Dendrimers functionalized with a single pyrene label: Synthesis, photophysics, and fluorescence quenching

Article abstract of DOI:10.1021/jp020862h

We have used a wide variety of molecular (i.e., nitromethane, acrylamide, N,N'-dimethylaniline, and methyl iodide) and ionic (iodide and cupric ions) quenchers to assess the relative structural permeabilities of a single, pyrenyl residue attached to the tertiary amine within a series of asymmetric poly(amido) dendrimers possessing carboxylate moieties at their periphery. From these quenching experiments, chain segmental densities and pyrene accessibility are probed as a function of dendrimer generation number (Pn, n = 1, 2, or 3), providing insight into the roles of size and electrostatics in this process. With the exception of dendrimer quenching by Cu²⁺, we observe classic Stern-Volmer behavior for Pn fluorescence quenching by all quenching agents. The recovered Stern-Volmer quenching constants (K_sv) and bimolecular quenching rates (k_q) generally decrease as n increases. This result is explained by a blocking of the pyrenyl residue by the growing dendrimer network. The decrease is particularly dramatic for the anionic heavy atom quencher I^-. This observation is rationalized in terms of pronounced electrostatic repulsion between the I^- quencher and the terminal COO^- residues of the dendrimer combined with an increase in the molecular network density surrounding the pyrenyl moiety as n increases. The Cu²⁺ quenching of the dendrimers is inconsistent with a diffusion-controlled reaction. Binding between the dendrimer and the Cu²⁺ is demonstrated.

Full text of DOI:10.1021/jp020862h

J. Phys. Chem. B 2002, 106, 8649-8656
8649
Dendrimers Functionalized with a Single Pyrene Label: Synthesis, Photophysics, and
Fluorescence Quenching
†
†
†
†
Claudia M. Cardona, Tom Wilkes, Winston Ong, Angel E. Kaifer,
‡
§
¶
¶
Tracy Donovan McCarley, Siddarth Pandey, Gary A. Baker, Maureen N. Kane,
Sheila N. Baker, and Frank V. Bright *
¶
¶,
Center for Supramolecular Science and Department of Chemistry, UniVersity of Miami,
Coral Gables, Florida 33124-0431, Choppin Laboratories of Chemistry, Louisiana State UniVersity,
Baton Rouge, Lousiana 70803-1804, Department of Chemistry, New Mexico Institute of Technology,
Socorro, New Mexico 87801, and Department of Chemistry, Natural Sciences Complex, UniVersity at Buffalo,
The State UniVersity of New York, Buffalo, New York 14260-3000
ReceiVed: April 2, 2002; In Final Form: June 10, 2002
We have used a wide variety of molecular (i.e., nitromethane, acrylamide, N,N′-dimethylaniline, and methyl
iodide) and ionic (iodide and cupric ions) quenchers to assess the relative structural permeabilities of a single,
pyrenyl residue attached to the tertiary amine within a series of asymmetric poly(amido) dendrimers possessing
carboxylate moieties at their periphery. From these quenching experiments, chain segmental densities and
pyrene accessibility are probed as a function of dendrimer generation number (Pn, n ) 1, 2, or 3), providing
insight into the roles of size and electrostatics in this process. With the exception of dendrimer quenching by
2+
Cu , we observe classic Stern-Volmer behavior for Pn fluorescence quenching by all quenching agents.
The recovered Stern-Volmer quenching constants (KSV) and bimolecular quenching rates (k
q
) generally decrease
as n increases. This result is explained by a blocking of the pyrenyl residue by the growing dendrimer network.
-
The decrease is particularly dramatic for the anionic heavy atom quencher I . This observation is rationalized
-
-
in terms of pronounced electrostatic repulsion between the I quencher and the terminal COO residues of
the dendrimer combined with an increase in the molecular network density surrounding the pyrenyl moiety
as n increases. The Cu2 quenching of the dendrimers is inconsistent with a diffusion-controlled reaction.
+
2
+
Binding between the dendrimer and the Cu is demonstrated.
Introduction
Our work in this area has focused on dendritic molecules
built through the covalent attachment of a single functional
group to dendritic building blocks inspired by the work of
Newkome and co-workers. The resulting unsymmetric den-
Research on functionalized dendrimers has flourished during
the past few years. One of the most important reasons driving
this extensive research work is that the attachment of functional
probes’ to these macromolecular structures facilitates the
1
8
drimers have the functional group or probe attached “off center”,
‘
9,10
that is, at the first branching point in these macromolecules.
detailed investigation of their properties. In this regard, pyrene
is an obvious choice as a probe because its fluorescence
properties are well-known and strongly dependent on the
In this paper, we report the synthesis of a new series of water-
soluble, pyrene-labeled dendrimers (first, second, and third
generation, see Figure 1), as well as their photophysical and
quenching properties.
2
microenvironment. Therefore, the noncovalent incorporation
of pyrene into the cavities present inside dendrimers has received
some attention, as illustrated by recent work reported by the
Experimental Section
3
Malliaris-Paleos group. A different approach relies on the
covalent attachment of pyrene residues to dendrimers. For
instance, Frank et al. investigated the properties of pyrene-
labeled arborescent polystyrenes. Baker and Crooks prepared
Reagents for Synthesis. All chemicals were reagent grade
Aldrich, Sigma, or Fisher) and were used as received. Tetra-
hydrofuran (THF) was freshly distilled over sodium. Column
chromatography was performed with Scientific Adsorbents silica
gel (63-200 µm).
(
4
and studied a series of poly(propylene imine) dendrimers
5
functionalized with multiple, peripheral pyrene groups. As we
were completing our own investigation, the groups of Shapiro
and Hanson reported the synthesis, characterization, and pho-
3
Cascade: 3-(1-Pyrenyl)-propane [1]: (3-Oxo-2-azapro-
pyldyne): Propanoic (P1). 4-(1-Pyrenyl)butanoic acid (0.206
g, 0.714 mmol) was dissolved in 3.6 mL of 0.2 M NaOH. The
water was removed under vacuum and the salt was completely
dried in a vacuum oven at 100 °C. The sodium salt was then
stirred under N2 in 50 mL of dry benzene and 10 mL of oxalyl
chloride at room temperature was added for an hour. After
complete removal of the excess reagent and the solvent under
reduced pressure, the residue was stirred with Behera’s amine8
(0.320 g, 0.771 mmol) and Et3N (0.15 mL, 1.1 mmol) in 25
6
tophysics of pyrene-labeled poly(aryl ether) monodendrons.
Fr e´ chet and co-workers have also recently reported star polymers
with a pyrene core.7
*
Corresponding author. E-mail: chefvb@acsu.buffalo.edu.
University of Miami.
Louisiana State University.
New Mexico Institute of Technology.
