A Novel Fluorescent Monomer for the Selective Detection of Phenols and Anilines

Full text of DOI:10.1021/jo982105e

J . Org. Chem. 1999, 64, 4191-4195
4191
probe molecules onto the end of a fiber optic cable.1
2-15
A Novel F lu or escen t Mon om er for th e
Selective Detection of P h en ols a n d An ilin es
Alternatively, solid-state alcohol and amine sensors have
also been fabricated by embedding the desired fluoro-
phores into a porous solid material. For example, sol-
gel chemistry has been used to immobilize synthetic dyes
as well as fluorescent biological sensing systems to
Mary A. Reppy, Martin E. Cooper,
J uston L. Smithers, and Douglas L. Gin*
Department of Chemistry, University of California,
Berkeley, California 94720-1460
fabricate silica-based sensors capable of detecting alco-
hols in water.1
6,17
In terms of organic solid-state sensors,
a number of researchers have designed polymer-based
alcohol sensors by doping poly(vinyl chloride) films with
fluorescein-, stillbene-, and azobenzene-based dyes.
Received October 19, 1998
The detection of analytes utilizing the fluorescence
response of synthetic molecular probes is a growing area
of academic and industrial research.1-3 The ultimate goal
of this work is to obtain a chemosensor with quantitative,
real-time optical response and exceptional selectivity for
18-20
This same technique has been used to produce polymer-
21
based thiamine sensors. Finally, Orellana and co-
workers have recently synthesized a series of fluorescent,
pyrazine-based dyes that are not only alcohol sensitive
but can also be easily attached to fibers or solid sup-
1
a particular analyte or class of analytes. Alcohols and
amines are two classes of analytes that have recently
received a great deal of attention in the area of fluores-
cence detection because of the prevalence of hydroxy- and
amine-containing molecules in biology and the extensive
use of alcohols and amines in industrial applications.
A number of novel solution-based fluorescent probes
with sensitivity for alcohols have recently been reported.
For example, Ueno and co-workers4 have developed a
series of fluorescent cyclodextrins for the host-guest
sensing of long-chain and biologically relevant alcohols.
Kumar and co-workers have developed a pyridine/pyrene-
based fluorescent probe that is quenched by protic
analytes such as phenol, aniline, and water via a hydro-
gen-bonding mechanism.7 Although amines are gener-
ally known to quench the fluorescence of a large number
of commercial dyes, a number of novel fluorophores with
sensitivity to amine-containing analytes have also been
recently developed. For example, Shinkai and co-workers
developed a pyrene-based receptor for the detection of
22
ports. In this system, fluorescence quenching is ac-
complished via a hydrogen-bonding interaction with the
alcohols.
We have developed a new polymerizable fluorescent
probe (1) that is quenched selectively by aromatic alcohols
and amines, even in the presence of their aliphatic
analogues, oxygen, and water. This selective quenching
occurs with 1 dissolved in nonpolar solvents such as
benzene or cross-linked inside a polymethacrylate matrix.
Monomer 1 contains a central pyridine ring similar to
Kumar’s fluorophore;7 however, it has a different con-
jugated core architecture and can also participate in
radical copolymerizations with conventional monomers.
This novel fluorophore architecture leads to a different
mechanism of fluorescence quenching from that of Ku-
mar’s fluorophore and also to a high degree of analyte
selectivity.
-6
,8
,8
Monomer 1 was synthesized as shown in Scheme 1.
2,6-Pyridinedimethanol was brominated with HBr in
acetic acid23 to afford 2,6-bis(bromomethyl)pyridine (2).
This compound was subsequently reacted with triethyl
phosphite to yield the corresponding phosphonate ester
9
barbital via hydrogen-bonding interactions. Sun et al.
used methylene diphenylene diisocyanate as a fluorescent
24
probe to monitor the conversion of amines and alcohols
during polyurethane formation.10 More recently, Fab-
2
5
(
3), which then underwent a Wadsworth-Emmons
brizzi and co-workers developed a fluorescent dizinc
26
condensation with 2 equiv of TBDMS-protected syring-
anthracene-octamine complex that is quenched by the
aldehyde (4).27 Removal of the silyl ether protecting
binding of imidazole and histidine.11
Although these novel sensing molecules have been
shown to operate well in solution, in terms of practicality
a solid-state sensor is often more desirable. A reversible
solid-state chemosensor minimizes sample contamination
(
12) Wolfbeis, O. S.; Posch, H. E.; Fresenius’ Z. Anal. Chem. 1998,
32(3), 255.
