Dendrimer disassembly by benzyl ether depolymerization

Article abstract of DOI:10.1021/ja0349960

The disassembly of dendritic structures was realized by a cascade cleavage reaction triggered by an initially stimulated group in the dendrimer periphery. A depolymerizable backbone was engineered into prototypical dendritic structures. Evidence for the completion of the disassembly process is provided by the absorbance peak of the p-nitrophenoxide ion that was intentionally installed at the focal point of the dendrons. Observation of the UV spectra during the disassembly process supports a stepwise cascade cleavage proceeding from the periphery into the core. Copyright

Full text of DOI:10.1021/ja0349960

Published on Web 08/09/2003
Dendrimer Disassembly by Benzyl Ether Depolymerization
Sheng Li, Michael L. Szalai, Robert M. Kevwitch, and Dominic V. McGrath*
Department of Chemistry, UniVersity of Arizona, Tucson, Arizona 85721
Received March 5, 2003; E-mail: mcgrath@u.arizona.edu
Chart 1
Degradable polymeric systems are the basis of several established
and emerging technologies such as controlled-release systems for
drug delivery1,2 and photoresist methodology for microlithography3-5
among other applications.6 However, as the demands on these and
other technologies have increased, higher levels of control over
the structure, properties, and performance of degradable materials
are necessary. Extremely few reports have addressed strategies for
the controlled degradation of dendritic structures7-12 or the implica-
tions of the development of such systems. Certainly, degradable
subunits could easily be engineered into the dendritic architecture,
and a limited number of reports have demonstrated selective
fragmentation of dendrimers at the periphery, core, and interior
residues of the architecture.7-12 However, we envision a more
efficient process whereby an initial stimulus causes a change in a
dendritic structure that triggers a subsequent cascading destruction
of the material into a number of smaller fragments, i.e., a chemically
amplified13 dendrimer degradation. Herein we introduce a new,
controlled method of dendrimer degradation that takes advantage
of the ubiquitous benzyl aryl ether dendritic subunit as a cleavable
backbone. This degradation method, or “dendrimer disassembly”
process, fragments a dendrimer through a cascade of cleavage
reactions initiated by a single triggering event.
the standard 3,5-branched subunits. Where 3,4-branched subunits
were not necessary for maintaining the cleavage pathway, 3,5-
branched units were used. When deprotection of the allyl group
occurs as the initial event, subsequent cleavages should result in a
cascade disassembly of the dendron toward the central core. A
p-nitrophenoxy moiety was intentionally installed at the focal point
of each dendron so that complete cleavage would be indicated by
the UV absorbance of liberated p-nitrophenoxide ion (∼400 nm).
To investigate the disassembly process, compounds 1-3 were
subjected to several variations on standard allyl deprotection
conditions.16 Optimum conditions for rapid disassembly proved to
be in DMF solution with a mixture of Pd(PPh3)4 and NaBH4.19
The absorption spectrum of the reaction mixture was monitored at
intervals. Smooth disassembly was observed, after an initial
