Minimal Self-Immolative Probe for Multimodal Fluoride Detection

Article abstract of DOI:10.1021/acs.joc.6b01787

Two single-molecule, self-immolative fluoride probes, namely tert-butyldimethylsilyl-protected 2- and 4-difluoromethylphenol, are described. Compared to similar systems previously described, the probes are characterized by a simpler structure and straightforward, two-step preparation. Nevertheless, they allow the detection of fluoride ions at micromolar concentration by the naked eye, UV-vis absorption, and fluorescence. A detailed investigation of the self-immolative reaction reveals that the rate-limiting step is the release of the first fluoride ion from the difluoromethylphenolate intermediate. Moreover, the mutual position of the difluoromethyl- and tert-butyldimethylsilyl-protected residues has a relevant effect on the reactivity. Likely, a CF₂H-O hydrogen bond in the 2-isomer increases the reactivity of the silyl ether toward hydrolytic cleavage but also stabilizes the phenolate intermediate, slowing the release of fluoride ions.

Full text of DOI:10.1021/acs.joc.6b01787

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Minimal self-immolative probe for multimodal fluoride detection
Luca Gabrielli, and Fabrizio Mancin
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.6b01787 • Publication Date (Web): 06 Oct 2016
Downloaded from http://pubs.acs.org on October 8, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 7
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Minimal self-immolative probe for multimodal fluoride detec-
tion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Luca Gabrielli* and Fabrizio Mancin.
Università di Padova, Dipartimento di Scienze Chimiche, via Marzolo 1, 35125 Padova, Italy
ABSTRACT: Two singleꢀmolecule selfꢀimmolative fluoride probes, namely tertꢀbutyldimethylsilyl protected 2ꢀ and 4ꢀ
difluoromethylphenol, are described. Compared to similar systems previously described, the probes are characterized by a simpler
structure and straightforward, twoꢀsteps preparation. Nevertheless they allow the detection of fluoride ions at micromolar concenꢀ
tration by nakedꢀeye, UVꢀvis absorption and fluorescence. A detailed investigation of the selfꢀimmolative reaction reveals that the
rate limiting step is the release of the first fluoride ion from the difluoromethylphenolate intermediate. Moreover, the mutual posiꢀ
tion of the difluoromethyl and tertꢀbutyldimethylsilyl protected residues has a relevant effect on the reactivity. Likely, a CF2H ꢀ O
hydrogenꢀbond in the 2ꢀisomer increases the reactivity of the silyl ether toward hydrolytic cleavage but also stabilizes the phenolate
intermediate, slowing down the release of fluoride ions.
OTBS
OTBS
INTRODUCTION
OTBDPS
F
F
Signal amplification is a valuable strategy for the realization
of effective chemosensors.1 Indeed it enables detection of very
low analyte amounts by converting the recognition event into
multiple signalling events. Several approaches have been exꢀ
plored, mainly involving catalysis and multivalency. Among
them, molecular systems featuring autoinductive signal ampliꢀ
fication2ꢀ9 represent a very promising strategy. Such probes
feature a specific triggering group, which is selectively actiꢀ
vated by the target analyte. This event starts a reaction where
the activated probe decomposes liberating both signal transꢀ
duction molecules, which propagate the reaction by activating
new trigger groups, and reporter molecules, which generate
the readout signal.5 Such cyclic reaction process results in sigꢀ
nal amplification and consequent high sensitivity. Scientists
call these probes “selfꢀimmolative”, since they structurally
“sacrifice” themselves in order to perform their designated
function.10 Several advantages are obtained with such systems:
selfꢀimmolative probes grant high selectivity along with relaꢀ
tively low costs and high shelf stability. For this reason, they
represent a valuable alternative to the most popular detection
techniques which rely on antibodies, enzymes and other bioꢀ
molecules.1a
O
O
F
HN
O
HN
O
O
F
O
O
F
c
b
a
NO2
F
O
OTBS
O
TMS
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
e
d
OTBS
CHF2
OTBS
CHF2
OTBS
1
2
3
Figure 1. Upper: Selfꢀimmolative probes a-e for detection of fluꢀ
orine reported in literature2, 12, 13, selfꢀimmolative probes 1 and 2
presented in this work, and reference compound 3.
reported over the years.11 Among them, the number of selfꢀ
immolative probes featuring signal amplification is quite
Among others, fluoride is a relevant target that can be detected
with selfꢀimmolative probes with signal amplification. Fluoꢀ
ride content in drinking water should be below 4 ppm to avoid
long term exposure effects on teeth and bones. For this reason,
several chemosensors and probes for fluoride have been
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 7
Scheme 2. Synthesis of the selfꢀimmolative chemosensors
OH
O-
O--Trigger Cap
O
1
2
3
4
5
6
7
8
OBn
OBn
OTBS
a) Pd/C, H2
MeOH, AcOH
DAST,
DCM
F
F
F
F
X-
H+
O
b) TBDMSCl,
TEA, DCM
OH
X
X
1. 63%
2. 70%
3o. 48%
3p. 50%
H2O
Scheme 1. Mechanism for the selfꢀcleavage of 4ꢀmethylhalide
phenolate derivatives (the 2ꢀderivatives undergo a similar proꢀ
cess).
derivatives (Figure 1). Following previous examples, we seꢀ
lected the tertꢀbutyldimethylsilyl residue (TBS) as trigger cap
to protect the phenol moiety, which is known to be selectively
cleaved by fluoride ion.4 The transducer unit is the difluoromeꢀ
thyl group in position 2 or 4. This ensures two advantages.
