The Synthesis and Photophysical Analysis of a Series of 4-Nitrobenzochalcogenadiazoles for Super-Resolution Microscopy

Article abstract of DOI:10.1002/chem.201702289

A series of 4-nitrobenzodiazoles with atomic substitution through the chalcogen group were synthesised and their photophysical properties analysed with a view for use in single-molecule localisation microscopy. Sub-diffraction resolution imaging was achieved for silica nanoparticles coated with each dye. Those containing larger atoms were favoured for super-resolution microscopy due to a reduced blink rate (required for stochastic events to be localised). The sulfur-containing molecule was deemed most amenable for widespread use due to the ease of synthetic manipulation compared to the selenium-containing derivative.

Full text of DOI:10.1002/chem.201702289

A Journal of
Accepted Article
Title: The synthesis and photophysical analysis of a series of 4-
nitrobenzochalcogenadiazoles for super-resolution microscopy
Authors: Daniel Ray Jenkinson, Ashley Cadby, and Simon Jones
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.201702289
Link to VoR: http://dx.doi.org/10.1002/chem.201702289
Supported by

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
The synthesis and photophysical analysis of a series of 4-
nitrobenzochalcogenadiazoles for super-resolution microscopy
D. R. Jenkinson,[a] A. J. Cadby,*[b] and S. Jones*[a]
Abstract: A series of 4-nitrobenzodiazoles with atomic substitution
through the chalcogen group were synthesised and their
photophysical properties analysed with a view for use in single-
molecule localisation microscopy. Sub-diffraction resolution imaging
was achieved for silica nanoparticles coated with each dye. Those
containing larger atoms were favoured for super-resolution
microscopy due to a reduced blink rate (required for stochastic
events to be localised). The sulfur containing molecule was deemed
most amenable for widespread use due to the ease of synthetic
manipulation compared to the selenium containing derivative.
technique in modern biology, it has become imperative to
investigate the structure-function relationship for the
photoblinking activity of dyes.
Atomic substitution down the chalcogen group has previously
been used as a method of tailoring the fluorescence properties
of oxygen containing dyes to induce a bathochromic shift in the
absorption and emission spectra.[11–14] Though red-shifting is
achieved, quantum yield and extinction coefficient are vastly
reduced owing to more stable triplet states.[15] Such atomic
substitution also leads to atoms with larger and more diffuse
orbitals, which has interesting effects on the aromatic nature of
these species.[16] Substitution of oxygen for sulfur and selenium
in 2,1,3-chalcogenodiazoles has been shown theoretically and
experimentally to increase the degree of aromaticity in the
heteroaromatic systems.[17–19] Sulfur containing compounds
appear to be almost twice as aromatic using Bird’s unified
aromaticity index (IA) than the oxygen containing analogues due
to a vastly decreased electronegativity. This effect is, however,
negated in 2,1,3-selenadiazole since its much larger atomic size
and more diffuse orbitals diminish the ability to delocalize
electrons, meaning that the selenium containing compound has
a much lower aromaticity than the sulfur containing compound,
but is still slightly higher than the oxygenated derivative (IA = 53,
104, and 58 for X = O, S, and Se, respectively).[19] By making
valence isoelectronic substitutions down a period it was hoped
to have an influence on the strength of the aromaticity, which
could be measured by observing differences in key
photophysical properties. Thus, in order to investigate atomic
substitution was the sole cause of any photophysical difference,
4-nitrobenoxadiazole (NBD) frameworks were investigated
where the planar structure should negate any flexibility in the
molecule that may arise from torsion or twisting that may also
Introduction
Fluorescence microscopy is commonly used in biology for non-
invasive imaging as the biochemical specificity of fluorescent
probes is often easily tuned.[1–3] Fluorophores add contrast to
images and have been used to visualise biological systems at a
microscopic level, however optical resolution is limited by
diffraction.[4] The development of single-molecule localisation
microscopy (SMLM) techniques such as STORM,[5] PALM,[6] and
FPALM[7] has enabled the analysis of biological structures and
processes with lateral resolutions on the nanometre scale.
These techniques exploit triggered emission switching
(photoblinking), an effect thought to arise as the result of a
stable dark state. This allows only a small subset of fluorophores
to exist in an emissive “on-state,” and each can be localised to
reconstruct a high-resolution image.[5] For example, two dark
states have been identified for cyanine 5 (Cy 5), of which one is
a
triplet state accessed via inter-system crossing and is
accessed considerably quicker in the presence of iodide due to
heavy-atom-induced spin-orbit coupling.[8] The other was
recognised as a trans-cis isomerisation to a non-fluorescent
isomer. Cy 5 can be reversibly switched between its on- and off-
states with a high degree of control.[9] The induced dark state
has a lifetime in the order of hours in the absence of oxygen that
suggested the existence of another energy level accessed via
the triplet state since this lifetime is much longer than other
reported triplet states compared in this work (on the order of 100
milliseconds). A similar observation was also made in a separate
study using rhodamine dyes.[10] Given the importance of this
disturb aromaticity. This moiety is commonly used as
a
fluorescent probe due to its small molecular mass, high cell
permeability and large stokes shift but has not yet been used in
super-resolution techniques.[20]
Results and Discussion
Molecular modelling was first conducted to examine the
structural framework of the target systems. This confirmed the
hypothesis that when oxygen is substituted for sulfur and
selenium, the aromatic system remains planar but the ring in
which the chalcogen is contained distorts within the plane in
order to accommodate the larger atom (Table 1, see
supplementary information for geometrically optimised
[a]
[b]
Mr. D. R. Jenkinson, Prof. S. Jones*
Department of Chemistry
University of Sheffield
Dainton Building, Brook Hill, Sheffield, S3 7HF, United Kingdom
E-mail: simon.jones@sheffield.ac.uk
Dr. A. J. Cadby*
structures).[21]
On
this
basis,
a
series
of
4-
nitrobenzochalcogenadiazoles were targeted that included a
group capable of being ligated to a nanoparticle for later
visualization. This was planned to be introduced via SNAr
reaction of (3-aminopropyl)triethoxysilane (APTES) with the
corresponding aryl-halide to give a siloxane terminus leading to
compounds 1 (NBD-APTES), 2 (NPT-APTES) and 3 (NPSe-
Department of Physics and Astronomy
University of Sheffield
Hicks Building, Hounsfield Rd., Sheffield, S3 7RH, United Kingdom
E-mail: a.cadby@sheffield.ac.uk
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
NO2
NO2
NO2
APTES) in preparation for ligation to silica nanoparticles which
are optically inert and easy to functionalise.