13) White, J .; Kauer, J . S.; Dickinson, T. A.; Walt, D. R. Anal. Chem.
1996, 68(13), 2191.
14) Dickinson, T. A.; White, J .; Kauer, J . S.; Walt, D. R. Nature
996, 382, 697.
15) Charlesworth, J . M.; McDonald, C. A. Sens. Actuators B 1992,
B8(2), 137.
3
(
(
(especially for in vivo applications) and, after regenera-
1
tion of the sensing element, can be used again. One of
the most common approaches for fabricating solid-state
alcohol and amine sensors is to immobilize fluorescent
(
(
16) Simon, D. N.; Czolk, R.; Ache, H. J . Thin Solid Films 1995,
2
60(1), 107.
(
17) Williams, A. K.; Hupp, J . T. J . Am. Chem. Soc. 1998, 120(18),
4366.
(18) Zeng, H.-H.; Wang, K.-M.; Li, D.; Yu, R.-Q. Talanta 1994, 41(6),
969.
(19) Mohr, G. J .; Lehmann, F.; Grummt, U.-W.; Spichiger-Keller,
(
(
(
(
(
(
1) Czarnik, A. W. Chem. Biol. 1995, 2(7), 423.
2) Mahutte, C. K. Clin. Biochem. 1998, 31(3), 119.
3) J ohnson, I. Histochem. J . 1998, 30(3), 123.
4) Ueno, A.; Minato, S.; Osa, T. Anal. Chem. 1992, 64, 1154.
5) Ueno, A.; Minato, S.; Osa, T. Anal. Chem. 1992, 64, 2562.
6) Hamasaki, K.; Ueno, A.; Toda, F. J . Chem. Soc., Chem. Commun.
U. E. Anal. Chem. Acta 1997, 344(3), 215.
(20) Mohr, G. J .; Spichiger-Keller, U. E. Anal. Chim. Acta 1997,
351(1-3), 189.
(21) He, H.; Uray, G.; Wolfbeis, O. S. Anal. Lett. 1992, 25(3), 405.
(22) Orellana, G.; Gomez-Carneros, A. M.; de Dios, C.; Garcia-
Martinez, A. A.; Moreno-Bondi, M. C. Anal. Chem. 1995, 67(13), 2231.
(23) Baker, W.; Buggle, K. M.; McOmie, J . F. W.; Watkins, D. A. M.
J . Chem. Soc. 1958, 3594.
(24) Offerman, W.; Voegtle, F. Synthesis 1977, 272.
(25) Kabachnic, M. I.; Polikarpv, Y.; Bodrin, G. V.; Bel’skii, F. I. Bull.
Acad. Sci. USSR Div. Chem. Sci. (Eng.) 1988, 37, 800.
(26) Wadsworth, W. S. Org. React. 1977, 25, 73.
1
1
993, 331.
(
(
7) Kumar, C. V. J . Chem. Soc., Chem. Commun. 1993, 772.
8) Kumar, C. V.; Tolosa, L. M. J . Photochem. Photobiol. A: Chem.
994, 78, 63.
9) Aoki, I.; Harada, T.; Sakaki, T.; Kawahara, Y.; Shinkai, S. J .
Chem. Soc., Chem. Commun. 1992, 1341.
10) Sun, X.-D.; Sung, C. S. P. Polym. Prepr., Am. Chem. Soc. Div.
Polym. Chem. 1994, 35(1), 435.
11) Fabbrizzi, L.; Francese, G.; Licchelli, M.; Perotti, A.; Taglietti,
A. Chem. Commun. 1997, 581.
(
(
(
1
0.1021/jo982105e CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/30/1999

4
192 J . Org. Chem., Vol. 64, No. 11, 1999
Notes
Sch em e 1
Ta ble 1. Qu en ch in g of 1 (90 µm in Ben zen e) by Va r iou s
Alcoh ols a n d Am in es
rel oxidation
E1/2 values (V) vs SCE
Ksva (M-1
)
b
analyte
N,N-dimethylaniline
N-methylaniline
m-toluidine
18
16
15
13
0.68
0.78
aniline
0.86
phenol
m-cresol
p-cresol
17
17
15
1.37
1.28
1.17
indole
8.9
benzyl alcohol
ethanol
cyclohexanol
t-butanol
1.6
1.1
0.7
0.6
>2.33
2.87
>2.33
2.86
triethylamine
dibutylamine
butylamine
1.3
0.5
0.2
0.99
1.20
1.52
F igu r e 1. Stern-Volmer plots of solution quenching of 1 (90
µM, benzene): 0, phenol; 4, dibutylamine; +, aniline; ×,
methanol.