Firstly, two fluoride anions are released by each probe moleꢀ
cule, in order to propagate and amplify the signal transduction.
Secondly, the species formed by the disassembling reaction
are 2ꢀ or 4ꢀhydroxybenzaldeide, that can be detected either by
their absorbance in the UVꢀvisible region or fluorescence
emission.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
limited. The first fluoride selfꢀimmolative probe was reported
by Shabat on 2011 (Figure 1, a). The phenolate derivative
formed by the fluorideꢀinduced cleavage of the tertꢀ
butyldimethylsilyl (TBS) protecting group undergoes a series
of quinoneꢀmethide rearrangement disassembling reactions (as
in Scheme 1) that lead to the release of two fluoride anions
and one 4ꢀnitroaniline reporter, which absorbs light in the visꢀ
ible region (400 nm). The formation of two fluoride ions starts
a cascade reaction that leads to the complete disassembling of
the probe molecules present in the sample and the consequent
signal amplification. Later on, Philips and coꢀworker deꢀ
scribed probe b (Figure 1) based on a similar sensing scheme.
In this case, cleavage of the TBS group initiate a disassemꢀ
bling reaction that generates 4ꢀamino benzaldehyde as reporter
group and two fluoride ions. Also probe c, recently reported
by Huang and coꢀworkers, uses the same working scheme. In
this case, beside the two fluoride ions, a coumarin molecule is
formed. The use of fluorescence emission as detection method
led to a substantial improvement of the limit of detection with
respect to the previous systems based on absorbance.
We could develop a facile and straightforward twoꢀsteps synꢀ
thesis of the target compounds 1 and 2 (Scheme 2). The conꢀ
version of an hydroxybenzaldehyde into a derivative featuring
a difluoromethyl moiety and a silyl ether protected phenol
group, represents a nonꢀtrivial synthetic challenge. On one
hand, the TBS ether is not stable under the conditions for carꢀ
bonyl group fluorination. On the other, fluorination followed
by silyl ether protection of the phenol group could be probꢀ
lematic due to the intrinsic instability of the intermediate.
Hence we decided to use a temporary protecting group for the
phenol moiety. Commercially available 2ꢀ and 4ꢀ
benzyloxybenzaldehydes were converted into difluoroꢀmethyl
derivatives with diethylaminosulfur trifluoride (DAST) in diꢀ
chloromethane (DCM). The obtained fluorinated compounds
(3o and 3p, Scheme 2) were deꢀbenzylated by hydrogenolysis
under acidic conditions, in order to minimize the disassembly
of the unstable phenolate, and then reacted with the tertꢀ
butyldimethylsilyl chloride (TBSCl) obtaining the desired
compounds 1 and 2. Compound 3 (Figure 1) was also preꢀ
pared, as control molecule, by reacting phenol with TBSCl in
DCM (see SI for details). Derivatives 1 and 2 are stable at
room temperature and, when stored at 4°C, they didn’t show
any hydrolysis for at least one year.
Also probes d and e (Figure 1) detect fluoride by a selfꢀ
immolative reaction using related working schemes. In the
case of probe d, cleavage of the TMS group leads to the forꢀ
mation of coumarin phthalate monoester which rapidly hyꢀ
drolizes to form coumarin. As in the previous case, the signal
produced is an increase of the luminescence but no signal amꢀ
plification is possible. In the case of probe e, the rearrangeꢀ
ment reaction following the cleavage of the protecting group
leads to the release of two adamantylꢀ3ꢀhydroxyphenyl 1,2ꢀ
diethoxyethane derivatives. These in turn decompose to form
electronically excited 3ꢀcaboxymethyl phenolates that emit at
466 nm. In this way, chemiluminescence based fluoride detecꢀ
tion was possible but signal amplification was limited to the
generation of two reporter molecules per fluoride anion.
In a series of experiments aimed at testing their capability to
detect fluoride, using tetrabutylammonium fluoride (TBAF) as
a Fꢀ source, we found that the best solvent for the selfꢀ
In general, the probes so fare reported are quite complex, beꢀ
ing in any case constituted by two components, namely a trigꢀ
ger and a signal generator,4, 6 connected to form a single moleꢀ
cule.2, 12 In this work we designed, synthesised and studied the
new singleꢀmolecule selfꢀimmolative probes for fluoride aniꢀ
ons 1 and 2, (Figure 1) based respectively on the 2ꢀ and 4ꢀ
hydroxybenzaldehyde scaffolds. At difference from previous
examples, detector and amplifier are not connected as subunits
of a large molecule but are integrated in the same scaffold.