University at Buffalo.
†
‡
§
¶
1
0.1021/jp020862h CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/06/2002

8650 J. Phys. Chem. B, Vol. 106, No. 34, 2002
Cardona et al.
Figure 1. Chemical structures of the dendrimers studied (left) and space-filling structures after using the SYBYL force field (right).
mL of dry THF for 4 days under N2 at room temperature.
Purification of the obtained greenish-brown residue was ac-
complished by column chromatography (SiO2, 2:1 CH2Cl2/
EtOAc) to produce the lipophilic (tert-butyl ester) form of the
dendrimer (0.380 g, 78%). The obtained beige product was
hydrolyzed (0.190 g, 0.278 mmol) by stirring it in 10 mL of
and the residue was stirred with the second-generation analogue
of Behera’s amine (0.522 g, 0.362 mmol) and Et3N (0.10 mL,
9
0.72 mmol) in 25 mL of dry THF under N2 for 4 days at room
temperature. Purification entailed column chromatography (SiO2,
2:1 CH2Cl2/EtOAc) and a pale yellow powder, the lipophilic
dendrimer was obtained (0.313 g, 53%). Hydrolysis of this
product (0.211 g, 0.123 mmol) was performed as described
above and a beige powder, P2, was obtained (0.139 g, 94%).
9
6% formic acid for 15 h. After removal of the formic acid,
the residue was washed with hot hexane and filtered, and a beige
powder, P1 (0.136 g, 95%), was obtained after drying in a
vacuum oven at 100 °C overnight. Fourier transform (FT)-IR
-
1
FT-IR (KBr): 1716 [ν(CdO)] cm ; 1652 [ν(AMIDE I, Cd
-
1
-1
-1
O)] cm ; 1522 [ν(AMIDE II)] cm ; 1313 [ν(C-O)] cm .
-
1
-1
1
(
1
(
2
KBr): 1716 [ν(CdO)] cm ; 1652 [ν(AMIDE I, CdO)] cm ;
H NMR (DMSO-d6): δ 1.7-2.2 (m, CH2CH2 gen 2, 36 H);
-
1
-1 1
540 [ν(AMIDE II)] cm ; 1286 [ν(C-O)] cm . H NMR
1.9-2.1 (m, CH2CH2 gen 1, 12H); 2.188 (t, CH2, 2H); 7.228
(bs, NH, 4H); 7.9-8.5 (m, CH, pyrene, 9H); 12.049 (bs, COOH,
DMSO-d6): δ 1.75-2.20 (m, CH2CH2, 12 H); 1.971 (q, CH2,
1
3
H, J ) 7.7 Hz); 2.234 (t, CH2, 2H, J ) 7.1 Hz); 3.289 (t,
9H). C NMR (DMSO-d6): δ 27.540 (CH2); 28.090, 29.103
(CH2CH2 gen 2); 30.263, 30.707 (CH2CH2 gen 1); 32.198
(CH2); 35.680 (CH2); 56.240 [C(CH2CH2)3 gen 2]; 56.759
[C(CH2CH2)3 gen 1]; 123.578, 124.139, 124.196, 124.700,
124.780, 124.886, 126.024, 126.377, 127.151, 127.424, 127.614,
128.152, 129.207, 130.434, 130.838, 136.630 (pyrene); 171.532
(CONH gen 1); 172.200 (CONH gen 2); 174.415 (COOH).
CH2, J ) 8 Hz); 7.268 (bs, NH, 1H); 7.9-8.4 (m, pyrCH, 9H);
13
1
2.07 (bs, COOH, 3H). C NMR (DMSO-d6): δ 27.680 (CH2);
28.139, 29.080 (CH2CH2); 32.228 (CH2); 35.574 (CH2); 56.376
[C(CH2CH2)3]; 123.392, 124.097, 124.681, 124.837, 126.028,
1
26.407, 127.162, 127.378, 128.084, 129.226, 130.367, 130.826,
max
136.543 (pyrene); 172.022 (CONH); 174.392 (COOH). [λabs
,
4
-1
-1
max
4
-1
-1
nm (ꢀ, 10 M cm ), phosphate buffer pH 7]: 342.0 (3.7).
[λabs , nm (ꢀ, 10 M cm ), phosphate buffer pH 7]: 343.0
+
Matrix-assisted laser desorption ionization-time of flight (MALDI-
(3.3). MALDI-TOF MS (DHB matrix): 1204.91 (M , M calcd
1205.28); 921.68 (MH - PyrCOOH).
+
+
TOF) MS (neat): 517.856 (M , M calcd 517.58).
9
Cascade: 3-(1-Pyrenyl)-propane [1]: (3-Oxo-2-azapro-
27 Cascade: 3-(1-Pyrenyl)-propane [1]: (3-Oxo-2-aza-
propylidyne): (3-Oxo-2-azapentylidyne): (3-Oxo-2-azapen-
tylidyne): Propanoic (P3). The sodium salt of the 4-(1-
pyrenyl)-butanoic acid (0.038 g, 0.132 mmol) was again
prepared as described above. The anhydrous salt was stirred in
50 mL of dry benzene and 10 mL of oxalyl chloride for an
pylidyne): (3-Oxo-2-azapentylidyne): Propanoic (P2). 4-(1-
Pyrenyl)-butanoic acid (0.100 g, 0.347 mmol) was again
converted to the sodium salt as described above. The anhydrous
salt was stirred in 50 mL of dry benzene and 10 mL of oxalyl
chloride for an hour. The solvent was then removed completely,

Dendrimers Functionalized with Single Pyrene Label
J. Phys. Chem. B, Vol. 106, No. 34, 2002 8651
TABLE 1: Photophysical Parameters of the Pyrene-labeled Dendrimers (P1, P2, and P3) Dissolved in PB
max
abs
(nm)a
ꢀ (dm mol cm-1)b
3
-1
(ns)c
(106 s-1
k )
nr (106 s-1
dendrimer
λ
Φ
f
<τ>
o
k
r
)
4
P1
P2
P3
342.0
343.0
344.5
3.57 × 10
0.651 ( 0.003
0.698 ( 0.002
0.819 ( 0.002
128.9 ( 2.5
137.2 ( 2.5
144.0 ( 2.5
5.05 ( 0.10
5.09 ( 0.09
5.69 ( 0.10
2.71 ( 0.06
2.20 ( 0.04
1.26 ( 0.03
4
3.27 × 10
4
2.16 × 10
a
Imprecision is 1 nm. b Imprecision is <5%. c <τ> ) (ΣR
2)/(ΣR
i
τ
i
i i
τ ); for a multiexponential decay scheme.
hour. The solvent was then removed completely and the residue
was stirred with the third-generation analogue of Behera’s
amine (0.507 g, 0.112 mmol) and Et3N (0.05 mL, 0.36 mmol)
with a N2-filled flashlamp operating at 40 kHz. The fluorescence
was detected under magic angle conditions. An emission
monochromator was used for wavelength selection (λem ) 376
nm, ∆λ ) 8 nm). The typical time resolution for an experiment
was 0.47 ns/channel and we used 1024 total channels within
the multichannel analyzer. To avoid pulse pile-up, the count
rate at each detector was always <800 Hz. Data were acquired
9
in 25 mL of dry THF under N2 for 7 days at room temperature.
