Site-selective abstraction in the reaction of 5-20 eV O+ with a self-assembled monolayer

Article abstract of DOI:10.1021/ja045475v

Site-specific reaction of hyperthermal O⁺ with a self-assembled monolayer is described. Isotopic labeling experiments reveal the percentage of abstraction products formed from hydrogen atoms bound originally to the top three carbon atoms in the chain. Copyright

Full text of DOI:10.1021/ja045475v

Published on Web 09/25/2004
Site-Selective Abstraction in the Reaction of 5-20 eV O+ with a
Self-Assembled Monolayer
Xiangdong Qin,† Tochko Tzvetkov,† Xin Liu,† Dong-Chan Lee,‡ Luping Yu,‡ and Dennis C. Jacobs*,†
Department of Chemistry and Biochemistry, UniVersity of Notre Dame, Notre Dame, Indiana 46556, and
Department of Chemistry and The James Franck Institute, The UniVersity of Chicago, Chicago, Illinois 60637
Received July 27, 2004; E-mail: jacobs.2@nd.edu
Steric effects exert significant control over many chemical
reaction pathways, yet it remains an experimental challenge to study
a bimolecular reaction under conditions where one reactant
encounters the other with a specific angle of attack. In both gas-
phase and solution environments, the relative orientation of two
approaching molecules is isotropic. On the other hand, when a beam
of reactants is targeted at an array of molecules adsorbed to a solid
surface, the approach geometry is severely constrained. In the
present study, an O+ beam is directed at an alkanethiolate self-
assembled monolayer (SAM), and the relative rate of hydrogen
abstraction is measured as a function of the position, along the
Figure 1. Outgoing energy distributions of scattered OH- and OD-
products resulting from 5 eV O+ reacting with C-12 SAM (incident angle
) 45°; exit angle ) 65°). Deconvolution of the instrument response function
from the experimental data (solid points) yields corrected energy distribu-
tions (curves).
hydrocarbon chain, from which the hydrogen atom is removed.
When a hyperthermal energy (3-500 eV) particle strikes a
surface, it can activate a wide variety of chemical process at the
gas/surface interface, including abstraction, elimination, insertion,
implantation, and oxidation.1 In low-earth orbit (LEO), a spacecraft
is continuously exposed to a flux of energetic neutral atoms/
molecules, ions, electrons, and photons that threaten to erode and
degrade the vehicle’s external surfaces. The most abundant neutral
and ionic species in LEO are O and O+; not only are they
chemically reactive, but they typically impact the RAM surfaces
of a spacecraft with 5 eV of translational energy. Next-generation
satellites could be partially fabricated with lightweight, low-cost
polymeric materials, if a strategy is developed to inhibit the
degradation of organic films in the corrosive environment of LEO.
Alkanthiolate molecules bound to Au(111) comprise the most
thoroughly characterized SAM system.2,3 The densely packed, stand-
up phase serves as a model for polymeric films such as poly-
ethylene. Studying the reaction of hyperthermal energy O+ with a
hydrocarbon SAM may uncover some of the erosion mechanisms
suffered by polymers in LEO.
Qin et al. monitored ion-induced modifications of a decane-
thiolate SAM by employing X-ray photoelectron spectroscopy
(XPS) in vacuo at different stages of exposure to 5 eV O+.4 The
etch rate caused by 5 eV O+ was four orders-of-magnitude higher
than Fairbrother and co-workers observed using only thermal
oxygen atoms.5 George and co-workers utilized an oxygen plasma
to erode a decanethiolate SAM and concluded that oxygen ions,
rather than the majority neutral species in the plasma, were
principally responsible for the film’s decay.6
determine the detailed site-dependence to abstraction in the present
system, we utilize labeled SAMs in which all the hydrogen atoms
bound to the 10th, 11th, or 12th carbon atom of 1-dodecanethiol
are selectively substituted with deuterium atoms.