This makes the synthesis of the probe easy and costꢀeffective.
Moreover different detection modes are enabled using the
same probe, including nakedꢀeye, luminescence and UVꢀvis
spectroscopy.
6.00x104
0.05
B
A
4.50x104
3.00x104
0.04
0.03
0.02
0.01
0.00
1.50x104
0.00
0.0
2.0x104 4.0x104
Time (sec)
2.0x104 4.0x104 6.0x104
Time (sec)
0.0
Figure 2. UVꢀVis absorbance (A, 400 nm) and fluorescence
emission (B, λexc = 400 nm, λem = 483 nm) traces at 400 nm for
the reaction of probe 1 (500 ꢀM) with different amounts of tetꢀ
rabuthylammonium fluoride (black: 250 ꢀM, red: 100 ꢀM, blue:
50 ꢀM, green: 25 ꢀM, dark red: 10 ꢀM, magenta: background)
Conditions: DMSO: ACN:H2O:triethylamine 47:47:5:0.5, 25°C.
RESULTS AND DISCUSSION
Probes 1 and 2 are based, as in the example previously disꢀ
cussed, on the selfꢀcleavage of 2ꢀ and 4ꢀhalomethyl phenolate
ACS Paragon Plus Environment

Page 3 of 7
The Journal of Organic Chemistry
0.3 B
0.3 A
1
2
0.2
0.2
3
4
0.1
0.1
1
2
3
4
5
6
7
0.0
0
0.0
0.0
2.0x104 4.0x104 6.0x104
Time (sec)
10 20 30 40 50 60
Time (sec)
8
9
Figure 4. UVꢀVis absorbance traces at 345 nm for the reaction of
probe 2 (500 ꢀM) with different amounts of tetrabuthylammoniꢀ
um fluoride (black: 250 ꢀM, red: 100 ꢀM, blue: 50 ꢀM, green: 25
ꢀM, dark red: 10 ꢀM, magenta: background). Conditions: A)
DMSO:ACN:H2O:triethylamine 47:47:5:0.5, 25°C; B) DMSO:
ACN:H2O 37:37:25, 50°C.
10
11
12
13
Figure 3. Absorbance improvement at 400 nm, due to addition of
Cs2CO3 (500 ꢀM) to probe 1 (500 ꢀM) at different Fꢀ concentraꢀ
tions. Inset: picture of the samples after Cs2CO3 addition; 1: conꢀ
trol; 2: 50 ꢀM Fꢀ; 3: 100 ꢀM Fꢀ; 4: 250 ꢀM Fꢀ. Conditions:
DMSO:ACN:H2O:triethylamine 47:47:5:0.5, 180 min, 25°C.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
amplified reaction is a mixture of organic polar solvents, waꢀ
ter, and trimethylamine as a base (DMSO:ACN:H2O: triethylꢀ
amine 47:47:5:0.5). Consequently, we investigated the fluoꢀ
ride dependant signal generation, by incubating probe 1 (500
ꢀM) with different concentrations of fluoride and monitoring
the increase of absorbance at 400 nm (Figure 2A), which corꢀ
responds to the absorption maximum of 2ꢀhydroxyaldehyde at
the highest wavelength. Kinetic profiles recorded follow difꢀ
ferent trends. At high fluoride concentration (250 ꢀM) the abꢀ
sorbance rapidly increases to level off at a plateau value after
about 100 min. At lower concentrations, the absorbance inꢀ
crease follows initially an exponential profile to level off later.
As it will be discussed in detail later, such a behaviour is due
to the presence of different species absorbing at this waveꢀ
length. The exponential absorbance increase, observed in the
first phase of the experiments at low fluoride concentration, is
coherent with the amplification process. When the kinetic proꢀ
files are compared with the background degradation of 1 in the
same conditions (Figure 2A), a detection limit of 25 ꢀM can
be established, which compares well with those obtained by
similar systems.2,12,13 Interestingly, we found that the signal
produced, i.e. the absorbance of the solution after the degradaꢀ
tion of 1, can be increased by adding Cs2CO3 to the mixture.
Figure 3 reports the values of the absorbance at 400 nm recꢀ
orded for 500 ꢀM solutions of 1 incubated for 180 min with
different amounts of fluoride, before and after the addition of
the cesium salt. In each case the absorbance is 2ꢀfold increased
and fluoride can be detected by naked eye at 100 ꢀM concenꢀ
tration. Fluoride detection was also investigated by monitoring
fluorescence emission of 2ꢀhydroxyaldheide at 483 nm. (Figꢀ
ure 2B). Profiles observed are similar to the one recorded by
UVꢀVis spectroscopy and the detection limit remains 25 ꢀM.