Purification entailed column chromatography (SiO2, 5:1 CH2-
Cl2/EtOAc, 2:1 CH2Cl2/EtOAc, and EtOAc) and a pale yellow
powder, the lipophilic dendrimer was obtained (0.156 g, 29%).
Hydrolysis of this product (0.146 g, 0.0305 mmol) was
performed as described above and a beige powder, P3, was
4
until there were at least 10 counts in the peak multichannel
analyzer channel. The excited-state fluorescence lifetimes were
recovered from the time-resolved intensity decay profiles by
using a commercially available nonlinear least-squares software
package (Globals Unlimited).
-1
obtained (0.0980 g, 98%). FT-IR (KBr): 1717 [ν(CdO)] cm ;
-
1
-1
1
1
652 [ν(AMIDE I, CdO)] cm ; 1541 [ν(AMIDE II)] cm ;
-
1 1
286 [ν(C-O)] cm . H NMR (DMSO-d6): δ 1.60-2.35 (m,
CH2CH2, 156 H); 7.2 (bs, NH, 13H) 7.9-8.4 (m, CH, pyrene,
9
2
5
1
H); 12.0 (bs, COOH, 27H). 13C NMR (DMSO-d6): δ 28.125,
9.095 (CH2CH2 gen 3); 30.252, 30.764 (CH2CH2 gen 2);
6.338 [C(CH2CH2)3 gen 3]; 56.744 [C(CH2CH2)3 gen 2];
Results and Discussion
Synthesis. The pyrene dendrimers reported here were pre-
max
4
72.532 (CONH gen 3); 175.200 (COOH). [λabs , nm (ꢀ, 10
cm ), phosphate buffer, pH 7]: 344.5 (2.2). MALDI-
TOF MS (DHB matrix): 3270.72 (MH , M calcd 3268.37);
pared by using the same general methodology used for the
-
1
-1
9
10
M
synthesis of structurally related ferrocene- and dansyl-labeled
+
dendrimers. In brief, pyrenebutyric acid was converted to its
acid chloride and reacted with the appropriate amine dendrons
to produce the lipophilic forms of the dendrimers with 3, 9, or
+
6
532.16 (dimerH ).
Materials for Photophysical and Fluorescence Quenching
2
7 tert-butyl esters on their peripheries (first, second, and third
Studies. The following chemicals were used as received: 1,4-
bis[5-phenyloxazole-2-yl)]benzene (POPOP); sodium hydrogen
phosphate heptahydrate, and sodium dihydrogen phosphate
monohydrate (J. T. Baker, Inc.); potassium iodide (99.99+%),
copper(II) sulfate pentahydrate (99.999%), nitromethane (99+%),
acrylamide (99+%), N,N′-dimethylaniline (redistilled, 99.5%),
and iodomethane (99%) (Aldrich Chemical Co.). Water was the
solvent. It was distilled in quartz and deionized by using a
Barnstead NANOpure II four-stage water purification system
generation, respectively). Their hydrolysis with formic acid led
to the corresponding hydrophilic forms having 3, 9, or 27 surface
carboxylic acid residues. These water-soluble dendrimers were
1
13
fully characterized using FT-IR, H and C NMR spectroscopic
data, and their molecular weights were verified by MALDI-
TOF mass spectrometry (see Experimental Section).
Steady-State Photophysical Studies. The absorbance maxima
max
(λabs ) and corresponding molar absorptivities (ꢀ) for den-
-
1
until the specific conductivity fell below 0.06 µS cm . Most
drimers P1, P2, and P3 (1 µM) dissolved in PB are summarized
in Table 1. A small decrease in ꢀ is evident as we progress
from dendrimer P1 to P2 followed by a more pronounced
decrease between P2 and P3. Our recovered ꢀ values are in
line with values previously reported by others for pyrene and
pyrene-labeled macromolecules dissolved in aqueous solution
experiments were performed in 100 mM phosphate buffer, pH
2+
7
.0 (PB). However, all Cu -quenching experiments were
performed in distilled-deionized water to avoid precipitation
-
37
11a
with phosphate anion [KSP ∼ 1.3 × 10
All stock solutions were refrigerated at 4 ( 2 °C in amber poly-
tetrafluoroethylene)-lined borosilicate vials (VWR Scientific).
for Cu3(PO4)2].
1
4,15
(
or organic solvents.
For example, Hara et al. report values
1
13
4
3
-1
-1
4
3
-1
-1
Instrumentation. H and C NMR spectra were recorded
of 3.30 × 10 dm mol cm and 2.13 × 10 dm mol cm
on a Varian VXR-400 spectrometer; all the chemical shift (δ)
values are reported in ppm. MALDI-TOF MS spectra were
recorded with a Bruker proFLEX III MALDI-TOF instrument.
for pyrene dissolved in acetonitrile and 2-propanol, respec-
1
4
tively. Similarly, Winnik and co-workers report an ꢀ value of
4
3
-1
-1
4.5 × 10 dm mol cm for 4-(1-pyrenyl)butanol dissolved
15
All UV-visible spectroscopic measurements were performed in
in methanol. These authors also noted that ꢀ for pyrene-labeled
2
4
3
-1
1
-cm quartz cuvettes at 22 °C. Electronic absorbance spectra
(hydroxypropyl)cellulose decreases from 4.5 × 10 dm mol
-1 4 3 -1 -1
were recorded on a Shimadzu model 2101 or a Milton Roy
model 1201 (Spectronic Instruments) UV-visible spectropho-
tometer. Static fluorescence experiments were performed with
an SLM-AMINCO model 48000 MHF fluorometer (Spectronic
Instruments). The excitation and emission spectral band-passes
were kept at 4 and 1 nm, respectively. The excitation wavelength
was set to 337 nm. All spectra were background corrected by
using appropriate blanks. Fluorescence quantum yields were
determined by using POPOP dissolved in cyclohexane (Φs )
cm to 2.5 × 10 dm mol cm when the solvent is switched
from methanol to water.
The dendrimer fluorescence quantum efficiencies (Φ ) were
1
5
f
16
determined by using the Parker-Rees method. The recovered
values are also collected in Table 1. The Φ values parallel the
f
ꢀ results and increase in the order P1 < P2 , P3. Our Φ values
f
are also well within the range of values reported by other
1
4,17,18
researchers.
