajagos, D. Kishpaugh, Q. Pei, Nuc. Instr. Meth. A 2016, 825, 40-50; b) H. P. Martinez, I. Pawelczak,
A. M. Glenn, M. L. Carman, N. Zaitseva, S. Payne, Nuc. Instr. Meth. A 2015, 771, 28-31.
[11] a) J. Iwanowska, L. Swiderski, M. Moszynski, T. Szczesniak, P. Sibczynski, N. Z. Galunov, N. L. Karavaeva,
J. Instr. 2011, 6, P07007; b) D. McKinsey, C. Brome, J. Butterworth, R. Golub, K. Habicht, P. Huffman, S.
Lamoreaux, C. Mattoni, J. Doyle, Nuc. Instr. Meth. B 1997, 132, 351-358.
[12] a) I. Berlman, Handbook of Florescence Spectra of Aromatic Molecules, Elsevier, 2012, p; b) R. L. Joseph and
R. Lakowicz in Principles of Fluorescence Spectroscopy, Vol. Kluwer Academic/Plenum Publishers, New York,
1999.
[13] J. B. Birks, Scintillation Counters, Pergamon Press, 1960.
16
This article is protected by copyright. All rights reserved.

10.1002/chem.201700877
Chemistry - A European Journal
[14] a) I. O. Huyal, U. Koldemir, T. Ozel, H. V. Demir, D. Tuncel, J. Mater. Chem. 2008, 18, 3568-3574; b) Y.-E.
Jin, J.-W. Kim, S.-H. Park, H.-J. Kim, K.-H. Lee, H.-S. Suh, Bull. Korean Chem. Soc. 2005, 26, 1807-1818; c) H. A.
Yemam, A. Mahl, U. Koldemir, T. Remedes, S. Parkin, U. Greife, A. Sellinger, Sci. Rep. 2015, 5.
[15] S. Fery-Forgues, D. Lavabre, J. Chem. Educ 1999, 76, 1260.
[16] a) Z. Yu, H. Tu, X. Wan, X. Chen, Q. Zhou, Mol. Cryst. Liq. Cryst. 2003, 391, 41-55; b) J. Andersch and C.
Tschierske, Liq. Cryst. 1996, 21, 51-63.
[17] M. C. Pirrung, The Synthetic Organic Chemist's Companion, John Wiley & Sons, 2007, p.
[18] S. Ghosh, X. Q. Li, V. Stepanenko, F. Würthner, Chem. Euro J. 2008, 14, 11343-11357.
[19] M. S. Kwon, D. Lee, S. Seo, J. Jung, J. Kim, Angew. Chem. Int. Ed. 2014, 53, 11177-11181.
[20] a) L. Ellinger, J. Appl. Polym. Sci. 1965, 9, 3939-3948; b) J. M. Watson in Polymerization Inhibitor for
Vinyltoluene, Vol. Google Patents, 1982; c) R. Neilson, I. Soutar, W. Steedman, J. Poly. Sci. Poly. Chem. 1976, 14,
1005-1013.
[21] a) Q. Fang, B. Xu, B. Jiang, H. Fu, W. Zhu, X. Jiang, Z. Zhang, Synth. Met. 2005, 155, 206-210; b) S. Tao, Z.
Peng, X. Zhang, P. Wang, C. S. Lee, S. T. Lee, Adv. Funct. Mater. 2005, 15, 1716-1721.
[22] C. Pietsch, J. Schäfer, R. Menzel, R. Beckert, J. Popp, B. Dietzek, U. S. Schubert, J. Polym. Sci., Part A:
Polym. Chem. 2013, 51, 4765-4773.
[23] T. Förster, Rad. Res. Suppl. 1960, 326-339.
[24] N. Zaitseva, A. Glenn, L. Carman, R. Hatarik, S. Hamel, M. Faust, B. Schabes, N. Cherepy, S. Payne, Nuc. Sci.
IEEE Trans. 2011, 58, 3411-3420.
[25] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
[26] Ł. Kapłon, A. Kochanowski, M. Molenda, P. Moskal, A. Wieczorek, T. Bednarski, P. Białas, E. Czerwiński, G.
Korcyl, J. Kowal, Bio. Alg. Med. Syst. 2014, 10, 27-31.
17
This article is protected by copyright. All rights reserved.