In these conditions, probe sensitivity appears to be mainly
controlled by the background degradation rate and not by the
detection technique used.14
Figure 5. Fluorescence emission intensity (λexc = 400 nm, λem
=
483 nm) upon incubation of probe 1 (500 ꢀM) with different aniꢀ
ons. Conditions: DMSO: ACN:H2O:triethylamine 47:47:5:0.5,
25°C.
initial absorbance increase appears to be faster. Also in this
case the minimal detectable fluoride concentration is 25 ꢀM.
The probes selectivity was investigated by probe 1 with differꢀ
ent anions each at 100 ꢀM concertation (Figure 5).15 In any
case, signal produced by all the ions but fluoride was superimꢀ
posable with the background one.4,16,17 On the other hand, in
the presence of fluoride the signal measured is 4ꢀfold more
intense.
A deeper understanding of the process was obtained by invesꢀ
tigating the probes degradation process by NMR spectroscopy,
in the same conditions of the UVꢀvis and fluorescence experꢀ
iments (Figure 5). According to the commonly accepted
mechanisms for the selfꢀimmolative reaction, several species
were detected in 1HꢀNMR and 19FꢀNMR experiments (see secꢀ
tion 3, SI). Figure 5 reports the time dependant concentration
changes of the most relevant species detected in the case of
probes 1 and 2 incubated at 500 ꢀM concentration with 25 ꢀM
fluoride. The profiles could be fitted according to the mechaꢀ
nism reported in Figure 6. The first step of the reaction is the
cleavage of the silyl ether, which is completed within 10 hours
in the case of 1 and within 7 hours in the case of 2. As a result
of this reaction, three products are formed: a) 2ꢀ or 4ꢀ
(difluoromethyl)phenolate (I1), b) tertꢀbutyldimethylsilyl fluoꢀ
ride (TBSF), c) tertꢀbutyldimethylsilanol (TBSOH). The latter
two products are detected in the spectral region between ꢀ0.1
and 0.3 ppm, as expected for alkylsilane derivatives. The
amount of TBSF formed at the end of the reaction (about 250ꢀ
400 ꢀM) is much greater than the amount of fluoride initially
The behaviour of probe 2 is similar with few remarkable difꢀ
ferences. Figure 4A reports the absorbance profiles obtained
by incubation of 2 with different amounts of fluoride, under
the same conditions used for 1. Here the highest absorption
maximum of the degradation products (mainly 4 hydroxybenꢀ
zaldheide) is at 345 nm. The kinetic profiles are similar to
those observed for probe 1, with the only difference that the
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 7
reaction mixture. Interestingly, the formation of TBSOH is
F-
H2O
Si
Si
1
2
3
more relevant in the case of 1. Indeed, the rates for the hydrolꢀ
ysis of the protecting group are respectively 4.2×10ꢀ5 secꢀ1 and
2.4×10ꢀ5 secꢀ1 for 1 and 2.
F
OH
TBSF
TBSOH
H2O
F-
Si
O
O-
O
OH
OH
4
5
6
7
8
9
The concentration of I1 increases as the result of the TBS
deprotection, reaching a maximum after 7ꢀ10 hours. The folꢀ
lowing decrease is due to the selfꢀcleavage of the CꢀF bonds to
form the final salicylates. Notably neither of the subsequent
intermediates (hydroquinone I2 and the α−hydroxyfluoro inꢀ
termediate I3, Figure 5) were detected in the selfꢀcleavage of
1, suggesting that these species are quite reactive and the rate
limiting step is the first CꢀF cleavage. On the other hand, deꢀ
composition of I1 is much faster for 2 (3.3×10ꢀ5 secꢀ1) than for
1 (8.8 ×10ꢀ6 secꢀ1) and this allows, in the case of 2, the accuꢀ
mulation of the hydroquinone intermediate I2, whose decomꢀ
position is still very fast (5×10ꢀ3 secꢀ1), as expected. Nicely
enough, the reactions rates obtained by the fit of the NMR exꢀ
periments allowed to interpolate the UVꢀvis kinetic experiꢀ
ments, with both I1 and the final hydroxyaldheides contribꢀ
uting to the absorption increase.
F
O
F-
F-
F
F
F
F
F
HO
H2
O
S
I1
I2
I3
P
7x103
6x103
5x103
4x103
3x103
2x103
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I1
S
TBSOH
TBSF
P
1x103
In order to further investigate deprotection process, we perꢀ
formed 1HꢀNMR kinetic experiments also with reference comꢀ
pound 3 (Figure 1), which is devoid of the difluoromethyl
group. In the presence of Fꢀ 25 ꢀM (SI, Figures S7, S8), as exꢀ
pected, we did not observed any amplification and only the
5% of 3 reacted to give phenol, since no more fluoride was
available to propagate the reaction. Hence, signal amplication
allowed by probes 1 and 2 with respect to 3 in this conditions
is 20ꢀfold.18
0
2.0x104
4.0x104
6.0x104
8.0x104
0.0
Time, sec
1.8x104
1.5x104
1.2x104
9.0x103
6.0x103
S
P
TBSF
Several relevant information are provided by the above experꢀ
iments: 1) the background reaction that determines the detecꢀ
tion limit of the probes is the hydrolysis of the TBS group, 2)
the presence of the difluoromethyl group increases the lability
of the TBS protected phenol; likely this is due to the electronꢀ
withdrawing ability of the substituent that decreases the basiciꢀ
ty of the resulting phenolate; 3) the effect is stronger when the
difluoromethyl group is in the orto position, and the decompoꢀ
sition of the resulting I1 intermediate is slower. This behavꢀ
iour suggests an additional phenolate stabilization, which can
likely arise by an intramolecular CF2H – Oꢀ hydrogen bond.