1
3
H NNMe I
2
3
tPentONa
eq.
eq.
NH2
NH2
N
N
SeO2
Se
DMSO
rt,
H2O/EtOH
reflux, 16 h
NH2
16
h
Table 1. Results of energy calculations and geometry optimisation. [a]
Br
Br
Br
30%
85%
5
6
LUMO–HOMO
difference (eV)
N-X-N bond
angle (°)
N-X bond
X
λmax (nm)
length (Å) [b]
NO2
NO2
1.2
APTES
Ph2P
eq.
2 mol% Pd2dba3
6 mol% S
N
N
N
X
( )-(+)-PPFA
O
3.2014
3.2509
3.0403
386
380
406
113.0
99.4
93.5
1.365
1.638
1.801
Fe
X
Me2N
1.5
tBuOK
eq.
N
S
THF
75 o
C
(0.25 M),
S
( )-(+)-PPFA
Y
H
15 min, microwave
N(CH2)3Si(OEt)3
Se
=
=
=
=
=
=
=
=
=
1
X
2 X
3 X Se
O
S
X
X
X
Y
Y
Y
48%
Br, 45%
Br, 7%
Cl,
O,
S,
Se,
[a] Calculated using Gaussian 09.[22] [b] Taken as an average of the two N-X
bonds.
Scheme 2. Preparation of precursor 6 and aromatic amination reactions to
prepare targets 1-3.
NBD-APTES 1 was prepared in one step from the commercially
available 4-chloro-7-nitrobenzo-2,1,3-oxadiazole (NBD-Cl) and
APTES as previously reported in the literature (Scheme 1).[23]
The chemical shifts of the aromatic protons for each of the three
compounds support an increased degree of aromaticity for those
with heavier atoms.[26] Peaks attributed to the aromatic protons
were shifted most downfield in compound 2, with the protons in
compound 1 being the most upfield, suggesting a lower degree
NO2
NO2
N
N
N
N
1.1
EtOH, 0 o
APTES
- rt
eq.
O
S
O
of deshielding and hence a lower degree of aromaticity (δArH
=
C
24 dark
h,
8.48 and 6.17, 8.67 and 6.41, 8.62 and 6.17 for 1, 2 and 3
respectively). The solution phase absorption and emission
spectra for NBD-APTES 1 and NPT-APTES 2 were found to be
comparable, but NPSe-APTES 3 was highly red-shifted (Figure
1). This effect has been observed with selenium analogues of D-
luciferin and attributed to the polar effect.[12,27] It is noted that,
although the measured absorption maxima are not exactly in line
with the predicted values, the energy calculations did predict that
the maxima for 1 and 2 would be close to each other, and that
λmax for compound 3 would be red shifted. Compound 2
absorbing at a lower wavelength than compound 1 was also
predicted.
Cl
Br
H
N(CH2)3Si(OEt)3
48%
1
NO2
Br
NO2
1.1
0.1
APTES
K CO
eq.
eq.
EtOH, 0 o
N
N
N
N
N
69% HNO3
2
3
S
S
h
reflux, 2
- rt
C
N
24 dark
h,
Br
4
H
N(CH2)3Si(OEt)3
40%
22%
2
Scheme 1. Synthesis of dyes 1 and 2.
NPT-APTES
2 was synthesised in two steps from 4,7-
dibromobenzo-2,1,3-thiadiazole and required the addition of a
base to improve the yield of the SNAr reaction (Scheme 1).
However, the selenium containing target 3 was more challenging
since the 1,4-nitrohalide 6 was not commercially available. This
was accessed by adapting a reported procedure via vicarious
nucleophilic substitution and selenium insertion (Scheme 2).[24]
However, a series of SNAr conditions failed to produce the
desired target. NPSe-APTES 3 was instead accessed via a
palladium catalysed aromatic amination,[25] which was
successful for the halo derivatives of all three heteroaromatic
systems when using THF as the solvent. Although the yield of
this reaction was poor, sufficient material was obtained for the
necessary measurements.
Figure 1. Normalised absorption (solid lines) and emission (dashed lines) for
compounds 1, 2, and 3.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
A decrease in molar extinction coefficient was observed when
heavier atoms were incorporated, in broad agreement with other
dyes with substitution patterns down the chalcogen group.[11–14]
(Table 2). The molecules containing heavier atoms have vastly
decreased extinction coefficients. According to Nijegorodov and
Mabbs, the introduction of a heavier atom may decrease the
oscillator strength for the S0 → S1 transition due to changes in
the parameters of matrix elements in the molecular orbital,
therefore making absorption less likely.[28,29]
in aqueous solution this too is unlikely. It is possible that under
laser irradiation the compound undergoes a photocatalysed
transformation to
Although there was no change in the observed emission
maximum, this should not be ruled out. Increases in
fluorescence intensity have been induced using pulsed lasers by
Hell and coworkers,[33] but in this case the laser irradiation was
continuous. A further study is required to determine the precise
cause of this phenomenon.
a
more intensely fluorescent species.[32]
Table 2. Measured photophysical properties of compounds 1, 2, and 3.
[a]
λmax
Dye
ε (M-1 cm-1) [b]
Φ (%) [c]
τ (ns)
Abs
465
460
500
Em
542
550
615
1
2
3
24,000
1,400
3,900
0.8
0.2
0.2
2.6
1.1
1.7
[a] Absorption and emission were measured using 100 µM solutions. [b] Molar
extinction coefficients were extrapolated from calibration curves plotted using
solutions with concentrations between 10-100 µM and are correct to two
significant figures. [c] Quantum yield measurements taken using an integration
sphere and compared to a highly fluorescent dye with a known quantum yield.