a
Measured in benzene solution. b Derived from tables of re-
ported oxidation E1/2 values measured in acetonitrile. See refs 30-
32. It was not possible to obtain oxidation potentials of the
compounds in benzene solution.
groups on the resulting intermediate using tetra(n-butyl)-
2
8
ammonium fluoride gave the bisphenol 5. Finally,
acrylation of 5 with acryloyl chloride29 afforded 1 in 19%
overall yield.
an electron-transfer quenching mechanism as described
by Rehm and Weller, in which the excited state of 1 is
able to spontaneously oxidize the quenching species and
Initial spectroscopic studies of 1 dissolved in benzene
(90 µM) revealed that the monomer has characteristic
33
undergo subsequent nonradiative decay. Within this
absorption maxima at 309 and 348 nm, intense emission
maxima at 395 and 420 nm when excited at 310 or 370
nm, and a fluorescence lifetime of 1.45 ns. The intensity
ratio of the 395 and 420 nm emission bands (1.06:1.00)
is largely invariant to monomer concentration below the
millimolar level.
trend, it should be noted that the phenols and the
aliphatic amines are comparable in oxidizability; how-
ever, the phenols are extremely good quenchers whereas
the aliphatic amines are almost nonquenching. This
apparent anomaly can be rationalized by the fact that
phenols are somewhat acidic (pK
a
≈ 10) whereas aliphatic
A wide variety of alcohols and amines were added to
the 90 µM benzene solution of 1, in concentrations of 10-
(30) For tabulated oxidation E1/2 values of amines, see: Reed, R.
1
00 mM, to test their quenching behavior. Anilines and
C.; Wightman, R. M. In Encyclopedia of Electrochemistry of the
Elements, Organic Section; Bard, A. J ., Lund, H., Eds.; Marcel
Dekker: New York, 1978; Vol. XV, Chapter XV-1, pp 4-30.
phenols were found to act as strong quenchers with
-1
Stern-Volmer constants (Ksv) ranging from 13 to 18 M
,
(31) For tabulated oxidation E1/2 values of alcohols, see: Parker, V.
whereas aliphatic amines and alcohols were found to be
very weakly quenching or effectively nonquenching (Fig-
ure 1). The quenching efficiencies of the added species
can be correlated to their relative ease of oxidation, as
D.; Sundholm, G.; Svanholm, U.; Rol a´ n, A.; Hammerich, O. In
Encyclopedia of Electrochemistry of the Elements, Organic Section;
Bard, A. J ., Lund, H., Eds.; Marcel Dekker: New York, 1978; Vol. XI,
Chapter XI-2, pp 182-183 and 188.
(32) E1/2 values reported in the literature are typically those
3
measured in water, CH CN, or propylene carbonate solution using 0.1
ranked by their oxidation potentials (E1/2) measured in
or 0.15 M supporting electrolyte and a Pt anode. E1/2 values reported
for compounds in the same solvent are also often cited against many
different reference electrodes. To provide a qualitiative ranking of the
a common solvent3
0-32
and compared against a common
reference electrode (Table 1). This trend is consistent with
3
oxidizability of the various analytes, oxidation E1/2 values in CH CN
of the compounds were obtained from the literature and converted to
values reported against a common reference electrode (SCE). The
oxidation E1/2 value for 1 in acetonitrile was measured to be 1.92 V vs
SCE. See the Supporting Information for details.
(33) (a) Rehm, D.; Weller, A. Israel J . Chem. 1970, 8, 259. (b) Lowry,
T. H.; Richardson, K. S. Mechanism and Theory in Organic Chemistry,
3rd ed.; Harper Collins: New York, 1987; pp 231-232.
(
27) Cushman, M.; Nagatarathnam, D.; Gopal, D.; He, H.; Lin, C.
M.; Hamel, E. J . Med. Chem. 1992, 35(12), 2293.
28) Corey, E. J .; Venkateswarlu, A. J . J . Am. Chem. Soc. 1972, 94,
190.
(
6
(29) Broer, D. J .; Boven, J .; Mol. G. N.; Challa, G. Makromol. Chem.