incubation period,20 by a simultaneous rapid decrease in the initial
absorption at 312 nm with a concomitant increase in the absorption
of p-nitrophenoxide ion (∼432 nm). This spectral evolution takes
place within 15 min under the conditions used (Figure 1a). The
final absorbance values observed at 432 nm indicated complete
disassembly (i.e., disassembly proceeding to the focal point) of 1-3
in 85-100% yield based on the measured absorptivity of p-
nitrophenoxide under the reaction conditions (Figure 1b).
The fragments liberated during the disassembly of compounds
1-3 were characterized by a combination of GC-MS and LC-
MS. Molecular weights were consistent with the trapping of quinone
methide intermediates by excess hydride in solution. For example,
fragments identified in the disassembly of 3 were 4-methylphenol,
2-benzyloxy-4-methylphenol,2-(3,5-dibenzyloxy)benzyloxy-4-meth-
ylphenol, and p-nitrophenol. Triphenylphosphine and triphenylphos-
phine oxide were also observed from catalyst decomposition.
The disassembly process was also investigated in tetrahydrofuran
(THF) solvent, in which the reaction proceeded more slowly than
in DMF. The spectral evolution during the disassembly reactions
of 1-3 in tetrahydrofuran also indicated decreasing starting material
absorption at ca. 305 nm and increasing p-nitrophenoxide absorption
at ca. 405 nm over the course of the disassembly (Figure 1c and
Supporting Information). However, a new absorption band emerged
during the disassembly of 2 and 3 in THF. Specifically, the
absorption band at ca. 305 nm decreased within 50 min for 2 and
3, while during the same time period a band at 340 nm increased
It was recently reported that 4-O-substituted benzyl ethers used
as protecting groups could be removed in nearly quantitative yield
either thermally or by mild oxidation after removal of the
O-substituent (eq 1, R ) Ac, SEM).14,15 We also recently discovered
that the p-allyloxybenzyl ether group could be similarly removed
under standard conditions16 for allyl deprotection (eq 1, R ) allyl).17
The p-benzyl ether phenoxide generated by removal of the
O-substituent R in eq 1 is the key intermediate in these deprotec-
tions. This phenoxide, a vinylogous hemiacetal anion, cleaves to
yield an alkoxide (-OR′ in eq 1), which is the conjugate base of
the protected alcohol HOR′, as well as p-quinone methide, which
is presumably trapped by a weak nucleophile under the reaction
conditions, consistent with the electrophilic nature of quinone
methides.18 We reasoned that if the alkoxide -OR′ was also a
p-benzyl ether phenoxide, then a subsequent cleavage would occur.
Multiple cleavages would constitute a depolymerization, and this
became our strategy for disassembling benzyl ether-based dendritic
structures.
To test this dendrimer disassembly strategy, we prepared zeroth-
through second-generation dendrons 1-3 (Chart 1). Compounds
1-3 consist of a single 4-allyloxy residue at the periphery and a
continuous chain of p-benzyl ether linkages extending to the focal
point. Since p-benzyl ether linkages are necessary to maintain the
cleavage pathway through each generational shell, 3,4-branched
subunits were used in the required parts of the dendrons rather than
9
10516
J. AM. CHEM. SOC. 2003, 125, 10516-10517
10.1021/ja0349960 CCC: $25.00 © 2003 American Chemical Society