[d] Excited state lifetime measurements were taken using a time-correlated
single photon counting method for time-resolved photoluminescence.
Similarly, there was a drop in the quantum yield when larger
atoms were incorporated due to a higher probability of transfer
to the triplet state via intersystem crossing (ISC), though the
method used to measure the excited state lifetime was not
accurate enough to allow more precise quantities to be taken.
Though there was little variation between the three molecules it
was clear that oxygen containing NBD-APTES 1 had the longest
excited state lifetime, indicating that fluorescence may be
favoured rather than undergoing ISC to the dark state. The rate
of photobleaching was measured using a method described by
Sauer et al.[30] in which solutions of dyes are bleached using a
laser and reactivated by reincorporation of oxygen by agitation in
the presence of different buffer solutions. When dissolved in
distilled water there was negligible decay in fluorescence
intensity of the sulfur and selenium containing dyes, even when
using high laser powers. Interestingly, it was observed that NBD-
APTES 1 showed an increase in fluorescence intensity to a
plateau with continued laser irradiation over 40 minutes at all
concentrations and laser intensities used (Figure 2). This may
be the result of an intermolecular FRET like mechanism arising
from the slight overlap in the absorption and emission spectra
between closely associated molecules. However, since
compounds 2 and 3 do not exhibit this behavior but also have
overlapping absorption and emission spectra this is less
probable. Similar observations have been made with other
fluorescent molecules whose fluorescence intensity increases
on binding to a substrate,[20,31] but since only the dye is present
Figure 2. Bleaching and reactivation graphs for 1 (A), 2 (B), and 3 (C).
The introduction of additives (MEA and GLOX) triggered
photobleaching and the average rate of recovery dramatically
increased under GLOX buffered conditions (Table 3) confirming
GLOX to be a superior buffer system for reversible photoblinking
as required for STORM imaging.[34]
Table 3. Results of reversible bleaching experiments.
Decay rate
(× 10-3 s-1)
Avg. recovery
(%)
Avg. PEL
(%) [c]
Dye [a]
Buffer [b]
A
B
A
0.265
1.21
12
78
12
-
1
2
15
-
0.555
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
B
A
B
0.439
2.06
10.0
48
15
81
15
-
3
24
[a] All dye solutions were diluted to 1 × 10-5 M. Recorded using OceanView
and analysed using Matlab.[35] [b] A: Dye solution with MEA (0.1 M); B: Dye
solution with MEA (0.1 M) and GLOX buffer. [c] Calculated as a percentage of
fluorescence intensity decay before the rate of decay slowed and became
exponential.
It was noted for each dye that when using GLOX buffer there
was a sharp drop in fluorescence intensity prior to exponential
decay (pre-exponential loss, PEL).
After solution phase analysis was complete, the blinking activity
and hence suitability for super resolution microscopy was
assessed by adopting a method recently described by Gibbs, et
al. by immobilisation of the dyes in a polyvinyl alcohol (PVA)
matrix.[36] The properties analysed for each dye were the
average number of blinking events per µm2 area, the average
length of continuous emission (on time), and the number of
times blinking molecules switched between their ‘on’ and ‘off’
states (number of cycles, Table 4). On time and number of
cycles were determined by first plotting z-axis profiles of pixel
intensity for individual blinking molecules (Figure 3).
Figure 4. Representative on/off trajectories for individual photoblinking
molecules of 1 (top), 2 (centre), and 3 (bottom). Frame rate = 20 Hz.
Dyes 1–3 displayed no trend in blink density between buffer
systems. NBD-APTES 1 displayed a higher blinking density in
the presence of MEA and without GLOX buffer, and this result
was reversed in molecules containing heavier atoms.
Table 4. Results of blink analysis for each dye for compounds 1, 2, and 3.
Avg. blink
Avg. on
Avg. Number of
cycles [d]
Dye [a]
Buffer [b]
density [µm-1] [c]
time [s] [d]
A
B
A
B
A
B
1.40
1.16
1.30
2.00
1.01
1.46
0.19
0.39
0.22
0.33
0.11
0.34
2.9
4.9
1.4
3.0
1.5
2.9
1
2
3
Figure 3. Representative z-axis profile for an individual photoblinking molecule
of 1. Frame rate = 20 Hz.
[a] All dye solutions were diluted to 3 × 10-8 M. Andor Solis was used with
Fiji,[37,38] and the ThunderSTORM plugin to record and analyse blinking
data.[39] [b] A: Dye solution with MEA (0.1 M); B: Dye solution with MEA
(0.1 M) and GLOX buffer. [c] Calculated as an average of the number of
blinking events captured within a specified area over the duration of the
experiment. [d] Calculated by plotting z-axis profiles of pixel intensity for
individual molecules.
The z-axis profiles can in turn be simplified to so-called ‘on/off
trajectories’ in which any signal above a specified threshold is
denoted ‘on’ giving a binary trace. In this case, the threshold
was set to detection of pixel intensity of 5, and this was used to
determine the on time and number of cycles for each molecule.
Representative on/off trajectories are provided for each
compound (Figure 4).
For 2, this is believed to be because sulfur atoms enable the
stabilisation of electronic ‘dark’ states,[40] however, although this
area has not been investigated with organoselenides, it is
feasible that selenium may offer the same effect. In all three
cases, both the average ‘on’ time average number of cycles
were increased in the presence of GLOX buffer. NBD-APTES 1
was vastly more active under both sets of conditions than NPT
and NPSe-APTES (2 and 3 respectively), which showed similar
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
results for both variables. These results correlate with the
excited state lifetimes observed. The enhanced blink rate may
also explain why the oxygen containing NBD moiety has not
successfully been used for SMLM imaging. If the oxygen
analogue 1 is used to densely label a sample (as is required for
SMLM imaging), too many fluorophores in the sample blink
simultaneously and localisation cannot occur. However, it may
have better compatibility with other techniques such as super-
resolution radial fluctuation or 3B, an SMLM software package
capable of localising overlapping fluorophores.
distinguished from background noise and subsequent
localisation of individual molecules. Although the quantum yields
are low, this is not problematic for SMLM imaging, as many of
the commonly used dyes have wide ranging quantum yields.[42]
While under ideal SMLM conditions (with MEA and GLOX
buffer) 1 and 2 performed similarly, 3 appeared to emit far fewer
photons, and this may be due to the poor overlap of available
laser wavelengths with the absorption spectrum of 3. Molecules
stochastically blinked throughout the whole duration of these
experiments.
