Dissociative and reactive hyperthermal ion - Surface collisions with Langmuir-Blodgett films terminated by CF3(CH2)n-, n-perfluoroalkyl, or n-alkyl groups

Article abstract of DOI:10.1021/ja990719a

Langmuir-Blodgett (L-B) films prepared from 18,18,18-trifluorooctadecanoic acid [CF₃(CH₂)₁₆-COOH], perfluorotetradecanoic acid [n-C₁₃F₂₇COOH], or octadecanoic acid [n-C₁₇H₃₅COOH]

Full text of DOI:10.1021/ja990719a

10554
J. Am. Chem. Soc. 1999, 121, 10554-10562
Dissociative and Reactive Hyperthermal Ion-Surface Collisions with
Langmuir-Blodgett Films Terminated by CF3(CH2)n-,
n-Perfluoroalkyl, or n-Alkyl Groups
Chungang Gu,† Vicki H. Wysocki,*,† Atsuhiro Harada,‡ Hiroaki Takaya,‡ and
Itsumaro Kumadaki‡
Contribution from the Department of Chemistry, UniVersity of Arizona, Tucson, Arizona 85721, and
Faculty of Pharmaceutical Sciences, Setsunan UniVersity, Osaka 573-01, Japan
ReceiVed March 5, 1999
Abstract: Langmuir-Blodgett (L-B) films prepared from 18,18,18-trifluorooctadecanoic acid [CF3(CH2)16-
COOH], perfluorotetradecanoic acid [n-C13F27COOH], or octadecanoic acid [n-C17H35COOH] were used as
target surfaces for hyperthermal ion-surface collisions. The synthesis of 18,18,18-trifluorooctadecanoic acid
and the preparation of the L-B film from this compound using a subphase containing Al3+ are reported in
this paper. The Ion-surface collision results show that the outermost surface group, namely CF3 vs CH3, is
the main determinant for the efficiency of both energy transfer and electron transfer during low-energy (e.g.,
10-100 eV) polyatomic ion-surface collisions. For atomic projectile ions, there is evidence of penetration
into a depth of the films.
Introduction
etry (MS/MS)9,10 with the gaseous target (in CID) replaced by
a surface target (in SID) for ion activation, but SID also brings
some special features to MS/MS studies. In particular, SID
provides higher transfer of kinetic energy into internal modes
of polyatomic projectile ions (i.e., T f V), and provides
narrower distributions of the internal energies deposited into
the ions (compared to CID).6 These two features result in easily
tunable internal energies of polyatomic projectile ions activated
by SID. The ion chemistry implications of these characteristics
of SID include the following: (i) effective fragmentation for
some refractory ions (either large11,12 or very stable13 ions), (ii)
mechanistic insights drawn from energetic studies,14,15 and (iii)
extraction of activation energies for some ion fragmentation
pathways.16,17
Gas-phase ion-surface interactions are of interest because
of both their practical applications and the fascinating funda-
mental chemistry and physics associated with the collision
processes. Investigations of these interactions may be viewed
as falling into two categories if based on the range of collision
energies applied: low energy (e.g., a few eV to a couple of
hundred eV) and high energy (e.g., 1 keV to several hundred
keV). For example, keV projectile ions (often atomic ions) can
impart their energy and momentum to the surface species
causing sputtering, or penetrate into the solid surface and
