Oxidative degradation of oligo(ethylene glycol)-terminated monolayers

Article abstract of DOI:10.1039/b911155g

The observation that oligo(ethylene glycol) (OEG)-terminated monolayers remained highly protein-resistant for a month at 37 °C in PBS buffer while they degraded faster in air can be rationalized by a proposed mechanism formulated from a model study using

Full text of DOI:10.1039/b911155g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Oxidative degradation of oligo(ethylene glycol)-terminated monolayersw
Guoting Qin and Chengzhi Cai*
Received (in Berkeley, CA, USA) 8th June 2009, Accepted 9th July 2009
First published as an Advance Article on the web 30th July 2009
DOI: 10.1039/b911155g
The observation that oligo(ethylene glycol) (OEG)-terminated
monolayers remained highly protein-resistant for a month at
37 1C in PBS buffer while they degraded faster in air can be
rationalized by a proposed mechanism formulated from a model
study using the first internal hydroperoxide of OEG.
(Inꢀ, Scheme 1b) on the film surface I to generate a carbon
radical in II. For the methyl-terminated films (R = Me), hydride
abstraction at the accessible –CH2O– group is energetically more
favorable than at the H3CO– group.8 Rapid trapping of the
radicals by O2 generates the peroxy radicals in III. For the
OH-terminated OEG films (R = H), the peroxy radical IV
should rapidly fragment to an aldehyde V and a hydroperoxy
radical,3,6 the calculated activation energy for similar reactions
being B9 kcal molÀ1.6 On the other hand, the peroxy radical in
III should abstract a hydride from an adjacent molecule, leading
to VI, rather than intramolecular hydride abstraction which has
a higher energy barrier (B25 vs. 20 kcal molÀ1).6 Repeating
the process accumulates hydroperoxides as the primary
oxidation products on the film surface VII. We studied their
decomposition using the model compound 1.
Monolayers presenting poly- or oligo(ethylene glycol)
(PEG/OEG) have attracted extensive research activities.1
However, PEG/OEG derivatives are susceptible to degradation
via autoxidation.2–6 Although the long-term stability of
PEG/OEG films has been investigated,7 their oxidative
degradation and the composition of the degraded surfaces are
not well understood. Herein, we present a model study using the
first internal hydroperoxide of an OEG derivative to represent
the primary products of autoxidation of PEG/OEG. The
availability of this model compound allowed for verification
of the decomposition products for the first time. A mechanism
differing from the general autoxidation process for oxidative
degradation of OEG films is proposed. It suggests that
the degraded surfaces are dominated by alcohols that can
terminate the autoxidation process, in agreement with the
experimental observations.
The hydroperoxide 1 was synthesized from the alcohol 2
via ozonization of the chlorovinyl derivative 3 followed by
selective cleavage of the primary ozonide to form the desired
carbonyl oxide9 that is quenched with methanol (Scheme 1c).
Thermo gravimetric analysis and differential thermal analysis
(TGA-DTA) of 1 showed that its exothermic decomposition
occurred rapidly at 80 1C (see ESIw). The decomposition of 1
Autoxidation of OEG/PEG has been described to proceed
via internal hydroperoxides A (Scheme 1a).2–6 They were
assumed to decompose via three routes: (1) dehydration to
an ester, (2) rearrangement to a hemiformate that decomposes
to a formate, an alcohol and formaldehyde, and (3) homolytic
cleavage followed by chain scissoring leading to a formate and
reactive radicals.3–5 Route 2 was proposed to be the major
pathway, accompanied by route 1.5 Both routes do not involve
radical intermediates. Methyl ethers were assumed to form via
route 3, which also generates radicals and the products of
route 2.3–5 In these studies, PEG/OEG derivatives were
exposed to air for days at 150 1C or for months and even
years at room temperature. Regardless of the conditions, the
hydroperoxides A were detected in the mixtures,3,5 but have
never been obtained in a pure form. Hence, their proposed
degradation products have not been validated. In this work we
synthesized 1 as the first internal hydroperoxide of OEG,
which closely represents the hydroperoxides initially formed
on methoxy-terminated OEG films.