Since the lengths of the dark states for each dye are long-lasting
(several seconds) it is likely that the blinking mechanism is redox
based.[41] Such reversible redox reactions of dyes with MEA via
photoinduced electron transfer that result in non-fluorescent
radical ions are more likely to occur from the triplet state.[10,41]
The heavier atoms in the sulfur and selenium containing dyes
stabilize the triplet state, promoting transition to the dark state,
that in turn results in longer dark states and fewer observed
localisation events. Further analyses were undertaken to
establish the number of photons detected per blinking event and
the average on/off duty cycle (fraction of time spent in the ‘on’
state after equilibrium is reached) over the duration of the 150 s
experiment (Table 5).
Armed with the knowledge of how individual molecules in
sparsely populated samples behave, it was then prudent to see
if these results could be used to predict the suitability of a dye to
take images of small structures using SMLM techniques. Silica
nanoparticles were synthesised via Stӧber synthesis in which
the silane functionalised dyes were added at a later stage to
coat the surface (Scheme 3).
NH
rt, 24 h
H
EtOH
,
2O,
(i)
3
O
O Si
O
H
"
"SiO2
N
NO2
Si(OEt)4
(CH2)3
or
1 2
,
3,
(ii)
Si(OEt)4
rt, 5 h
N
N
X
Scheme 3. Synthesis of fluorescent nanoparticles. X = O (1), S (2), or Se (3).
Table 5. Further blink analysis for compounds 1, 2, and 3.
Avg. photons
Avg. on/off
duty cycle
Dye [a]
Buffer [b]
detected per blink [c]
As well as the nanoparticles being visibly coloured after
functionalisation, coating was confirmed using fluorimetry. The
average size and spherical shape of the particles were
confirmed using dynamic light scattering (see supplementary
information) and scanning electron microscopy (Figure 5). The
average diameter of nanoparticles coated with 1, 2 and 3 were
239.8, 223.3 and 256.4 nm, respectively.
A
B
A
B
A
B
182
889
514
875
164
468
0.0013
0.0027
0.0015
0.0023
0.00078
0.0023
1
2
3
[a] All dye solutions were diluted to 1 × 10-5 M. FITC filter cube
was used. Andor Solis was used with Fiji,[37,38] and the
ThunderSTORM plugin to record and analyse blinking data.[39] [b]
A: Dye solution with MEA (0.1 M); B: Dye solution with MEA (0.1
M) and GLOX buffer. [c] Calculated based on the analog-to-digital
ratio for the camera used (Andor Zyla sCMOS) being 0.28 as
stated in the manufacturers specifications.
The average on/off duty cycles (fraction of time in the on state)
for each of the dyes in this study are comparable and similar to
the top performing dyes currently used for STORM, as analysed
by Dempsey and coworkers.[42] The average numbers of
photons observed per blinking event is lower than, but of the
same magnitude, as other dyes that absorb in the blue region of
the visible spectrum.[42] The dyes also perform similarly in this
respect when compared to commonly used SMLM dyes
suspended in PVA matrices.[36] This study has shown that
sufficient photons can be detected for fluorescence to be
Figure 5. Scanning electron micrograph of silica nanoparticles coated with 1.
Scale bar 1 µm.
When viewed under an epifluorescence microscope, the
stochastic photoswitching process easily allowed for loose
molecules to be localised and ignored as noise whereas when a
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
large number of fluorophores were present on a nanoparticle it
was possible to detect them as separate entities (Figure 6).
the increase in fluorescence intensity with prolonged laser
exposure, makes derivatives of 1 unsuitable as a dye for
capturing SMLM images. The slower blink rate of NPT- and
NPSe-APTES (2 and 3) make them more suited for super-
resolution microscopy, however due to the relative ease in
synthetic manipulation of the sulfur containing dye 2, and given
its performance compared to existing dyes used for SMLM, the
organosulfur compound is much more amenable for widespread
use as an SMLM dye.
Experimental Section
All starting materials were purchased from Sigma-Aldrich, Alfa Aesar or
Fluorochem and used as purchased. Air sensitive reactions were
conducted in flame-dried glassware under an inert atmosphere of an inert
gas. 1H and 13C NMR spectra were recorded in CDCl3, MeOD, or
[D6]DMSO as specified on a Bruker 400 MHz spectrometer. Chemical
shifts are recorded on the δ scale relative to the residual solvent peak
and coupling constant (J) values are reported in Hz. Liquid
chromatographic mass spectrometry (LCMS) was performed via electron
ionisation (EI), electrospray ionisation (ES), affinity purification (AP) or
direct infusion using either a Walter ITC Premier XE instrument or a
Micromass LCT instrument. Infrared spectra were obtained using a
Figure 6. Reconstructed wide field (1.0 µm scale, A – C) and close up images
of silica nanoparticles (0.2 µm scale, D – F) coated with NBD-APTES 1 (A, D),
NPT-APTES 2 (B, E), and NPSe-APTES 3 (C, F).
As observed in the solution phase bleaching experiments,
nanoparticle bound NBD-APTES 1 stopped photoswitching and
showed a dramatic increase in fluorescence intensity with
continued laser irradiation throughout data acquisition, even at
low laser power. Similarly to the solution phase experiments, this
was not observed with nanoparticle bound NPT-APTES 2 or
NPSe-APTES 3.