Thus, although the data in Table 1 clearly
reflect the environmental influence of the dendritic substituents,
the trend is surprising for two reasons. First, one might expect
the introduction of amido functionalities near the pyrene residue
to enhance the dipolarity surrounding the pyrene residue
resulting in a lower Φf. Second, aromatic and aliphatic amines
are well-known quenchers of aromatic hydrocarbon fluores-
1
2
0
.97) as a standard. All sample solutions were freeze-pump-
1
3
thaw degassed at least four times to remove residual O2.
Time-resolved fluorescence measurements were performed
by using an IBH model 5000W SAFE time-correlated single-
photon-counting fluorometer. Samples were excited at 337 nm

8652 J. Phys. Chem. B, Vol. 106, No. 34, 2002
Cardona et al.
in deaerated water which decays with a single fluorescence
2
2
23
24
lifetime (τ) (175 ns, 200 ( 20 ns, and 200.8 ( 0.9 ns ).
Similarly, the excited-state fluorescence intensity decay of
1
-pyrene butyric acid (PBA) dissolved in toluene is single
25
exponential (τ ) 61.5 ns). However, when PBA is incorporated
within a silicone film, its intensity decay is no longer single
2
5
exponential. The Berglund group showed that the time-
resolved intensity decay of PBA-labeled hen egg-white lysozyme
is also triple exponential. Finally, the Winnik group has
reported that the time-resolved fluorescence intensity decays
from pyrene-labeled (hydroxypropyl)celluloses are multiexpo-
2
6
1
5,27
nential.
Considering the complexity of the dendrimer time-resolved
intensity decay kinetics we have opted to use the mean
fluorescence lifetimes <τ> (Table 1) to further assess the
dendrimer photophysics. Inspection of the data presented in
Table 1 shows that the trend in <τ> with dendrimer generation
parallels the Φf and ꢀ results, demonstrating that the pyrene
residue is progressively shielded from the solvent as the
dendrimer generation increases. From the Φf and <τ> data we
calculated the average radiative (kr) and nonradiative decay rates
Figure 2. Time-resolved intensity decay traces for P1, P2, and P3
dissolved in PB. The instrument response function (IRF), data (points),
and fit to triple exponential decay models (s) are shown. The
corresponding residuals are shown in the lower panels.
-
1
-1
(knr) (kr ) Φf <τ> ; knr ) kr(Φf - 1) for P1, P2, and P3
dissolved in PB (Table 1). These results reveal the underlying
reasons for the increase in Φf and, to a lesser extent <τ>, as
we proceed from P1 to P2 to P3 we note the following: (1) a
small (∼13%) increase in kr and (2) a 2-fold decrease in knr.
Thus, the pyrenyl residue’s nonradiative deactivation pathway
is influenced significantly by the dendritic matrix; however, the
radiative deactivation pathway is effectively independent of the
dendrimer generation. Our research groups observed similar
TABLE 2: Parameters Extracted from the Excited-State
Fluorescence Intensity Decay of Pyrene-labeled Dendrimers
(
P1, P2, and P3) Dissolved in PB at 376 nma
ø2
dendrimer
τ
1
τ
2
τ
3
R
1
/R
2
1
R /R
3
P1
127.0
12.8
1
1
27.7
56.4
1.0
115.0
1.011
0.398
1.97
1.02
15.6
2.45
1.10
13.3
3.40
1.06
0.3
0.4
1.0
0.146
0.469
7.068
P2
P3
131.4
1
1
34.2
45.0
6.5
69.5
4.252
4.631
10
behavior for a series of dansyl-based dendrimers.
Fluorescence Quenching of Pn. Although radiative rates for
133.2
an excited-state fluorescent species are generally insensitive to
1
1
40.9
44.8
13.8
30.9
2.443
35.394
21,28,29
its molecular environment,
the efficiencies of nonradiative
processes are strongly dependent upon perturbations caused by
the presence of external agents such as quenchers. For example,
a
All excited-state lifetimes are in nanoseconds. All <τ
i
> and R
i
/R
j
have imprecisions of 2.5 ns and 0.1, respectively.
excited-state pyrene is readily quenched by a host of molecules,
cence.19 As a recent example, Winnik et al. reported that
secondary and tertiary amino groups within the actual polymer
backbone are responsible for the observed quenching in pyrene-
labeled branched polyethylenimines dissolved in aqueous solu-
tion; however, the tertiary amines were reportedly far more
efficient quenchers.20 Gauthier and co-workers have also
reported a “backbone quenching” effect for highly branched
polystyrenes labeled with pyrene relative to the soluble pyrene
monomer 1-pyrenemethanol. The results of our experiments
demonstrate that these types of “backbone” quenching phe-
nomena are absent in our dendrimers.
Time-Resolved Fluorescence. To further assess the photo-
physical properties of these dendrimers, we recorded the time-
resolved fluorescence intensity decay traces for each dendrimer
21,28,29
ions, and atoms.
In homogeneous solutions the bimolecular
quenching rate constants (k ) that describe the pyrene/quencher
q
10
-1 -1
system approach the diffusion limit (k ∼ 10
q
M
s
in water
2
1
at room temperature) ; however, the magnitude of k and even
q
the quenching process itself can change significantly when
pyrene is sequestered within organized assemblies such as
3
0-33
micelles, vesicles, or membranes.
Thus, quencher studies
can be used to provide additional information regarding the
location or accessibility of intrinsic or extrinsic fluorophores
within a macromolecular structure or conformational fluctuations
4
3
3
therein. In addition, by using a carefully selected set of
quenching agents, one can discern the role of electrostatics and
solute (quencher) type on the pyrene residue’s accessibility as
a function of dendritic branching.
(
Figure 2). In freeze-pump-thaw degassed PB, dilute solutions
The Stern-Volmer expression (eq 1) provides a link between
the experimental measurables and the physical aspects of the
of each dendrimer (<1 µM) exhibited multiexponential intensity
decays. On the basis of an assessment of the residuals, auto
correlation functions, and ø2 values, we conclude that Pn
intensity decays are well described by a triple exponential decay
model (Figure 2, Table 2).
2
1,33
quenching process:
F /F ) 1 + K [Q] ) 1 + <k > <τ> [Q]
(1)
0
SV
q
o
The intensity decay for pyrene when it is attached to another
In this expression F0, F, KSV, <kq>, <τ>0, and [Q] are the
fluorescence intensity in the absence and presence of quencher,
Stern-Volmer quenching constant, average bimolecular quench-
ing constant, mean fluorophore lifetime in the absence of
quencher, and the analytical concentration of quencher, respec-
tively. In situations where the Stern-Volmer plots (F0/F vs [Q])
are not linear, researchers have used a variety of more complex
4,15,21-27
species is often multiexponential.