1
989, 190, 2255.

Notes
J . Org. Chem., Vol. 64, No. 11, 1999 4193
Ta ble 2. Qu en ch in g of 1 (90 µm in Ben zen e) by Va r iou s
P r otic Com p ou n d s a n d Com m on Solven ts
rel oxidation
analyte
Ksva (M-1
)
E1/2 values (V) vs SCEb
hexanethiol
acetic acid
0.5
2.4
∼1.82
1.58
anisole
DMSO
acetone
0.2
1.0
0.2
0.3
2
-butanone
a
Measured in benzene solution. b Derived from tables of re-
ported oxidation E1/2 values measured in acetonitrile. See refs 32,
6, and 37. It was not possible to obtain oxidation potentials of
3
the compounds in benzene solution.
amines are effectively nonacidic (pK
a
≈ 35). Thus, phenols
can partially donate a proton (e.g., hydrogen bond) to the
basic pyridine nitrogen on the fluorophore. This binding
of the phenolic hydrogen by the pyridine moiety of the
fluorophore will lead to more partial negative charge
character on the phenol oxygen and thus lower the
effective oxidation potential. It has been shown that
F igu r e 2. Stern-Volmer plots of quenching of poly(1-co-
EGDMA), cyclohexane solutions: 0, phenol; 4, ethanol; +,
aniline; ×, cyclohexylamine.
Ta ble 3. Qu en ch in g of P oly(1-co-EGDMA) Bea d s
Im m er sed in Cycloh exa n e
analyte
aniline
N-methylaniline
N,N-dimethylaniline
indole
Ksv (M-1)
deprotonation of phenols lowers their oxidation potentials
by 0.5-0.7 V.34 Consequently, phenols act as better
70
33
21
51
0
oxidative quenchers than would be expected from their
neutral oxidation potentials.
To determine whether protonation or hydrogen bond-
ing alone can lead to quenching of 1, a series of control
cyclohexylamine
phenol
15
16
1.0
0
a
experiments with protic species of differing pK values
m-cresol
methanol
ethanol
and hydrogen-bonding abilities were tested (Table 2).
Hexanethiol, which is of comparable acidity to phenol but
3
5
a poor hydrogen-bond donor with a much higher oxida-
tion potential,3
2,36
showed negligible quenching. The
solvents that are miscible with benzene (e.g., acetone,
DMSO, 2-butanone) were also tested and found to be
relatively nonquenching with respect to 1.
relatively low Ksv values measured for the aliphatic
3
5
alcohols, which are good hydrogen-bond donors but not
very acidic, suggest that hydrogen bonding is only a very
minor contributor to quenching. Acetic acid, a much
To demonstrate that 1 could be used in a solid-state
fluorescent sensor, 1 (0.6 wt %) was cross-linked into
macroporous poly(EGDMA) beads (30 µm). The beads
were then adhered onto quartz slides using a thin film
of transparent cyanoacrylate resin (i.e., SuperGlue) for
systematic solid-state fluorimetry studies (Table 3). It
was not possible to use benzene for these studies because
benzene tended to swell the glue resin, causing the beads
to detach from the plates. Instead, the initial solid-state
studies were performed primarily with the adhered
fluorescent beads immersed in cyclohexane. Later studies
were performed with the beads held on to the end of a
fiber optic bundle. Similar Ksv values were found with
the slide and the fiber optic methods.
stronger acid (pK
donor than the phenols, was found to be a weak
a
≈ 4.7) and a better hydrogen-bond
35
-
1
quencher with a slightly larger Ksv (2.4 M ) than the
aliphatic alcohols. It should be noted that protonated and
deprotonated acetic acid have approximately the same
oxidation potential,37 which is much higher than that of
the other effective quenchers. In fact, it is comparable to
the oxidation potential of nonquenchers such as butyl-
amine. This suggests that for acetic acid the slight
quenching observed likely arises from protonation of the
fluorophore rather than electron transfer from the quench-
er. In contrast, the strong quenching shown by N,N-
dimethylaniline, which has a low oxidation potential30
and no labile protons, demonstrates that electron transfer
is a much more effective quenching mechanism than
proton transfer or hydrogen bonding. Thus, it appears
that the mechanism for the selective quenching of 1 is
primarily dependent on the ease of oxidation of the
quenching species, with proton donation from sufficiently
acidic analytes providing a small degree of quenching.