Perkin-Elmer RX1 FT-IR spectrometer with
a Sinsir Technologies
DuraSimlir II ATR attachment. UV/vis absorption spectra were measured
using a Cary Bio 3 spectrophotometer. Excitation and emission spectra
were recorded a Fluoromax Fluorometer. Dynamic light scattering data
was collected using a Malvern Zetasizer NanoZS instrument. Confocal
microscopy images were taken using a wide-field Delta Vision confocal
microscope. Scanning electron microscopy images were taken using a
Philips XL-20 SEM. Melting points were measured using a Gallenkamp
melting point apparatus. Dye photobleaching data was recorded using
Ocean Optics STS-VIS Miniature Spectrometer. Blink rate analysis was
undertaken using a Nikon Eclipse Ti microscope coupled to a Andor Zyla
sCMOS camera.
Conclusions
In summary, modification of the aromatic systems by the
introduction of larger atoms in this series of dye molecules gave
rise to minor perturbations in the internal angle of the five-
membered ring (however, the molecules do remain planar)
which, when combined with an increased degree of aromaticity,
resulted in significant changes in their photophysical properties.
Integral properties such as extinction coefficient, quantum yield
and excited state lifetime were all reduced in dyes containing
heavier atoms. These properties correlated with the observed
photoblinking behaviour. Blinking density in the two buffers
tested was lower in GLOX buffer for NBD-APTES 1 but higher
for NPT- and NPSE-APTES (2 and 3) since the heavy atom
stabilisation afforded by the presence of sulfur and selenium
meant that those dye molecules were less likely to blink unless
in an environment which encourages them to do so. The oxygen
containing dye has the most stable excited state lifetime,
meaning it is more likely to fluoresce than exist in its dark state,
unless in the presence of GLOX buffer.
As for the suitability of each dye for super-resolution microscopy,
the faster blink rate of the oxygen containing NBD-APTES 1 is
actually its downfall. Although in this study it was possible to
analyse the blinking of each dye since the molecules were very
sparsely distributed, in reality when attempting to image densely
labelled specimens (as is required for SMLM) too many dye
molecules exist in the ‘on’ state at any given time and as such
individual molecules cannot be localised. This, combined with
Geometry optimisations were calculated using Gaussian 09 using density
functional theory (DFT) with a B3LYP 6311g (d,p) basis set. Energy
levels were calculated using a time-dependant self-contained field (TD-
SCF) DFT with a B3LYP 6311g (d,p) basis set and was solved for N =
100 states.
Data for time-resolved photoluminescence lifetimes was obtained using
the time-correlated single photon counting (TCSPC) method. The
samples were excited with a pulsed laser of 510 nm, repetition rate 2.5
MHz, average power <4 mW, pulse width <60 ns. The PL was detected
with a single photon avalanche diode, which has an instrument response
function that was measured to be around 1 ns.
7-Nitro-N-[3-(triethoxysilyl)propyl]-2,1,3-benzoxadiazol-4-amine (1):
A sample of 4-chloro-7-nitrobenzo-2,1,3-oxadiazole (0.10 g, 0.5 mmol)
was stirred in ethanol (10 mL). Once dissolved, the solution was cooled
to 0 °C and 3-aminopropyltriethoxysilane (0.14 mL, 0.6 mmol, 1.2 eq)
was added. The reaction mixture was stirred in the dark for 2 h at 0 °C
then for 24 h at room temperature during which time the mixture turned
from yellow to dark green. The solvent was removed in vacuo to leave a
dark green residue which was purified by column chromatography on
silica, eluting with petroleum ether/ethyl acetate 2:1 to yield the product
as an orange/red solid (0.10 g, 48%); m.p. 110–114 °C; Rf = 0.73
(petroleum ether: ethyl acetate 2:1); 1H NMR (400 MHz, CDCl3): 8.48 (1H,
d, J 8.7, ArCH), 6.95 (1H, s, NH), 6.17 (1H, d, J 8.7, ArCH), 3.86 (6H, q,
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
J 7.0, 3 × OCH2CH3), 3.61–3.50 (2H, m, NHCH2), 2.04–1.93 (2H, m,
CH2), 1.24 (9H, t, J 7.0, 3 × OCH2CH3), 0.77 (2H, t, J 7.6, SiCH2); m/z
(ES+) 439 (8%), 407 (100, M+Na+), 393 (8), 339 (22, M+-OEt), 235 (5),
210 (5); HRMS: found 407.1379, M+Na+. C15H24N4O6NaSi requires
407.1363). Data was in accordance with the literature.[23]
4-Bromo-7-nitro-2,1,3-selenadiazole (6): Diamine 5 (0.16 g, 0.71
mmol) was dissolved in ethanol (40 mL) and heated at reflux. A solution
of selenium dioxide (0.39 g, 3.55 mmol, 5 eq) in boiling water (20 mL)
was added over 5 mins to the reaction mixture and on complete addition
a yellow precipitate began to form. The mixture was maintained at reflux
for a further 16 h before cooling to rt. The precipitate was isolated via
vacuum filtration and washed with ice-cold water (2 × 10 mL) to give the
title product as a mustard yellow solid which was used without further
purification (0.19 g, 85%); m.p. 282–284 °C (decomp.), lit.[45] >300 °C;
Found: C, 23.4; H, 0.95; N, 13.55; Br, 26.1. C6H2N3O2SeBr requires C,
23.5; H, 0.7; N, 13.7; Br, 26.0%; νmax (ATR)/cm-1 3461 (w), 3422 (w),
3357 (m), 3255 (w), 1618 (s), 1506 (s, NO2); 1H NMR (400 MHz,
[D6]DMSO): 8.39 (1H, d, J 8.0, ArCH), 8.12 (1H, d, J 8.0, ArCH); 13C
NMR (100 MHz, [D6]DMSO): 157.7 (ArC), 149.3 (ArC), 140.3 (ArC),
129.5 (ArCH), 127.4 (ArCH), 123.8 (ArC); m/z (EI+) 306.8486 (76%, M+
C6H2N3O280Se79Br requires 306.8490), 277 (100), 248 (50), 182 (62), 159
(29), 131 (19), 101 (33), 80 (34). 1H NMR and IR spectra in broad
agreement with the literature – not other data is reported.[45]
4-Bromo-7-nitro-2,1,3-benzothiadiazole (4): 4,7-Dibromobenzo-2,1,3-
thiadiazole (0.10 g, 0.34 mmol) was dissolved in nitric acid (68%, 5 mL).