For example, Gauthier
and co-workers found that a triple exponential decay model was
required to describe the time-resolved intensity decay data for
4
pyrene-labeled arborescent polystyrene. It has also been sug-
gested that the excited-state intensity decay profiles of pyrenyl
21
derivatives bound to any macromolecule are multiexponential.
This is in striking contrast to the behavior of pyrene dissolved
2
1,33
models.

Dendrimers Functionalized with Single Pyrene Label
TABLE 3: Parameters Recovered from Fluorescence Quenching of P1, P2, and P3 by Various Quenching Agents
P1 P2 P3
<k <k <k )
> (M-1 s-1
J. Phys. Chem. B, Vol. 106, No. 34, 2002 8653
K
SV (M-1)
q
> (M-1 s-1
)
K
SV (M-1
)
q
> (M-1 s-1
)
K
SV (M-1
)
q
CH
3 2
NO
9
9
9
9
9
8
8
9
9
9
8
7
3
8
29.2 ( 11.0
38.1 ( 4.2
04.7 ( 35.5
(5.66 ( 0.14) × 10
(2.62 ( 0.06) × 10
(6.24 ( 0.30) · 109
(1.39 ( 0.17) × 10
(4.17 ( 0.21) × 10
552.3 ( 7.4
253.2 ( 3.1
520.8 ( 20.8
10.0 ( 1.4
22.1 ( 0.8
(4.03 ( 0.09) × 10
(1.85 ( 0.04) × 10
(3.80 ( 0.17) × 10
(0.73 ( 0.10) × 10
(1.61 ( 0.07) × 10
220.1 ( 9.4
71.2 ( 1.0
127.4 ( 15.1
5.1 ( 0.9
(1.53 ( 0.07) × 10
(0.49 ( 0.01) × 10
(0.88 ( 0.11) × 10
(0.35 ( 0.06) × 10
CH2 dCH CON H
2
C
6
H
5
N(CH
3
)
2
CH
3
I
8
8
1
7.9 ( 2.1
3.8 ( 2.5
KI
8
5
2.7 ( 0.1
(0.19( 0.08) × 10
Neutral Charge-Transfer Quenching. We initiate our
discussion of the Pn dendrimer quenching by using nitromethane
emission is not observed, because radical ions are formed with
high efficiency from the electron-transfer quenching.28 This
phenomenon has been explained in terms of a dramatic increase
(
CH3NO2) and acrylamide (CH2dCHCONH2), both well-
33-36
28
known, molecular charge-transfer quenchers.
During charge-
in the probability of radiationless transitions.
transfer quenching, transfer of an electron (charge) takes place
from the pyrene excited singlet state to the quencher with the
formation of a transient charge-transfer complex. The Stern-
Volmer plots for nitromethane and acrylamide quenching of the
To better assess the effects of the dendrimer structure on
quencher accessibility, we investigated the Pn dendrimer
fluorescence as a function of added DMA. In all cases, DMA
2
quenching followed Stern-Volmer behavior (r > 0.990) up
2
11b
Pn dendrimers are linear (nitromethane: r > 0.999; acryla-
to the DMA solubility limit in water. In parallel with our
2
mide: r > 0.992) and the KSV and <kq> values are compiled
results for nitromethane and acrylamide, KSV and <kq> for the
Pn/DMA system decreases as we proceed from P1 to P2 to P3
(Table 3) with a greater than 7-fold decrease in <kq> as the
dendrimer generation is increased. By way of comparison the
reported kq values for the DMA/perylene system dissolved in
in Table 3. Examination of these data shows a clear decrease
in KSV and <kq> as we proceed from P1 to P2 to P3.
Specifically, an almost 4-fold decrease in <kq> is apparent
between P1 and P3 quenched by nitromethane. Similarly, <kq>
is more than five times lower for P3 than for P1 when quenched
by acrylamide.The acrylamide bimolecular quenching rate for
P1 is less than half the diffusion-limited rate observed for
9
-1 -1
several pure liquid solvents are 17.8 × 10 M
s
(acetoni-
9
-1 -1
9
-1
trile), 8.0 × 10 M
s
(dimethylformamide), 1.9 × 10 M
-
1
9 -1 -1
9
s
(formamide), 12.2 × 10 M
s
(methanol), 1.5 × 10
-1 -1
9
-1 -1 33
-1 -1
9
tryptophan quenching in solution (5.9 × 10 M s ). In cases
where quenchers can only access the fluorescent residue from
a limited region (e.g., a fluorescently labeled protein where the
protein partially blocks access to the fluorophore), the maximum
kq is expected to be less than the value for a purely diffusion-
controlled reaction.37 Our results on Pn clearly suggest that:
M
s
(benzene), and 0.33 × 10 M
s
(methylcyclohex-
2
8
ane). Values of similar magnitude have also been reported
9
-1 -1
23
for the pyrene/DMA system [13 × 10 M
s
(acetonitrile),
9
-1 -1
44
9
-1 -1
12 × 10 M
s
(acetonitrile), and 4.5 × 10 M
s
(50:
2
3
50 water/methanol, v/v) ]. We interpret the decrease in <kq>
with increasing dendrimer generation to arise from the convolu-
tion of two factors. First, a decreased dipolarity within the
(
1) even the lowest generation dendrimer, P1, effectively
impedes quencher diffusion/access to the pyrenyl moiety and
2) the approach of neutral quenching agents toward the pyrenyl
10
cybotactic “pocket” in which the pyrenyl residue resides giving
rise to ever lower <kq> values from P1 to P2 to P3. Second,
in line with the nitromethane and acrylamide results, the
increased density afforded by the increasingly branched arboreal
chains (Figure 1) effectively protects the excited pyrenyl residue
from the approaching DMA molecules. No exciplex emission
was observed for our Pn/DMA system in water.
(
moiety is increasingly hindered as the dendrimer generation
increases. We attribute this latter feature to the greater segmental
density within the dendrimer network surrounding the pyrenyl
moiety, with successive generation, that effectively impedes
quencher access to the pyrenyl moiety (Figure 1). A similar
effect was seen for a series of dansyl-based dendrimers
Neutral Heavy-Atom Quenching. Another important class
of molecular quenching agents exhibits external heavy-atom-
1
0
complexing with cyclodextrins and anti-dansyl antibodies ;
however, the size of the quenchers used in the current study
are 10- to 1000-fold smaller than a cyclodextrin or anti-dansyl
antibody, respectively. The lower <kq> seen for acrylamide
relative to nitromethane can be rationalized by nitromethane’s
smaller van der Waals radius (1.7 vs 2.6 Å)38 and thus higher
diffusion coefficient. Finally, it is important to mention that the
4
quenching efficiencies that are roughly proportional to Z where
4
5-51
Z is the atomic number.