The three isotopically labeled dodecanthiols (C-10, C-11, and
C-12) are synthesized through methods described in Supporting
Information. An isotopomeric SAM is prepared by immersing a
Au(111) crystal in the corresponding 1 mM dodecanethiol ethanolic
solution for 24 h. The surface is transferred to a customized ultra-
high vacuum (UHV) chamber where it is held at ground potential
and room temperature.9 A monoenergetic beam of mass-filtered
O+ is directed at the SAM, 45° to the surface normal. Incident O+
efficiently neutralizes close to the SAM. When abstraction occurs,
the nascent OH species adopts a partial negative charge on the low
work-function surface (Φ ) 3.7 eV).10 If the product departs rapidly
from the surface, there is a finite probability that it will escape
nonadiabatically and appear as OH-. For each of the three
isotopomeric SAMs, the yields of OD- and OH- are compared.
Figure 1 shows the energy distributions of scattered OH- and
OD- products resulting when 5 eV O+ is incident on a C-12-labeled
SAM. The detector is positioned 65° from the surface normal at
the angle where the scattered signal is at its maximum. The overall
intensity of OD- is approximately double that of OH-. This result
indicates that 5 eV O+ preferentially abstracts D from the labeled
position (the topmost carbon of the SAM) in comparison to all other
unlabeled H positions combined.
Hydrogen abstraction near the vacuum/SAM interface is a major
pathway by which SAMs are degraded. Once a chain is compro-
mised, further reactions with O+ can quickly consume the remaining
molecule.7 Of interest is the precise location where an impinging
O+ will first remove a hydrogen atom from the hydrocarbon chain.
Gu and Wysocki found that 30 eV pyrazine and acetone ions
preferentially abstract hydrogen atoms from the terminal methyl
group of an isotopically labeled Langmuir-Blodgett film.8 To
The integrated intensities, IOH and IOD, are calculated by
integrating the OH- and OD- signal intensities, respectively, over
all outgoing energies depicted in Figure 1. Equation 1 defines POD
,
the percentage of all abstracted H/D that originates from the labeled
position in the SAM.
IOD
POD
)
× 100%
(1)
† University of Notre Dame.
‡ The University of Chicago.
IOH + IOD
9
13232
J. AM. CHEM. SOC. 2004, 126, 13232-13233
10.1021/ja045475v CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
molecular mechanical model to realistically simulate the reaction
of 5 eV O(3P) with an octanethiol SAM.12 A comparison between
the simulation of neutral scattering at 45° incidence and our ion-
scattering experiment is not completely unreasonable, because in
the latter, the incident O+(4S) ions neutralize to become O(3P)
immediately prior to impact with the SAM.13 The simulations
predict that hydrogen abstraction is the dominant reaction channel
and that abstraction occurs only at the top three carbon atoms in
an alkanethiol chain, in qualitative agreement with the results
presented in Figure 2b. Furthermore, the trajectories show a 4-fold
preference for abstraction from the methylene rather than the methyl
hydrogens, when O(3P) approaches the SAM in the direction
opposite to the SAM tilt orientation. In contrast, the experimental
data indicate a much stronger preference for abstraction occurring
from primary rather than secondary carbon sites. Although the
experiments cannot resolve the scattering from a single orientational
domain, the discrepancy in site selectivity between simulation and
experiment is quantitatively significant. The simulations track only
the adiabatic channel leading to neutral OH, whereas the experiment
detects OH-/OD- emerging nonadiabatically. The nascent OH-
product formed deep within the SAM layer will have a difficult
time retaining its negative charge as it passes between the chains
on escape. Hence, the detection of only negative ions in the
experiment may weight the distribution shown in Figure 2b in slight
favor of abstraction sites close to the vacuum interface.
Figure 2. (a) Dependence of POD on the incident O+ energy and on the
labeled position (triangles, C-12; squares, C-11; diamonds, C-10) of
1-dodecanethiol. (b) Relative contribution of scattered hydroxide signal,
arising from abstraction at the top three carbon atoms, as a function of the
incident O+ energy. The latter data have been corrected for velocity-
dependent charge-transfer rates along the outgoing trajectory.
POD depends strongly on the O+ dose, because ion-induced
damage of the SAM alters the layer’s structure and chemical
composition. Consequently, we present only POD values recorded
at low doses (<1013 ions/cm2) before the SAM becomes damaged.