The mixture was stirred at reflux for 2 h during which time the solution
changed colour from orange to pale yellow. The hot solution was poured
onto ice-water (10 mL) and produced a pale yellow precipitate. Filtration
and drying afforded the product as a pale yellow crystalline solid which
was used without further purification (0.03 g, 40%); m.p. 218–220 °C
(lit.[43] 218–220 °C); 1H NMR (400 MHz, CDCl3): 8.47 (1H, d, J 8.0, ArCH),
8.03 (1H, d, J 8.0, ArCH); 13C NMR (100 MHz, [D6]DMSO): 154.4 (ArC),
145.8 (ArC), 139.1 (ArC), 131.4 (ArCH), 128.8 (ArCH), 122.3 (ArC); m/z
(EI+) 261 (48%, C6H2N3O2S81Br), 259 (47), 231 (81), 229 (80), 203 (46),
201 (43), 134 (100), 83 (88). Data in accordance with the literature.[44]
7-Nitro-N-[3-(triethoxysilyl)propyl]-2,1,3-benzoselenadiazol-4-amine
(3): tris-(Dibenzylideneacetone)dipalladium (0.0046 g, 0.005 mmol, 2
mol%) and (S)-(+)-PPFA (0.0062 g, 0.014 mmol, 6 mol%) were dissolved
in dry THF (1 mL) in a microwave vial. Bromide 6 (0.070 g, 0.228 mmol)
and APTES (0.061 g, 0.274 mmol, 1.2 eq) were added to the solution
before charging the solution with potassium tert-butoxide (0.033 g, 0.342
mmol, 1.5 eq). The solution was degassed with argon for 15 mins. The
tube was sealed and heated in a microwave reactor to 75 °C over 5 mins
at 150 W, then held at this temperature for a further 9 mins at 100 W. On
completion, the reaction mixture was diluted with ethyl acetate (10 mL)
and filtered. The filtrate was evaporated to dryness before purifying the
crude product on silica gel (1:1 petroleum ether/ethyl acetate) to give the
product as a red/orange solid. (0.0015 g, 7%); Rf = 0.56 (petroleum
ether/ethyl acetate); m.p. 68–71oC; νmax (ATR)/cm-1 3465 (w, NH), 3361
(m, NH), 2926 (m), 1621 (s, C=N), 1511 (s, NO2); 1H NMR (500 MHz,
CDCl3): 8.62 (1H, d, J 8.8, ArCH), 6.65–6.61 (1H, m, NH), 6.17 (1H, d, J
8.8, ArCH), 3.78 (6H, q, J 7.4, 3 × CH2CH3), 3.42 (2H, q, J 6.4, NCH2),
1.85 (2H, q, J 7.5, CH2), 1.18 (9H, t, J 7.0, 3 × CH3), 0.71–0.68 (2H, m,
SiCH2); 13C NMR (125 MHz, CDCl3): 152.7 (ArC), 152.5 (ArC), 147.5
(ArC), 135.7 (ArCH), 129.9 (ArC), 97.2 (ArCH), 58.6 (3 × OCH2), 45.7
(NCH2), 22.3 (CH2), 18.3 (3 × CH3), 7.9 (SiCH2); m/z (ESI+) 449.0757
(100%, MH+, C15H25N4O5Si80Se requires 449.0754).
7-Nitro-N-[3-(triethoxysilyl)propyl]-2,1,3-benzothiadiazole-4-amine
(2): A sample of bromide 4 (0.030 g, 0.12 mmol) was dissolved in ethanol
(10 mL) with (3-aminopropyl)triethoxysilane (0.034 g, 0.15 mmol, 1.25
eq) and cooled with stirring to 0 °C in the dark. Solid potassium
carbonate (0.017 g, 0.012 mmol, 0.1 eq) was added to the stirred
solution which was kept at 0 °C for 2 h before allowing the reaction to
proceed at room temperature for a further 22 h, during which time the
reaction mixture turned from yellow to green. The solvent was removed
under reduced pressure and the residue re-dissolved in ethyl acetate (25
mL). This solution was washed with water (2 × 25 mL) and brine (2 × 25
mL), dried over MgSO4, filtered and concentrated in vacuo, followed by
purification using column chromatography on silica gel to give the title
compound as a dark orange solid; (0.022 g, 45%); m.p. 100–102 °C; Rf =
0.51 (petroleum ether: ethyl acetate 3:1); νmax (ATR)/cm-1 3339 (br, NH),
2972 (w, CH), 2885 (w, CH), 1563 (s, NO2), 1493 (s); 1H NMR (400 MHz,
CDCl3): 8.67 (1H, d, J 8.8, ArCH), 6.65 (1H, br t, J 5.1, NH), 6.41 (1H, d,
J 8.8, ArCH), 3.86 (6H, q, J 7.0, 3 × OCH2CH3), 3.53 (2H, q, J 7.1,
NHCH2), 1.95 (2H, pent, J 7.1, CH2), 1.24 (9H, t, J 7.0, 3 × OCH2CH3),
0.81–0.77 (2H, m, SiCH2); 13C NMR (100 MHz, CDCl3): 148.3 (ArC),
146.7 (ArC), 146.5 (ArC), 134.0 (ArCH), 128.3 (ArC), 98.9 (ArCH), 58.6
(3 × OCH2), 45.6 (CH2), 22.2 (CH2), 18.3 (3 × CH3), 7.8 (CH2); m/z (ESI+)
446 (6), 401.1313 (100%, MH+ C15H25N4O5SSi requires 401.1309), 355
(16).