To establish conditions under which the fluorophore
might be used for sensor applications, common organic
As can be seen from Table 3 and Figure 2, the
quenching trends observed with 1 in solution (Figure 1)
also hold for 1 in the cross-linked polymer, showing again
the selective quenching of 1 with phenols and anilines
over aliphatic alcohols and amines. Comparison of the
quenching of 1 dissolved in benzene and of cross-linked
1
immersed in benzene shows that the response in the
poly(EGDMA) matrix is much lower than that in solution
but still quite detectable. Furthermore, the response of
fluorescent polymer to the marginally quenching or
nonquenching analytes drops to zero; thus the analyte
selectivity of 1 is enhanced. In the case of the anilines,
the fluorescent polymer shows increased selectivity by
being able to better distinguish between aniline, N-
methylaniline, and N,N-dimethylaniline. The trend in the
(
(
(
34) Weinberg N. L.; Weinberg H. R. Chem. Rev. 1968, 68, 449.
35) Epshtein, L. M. Russ. Chem. Rev. 1979, 48(9), 854.
36) Chambers, J . Q. In Encyclopedia of Electrochemistry of the
Elements, Organic Section; Bard, A. J ., Lund, H., Eds.; Marcel
Dekker: New York, 1978; Vol. XII, Chapter XII-3, p 415.
(37) Eberson, L.; Nyberg, K. In Encyclopedia of Electrochemistry of
K
sv values of the three aniline derivatives for the fluo-
the Elements, Organic Section; Bard, A. J ., Lund, H., Eds.; Marcel
Dekker: New York, 1978; Vol. XII, Chapter XII-2, p 266.
rophore in the cross-linked polymer was found to be

4
194 J . Org. Chem., Vol. 64, No. 11, 1999
Notes
Ta ble 4. Qu en ch in g of P oly(1-co-EGDMA) Bea d s by
fluorescence detection appears to be primarily based on
the ease of electron transfer from the organic analyte,
with partial proton donation from sufficiently acidic
analytes providing a small degree of quenching. Incor-
poration of the fluorescent monomer into a polymer
allows for fluorescence detection in a wide variety of
solvents including alcohol/water. The ability to sense in
aqueous and protic environments makes this fluorophore/
polymer combination potentially useful for a wide range
of applications from probing of contaminated wastewater
to monitoring of biological systems. Ongoing research
with 1 includes incorporation into an imprinted polymer
matrix to further increase analyte selectivity.
P h en ols in Differ en t Solven ts
analyte
solvent
benzene
cyclohexane
1:1 ethanol/water
Ksv (M-1)
phenol
phenol
phenol
5.0
15
3.0
m-cresol
m-cresol
cyclohexane
ethanol
17
3.0
opposite to that found with 1 in solution and also the
opposite to that predicted by their oxidation potentials
according to the Rehm-Weller equation.33 As shown in
the first three entries of Table 3, aniline was found to be
a more efficient quencher of 1 in the polymer than either
N-methylaniline or N,N-dimethylaniline; however, N,N-
dimethylaniline is more easily oxidized than either
N-methylaniline or aniline.30 This reversed trend may be
due to the more substituted, and thus sterically hindered,
amine groups being less able to approach the fluorophore
in the micropores of the cross-linked network, thereby
reducing the efficiency of electron transfer. The fluores-
cence quenching of 1 in the polymer matrix was found
to be reversible upon exposure to fresh solvent, and the
recovered polymer beads can be used for analyte detec-
tion with sensitivity comparable to that of fresh beads.
The nature of the solvent was found to have a strong
effect on the quenching behavior of 1 in the polymer. The
quenching of the fluorescence from the polymer beads by
phenols was studied in a series of solvents including
benzene, cyclohexane, ethanol, and 1:1 ethanol/water
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All synthetic manipulations were
performed using standard Schlenk line techniques. Nitrogen was
purified by passage through columns of Q-5 catalyst (Engelhard)
and 13X molecular sieves (Aldrich). All reactions were performed
under nitrogen flush. All reagents and anhydrous N,N-dimeth-
ylformamide (ACS certified) were purchased from the Aldrich
Chemical Co., with the exception of imidazole, which purchased
from Fisher Scientific. Tetrahydrofuran (THF) was purchased
from Fisher and distilled from sodium/benzophenone ketyl prior
to use. Reaction mixtures and chromatography fractions were
monitored with EM Science 250 µm silica gel F254 plates. Column
chromatography was performed using 40 µm silica gel purchased
from J . T Baker and Optima grade solvents from Fisher
Scientific. Fluorescence quenching experiments were performed
in air, using ACS Spectroanalyzed grade benzene and cyclohex-
ane purchased from Fisher. Analytes tested as quenchers were
all reagent grade or higher. 1H and C NMR spectra were
13
(
Table 4). With the same analyte, the Ksv values of the
6
acquired at 500 and 125 MHz, respectively, using acetone-d as
poly(1-co-EGDMA) beads immersed in cyclohexane are
approximately three times larger than those observed in
ethanol or benzene and five to eight times greater than
those observed in ethanol/water. The initial reductions
in absolute fluorescence intensity upon immersing the
polymer in pure cyclohexane and in pure water were
comparable, however (80% and 84% drops, respectively).