I1
3.0x103
TBSOH
I2
0.0
2.0x104
4.0x104
6.0x104
8.0x104
0.0
Time, sec
One of the main limitation of selfꢀimmolative probes is the
long time required for the analysis, which is generally in the
range of hours. Our results demonstrate that the rate limiting
step is the first quinoneꢀmethide rearrangement. For this reaꢀ
son phenolate intermediate accumulate in the early reaction
time and both signal producing species and signal amplifying
fluoride are produced in a small extent. The stabilization of the
phenolate oxygen may further decrease the reaction rate.
However, we found that reaction times can be sensibly deꢀ
creased by increasing the temperature. At 50° C, maximum
absorbance is reached 30 minutes with both the probes. Nicely
enough, in the case of probe 2 the background reaction is
much less sensible to the temperature increase than the fluoꢀ
ride initiated reaction. This allowed decreasing of the detecꢀ
tion limit to 5 ꢀM.
Figure 6. 1HꢀNMR kinetic study of 500 ꢀM probes 1 (upper) and
2 (lower) reaction with TBAF (25 ꢀM). The normalised area of
characteristic compound’s peaks is plotted vs reaction time. Conꢀ
ditions: d6ꢀDMSO:CD3CN:D2O:triethylamine 47:47:5:0.5, 28°C.
present. This confirms the occurrence of the selfꢀimmolative
reaction with the additional fluoride yielded by the decomposiꢀ
tion of the phenolate I1. Moreover, the concentrations of
TBSF and TBSOH remain almost constant for at least 10
hours after the complete deprotection of 1 and 2. TBSOH is
hence a relatively stable specie in the reaction conditions, and
is formed only marginally by hydrolysis of TBSF in the time
span of the experiments. The pseudoꢀfirst order rate constant
for the hydrolysis of TBSF obtained by the interpolation of the
kinetic profiles is about 2.2×10ꢀ6 secꢀ1. TBSF and TBSOH are
consequently the products of two concurrent deprotection reꢀ
actions. In the first one the TBS group is cleaved as plannedby
fluoride. This reaction is very fast (diffusion controlled acꢀ
cording to the fittings) but is limited by the availability of fluꢀ
oride in the samples. In the second reaction, the TBS groups is
cleaved by hydroxide anions or water molecules present in the
Another relevant drawback is the requirement of an highly orꢀ
ganic solvent system for the selfꢀimmolative reactions. Such
problem can be addresses easily when the reaction is perꢀ
formed at higher temperature. Indeed, working at 50 °C we
could simplify the solvent system to ACN:DMSO:H2O
37:37:25, simultaneously increasing the amount of water preꢀ
sent in the reaction mixture. The detection limits in these opꢀ
ACS Paragon Plus Environment

Page 5 of 7
The Journal of Organic Chemistry
General procedure for sensing experiments: the reported
erating conditions is 25 ꢀM for both the probes while Fꢀ detecꢀ
1
2
3
4
5
6
7
tion time is reduced to about 10 minutes (Figure 4B and SI). It
is worth mentioning that, since the water content reaches the
25%, the detection limit in the analysed water sample reaches
1.9 ppm (100 ꢀM), which is below the US Environmental Proꢀ
tection Agency’s (EPA) recommended fluoride concentration
in water (2 ppm) as well as the EPA’s mandatory upper limit
(4 ppm).19
disassembly reactions were performed as follows. 5 ꢀL of a 50
mM DMSO stock solution of the desired probe were diluted
with 229 ꢀL of DMSO and 234 ꢀL of acetonitrile (ACN); then
water (25 ꢀL) and triethylamine (2,5 ꢀL) were added. Finally
to this solution we added 5 ꢀL of a freshly prepared TBAF
solution in THF. The mixture was vigorously shaken for few
seconds and then incubated at room temperature. For NMR
experiments deuterated DMSO, D2O and ACN were used.
Fresh TBAF solutions were prepared dissolving the required
amount of TBAF×3H2O in dry THF. The DMSO probe’s
stock solution were stored at ꢀ18°C. We performed the disasꢀ
sembly reactions in 1,5 mL Eppendorf® plastic vials, in order
to avoid glass interferences in the catalytic amount of fluoride.
Only for the NMR studies the mixture was further transferred
into a wellꢀsealed NMR tube.