General procedure for the preparation of dye coated silica
nanoparticles: Tetraethyl orthosilicate (TEOS, 1.7 mL, 7.6 mmol) was
added to a stirred mixture of water (4.5 mL) and ammonium hydroxide
(35%, 330 µL) in ethanol (43.5 mL). The mixture was stirred rapidly at
room temperature for 24 h, after which a solution of APTES conjugated
dye in ethanol (18 mM, 50 µL) was added with additional TEOS (10 µL).
The particles were left to stir with the dye for 5 h at rt. Coated particles
were isolated by centrifugation, re-suspended in ethanol (30 mL),
isolated by centrifugation, re-suspended again in water (30 mL), and
isolated by centrifugation.
3-Bromo-6-nitro-1,2-benzenediamine
(5):
1-Amino-2-bromo-5-
nitrobenzene (0.25 g, 1.15 mmol) was dissolved in dry dimethyl sulfoxide
(15 mL) at room temperature under an atmosphere of argon. 1,1,1-
Trimethylhydrazinium iodide (0.23 g, 1.15 mol, 1 eq) was added in one
portion before adding fresh sodium pentoxide (0.38 g, 3.45 mmol, 3 eq).
The solution turned from orange to dark blue on immediate addition of
the base. The reaction was stirred overnight at room temperature before
pouring onto ice water (50 mL). The product was extracted with
dichloromethane (2 × 25 mL) and the organic layers washed with
hydrochloric acid (1M, 2 × 25 mL) and brine (25 mL) before drying over
magnesium sulfate, filtering and removing solvent in vacuo to give the
Preparation of buffers for photophysical analysis: Aqueous stock
solutions of catalase from bovine liver (4 mg per 1 mL), glucose oxidase
from Aspergillus niger (5 mg per 1 mL) and mercaptoethylamine (MEA, 1
M) were prepared.
product as
a dark orange solid which was used without further
purification (0.078 g, 30%); m.p. 145–148 °C; νmax (ATR)/cm-1 3465 (w,
NH), 3422 (w, NH), 3358 m, NH), 1620 (s, NH), 1556 (m), 1505 (s, NO2);
1H NMR (400 MHz, CDCl3): 7.62 (1H, d, J 9.3, ArCH), 6.91 (1H, d, J 9.3,
ArCH), 6.16 (2H, br s, NH2), 3.95 (2H, br s, NH2); 13C NMR (100 MHz,
[D6]DMSO): 135.9 (ArC), 135.2 (ArC), 130.4 (ArC), 119.3 (ArCH), 113.9
(ArCH), 111.9 (ArC); m/z (ESI+) 233.9698 (95%, MH+, C6H7N3O281Br
requires 233.9696), 232 (100, C6H7N3O279Br). 1H NMR data was not in
accordance with that in the literature, and so full analysis was
undertaken.[24]
To make 10 mL MEA buffer: take 1 mL MEA stock and make up to 10 mL
with distilled water.
To make 10 mL GLOX buffer: take 1 mL MEA stock, 1 mL glucose
oxidase stock, 0.1 mL catalase stock and make up to 10 mL with distilled
water. Add solid glucose (1 g, 10% w/v).
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
[16]
[17]
A. T. Balaban, D. C. Oniciu, A. R. Katritzky, Chem. Rev. 2004, 104,
2777–2812.
A. R. Katritzky, I. P. Barczynski, G. Musumarra, D. Pisano, M.
Szafranll, J. Am. Chem. Soc. 1989, 111, 7–15.
A. R. Katritzky, P. Barczynski, J. Prakt. Chem. 1990, 332, 885–897.
C. W. Bird, Tetrahedron 1992, 48, 335–340.
Y. Shen, Y. Zhang, X. Zhang, C. Zhang, L. Zhang, J. Jin, H. Li, S.
Yao, Anal. Methods 2014, 6, 4797–4802.
Y. Koide, Y. Urano, K. Hanaoka, T. Terai, T. Nagano, ACS Chem.
Biol. 2011, 6, 600–8.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.
A. Petersson, et al., Gaussian, Inc. 2009.
B. Liu, F. Zeng, G. Wu, S. Wu, Chem. Commun. (Camb). 2011, 47,
8913–5.
General procedure for the preparation and imaging of PVA
suspensions: Fluorophores were diluted to a nanomolar concentration
in 1 wt% PVA (48,000 MW) in distilled water. 50 µL was added to each
slide before air-drying in a fume hood overnight. Samples were imaged
between 30 and 60 mins after buffer solution (MEA or GLOX, 15 µL) was
added to allow sufficient time for matrix penetration but to avoid
rehydration and degradation of the polymer. Slides sealed with a cover
slip and 740 clear Rimmel London 60 second nail polish.
[18]
[19]
[20]
[21]
[22]
Each sample was imaged twice in two different regions of the matrix
under each buffer condition. Two samples were imaged for each
dye/buffer combination. Samples were excited using a 470 nm laser with
a power of 550 mW. Imaging videos were captured at a rate of 20 frames
per second for 3000 frames. The ThunderSTORM plugin for Fiji was
used to analyse captured data.
[23]
[24]
U. Baettig, I. Bruce, N. J. Press, S. J. Watson, PCT Int. Appl. 2010,
WO2010063802A1.
T. Wang, D. R. Magnin, L. G. Hamann, Org. Lett. 2003, 5, 897–900.
J. Jona, W. Derbyshire, H. S. Gutowsky, J. Phys. Chem. 1965, 69, 1.
Y. Yamashita, K. Ono, M. Tomura, S. Tanaka, Tetrahedron 1997,
53, 10169–10178.
[25]
[26]
[27]
[28]
[29]
[30]
[31]
N. Nijegorodov, R. Mabbs, Spectrochim. Acta - Part A Mol. Biomol.
Spectrosc. 2001, 57, 1449–1462.
T. Y. Ohulchanskyy, D. J. Donnelly, M. R. Detty, P. N. Prasad, J.
Phys. Chem. B 2004, 108, 8668–8672.