One possible explanation for this phenomenon is that
different solvents may be able to penetrate and swell the
cross-linked poly(EGDMA) matrix of the beads better
than others. In polymer chemistry, it is well-known that
different solvents can swell polymer gels to different
degrees. This phenomenon would allow some solvents to
cause more dramatic changes in the local microenviron-
ment around the immobilized fluorophores than others
or to transport quenchers more efficiently to the fluoro-
phores than other solvents. Preliminary photophysical
studies also revealed that the fluorescence lifetime of 1
changes depending on the environment around the
fluorophore. For example, the fluorescence lifetime of 1
was found to be 1.45 ns in 90 µm benzene solution, 2.86
ns in the dry cross-linked polymer, and 2.49 ns when the
polymer beads were wetted by benzene. This is a reflec-
tion of the sensitivity of the excited state of the fluoro-
phore to its microenvironment. Unfortunately, monomer
the solvent. FT-IR spectra were taken on compounds cast from
chloroform solutions onto NaCl plates. UV-vis absorption
spectra were taken of the compounds as dilute benzene solutions.
Fluorescence studies were performed using an ISA Spex Fluo-
romax 2 photon-counting fluorimeter equipped with a magnetic
stirrer in the sample compartment and a fiber optic attachment.
Fluorescence lifetime measurements were performed at the
Applications Laboratory of ISA Spex, Inc., using a Fluorolog-
Tau3 spectrometer equipped with a double-grating monochro-
mator for excitation and a single-grating emission monochro-
mator. High-resolution and low-resolution mass spectral analyses
were performed at the Mass Spectrometry Facility in the
Chemistry Department at the University of California, Berkeley.
2
,6-Bis(br om om eth yl)p yr id in e (2).24 A solution of 2,6-
pyridinedimethanol (1.50 g, 10.8 mmol) in 30 wt % hydrobromic
acid in acetic acid (22 mL) was heated at 100 °C for 1.5 h and
then poured onto ice (20 mL) and neutralized with aqueous 1 M
NaOH. The resulting precipitate was collected and recrystallized
from 1:1 hexane/ethyl acetate to give 1.72 g of product (62%
yield). 1H NMR shifts were consistent with values previously
reported.24 C NMR: δ 156.97, 138.26, 122.85, 33.57.
13
2
5
2
,6-Bis(dieth oxyph osph or ylm eth yl)pyr idin e (3). A flask
was charged with 2 (6.00 g, 22.6 mmol) and triethyl phosphite
8.28 g, 49.8 mmol) and fitted with a Dean-Stark trap and
(
condenser. The solution was heated at 110 °C for 4 h while ethyl
bromide was generated. After being cooled to ambient temper-
ature, the reaction mixture was concentrated in vacuo to give
8
.80 g (85% yield) of 3 as a viscous orange oil. (A portion was
purified for analysis by flash chromatography using an increas-
1
is not soluble in cyclohexane or in ethanol/water, so it
ing gradient of 2-propanol (3-10%) in chloroform as the eluent.)
was not possible to determine whether this solvent
dependence of quenching sensitivity also holds true in
solution.
In summary, a novel polymerizable fluorophore has
been synthesized that is highly selective for the detection
of phenols and anilines. The selectivity of this fluorescent
probe is comparable in both solution and the solid
polymer. Sensing can be performed in the presence of air
and ambient moisture. The mechanism for the selective
1
H NMR: δ 7.65 (d, 1H), 7.29 (d, 2H), 4.03(q, 8H), 3.35 (d, 4H),
1.22 (t, 12H). 13C NMR: δ 152.92, 136.47, 122.06, 61.44 (d, J C-P
) 26 Hz), 35.96 (d, J C-P ) 535 Hz), 15.71.