8
9
CONCLUSION
In conclusion, we proved that salicylaldehyde and its para
isomer can be successfully used as scaffolds and reporters, for
buildingꢀup new selfꢀimmolative constructs, through a facile
and straightforward twoꢀstep synthesis, starting from cheap
compounds. Indeed, 1 and 2 are, at the best of our knowledge,
the simplest singleꢀmolecule fluoride chemosensors so far reꢀ
ported. The sensing system is able to detect fluoride anions
down to 25 ꢀM, both by UVꢀvis absorption and fluorescence,
through an autoinductive signal amplification reaction. Naked
eye detection is also possible with a threshold value of 100
ꢀM. For the first time the selfꢀimmolative mechanism was
proved and deeply studied by NMR kinetic experiments. Exꢀ
periments confirmed the amplification mechanism and reꢀ
vealed that the rate determining step of the reaction is the deꢀ
composition of the phenolate intermediate with concurrent reꢀ
lease of one fluoride ion. On the other hand, the limit of the
detection is controlled by the hydrolytic resistance of the trigꢀ
gering cup. In further evolutions, the probe sensitivity could
be improved by increasing the stability of the cap. Moreover,
these chemosensors could work as amplifiers in a twoꢀ
component analyte detector system as the simple synthetic
strategy setꢀup opens a wide range of possibilities. By modifyꢀ
ing only the last synthetic step and protecting the hydroxyl
moiety with an appropriate group, it could be in principle posꢀ
sible to prepare a probe for almost any desired analyte.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General procedure for intermediates 3o and 3p. Benꢀ
zyloxybenzaldehyde (430 mg, 2,026 mmol, 1 equiv) was disꢀ
solved in dry DCM (1 mL), then catalytic MeOH (1,5 ꢀL) and
DAST (495 ꢀl, 5,065 mmol, 2,5 equiv) were added. The reacꢀ
tion was stirred at room temperature, under N2 atmosphere for
3 days. The reaction was quenched adding a Na2CO3 saturated
solution, till neutralization, then the product was extracted in
DCM, the organic layers were collected, dried over MgSO4
and finally purified by flash column chromatography (from
9,5:0,5 to 8:2 Exane:EtOAc) affording the fluorinated product
(as a colorless oil) and the unreacted aldehyde. Although the
prolonged reaction time, the reaction is clean, since the desired
fluorinated product is the only formed and the unreacted startꢀ
ing compound can be easily recovered.
1-(benzyloxy)-2-(difluoromethyl)benzene (3o): Obtained:
228 mg, yield 48%; 1H NMR (200 MHz, CDCl3) δ 7.68 (d, J =
7.6 Hz, 1H), 7.56 – 7.37 (m, 6H), 7.18 – 7.02 (m, 2H), 7.12 (t,
JH,F = 55.7 Hz, 1H), 5.20 (s, 2H); 13C NMR (50 MHz, MeOD)
2
δ 156.9, 136.8, 132.2, 128.9, 128.4, 127.5, 126.6 (t, JCF = 5.8
Hz), 121.2, 112.6, 111.9 (t, JCF = 235.7 Hz), 70.6; 19F NMR
EXPERIMENTAL SECTION
(376 MHz, CDCl3) δ ꢀ115.68 (d, J = 55.7 Hz); GCꢀMS (EI)
+
m/z: 234 (M+), 91 (PhCH2 ), 51 (CHF2+).20
General: Solvents were purified by standard methods. All
commercially available reagents and substrates were used as
received. TLC analyses were performed using F254 precoated
silica gel glass plates. Column chromatography was carried
out on silica gel 60 (70ꢀ230 mesh). Pt/C was filtered by 25
mm syringe filter, with 0.45 ꢁm nylon membrane. NMR specꢀ
tra were recorded using a spectrometer operating at 500 MHz
1-(benzyloxy)-4-(difluoromethyl)benzene (3p): Obtained:
237 mg, yield 50%; 1H NMR (300 MHz, CDCl3) δ 7.58 – 7.32
(m, 7H), 7.07 (d, J = 8.5 Hz, 2H), 6.63 (t, JH-F = 56.7 Hz, 1H),
5.13 (s, 2H); 13C NMR (75 MHz, CDCl3) δ 160.6, 136.5,
128.7, 128.2, 127.5, 127.2 (t, 2JCF = 5.7 Hz), 115.0 (2C), 114.9
(t, JCF = 237.0 Hz), 70.1; 19F NMR (376 MHz, CDCl3) δ ꢀ
108.72 (d, J = 56.7 Hz), GCꢀMS (EI) m/z: 234 (M+), 91
for H, 125.8 MHz for 13C. Chemical shifts are reported relaꢀ
1
tive to internal Me4Si. 19FꢀNMR spectra were acquired using a
+
+
(PhCH2 ), 51 (CHF2 ).20
spectrometer operating at 400 MHz for H, 376 MHz for 19F,
1
General procedure for compounds 1 and 2. Compound
3o/3p (197 mg, 0,841 mmol, 1 equiv) was dissolved in MeOH
(2 mL) and AcOH (100 ꢀL) and Pd/C (180 mg) were added;
then the solution was degassed in vacuum under stirring. Fiꢀ
nally H2 was added and the mixture was vigorously stirred for
1,5 hours. After completion, Pd was filtered off with a syringe
filter, the solvent was evaporated under reduced pressure and
the product was dissolved in dry DCM (2 mL). TEA (470 ꢀl,
3,364 mmol, 4 equiv) and TBDMSCl (253 mg, 1,682 mmol, 2
equiv) were quickly added and the mixture was reacted under
N2. After 2 hours silica was added, the solvent was evaporated
and the product was finally purified by flash column chromaꢀ
tography (9,5:0,5 EP:EtOAc) affording the final compound as
a colorless oil.