C O M M U N I C A T I O N S
Figure 1. (a) UV spectra recorded during the disassembly reaction of 3 in DMF (40 s intervals). Inset: UV absorbance of the main absorption bands as a
function of time. (b) UV absorbance at 432 nm as a function of time in the disassembly of 1-3 in DMF. (c) UV spectra recorded during the disassembly
reaction of 3 in THF (time intervals are 0, 15, 30, 45, 60, 90, 120, 180, 240, 360, 480, and 600 min). Inset: UV absorbance of the main absorption bands
as a function of time.
in absorbance. The p-nitrophenoxide absorption then increased over
a longer time period at the expense of the 340 nm band. This can
References
(1) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M. Chem.
be seen readily in a plot of absorbance at these three wavelengths
ReV. 1999, 99, 3181.
versus time in the disassembly of second generation dendron 3
(Figure 1c, inset). We assign this 340 nm band to phenoxide
intermediates on the basis of independent generation and observation
of these species under similar spectral conditions.
(2) Griffith, L. G. Acta Mater. 2000, 48, 263-277.
(3) Reichmanis, E.; Nalamasu, O.; Houlihan, F. M. Acc. Chem. Res. 1999,
32, 659-667.
(4) Wallraff, G. M.; Hinsberg, W. D. Chem. ReV. 1999, 1801-1822.
(5) MacDonald, S. A.; Willson, C. G.; Fre´chet, J. M. J. Acc. Chem. Res. 1994,
27, 151.
The results in both DMF and THF solutions support our proposed
disassembly process of eq 1 as taking place in two stages. The
first stage is the removal of the allyl group and generation of the
initial, and subsequently generated, phenoxide intermediates. The
second stage is progression of the disassembly to the dendron focal
point and appearance of the p-nitrophenoxide absorption. In DMF,
the rate-limiting step in the process appears to be the removal of
the allyl triggering group, as disappearance of the starting material
band coincides with appearance of the p-nitrophenoxide absorption.
However, in THF, the rate of disassembly is slower than the rate
of removal of the allyl trigger, and phenoxide intermediates are
observed during the course of the disassembly.
In summary, we have demonstrated the disassembly of dendritic
structures by a benzyl ether depolymerization reaction. In accord
with previous work,14 we anticipate that the initiating site may be
constructed with any manner of protecting group and not just the
allyl group investigated here. Hence, the initial stimulus for
disassembly may be varied.
(6) Sawada, H. In Encyclopedia of Polymer Science and Engineering; Mark,
H. F., Bikales, N. M., Overberger, C. G., Menges, G., Kroschwitz, J. I.,
Eds.; Wiley: New York, 1985; Vol. 4, pp 719-745.
(7) Seebach, D.; Herrmann, G. F.; Lengweiler, U. D.; Bachmann, B. M.;
Amrein, W. Angew. Chem., Int. Ed. Engl. 1996, 35, 2795-2797.
(8) Smet, M.; Liao, L.-X.; Dehaen, W.; McGrath, D. V. Org. Lett. 2000, 2,
511-513.
(9) Photolabile and thermally labile hyperbranched polyesters have been
reported by B. Voit et al., 13th Biennial Carl S. Marvel Symposium,
Tucson, AZ, March 14-16, 1999.
(10) Jansen, J. F. G. A.; Meijer, E. W.; de Brabander-van den Berg, E. M. M.
J. Am. Chem. Soc. 1995, 117, 4417-4418.
(11) Tully, D. C.; Trimble, A. R.; Fre´chet, J. M. J. AdV. Mater. 2000, 12,
1118-1122.
(12) Watanabe, S.; Sato, M.; Sakamoto, S.; Yamaguchi, K.; Iwamura, M. J.
Am. Chem. Soc. 2000, 122, 12588-12589.
(13) Ito, H. In Desk Reference of Functional Polymers; Arshady, R., Ed.;
American Chemical Society: Washington, DC, 1997; pp 341-369.
(14) Jobron, L.; Hindsgaul, O. J. Am. Chem. Soc. 1999, 121, 5835-5836.
(15) For a deprotection strategy related to that in ref 14, see: Plante, O. J.;
Buchwald, S. L.; Seeberger, P. H. J. Am. Chem. Soc. 2000, 122, 7148-
7149.
(16) Beugelmans, R.; Bourdet, S.; Bigot, A.; Zhu, J.-P. Tetrahedron Lett. 1994,
35, 4349.
(17) Li, S.; McGrath, D. V. J. Am. Chem. Soc. 2000, 122, 6795-6796.
(18) Wan, P.; Barker, B.; Diao, L.; Fischer, M.; Shi, Y.; Yang, C. Can. J.
Chem. 1996, 74, 465-475.
(19) General procedure for the disassembly of compounds 1-3 in DMF: A
quartz cuvette is charged with 2.00 mL of a solution of NaBH4 in DMF
(1.0 mg/1.0 mL). To this is added 20 µL of a 3 mM solution of substrate
in DMF, followed by 20 µL of a solution of Pd(PPh3)4 in DMSO (1.0
mg/1.0 mL). Monitoring of the reaction begins exactly 75 s following
the final addition.
(20) During the incubation period for all runs in DMF, a decrease in the
Pd(PPh3)4 catalyst absorbance band at ca. 287 nm is observed (see
Supporting Information). Hence, although we find the reproducibility of
this incubation period to be only moderate under these particular
conditions, it is clearly related to catalyst activation. Further studies are
underway.
Acknowledgment. This work was generously supported by
grants from the National Science Foundation (DMR-9996003, CHE-
0213537), the Air Force Office of Scientific Research (F49620-
97-1-0161), and the Petroleum Research Fund (38421 AC-7).
D.V.M. is a Cottrell Scholar of Research Corporation and a Camille
Dreyfus Teacher-Scholar. We thank Seth R. Marder for helpful
discussions.
Supporting Information Available: Details of the syntheses,
characterization data, and complete UV spectra of the disassembly of
dendrons 1-3 (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
JA0349960
9
J. AM. CHEM. SOC. VOL. 125, NO. 35, 2003 10517