4
-ter t-Bu t yld im et h ylsilyloxy-3,5-d im et h oxyb en za ld e-
2
7
h yd e (4). To a solution of syringaldehyde (7.10 g, 39.0 mmol)
and imidazole (6.63 g, 97.4 mmol) in anhydrous DMF (30 mL)
at 0 °C was added tert-butyldimethylsilyl chloride (8.81 g, 58.5
mmol) in dry DMF (15 mL). The reaction mixture was then
heated at 60 °C for 4 h. After being quenched with water, the
mixture was extracted into diethyl ether, washed with saturated

Notes
J . Org. Chem., Vol. 64, No. 11, 1999 4195
aqueous sodium bicarbonate, and dried over anhydrous Na
2
SO
4
.
mixture was stirred at 65 °C for 12 h. The resulting glassy
Removal of the solvent in vacuo afforded 9.20 g (90% yield) of
3
polymer was ground by hand, passed as a CHCl suspension
through a 30 µm nylon mesh, and then dried in vacuo at 70 °C
for 24 h.
F lu or escen ce Qu en ch in g Exp er im en ts. Fluorescence data
were collected with the fluorophore excited at 310 and at 370
nm. Quenching behavior was not dependent upon the excitation
wavelength.
1
pure 4 as an off-white, crystalline solid. H NMR shifts were
2
7 13
consistent with values reported in the literature.
91.52, 171.17, 152.97, 131.00, 107.35, 56.30, 26.17, 1.45, -4.25.
,6-Bis(2-(4-h yd r oxy-3,5-d im et h oxyp h en yl)vin yl)p yr i-
C NMR: δ
1
2
d in e (5). To a nitrogen-filled flask was charged a 1 M THF
solution of lithium tert-butoxide (32.6 mL, 32.6 mmol) and dry
THF (20 mL). After the solution was cooled to 0 °C, crude 3 (6.17
g, 16.3 mmol) in dry THF (20 mL) was then added dropwise,
and the reaction mixture was stirred for 0.5 h at 0 °C. To this
mixture was then added a solution of 4 (9.20 g, 31.0 mmol) in
dry THF (30 mL). The resulting reaction mixture was stirred
at room temperature for 6 h and then cooled to 0 °C, upon which
time a solution of tetrabutylammonium fluoride (12.60 g, 40.0
mmol) in dry THF (50 mL) was added dropwise. The resulting
bright red solution was subsequently stirred at room tempera-
ture for 2 h and then quenched with aqueous 1 M HCl (100 mL).
The mixture was neutralized (pH 7) with aqueous 1 M NaOH
and extracted with dichloromethane (3 × 75 mL). The combined
organic layers were washed with water, dried over anhydrous
Solution measurements were performed on 90 µM benzene
solutions of 1 in quartz cuvettes containing a micro stirbar.
Analytes were added to the stirred cuvettes via syringe. The
fluorescence intensity at 395 nm was recorded 5-10 min after
each successive addition. Fluorescence measurements were
taken in the absence of analytes and then after analyte addition.
Solid-state quenching experiments were performed on 30 µm
beads of poly(1-co-EGDMA) adhered onto quartz slides with
SuperGlue. Each slide was wedged diagonally inside a quartz
cuvette in the fluorimeter sample chamber to ensure fixed
positions of the beads with respect to the detector. For each
experiment, the cuvette was filled with 2.5 mL of solvent and
stirred with a micro stirbar. Analyte addition and fluorescence
monitoring were performed as before. The observed fluorescence
behavior of the polymer beads was verified by later studies using
beads held to the tip of a fiber optic bundle by tissue paper and
a wire holder. Fluorescence measurements were taken with the
fiber tip immersed in pure solvent and then in stirred analyte
solutions of increasing concentration, allowing 10 min for
equilibration prior to each measurement.
Na
lization of the crude product from ethanol gave 6.30 g of pure 5
55% yield) as a yellow-green solid. Mp: 115-116 °C dec. 1
NMR: δ 7.73 (d, 2H), 7.68 (t, 1H), 7.29 (d, 2H), 7.17 (d, 2H),
2 4
SO , and concentrated under reduced pressure. Recrystal-
(
H
1
3
7
1
3
4
4
.00 (s, 4H), 3.90 (s, 12H), 2.8 (broad s, 2H). C NMR: δ 155.58,
47.98, 136.83, 132.92, 127.78, 126.75, 104.74, 55.72. IR (cm-1):
425, 1637, 1602, 1514, 1451, 1334, 1110, 750. EI LRMS: m/z
35, 167. EI HRMS: calcd for C25
35.1689.