The proposed mechanism for autoxidation of OEG films
starts with hydride abstraction by
a radical initiator
Department of Chemistry, University of Houston, Houston,
TX 77204-5003, USA. E-mail: cai@uh.edu; Fax: +1 713 743 2709;
Tel: +1 713 743 2710
w Electronic supplementary information (ESI) available: Synthesis,
TGA-DTA and decomposition of 1, NMR, GC-MS of the decom-
position mixture, and proposed mechanism of oxidative degradation
of OEG films. See DOI: 10.1039/b911155g
Scheme 1
ꢁc
This journal is The Royal Society of Chemistry 2009
5112 | Chem. Commun., 2009, 5112–5114

View Article Online
Fig. 1 GC-MS diagram after heating of 1 at 80 1C in air for 1 h.
was then performed at 80 1C in air for 1 h. The reaction mixtures
were analyzed by GC-MS (Fig. 1), showing the presence of the
products 2 and 4–8. The yields listed in Scheme 1c were obtained
by GC, indicating that 1 was decomposed mainly through route
2 (Scheme 1a) to the alcohol 2. 1H- and 13C-NMR spectroscopy
of the mixture showed that methyl formate and formic acid
were also formed (ESIw), the latter probably by oxidation of
formaldehyde in air.5 To a lesser extent, the ester 4 was formed
in 19% yield via route 1. The aldehyde 6 (3%) and formate 7
(3%) as minor products might be formed by reactions of 1
with formaldehyde, and 2 with formic acid, respectively. Note
that the methyl ether 5 was formed in 11% yield. Only trace
amounts of other byproducts, mainly the aldehyde 8 (2%),
were present. This result indicated that 5 was not generated
from route 3, which would also produce equal amounts of
highly reactive HOꢀ and carbon radicals leading to substantial
amounts of other products (see ESIw for a speculative,
non-radical route).
Fig. 2 Top: XPS N1s narrow scans of MeO-EG7-terminated mono-
layers on Si substrates after storage in air for 24 days, or in PBS buffer
at 37 1C for 28 days, followed by incubation in 0.1% fibrinogen for
1 h. A freshly prepared hydrogen-terminated silicon substrate with an
adsorbed monolayer of fibrinogen is used as a reference. Bottom: XPS
C1s narrow scans of MeO-EG7-terminated monolayers before and
after storage in air for 24 days, or in PBS buffer for 28 days followed
by treatment with fibrinogen.
The above model study establishes that the internal hydro-
peroxides on the OEG film VII (Scheme 1b) decompose mostly
to alcohols and formates as in VIII. The latter are hydrolyzed
to alcohols as in IX. Note that rather than undergoing hydride
abstraction to propagate the radical chain reaction, i.e. III to
VI (Scheme 1b), the a-peroxy radicals of alcohols rapidly
decompose to aldehydes and HOOꢀ, i.e. IV to V.3,6 If
the HOOꢀ radicals can be removed from the film, e.g. by
forming hydrogen bonds with water11 and diffusion into the
bulk solution (see below), the radical chain process is then
terminated. Otherwise, HOOꢀ radicals will generate the
carbon radicals II (Scheme 1b), and the process is repeated
from II to IX leading to mostly alcohols and aldehydes (via IV
to V) on the surface.
silicon (111) substrate was subjected to an identical fibrinogen
adsorption experiment, forming a monolayer of fibrinogen.