M. Heilemann, S. van de Linde, A. Mukherjee, M. Sauer, Angew.
Chem. Int. Ed. Engl. 2009, 48, 6903–8.
S. Stimpson, D. R. Jenkinson, A. Sadler, M. Latham, D. A. Wragg, A.
J. H. M. Meijer, J. A. Thomas, Angew. Chemie Int. Ed. 2015, 54,
3000–3003.
E. Laureto, M. A. T. Da Silva, R. V. Fernandes, J. L. Duarte, I. F. L.
Dias, H. De Santana, A. Marletta, Synth. Met. 2011, 161, 87–91.
G. Donnert, C. Eggeling, S. W. Hell, Nat. Methods 2007, 4, 81–86.
M. Heilemann, E. Margeat, R. Kasper, M. Sauer, P. Tinnefeld, J. Am.
Chem. Soc. 2005, 127, 3801–3806.
Acknowledgements
The authors would like to thank Dr. Anthony Meijer for his
guidance in geometry optimisation and energy calculations. Dr.
D. M. Coles and Claire Greenland measured quantum yields
and excited state lifetimes, respectively. DRJ is funded by the
EPSRC at the Molecular Scale Engineering Centre for Doctoral
Training (EP/J500124/1).
[32]
[33]
[34]
[35]
[36]
MathWorks Inc. 2016.
A. M. Bittel, A. Nickerson, I. S. Saldivar, N. J. Dolman, X. Nan, S. L.
Gibbs, Sci. Rep. 2016, 6, 29687.
C. A. Schneider, W. S. Rasband, K. W. Eliceiri, Nat. Methods 2012,
9, 671–675.
J. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair,
T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, et al.,
Nat. Methods 2012, 9, 676–682.
M. Ovesný, P. Křížek, J. Borkovec, Z. Švindrych, G. M. Hagen,
Bioinformatics 2014, 30, 2389–2390.
S. Mai, M. Pollum, L. Martínez-Fernández, N. Dunn, P. Marquetand,
I. Corral, C. E. Crespo-Hernández, L. González, Nat. Commun.
2016, 7, 13077.
T. Ha, P. Tinnefeld, Annu. Rev. Phys. Chem. 2012, 63, 595–617.
G. T. Dempsey, J. C. Vaughan, K. H. Chen, M. Bates, X. Zhuang,
Nat. Methods 2011, 8, 1027–36.
V. G. Pesin, A. M. Khaletskii, V. A. Sergeev, Zhurnal Obs. Khimii
1963, 33, 1759–1766.
Keywords: Dyes/Pigments • Nanoparticles • Atomic Substitution
• Super-Resolution Microscopy • Photoblinking
[37]
[38]
[1]
[2]
[3]
[4]
[5]
[6]
H. Popper, Physiol. Rev. 1944, 24, 205–224.
S. Strugger, Can. J. Res. 1948, 26, 188–193.
P. Ellinger, Biol. Rev. 1940, 15, 323–347.
E. Abbe, Arch. für Mikroskopische Anat. 1873, 9, 413–418.
M. J. Rust, M. Bates, X. Zhuang, Nat. Methods 2006, 3, 793–796.
E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S.
Olenych, J. S. Bonifacino, M. W. Davidson, J. Lippincott-Schwartz,
H. F. Hess, Science 2006, 313, 1642–1645.
[39]
[40]
[7]
[8]
S. T. Hess, T. P. K. Girirajan, M. D. Mason, Biophys. J. 2006, 91,
4258–4272.
D. N. Dempster, T. Morrow, R. Rankin, G. F. Thompson, J. Chem.
Soc. Faraday Trans. 2 Mol. Chem. Phys. 1972, 68, 1479–1496.
M. Bates, T. Blosser, X. Zhuang, Phys. Rev. Lett. 2005, 94, 108101.
R. Zondervan, F. Kulzer, S. B. Orlinskii, M. Orrit, J. Phys. Chem. A
2003, 107, 6770–6776.
[41]
[42]
[43]
[44]
[9]
[10]
M. Pagano, D. Castagnolo, M. Bernardini, A. L. Fallacara, I.
Laurenzana, D. Deodato, U. Kessler, B. Pilger, L. Stergiou, S.
Strunze, et al., ChemMedChem 2014, 9, 129–150.
R. Edwards, I. Cummins, P. Steel, PCT Int. Appl. 2009,
WO2009034396A2.
[11]
[12]
[13]
[14]
[15]
M. S. Afzal, J.-P. Pitteloud, D. Buccella, Chem. Commun. (Camb).
2014, 50, 11358–61.
N. R. Conley, A. Dragulescu-Andrasi, J. Rao, W. E. Moerner,
Angew. Chemie - Int. Ed. 2012, 51, 3350–3353.
M. R. Detty, P. N. Prasad, D. J. Donnelly, T. Ohulchanskyy, S. L.
Gibson, R. Hilf, Bioorganic Med. Chem. 2004, 12, 2537–2544.
Y. S. Park, T. S. Kale, C.-Y. Nam, D. Choi, R. B. Grubbs, Chem.
Commun. (Camb). 2014, 7964–7967.
A.-M. Kelterer, A. Mansha, F. J. Iftikhar, Y. Zhang, W. Wang, J.-H.
Xu, G. Grampp, J. Mol. Model. 2014, 20, 2344.
[45]
This article is protected by copyright. All rights reserved.

10.1002/chem.201702289
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
4-Nitrobenzodiazoles with atomic
substitution through the chalcogen
group were synthesised and
analysed for use in single-molecule
localisation microscopy. Reversible
photoswitching was achieved and
images of silica nanoparticles
coated with each dye were
D. R. Jenkinson, A. J. Cadby,* and S.
Jones*
Page No. – Page No.
The synthesis and photophysical
analysis of a series of 4-
nitrobenzochalcogenadiazoles for
super-resolution microscopy
captured. Dyes containing larger
atoms were favoured, but the sulfur
derivative was preferred due to
synthetic ease.
This article is protected by copyright. All rights reserved.