H
25NO
6
m/z 435.1682, found
The Stern-Volmer quenching constants, Ksv, of the analytes
were calculated by plotting the ratio of the initial fluorescence
2
,6-Bis(2-(4-a cr yloyloxy-3,5-d im eth oxyp h en yl)vin yl)p yr -
intensity (I
0
) at the emission maximum over the observed
id in e (1). A solution of acryloyl chloride (0.45 g, 5.5 mmol) in
dry THF (5 mL) was added dropwise to a flask containing a
solution of 5 (1.00 g, 2.3 mmol) and triethylamine (0.96 mL, 6.9
mmol) in dry THF (30 mL) cooled to 0 °C. The resulting reaction
mixture was then stirred at room temperature for 1 h, cooled to
fluorescence intensity in the presence of the added analyte (I),
7
,8
as a function of increasing analyte concentration.
Linear
regression gives the Ksv value as the slope.
F lu or escen ce Lifetim e Mea su r em en ts. Fluorescence life-
times of 1 in benzene solution (90 µm) and cross-linked in the
polymer beads (dry and wetted by benzene) were measured by
the Applications Laboratory at ISA Spex, Inc. The data were
well described by a double-exponential decay law.
0
°C, and quenched with 0.2 M aqueous HCl. After the solution
was adjusted to pH 7 with aqueous 1 M NaOH, it was extracted
with CH Cl
(3 × 30 mL). The combined organic layers were
then washed with water, dried over anhydrous Na SO , and
2
2
2
4
concentrated under reduced pressure. The crude product was
purified by flash chromatography on silica gel using 1:1 ethyl
acetate/hexane as the eluent to afford 0.80 g (64%) of 5 as a
Ack n ow led gm en t. This research was supported
primarily by a Cottrell Scholars Award of Research
Corp. (CS0376). Additional support was provided by
research gifts from the Raychem Corp. and the 3M Co.
The authors are indebted to Dr. Salvatore Atzeni of ISA
Spex, Inc., for measuring the fluorescence lifetimes of
1
yellow, crystalline solid. Mp: 110-111 °C (polym). H NMR: δ
7
6
.83 (d, 2H), 7.76 (t, 1H), 7.39 (d, 2H), 7.37 (d, 2H), 7.08 (s, 4H),
.54 (d of d, 2H), 6.42 (d, 2H), 6.08 (d of d, 2H), 3.89 (s, 12H).
C NMR: δ 164.03, 156.24, 153.59, 138.20, 136.40, 133.44,
32.95, 129.71, 129.64, 128.65, 121.85, 104.79, 56.63. FT-IR
1
3
1
in benzene and in the cross-linked polymer beads. We
1
-
1
(
cm ): 1743, 1631, 1594, 1452, 1249, 1130, 977, 806. EI
LRMS: m/z 543, 489, 435. EI HRMS: calcd for C31 m/z
43.1893, found 543.1892. Anal. Calcd for C31 N: C, 68.50;
are also grateful to Dr. Olivier Buriez and Mr. Benjamin
J . Gross for measuring the oxidation potential of 1 in
CH3CN.
H
29NO
8
5
29 8
H O
H, 5.38; N, 2.58. Found: C, 68.15; H, 5.68; N, 2.57.
Syn th esis of Cr oss-Lin k ed P oly(1-co-EGDMA) Bea d s. A
Su p p or t in g In for m a t ion Ava ila b le: 500 MHz 1H and
vial containing 1 (0.3 mg, 0.6 µmol) and CHCl
3
(0.55 mL) was
13
1
25 MHz C NMR spectra of compounds 1 and 5; reported
charged with ethylene glycol dimethacrylate (495.0 mg, 2.50
oxidation E1/2 values of the analytes, and the measured
oxidation E1/2 value of 1 in CH CN. This material is available
mmol), 2,2′-azobisisobutyronitrile (5.0 mg, 0.030 mmol), and
CHCl (0.45 mL). The mixture was then cooled to at 0 °C and
3
agitated by ultrasound while being purged with nitrogen for 5
min. The vial was sealed under nitrogen, and the polymerization
3
free of charge via the Internet at http://pubs.acs.org.
J O982105E