become implanted there. Ion sputtering from the surface is
exploited for the analysis of chemical composition of surfaces
in the established technique of secondary ion mass spectrometry
(SIMS).1 Ion implantation is investigated for fabrication of
semiconductor materials and optical films.2 The studies pre-
sented here belong to the low-energy category of ion-surface
interactions. The physical chemistry and physics associated with
fundamental processes involving low-energy collisions of small
ions (e.g., up to a few atoms) with crystal surfaces have been
active research areas.3 More recently, collisions of polyatomic
ions with organic thin film surfaces have been investigated.
Dramatic differences are seen between low-energy ion-surface
collisions and those in the keV range. For example, some
polyatomic projectiles may survive the collision at lower
energies (e.g., 5 eV) and “soft-land” as “molecular parachutes”
on a fluorinated organic surface.4 There are also several other
aspects of interest to chemists in this range of collision energies.
One is dissociative scattering of polyatomic projectile ions. This
led to the development of surface-induced dissociation (SID)
by Cooks and co-workers.5-7 SID is not only an alternative to
CID (collision-induced dissociation)8 in tandem mass spectrom-
Another interesting chemical aspect of low-energy ion-
surface interactions is reaction between the gas-phase projectile
ions and the target surfaces. For example, abstractions of CnHm
and/or H from hydrocarbon-terminated surfaces6,13,18-26 and F
(single or multiple) from fluorinated surfaces18,21,27,28 can occur
(1) (a) Czanderna, A. W.; Hercules, D. M. Ion Spectroscopies For Surface
Analysis; Plenum Press: New York, 1991. (b) Winograd, N. Anal. Chem.
1993, 65, 622A-629A. (c) Benninghoven, A.; Hagenhoff, B.; Niehuis, E.
Anal. Chem. 1993, 65, 630A-640A.
(2) (a) Rimini, E. Ion implantation: basics to deVice fabrication; Kluwer
Academic Publishers: Boston, 1995. (b) Townsend, P. D.; Chandler, P. J.;
Zhang, L. Optical effects of ion implantation; Cambridge University Press:
Cambridge, UK, 1994. (c) Shi, J.; Kikkawa, J. M.; Proksch, R.; Schaffer,
T.; Awschalom, D. D.; Medeirosribeiro, G.; Petroff, P. M. Nature 1995,
377, 707-710. (d) Narayan, J.; Godbole, V. P.; White, C. W. Science 1991,
252, 416-418.
(3) Rabalais, J. W. Low energy ion-surface interactions; John Wiley:
New York, 1994.
(4) Miller, S. A.; Luo, H.; Pachuta, S. J.; Cooks, R. G. Science 1997,
275, 1447-1450.
(5) Mabud, M. D. A.; Dekrey, M. J.; Cooks, R. G. Int. J. Mass Spectrom.
Ion Process. 1985, 67, 285-294.
(6) Cooks, R. G.; Ast, T.; Mabud, A. Int. J. Mass Spectrom. Ion Process.
1990, 100, 209-265.
* To whom correspondence should be addressed. Tel: (520)621-2628.
Fax: (520)621-8407. E-mail: vwysocki@u.arizona.edu.
† University of Arizona.
(7) Cooks, R. G.; Miller, S. A.; Ast, T.; Feng, B.; Pradeep, T. AdV. Mass
Spectrom. 1995, 13, 179-197.
‡ Setsunan University.
(8) McLuckey, S. A. J. Am. Soc. Mass Spectrom. 1992, 3, 599-614.
10.1021/ja990719a CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/30/1999