Quenchers of this type, as shown
by Wilkinson and co-workers, quench fluorescence by enhanc-
ing intersystem crossing by spin-orbit coupling which is
accompanied by a correspondingly enhanced triplet-state
49-51
population.
Of course, singlet-to-triplet intersystem crossing
<
kq> values observed here are all in good agreement with
may also be influenced by molecules containing heavy atoms
as exemplified by the work of Quina and Carroll on alkylhalide
quenching of polycyclic aromatic hydrocarbons.
30,39
values reported previously
for these same quenchers when
5
2
pyrene is located within microheterogeneous media and are
always less than the diffusion-controlled limits.
Electron-transfer quenching of molecular fluorescence can
occur when an electron is transferred from a donor species to
an appropriate excited-state acceptor species. The N,N′-di-
We explored the fluorescence quenching of our Pn dendri-
mers by methyliodide (CH3I), a neutral heavy-atom containing
agent. The quenching of our Pn dendrimers by CH3I is well
2
described by the Stern-Volmer expression (r > 0.982).
-
1
methylaniline (DMA)/pyrene system represents a classic
Although the KSV for P1 (17.9 M ) is similar to the value
case.2
1,23,28,40-43
In nonpolar solvents, the excited-state complex
reported for the anthracene/CH3I system dissolved in cyclo-
-1 28
(exciplex) resulting from intermolecular electron transfer from
hexane (18.8 M ), Table 3 reveals the considerable shielding
of the pyrenyl residue offered by the higher generation den-
drimers. That is, although the smallest dendrimer (P1) offers
minimal protection for the pyrenyl residue from CH3I (by far
DMA to excited-state pyrene molecule manifests itself as a
broad, low-energy, structureless band in the emission spectrum.
However, in solvents with high dielectric constants, exciplex

8654 J. Phys. Chem. B, Vol. 106, No. 34, 2002
Cardona et al.
the smallest neutral solute/quencher used in this study), higher
dendrimer generations offer a substantial diffusive barrier with
<
kq> values decreasing by a factor of about 4 between P1 and
P3. Thus, the molecular “netting” within P3 is relatively
impermeable to even small neutral quenchers/solutes such as
CH3I.
Charged Species Quenching. Because properties such as
binding, solubility, and molecular recognition rely heavily on
surface-charge effects, an understanding of such is essential to
the rational design of dendrimer-based materials as efficient
pharmaceutical vehicles, controlled release systems, reaction
media, and the like. To assess the role of electrostatics in the
molecular approach of charged quenchers/solutes toward a well-
defined site within dendrimers of varying generation, we
investigated the behavior of our Pn dendrimers in the presence
-
2+
of the quenchers iodide (I ) and cupric (Cu ) ions.
2
+
Figure 3. Stern-Volmer quenching plots for 1 µM P1 by Cu in the
Like the behavior we saw with its neutral heavy atom
2+
absence (closed symbols) and the presence of 1 mM Zn (open
-
counterpart, CH3I, the I -mediated quenching of our Pn
symbols).
2
dendrimers follows a Stern-Volmer mechanism (r > 0.981)
with KSV and <kq> that decrease with increasing generation
number (Table 3). However, while <kq> decreases by a factor
of 4 as we progress from P1 to P2 to P3 for CH3I, the <kq>
-
for I systematically decreases more than 20-fold as we progress
-
from P1 to P3. Given that I is considerably smaller than CH3I,
electrostatics clearly play the dominant role in the observed Pn
-
dendrimer quenching behavior by I . To fully appreciate the
subtleties of this process, it is important to realize that a large
fraction of the surface carboxylic groups will be ionized at pH
7
.0 (the acid dissociation constant for an isolated carboxylic
-
5 11c
acid Ka is 1.75 × 10 ). On the basis of previous work from
our groups on dansyl-based dendrimers,10 we estimate the P1,
P2, and P3 hydrodynamic radii to be 5.7, 8.3, and 12.2 Å,
respectively. Thus, considering that this series of dendrimers
contains 3 (P1), 9 (P2), and 27 (P3) carboxylic acid groups in
their peripheries, the surface-charge density is roughly twice
as high for P3 as it is for P1. At this point, it should be clear
that in addition to the increased molecular crowding of the
dendrimer chains surrounding the pyrenyl moiety (Figure 1),
Coulombic repulsion is largely responsible for protection of the
Figure 4. Stern-Volmer quenching plots (truncated quencher con-
2+
centration range) for 1 µM P2 and P3 by Cu in the absence (solid
2+
symbols) and presence of 1 mM Zn (open symbols). Note that the
plots without Zn2 are nonlinear.
+
-
excited pyrenyl residue from I within P3. Provided merely as
9
-1 -1
benchmarks, kq values of 12.0 × 10 M
s
(acetonitrile),
9
-1 -1
9
8
.0 × 10 M
s
(dimethylformamide), and 0.23 × 10
-1 -1
M s (methanol) have been reported for perylene quenching
-
28
by I .
We now turn our attention to Pn dendrimer quenching by
2
+
21
Cu , which acts as a heavy atom and electron scavenger,
both of which contribute to its quenching efficacy. (Note: It
2+
was necessary for us to perform the Cu -quenching studies in
2
+
distilled-deionized water to avoid Cu
precipitation by
Figure 5. Stern-Volmer quenching plots (full range) for 1 µM P2
phosphate (vide supra).1 However, a large fraction of the
dendrimer surface carboxylic acid groups should remain dis-
sociated under these conditions.) As a benchmark, Thomas and
1a
and P3 by Cu2 . Note the quenching saturates at the higher Cu
+
2+
concentrations and the plots are nonlinear.
23
co-workers have reported that the kq value for pyrene quenched
dendrimers actually seem to saturate at the higher concentrations
2
+
9
-1 -1
by Cu in water is 5.3 × 10 M s .
2+
of Cu investigated, and the Stern-Volmer plots are a function
2
+
The Cu quenching of our Pn dendrimers proved to be
of the dendrimer analytical concentration (Figure 6). These
2+
2+
complex (Figures 3-6). The results for the P1/Cu system were
characterized by a linear (r ) 0.999) Stern-Volmer plot (Figure
results are fully consistent with Cu binding to the Pn
2
53-57
dendrimers.