equipped with a probe QNP 5 mm. Multiplicity is given as folꢀ
low: s = singlet, d = doublet, t = triplet, q = quartet, qn = quinꢀ
tet, m = multiplet, br = broad peak. GCꢀMS mass spectra were
obtained using a (5%ꢀphenyl)ꢀmethylpolysiloxane capillary
column 30 cm long, I.D. 0.25, film 0.25 ꢀm, He as the carrier
gas, and the following settings: inlet 250°C, oven ramp: 3 min
50°C, 15°C/min, 10 min 250°C. UVꢀvis spectra were recorded
with a doubleꢀbeam spectrophotometer using an average time
of 0.0125s, data interval 5 nm, scan rate 24000 nm/min. Fluoꢀ
rescence spectra were recorded using a plate reader; excitation
400 nm, emission 483 nm, excitation bandwidth 10 nm, emisꢀ
sion bandwidth 5 nm, gain 93, n° of flashes 50, flash frequenꢀ
cy 400 Hz, integration time 20 ꢀs, lag time 0 ꢀs, settle time
100 ms.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 7
(7) AvitalꢀShmilovici, M.; Shabat, D. Bioorg. Med. Chem. 2010,
tert-butyl(2-(difluoromethyl)phenoxy)dimethylsilane (1):
Obtained: 137 mg, yield 63%; H NMR (200 MHz, CDCl3) δ
1
18, 3643.
1
2
3
4
5
6
7
8
(8) Sella, E.; Shabat, D. J. Am. Chem. Soc. 2009, 131, 9934.
(9) Zhu, L.; Anslyn, E. V. Angew. Chem., Int. Ed. 2006, 45,
1190.
7.61 (d, J = 7.5 Hz, 1H), 7.37 (d, J = 7.4 Hz, 1H), 7.08 (t, J =
7.4 Hz, 1H), 6.98 (t, JH,F = 55.8 Hz, 1H), 6.91 (d, J = 8.2 Hz,
1H), 1.08 (s, 9H), 0.32 (s, 6H); 13C NMR (50 MHz, MeOD) δ
154.0, 131.9 (2C), 126.7 (t, 2JC,F = 5.6 Hz), 121.4, 119.1, 112.1
(t, JC,F = 236.0 Hz), 25.8, 18.4 (3C), ꢀ4.1 (2C); 19F NMR (376
MHz, DMSO/ACN 1:1,D2O 5%, TEA 0,5%) δ ꢀ118.89 (d, JF,H
= 55.8 Hz); 29Si NMR (99 MHz, CDCl3) δ 22.68; GCꢀMS (EI)
m/z: 258 (M+), 201 ([MꢀtBu]+), 57 (tBu+).20
(10) Gnaim, S.; Shabat, D. Acc. Chem. Res. 2014, 47, 2970.
(11) (a) Zhou,Y.; Zhang, J. F.; Yoon, Y. Chem. Rev. 2014, 114,
5511 and references cited therein. (b) Xiong, Y.; Wang, C.;
Tao, T.; Duan, M.; Tan, J.; Wub, J.; Wang, D. Analyst, 2016,
141, 3041. (c) Basu, A.; Suryawanshi, A.; Kumawat, B.;
Dandia, A.; Guin D.; Ogale, S. B. Analyst, 2015, 140, 1837.
(12) (a) Gu, J. A.; Mania, V.; Huang, S. T. Analyst. 2015, 140,
346. (b) PerryꢀFeigenbaum, R.; Sella, E.; Shabat, D. Chem.
Eur. J. 2011, 17, 12123. (c) Baker, M. S.; Phillips, S. T. Org.
Biomol. Chem. 2012, 10, 3595.
(13) Mahoney, K. M.; Goswami, P. P.; Winter, A. H. J. Org.
Chem. 2013, 78, 702.
(14) AvitalꢀShmilovici, M.; Shabat, D. Soft Matter. 2010, 6, 1073.
(15) Similar results were obtained with probe 2, see Supporting Inꢀ
formation.
(16) Kim, S. Y.; Hong, J.ꢀI. Org. Lett. 2007, 9, 3109.
(17) Zhu, C.ꢀQ.; Chen, J.ꢀL.; Zheng, H.; Wu, Y.ꢀQ.; Xu, J.ꢀG.
Anal. Chim. Acta 2005, 539, 311.