Figure 2a shows the dependence of POD on the incident O+ energy
for SAMs labeled at the C-10, C-11, and C-12 positions. POD is
vanishingly small for abstraction originating from C-10 and
presumably is negligible for positions below C-10. Although
experimental protocols were employed to minimize contamination
from any unlabeled molecules, the sum of POD contributions for
C-10, C-11, and C-12 is 7-16% less than the expected value of
100%. One possible explanation for this deficit is a kinetic isotope
effect, through which D is abstracted less readily than H, all else
being equal. Classical trajectory calculations predict only a 3.5%
kinetic isotope effect for O(3P) abstracting H versus D from
propane;11 hence, there is little reason to believe that a kinetic
isotope effect will adequately account for the missing intensity in
the present experiment. A more plausible explanation for the
shortfall in the OD- signal relative to the OH- signal involves the
nonadiabaticity of charge transfer. Figure 1 illustrates that OH-
leaves the surface with slightly more kinetic energy than does OD-.
Overall, the mean velocity of OH- appears to be ∼10% faster than
that for OD-. Consequently, a nascent OD- product is more likely
to adiabatically transfer its electron back to the surface than is a
quicker OH- product. As described in Supporting Information, the
data in Figure 2a can be corrected for charge transfer along the
outgoing trajectory to yield the relative contributions to abstraction
shown in Figure 2b. These adjusted contributions sum to 100 (
3%. At least 3/4 of the scattered abstraction signal arises from the
terminal methyl group in the SAM, and <2.5% originates from
H/D bonded to the C-10 position. This site preference is strongest
at low incident energies.
In summary, the reaction of hyperthermal (5-20 eV) O+ with
an isotopically labeled self-assembled monolayer demonstrates that
hydrogen abstraction is strongly site-specific. At 5 eV incident
energy, abstraction of hydrogen from the terminal methyl group is
favored by a factor of 4.6 over abstraction from the methylene group
just below it.
Acknowledgment. This work was performed through the Center
for Materials Chemistry in the Space Environment, a MURI
supported by the AFOSR (F49620-01-1-0335).
Supporting Information Available: Synthesis procedures and
characterization of C-10-, C-11-, and C-12-labeled 1-dodecanethiols
and charge transfer corrections to anionic product yields (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Jacobs, D. C. Annu. ReV. Phys. Chem. 2002, 53, 379-407.
(2) Poirier, G. E. Chem. ReV. 1997, 97, 1117-1127.
(3) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151-256.
(4) Qin, X.; Tzvetkov, T.; Jacobs, D. C. Nucl. Instrum. Methods B 2003,
203, 130-135.
(5) Torres, J.; Perry, C. C.; Bransfield, S. J.; Fairbrother, D. H. J. Phys. Chem.
B 2002, 106, 6265-6272.
(6) Dai, X. J.; Elms, F. M.; George, G. A. J. Appl. Polym. Sci. 2001, 80,
1461-1469.
(7) Wagner, A. J.; Wolfe, G. M.; Fairbrother, D. H. J. Chem. Phys. 2004,
120, 3799-3810.
(8) Gu, C.; Wysocki, V. J. Am. Chem. Soc. 1997, 119, 12010-12011.
(9) Quinteros, C. L.; Tzvetkov, T.; Qin, X.; Jacobs, D. C. Nucl. Instrum.
Methods B 2001, 182, 187-192.
(10) Alloway, D. M.; Hofmann, M.; Smith, D. L.; Gruhn, N. E.; Graham, A.
L.; Colorado, R.; Wysocki, V. H.; Lee, T. R.; Lee, P. A.; Armstrong, N.
R. J. Phys. Chem. B 2003, 107, 11690-11699.
(11) Troya, D.; Schatz, G. C. Private communication.
(12) Troya, D.; Schatz, G. C. J. Chem. Phys. 2004, 120, 7696-7707.
(13) O(3P) is formed rather than O(1D) or O(1S), because the quartet to singlet
spin-flip would require an improbable two-electron rearrangement.
Troya and Schatz developed a hybrid quantum mechanical/
JA045475V
9
J. AM. CHEM. SOC. VOL. 126, NO. 41, 2004 13233