10
-1 -1
3
). However, the recovered <kq> value (9.5 × 10
M
s )
At least two possible binding sites exist for cations within
our dendrimers (Figure 1). First, one could envision electrostatic
attraction and complexation between the dendrimer carboxylates
is about 20-fold greater than the kq value for pyrene quenched
by Cu2 in water. This result suggests that charge-transfer and
heavy-atom quenching are operating in concert with one or more
+
2
+
and the Cu quenchers that could bring the quenchers into the
2+
additional quenching processes for the P1/Cu system. In P2
and P3, quenched by Cu2 (Figures 4-6), the Stern-Volmer
plots were nonlinear; the results for the higher generation
pyrenyl residue’s immediate proximity on the Pn dendrimer
+
2+
surface. A second possibility is the Cu quenchers actually
entering into and associating with parts of the dendrimer core

Dendrimers Functionalized with Single Pyrene Label
J. Phys. Chem. B, Vol. 106, No. 34, 2002 8655
up to a factor of 500 (for P3) greater than predicted for a
diffusion-controlled reaction. Thus, the addition of excess Zn2
+
(
1000-fold over the concentration of dendrimer) can decrease
2
+
the amount of Cu bound to the dendrimers, but it does not
eliminate the Cu binding. The exact nature of the Cu
2
+
2+
association to and quenching of these Pn dendrimers is unknown
and currently under investigation in our laboratories.
Conclusions
By using pyrene as a spectroscopic probe sensitive to a series
of water-soluble quenchers, we have evaluated several aspects
of the local morphology and accessibility within a family of
poly(amido) carboxylate dendrimers (Figure 1). Fluorescence
quantum yields and, to a lesser extent, excited-state fluorescence
lifetimes of the pyrenyl moiety increase and molar absorptivities
decrease as the dendrimer generation increases. This behavior
is largely a consequence of a decrease in the nonradiative decay
rate with increasing dendrimer generation. Approach of neutral
Figure 6. Effects of P3 dendrimer concentration on the Stern-Volmer
plot.
(
nitromethane, acrylamide, N,N′-dimethylaniline, and methyl
iodide) and anionic (iodide ion) fluorescence-quenching agents
toward the pyrenyl moiety is increasingly hindered with
dendrimeric generation. This effect is, however, markedly more
dramatic for the anionic quencher because electrostatic repulsion
between the anionic quencher and anionic dendrimer surface
increases as n increases. The quenching of our Pn dendrimers
2+
by Cu do not follow any of the basic quenching models, small
amounts of quencher (,1 mM) can lead to near total quenching
of the fluorescence, and the apparent quenching rates are orders
of magnitude greater than is possible for a diffusion-controlled
2
+
reaction. We attribute these results to association of the Cu
with our dendrimers. The quenching results with Cu2 also
suggest the potential of using these types of dendrimers as
amplifiers for sensing divalent metal cations that can quench
the pyrene residue fluorescence.
+
Acknowledgment. The authors thank the American Chemi-
cal Society for an Analytical Division Fellowship sponsored
by Eli Lilly (G.A.B.), the National Institutes of Health for a
Minority Graduate Fellowship (C.M.C.), the National Science
Foundation (to A.E.K., CHE-9982014; and to F.V.B., CHE-
Figure 7. Electronic absorbance spectra for 1 µM P3 in the absence
2
+
and presence of 0.2 mM Cu
.
0
078101), the Department of Energy (to F.V.B), and the
itself. It is well-known that Cu2+ has a particularly high
thermodynamic affinity for typical N-,O-, and -S-based chelate
University at Buffalo (to F.V. B.).
5
3
ligands and fast metal-to-ligand binding kinetics. In this
context Okada’s group demonstrated that Cu2 associated with
surface primary amines and internal tertiary amines in poly(l-
References and Notes
+
(1) For recent reviews, see: (a) Chow, H.-F.; Mong, T. K.-K.;
5
4
2+
Nomgrum, M. F.; Wan, V. W. Tetrahedron 1998, 54, 8543. (b) Newkome,
G. R.; He, E.; Moorefield, C. N. Chem. ReV. 1999, 99, 1689. (c) Archut,
A.; V o¨ gtle, F. Chem. Soc. ReV. 1999, 27, 233. (d) Hecht, S.; Fr e´ chet, J. M.
Angew. Chem., Int. Ed. Engl. 2001, 40, 74.
(2) (a) Kalyanasundaram, K. In Photochemistry in Organized and
Constraint Media; Ramamurthy, V., Ed.; VCH: Weinheim, 1991; p 39.
glutamic acid). Given this, it seems likely that Cu could
simultaneously coordinate within the skeleton of our dendrimers
and bind to surface carboxylates which would increase the local
concentration of Cu2+ surrounding the pyrenyl residue. The data
in Figures 3-6 are consistent with these scenarios. Additional
evidence for association/complexation is found in the dendrimer
electronic absorbance spectra in the absence and presence of
(
b) Carmichael, I.; Hug, G. L. Handbook of Photochemistry, 2nd ed.; Marcel
Dekker: New York, 1993.
3) (a) Pistolis, G.; Malliaris, A.; Paleos, C. M.; Tsiourvas, D. Langmuir
997, 13, 5870. (b) Pistolis, G.; Malliaris, G.; Tiourvas, D.; Paleos, C. M.
(
1
2
+
Cu (Figure 7) where we note shifts in the absorbance spectra
consistent with a static, ground-state complex.
Chem. Eur. J. 1999, 5, 1440.
(4) Frank, R. S.; Merkle, G.; Gauthier, M. Macromolecules 1997, 30,
5397.
In an electrolyte, Coulombic interactions can be screened by
(
(
5) Baker, L. A.; Crooks, R. M. Macromolecules 2000, 33, 9034.
6) Riley, J. M.; Alkan, S.; Chen, A.; Shapiro, M.; Khan, W. A.;
58
free ions. Thus, addition of an excess of a second divalent
cation (e.g., Zn2 added as ZnSO4) to our system should screen
the surface charge on the dendrimer resulting in reduced
+
Murphy, W. R.; Hanson, J. E. Macromolecules 2001, 34, 1797.
7) Hecht, S.; Vladimirov, N.; Fr e´ chet, J. M. J. J. Am. Chem. Soc. 2001,
23, 18.
8) Newkome, G. R.; Behera, R. K.; Moorefield, C. N.; Baker, G. R.
(
1
2+
electrostatic attraction by Cu and concomitantly reduced
(
2
+
28
quenching (Zn itself is not a molecular quencher of pyrene ).
The results in Figures 3 and 4 show that addition of a 1000:1
J. Org. Chem. 1991, 26, 4376.
(9) Cardona, C. M.; McCarley, T. D.; Kaifer, A. E. J. Org. Chem.
2000, 65, 1857.
2
+
2+
mole ratio of Zn :dendrimer reduces the extent of Cu
quenching. However, the recovered <kq> values, even in the
presence of 1 mM Zn , are at least a factor of 10 (for P1) and
(10) Cardona, C. M.; Alvarez, J.; Kaifer, A. E.; McCarley, T. D.; Pandey,
S.; Baker, G. A.; Bonzagni, N. J.; Bright, F. V. J. Am. Chem. Soc. 2000,
122, 6139.