(18) The extent of amplification depends both on the analyte and
probe concentrations. In general, the signal produced by a
notꢀimmolative system is proportional to the analyte concenꢀ
tration while the signal produced by a selfꢀimmolative system
is proportional to the probe concentration. Signal amplificaꢀ
tion is consequently the ratio of the two values.
(19) EPA National Primary Drinking Water Standards, see
https://www.epa.gov/dwstandardsregulations for more inforꢀ
mation.
(20) Highꢀresolution mass spectrum (HRꢀMS) were recorded both
on APIꢀTOF mass spectrometer and on a QꢀTOF (MeOH,
9
tert-butyl(4-(difluoromethyl)phenoxy)dimethylsilane (2):
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Obtained: 152 mg yield 70%; H NMR (300 MHz, CDCl3) δ
7.40 (d, J = 8.3 Hz, 2H), 6.91 (d, J = 8.4 Hz, 2H), 6.60 (t, JH,F
= 56.8 Hz, 1H), 1.01 (s, 9H), 0.23 (s, 6H); 13C NMR (75 MHz,
CDCl3) δ 157. 8, 127.1 (2C), 120.2 (2C), 114.9 (t, JC,F = 237.5
Hz), 25.6, 18.2, ꢀ4.4; 19F NMR (377 MHz, DMSO/ACN
1:1,D2O 5%, TEA 0,5%) δ ꢀ112.08 (d, JF,H = 56.8 Hz); 29Si
NMR (99 MHz, CDCl3) δ 21.84; GCꢀMS (EI) m/z: 201 ([Mꢀ
tBu]+), 77 (Ph+), 57 (tBu+).20
[[(1,1-dimethylethyl)dimethylsilyl]oxy]-Benzene (3). Phenol
(200 mg, 2,125 mmol, 1 equiv) was dissolved in dry DCM (10
mL); then under stirring TEA (323 ꢀl, 3,190 mmol, 1,5 equiv)
and TBDMSCl (481 mg, 3,190 mmol, 1,5 equiv) were added
and the mixture was reacted under N2 for 6 hours. Then silica
was added, the solvent was evaporated and the product was
finally purified by flash column chromatography (100% EP)
affording the protected phenol 3 as a clear oil (376 mg, 85%
yield): 1H NMR (200 MHz, CDCl3) δ 7.24 (t, J = 7.8 Hz, 2H),
6.96 (t, J = 7.4 Hz, 1H), 6.86 (d, J = 7.5 Hz, 2H), 1.01 (s, 9H),
0.22 (s, 6H); 13C NMR (50 MHz, CDCl3) δ 155.8, 129.6 (2C),
121.5 120.4 (2C), 25.9 (3C), 18.3, ꢀ4.19 (2C). (Data in agreeꢀ
ment with those reported in literature).21
0.5% formic acid). However, we could not detect the molecuꢀ
lar ion, because the analytes were not ionizable under these
analytical conditions. On the other hand the volatility and oily
nature of the reported molecules, together with the presence
of F, Si and O atoms, did not allow for meaningful elemental
analysis (CHN microanalyzer). Ionization and consequent deꢀ
tection of the compounds was possible only with a GCꢀMS
(EI) spectrometer. In particular the molecular ion and signifiꢀ
cant fragments (i.e. [MꢀtBu]+) were detectable. Considering
the simple nature of the compound and the fact that we have
obtained the 1H, 13C, 19F and 29Si NMR spectra as well as the
GC MS traces, we are confident of both the purity and identiꢀ
ty of the compounds reported.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
1
ACS Publications website. Additional H, 19F NMR and UV exꢀ
periments. (PDF)
AUTHOR INFORMATION
(21) Baker, M. S.; Phillips, S. P. J. Am. Chem. Soc. 2011, 133,
5170.
Corresponding Author
* Email: luca.gabrielli@unipd.it.
ACKNOWLEDGMENT
This work was supported by the ERC Starting Grants Project
MOSAIC (259014) and by Università di Padova (Progetto Strateꢀ
gico di Ateneo NAMECA).
REFERENCES
(1) (a) Scrimin, P.; Prins, L. J. Chem. Soc. Rev. 2011, 40, 4488.
(b) Lei, J; Ju, H. Chem. Soc. Rev. 2012, 41, 2122. (c) Maꢀ
khlynets, O. V.; Korendovych, I. V. Biomolecules 2014, 4,
402.
(2) Turan, I. S.; Akkaya, E. U. Org. Lett. 2014, 16, 1680.
(3) KartonꢀLifshin, N; Shabat, D. New J. Chem. 2012, 36, 386.
(4) Baker, M. S.; Phillips, S. T. J. Am. Chem. Soc. 2011, 133,
5170.
(5) Sella, E.; Lubelski, A.; Klafter, J.; Shabat, D. J. Am. Chem.
Soc. 2010, 132, 3945.
(6) Sella, E.; Weinstain, R.; Erez, R.; Burns, N. Z.; Baran P. S.;
Shabat, D. Chem. Commun. 2010, 46, 6575.
ACS Paragon Plus Environment

Page 7 of 7
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
ACS Paragon Plus Environment