This sample was used as a reference for the measurement of
the % monolayer adsorption of fibrinogen onto the OEG
films. Thus, after 4 weeks in PBS buffer at 37 1C, the OEG
films adsorbed only a B1.1% monolayer of fibrinogen, while
they adsorbed a 23% monolayer of fibrinogen after 24 days in
air, as derived from the N1s data (Fig. 2). The higher stability
in PBS buffer might be due to the decomposition of the OEG
hydroperoxides to alcohols that stop the propagation of
peroxy radicals, leading to HOOꢀ radicals that hydrogen-bond
to water11 and diffuse into the solution. The short singlet state
lifetime of O2 in water12 may also suppress the oxidation. The
alcohols accumulated on the monolayer surface can terminate
the propagation of OEG peroxy radicals on the surfaces if the
resultant HOOꢀ radicals can be removed. This advantage for
OEG/PEG monolayers is not shared by OEG/PEG in the
bulk. For the films maintained in air, HOOꢀ radicals will
remain in the film, and may initiate the degradation process as
discussed above. Indeed, after storage of the OEG films in PBS
buffer at 37 1C for 28 days, the number of EG units remained
the same, as indicated by the same intensity of the etheric
carbon peaks at 286.6 eV (Fig. 2). In comparison, as calculated
from the decrease of the etheric C1s peak, one out of seven EG
units was lost for films stored in air for 24 days. However,
EG6-terminated monolayers on Si are expected to adsorb only
3% instead of 23% monolayer of fibrinogen.10 The substantial
We found that films prepared by hydrosilylation of
MeO-EG7(CH2)8CHQCH2 on hydrogen-terminated silicon
(111) surfaces10 remained protein resistant in phosphate
buffered saline (PBS) for weeks, while they were easier to
degrade in air. Preparation of the OEG monolayers on silicon
substrates was described in ref. 10. The stability of the mono-
layers for protein-resistance were evaluated by incubation of
the samples in PBS buffer (pH 7.4, Sigma) at 37 1C for 28 days,
or placed in semiconductor wafer containers in air at room
temperature for 24 days. The samples were then immersed in a
solution of fibrinogen (1 mg mLÀ1) in PBS buffer for 1 h,
washed under a gentle flow of deionized water (Millipore) for
about 15 s, and dried with a flow of Ar. The samples were
immediately measured by X-ray photoelectron spectroscopy
(XPS, Fig. 2). The XPS N1s signal intensity is proportional to
the amount of adsorbed protein. A hydrogen-terminated
ꢁc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 5112–5114 | 5113

View Article Online
increase in protein binding is attributed to the formation of
imines with the aldehydes on the film surface,13 although
the C1s XPS signal for aldehydes at 288 eV was at the noise
level (Fig. 2) due to the low density and the presence of
22 other carbon atoms in the molecule. In fact, the C1s signal
(blue curve in Fig. 2) was obtained on the sample that
absorbed 1.1% fibrinogen. Yet, the carbonyl signals at
288 eV and 289 eV were also at the noise level. In a relevant
study, Leggett and co-workers performed nanopatterning with
photochemical oxidation of thiolate monolayers terminated
with a short OEG (EG3) on gold substrates. A high density
of aldehydes and OCQO species were observed by XPS, likely
as a result of exposure of the films to UV light in air which
generates a high concentration of oxy radicals.13
2 (a) M. Bergh, K. Magnusson, J. L. G. Nilsson and A. T. Karlberg,
Contact Dermatitis, 1998, 39, 14; (b) F. Currie, M. Andersson and
K. Holmberg, Langmuir, 2004, 20, 3835; (c) A. T. Karlberg,
A. Bodin and M. Matura, Contact Dermatitis, 2003, 49, 241.
3 A. Bodin, M. Linnerborg, J. L. G. Nilsson and A. T. Karlberg,
Chem. Res. Toxicol., 2003, 16, 575.
4 (a) C. Decker and J. Marchal, Makromol. Chem., 1973, 166, 155;
(b) G. Gallet, S. Carroccio, P. Rizzarelli and S. Karlsson, Polymer,
2002, 43, 1081; (c) S. Morlat and J. L. Gardette, Polymer, 2001, 42,
6071.
5 (a) L. Yang, F. Heatley, T. G. Blease and R. I. G. Thompson, Eur.
Polym. J., 1996, 32, 535; (b) O. A. Mkhatresh and F. Heatley,
Macromol. Chem. Phys., 2002, 203, 2273; (c) O. A. Mkhatresh and
F. Heatley, Polym. Int., 2004, 53, 1336.
6 (a) C. Backtorp, A. Borje, J. L. G. Nilsson, A. T. Karlberg,
P. O. Norrby and G. Nyman, Chem.–Eur. J., 2008, 14, 9549;
(b) T. S. Dibble, Chem. Phys. Lett., 2002, 355, 193.