2+

8656 J. Phys. Chem. B, Vol. 106, No. 34, 2002
Cardona et al.
(
11) Dean, J. A. Lange’s Handbook of Chemistry, 14th ed.; McGraw-
(36) Acree, W. E., Jr.; Pandey, S.; Tucker, S. A.; Fetzer, J. C. Polycyclic
Aromat. Compd. 1997, 12, 71.
Hill: New York, 1992; (a) Table 8.6, (b) Table 1.15, (c) Table 8.8.
(
(
(
(
12) Mardelli, M.; Olmsted, J. J. Photochem. 1977, 7, 277-285.
13) Lakowicz, J. R.; Weber, G. Biochemistry 1973, 12, 4161.14.
14) Hara, K.; Ware, W. R. Chem. Phys. 1980, 51, 61.
(37) Johnson, D. A.; Yguerabide, J. Biophys. J. 1985, 48, 949.
(38) Semiempirical calculations were carried out with HyperChem 4.5
(Hypercube Inc.) using PM3-optimized geometries.
15) Winnik, F. M.; Winnik, M. A.; Tazuke, S.; Ober, C. K. Macro-
(39) Thomas, J. K. Chem. ReV. 1980, 80, 283; see also references therein.
molecules 1987, 20, 38.
(40) Okamoto, M. J. Phys. Chem. A. 2000, 104, 5029; see also references
(
16) Parker, C. A.; Rees, W. T. Analyst 1962, 87, 83.
therein.
(17) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochem-
(
49.
41) Herbrich, R. P.; Schmidt, R. J. Photochem. Photobiol. A 2000, 133,
istry, 2nd ed.; Marcel Dekker: New York, 1993; p 44.
1
(18) Karpovich, D. S.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 3951.
(19) Naciri, J.; Weiss, R. G. Macromolecules 1989, 22, 3928.
(20) Winnik, M. A.; Bystryak, S. M.; Liu, Z.; Siddiqui, J. Macromol-
(
(
42) Mataga, N.; Ezumi, K.; Okada, T. Mol. Phys. 1966, 10, 203.
43) Schomburg, H.; Staerk, H.; Weller, A. Chem. Phys. Lett. 1973,
22, 1.
ecules 1998, 31, 6855.
21) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: New York, 1999.
22) Vethamuthu, M. S.; Almgren, M.; Mukhtar, E.; Bahadur, P.
Langmuir 1992, 8, 2396.
23) Hashimoto, S.; Thomas, J. K. J. Colloid Interface Sci. 1984, 102,
52.
24) De Feyter, S.; van Stam, J.; Boens, N.; De Schryver, F. C. Chem.
Phys. Lett. 1996, 249, 46.
(
44) Taylor, G. N.; Chandross, E. A.; Shiebel, A. H. J. Am. Chem. Soc.
(
1
974, 96, 2693.
(45) Kasha, M. J. Chem. Phys. 1952, 20, 71.
(
(46) McGlynn, S. P.; Reynolds, M. J.; Daigre, G. W.; Christodouleas,
N. D. J. Phys. Chem. 1962, 66, 2499.
47) McGlynn, S. P.; Azumi, T.; Kinoshita, M. Molecular Spectroscopy
of the Triplet State; Prentice Hall: Englewood Cliffs, NJ, 1969.
48) McGlynn, S. P.; Azumi, T.; Kasha, M. J. Chem. Phys. 1964, 40,
07.
(
(
1
(
(
5
(
(
(
25) Ishiji, T.; Kaneko, M. Analyst 1995, 120, 1633.
26) Pan, B.; Berglund, K. A. J. Crystal Growth 1997, 171, 226.
27) Yamazaki, I.; Winnik, F. M.; Winnik, M. A.; Tazuke, S. J. Phys.
(
(
49) Medinger, T.; Wilkinson, F. Trans. Faraday Soc. 1965, 61, 620.
50) Horrocks, A. R.; Wilkinson, F. Proc. R. Soc. London Ser. A 1968,
3
06, 257.
Chem. 1987, 91, 4213.
28) Birks, J. B. Photophysics of Aromatic Molecules; Wiley-Inter-
science: London, 1970.
29) Turro, N. J. Modern Molecular Photochemistry; University Science
Books: Sausalito, CA, 1991.
30) Chen, T. S.; Thomas, J. K. J. Polym. Sci.: Polym. Chem. Ed. 1979,
7, 1103.
31) Gratzel, M.; Thomas, J. K. In Modern Fluorescence Spectroscopy;
Wehry, E. L., Ed.; Plenum Press: New York, 1974; Vol. 2.
(
(
(
51) Wilkinson, F. AdV. Photochem. 1964, 3, 241.
(
52) Quina, F. H.; Carroll, F. A. J. Am. Chem. Soc. 1976, 98, 6.
53) For copper ion complexation by poly(amidoamine) dendrimers,
(
see: (a) Zhou, L.; Russell, D. H.; Zhao, M.; Crooks, R. M. Macromolecules
2
1
1
1
001, 34, 3567. (b) Zhao, M.; Sun, L.; Crooks, R. M. J. Am. Chem. Soc.
998, 120, 4877. (c) Zhao, M.; Crooks, R. M. Angew. Chem., Int. Ed. Engl.
999, 38, 364. (d) Balogh, L.; Tomalia, D. A. J. Am. Chem. Soc. 1998,
20, 7355. (e) Swager, T. M. Acc. Chem. Res. 1998, 31, 201.
(
1
(
(54) Tsutsumiuchi, K.; Aoi, K.; Okada, M. Polym. J. 2000, 32, 107.
(
32) Thomas, J. K. Acc. Chem. Res. 1977, 10, 133.
(33) Eftink, M. R. In Topics in Fluorescence Spectroscopy; Lakowicz,
(55) Wang, J.; Wang, D.; Miller, E. K.; Moses, D.; Bazan, G. C.; Heeger,
A. J. Macromolecules 2000, 33, 5153.
(56) Blatt, E.; Sawyer, W. H. Biochim. Biophys. Acta 1985, 822, 43.
(57) Blatt, E.; Chatelier, R. C.; Sawyer, W. H. Biophys. J. 1986, 50,
349.
J. R., Ed.; Plenum Press: New York, 1991; Vol. 2; see also references
therein.
(
34) Acree, W. E., Jr.; Pandey, S.; Tucker, S. A. Curr. Top. Solution
Chem. 1997, 2, 1.
35) Zander, M. In Chemical Analysis of Polycyclic Aromatic Com-
pounds; Vo-Dinh, T., Ed.; John Wiley: New York, 1989.
(
(58) Israelachvili, J. In Intermolecular & Surface Forces, 2nd ed.;
Academic Press Inc.: San Diego, 1991.