7 (a) M. C. Shen, L. Martinson, M. S. Wagner, D. G. Castner,
B. D. Ratner and T. A. Horbett, J. Biomater. Sci., Polym. Ed.,
2002, 13, 367; (b) N. T. Flynn, T. N. T. Tran, M. J. Cima and
R. Langer, Langmuir, 2003, 19, 10909; (c) D. W. Branch,
B. C. Wheeler, G. J. Brewer and D. E. Leckband, IEEE Trans.
Biomed. Eng., 2000, 47, 290; (d) K. Jans, K. Bonroy, R. De Palma,
G. Reekmans, H. Jans, W. Laureyn, M. Smet, G. Borghs and
G. Maes, Langmuir, 2008, 24, 3949; (e) E. Ostuni, R. G. Chapman,
R. E. Holmlin, S. Takayama and G. M. Whitesides, Langmuir,
2001, 17, 5605; (f) S. Y. Hou, E. A. Burton, R. L. Wu, Y. Y. Luk
and D. C. Ren, Chem. Commun., 2009, 1207; (g) Y. Y. Luk,
M. Kato and M. Mrksich, Langmuir, 2000, 16, 9604;
(h) C. M. Nelson, S. Raghavan, J. L. Tan and C. S. Chen,
Langmuir, 2003, 19, 1493.
In conclusion, we demonstrated for the first time that
internal OEG hydroperoxides are decomposed mainly through
non-radical pathways to alcohols and esters. When stored in
an aqueous solution (PBS buffer), OEG monolayers remained
highly protein-resistant for at least a month at 37 1C. They
were easier to degrade when stored in air. This observation can
be rationalized by a proposed mechanism suggesting that
oxidation on the surface of OEG monolayers by reactive oxy
radicals generates mostly alcohols and aldehydes. The mechanism
may be applied to other processes involving oxy radicals, such
as photooxidation,13 electro-chemical oxidation,14 and inter-
actions with polymorpho-nuclear leukocytes that secrete
reactive oxy radicals.7a
8 T. Ogura, A. Miyoshi and M. Koshi, Phys. Chem. Chem. Phys.,
2007, 9, 5133.
9 W. V. Turner and S. Gab, J. Org. Chem., 1992, 57, 1610.
10 C. M. Yam, J. M. Lopez-Romero, J. H. Gu and C. Z. Cai, Chem.
Commun., 2004, 2510.
11 K. Suma, Y. Sumiyoshi and Y. Endo, Science, 2006, 311, 1278.
12 A. A. Gorman, M. A. J. Rodgers, Singlet Oxygen, ed.
J. C. Scaiano, CRC Press, Boca Raton, Florida, 1989, vol. II,
p. 229.
We thank The Welch Foundation, the NSF CAREER
CTS-0349228 and DMR-0706627 for supporting this work.
GQ thanks the NIH Nanobiology Training Program
(5R90 DK71504-3) for a fellowship.
13 (a) R. E. Ducker, S. Janusz, S. Q. Sun and G. J. Leggett, J. Am.
Chem. Soc., 2007, 129, 14842; (b) M. Montague, R. E. Ducker,
K. S. L. Chong, R. J. Manning, F. J. M. Rutten, M. C. Davies and
G. J. Leggett, Langmuir, 2007, 23, 7328.
14 J. H. Gu, C. M. Yam, S. Li and C. Z. Cai, J. Am. Chem. Soc., 2004,
126, 8098.
Notes and references
1 (a) Z. Harris, ACS Symposium #680: Poly(ethylene Glycol):
Chemistry and Biological Applications, ACS, Washington, DC,
1998; (b) K. L. Prime and G. M. Whitesides, Science, 1991, 252,
1164.
ꢁc
This journal is The Royal Society of Chemistry 2009
5114 | Chem. Commun., 2009, 5112–5114