Ion-Surface Collisions with Langmuir-Blodgett Films
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10555
with some projectile ions. Certain projectiles react with orga-
nized organic thin films to provide spectral features that reflect
the film quality.22 The potential to probe the outermost surface
groups or atoms by certain polyatomic ion-surface reactions
has been shown.25 In addition to the reaction products scattered
off surfaces, reaction products are sometimes formed on the
target surface, and the potential of ion-surface reactions in the
modification of organic surfaces is being investigated29,30
One of the important developments in surface-induced disso-
ciation is the use of organic thin films on metals as the collision
targets (e.g., self-assembled monolayers (SAMs) of n-alkanethi-
ols on gold or silver).31,32 Organic thin films such as SAMs
provide control of the chemical properties of the target surface.
The thin film provides a barrier to electron transfer and thus
limits the neutralization of projectile ions that otherwise would
occur readily on a bare metal,18,21 without the charge accumula-
tion that would occur on an insulator surface.33 Fluorinated thin
films show some practical advantages as the collision targets.
It is known that increased T f V and decreased neutralization
of projectile ions occur on fluorinated surfaces relative to
hydrocarbon-terminated surfaces. These differences have been
reported for ordered thin films such as SAMs of fluoroal-
kanethiols [e.g., CF3(CF2)nCH2CH2SH, n ) 7, 11] vs n-
alkanethiols on gold,18,21 and thin films of a liquid perfluo-
ropolyether (PFPE) vs pump oil residue on metal.34,35 An
impulsive excitation is suggested in the literature as a dominant
mechanism for internal energy deposition into a polyatomic
projectile ion during its collision, in particular, with organic
thin films on metal.7,36,37 Charge transfer-induced electronic
+
excitation has also been reported (e.g., CF3 collisions with a
metal surface,38 in addition to many diatomic ion collisions with
crystal surfaces3). It has been suggested by several au-
thors13,18,31,34,36,39 that the difference in energy transfer occurring
with fluorinated vs hydrocarbon-terminated surfaces is due to
the different effective masses of target surface species. The
molecular beam scattering experiments with liquid films by
Nathanson and co-workers show that the energy transfer to the
film is dependent on the mass of both partners of the collision
(i.e., mass of the gaseous impinging molecule and effective mass
of the surface).39 A large influence of the azimuthal orientation
of a single-crystal surface indicated the importance of direct
momentum transfer.38 The dependence of internal energy
deposition ratio on the size and shape of projectile ions was
suggested as due to the effective mass of the surface contacted
at the moment of impact.23
The experiments in this study are designed to investigate (i)
whether the difference in internal energy deposition into
projectile ions with fluorinated and hydrocarbon terminated
surfaces is mainly caused by the rigidity of (CF2)n vs (CH2)n or
the mass (momentum) effect of terminal fluorine vs hydrogen
and (ii) whether the difference in neutralization of projectile
ions is mainly controlled by the exposed surface groups (e.g.,
terminal CF3 vs CH3). Answers to these two questions will add
to our fundamental understanding of ion-surface interactions.
The CF3-terminated model surface also raises an additional
interesting point. It provides a convenient model surface for
further experimental examination of the proposed mechanism
of multiple fluorine additions to an atomic transition metal
projectile ion.28
The present study compares the results obtained with a model
surface terminated by a CF3(CH2)n- group to those obtained with
n-perfluoroalkyl- or n-alkyl-terminated surfaces. Langmuir-
Blodgett (L-B) films40,41 were utilized in this study. In the
1930s, Irving Langmuir and Katharine B. Blodgett discovered
the technique of transferring monomolecular films of am-
phiphilic molecules (e.g., fatty acids or their dicovalent metal
salts) from an air-water interface to a solid substrate (e.g., a
glass or metal plate) to form a monolayer or a builtup multilayer
by dipping procedures. Since then, comprehensive investigations
made on L-B films for diverse purposes using a great variety
of amphiphilic molecules have been described in many books
and review papers, including several excellent recent ones.40,41
In a Communication to this journal in 1997, we reported, by
using isotopically labeled L-B films, that low-energy ion-
surface reactions of organic projectiles predominantly involve
(9) Busch, K. L.; Glish, G. L.; McLuckey, S. A. Mass Spectrometry/
Mass Spectromety: Techniques and Application of Tandem Mass Spec-
trometry; VCH Publishers: New York, 1988.
(10) de Hoffmann, E. J. Mass Spectrom. 1996, 31, 129-137.
(11) Chorush, R. A.; Little, D. P.; Beu, S. C.; Wood, T. D.; McLafferty,
F. W. Anal. Chem. 1995, 67, 1042-1046.
(12) Dongre´, A. R.; Somogyi, AÄ .; Wysocki, V. H. J. Mass Spectrom.
1996, 31, 339-350.
(13) Callahan, J. H.; Somogyi, AÄ .; Wysocki, V. H. Rapid Commun. Mass
Spectrom. 1993, 7, 693-699.
(14) Beck, R. D.; St. John, P.; Homer, M. L.; Whetten, R. L. Science
1991, 253, 879-883.
(15) Dongre´, A. R.; Jones, J. L.; Somogyi, AÄ .; Wysocki, V. H. J. Am.
Chem. Soc. 1996, 118, 8365-8374.
(16) Wainhaus, S. B.; Gislason, E. A.; Hanley, L. J. Am. Chem. Soc.
1997, 119, 4001-4007.
(17) Weis, P.; Rockenberger, J.; Beck, R. D.; Kappes, M. M. J. Chem.
Phys. 1996, 104, 3629-3637.
(18) Morris, M. R.; Riederer, D. E.; Winger, B. E.; Cooks, R. G.; Ast,
T.; Chidsey, C. E. D. Int. J. Mass Spectrom. Ion Process. 1992, 122, 181-
217.
(19) Williams, E. R.; Jones, G. C.; Fang, L.; Zare, R. N.; Garrison, B.
J.; Brenner, D. W. J. Am. Chem. Soc. 1992, 114, 3207-3210.
(20) Wu, Q. Y.; Hanley, L. J. Am. Chem. Soc. 1993, 115, 1191-1193.
(21) Somogyi, AÄ .; Kane, T. E.; Ding, J. M.; Wysocki, V. H. J. Am. Chem.
Soc. 1993, 115, 5275-5283.
(22) Kane, T. E.; Somogyi, AÄ .; Wysocki, V. H. Org. Mass Spectrom.
1993, 28, 1665-1673.
(23) Beck, R. D.; Rockenberger, J.; Weis, P.; Kappes, M. M. J. Chem.
Phys. 1996, 104, 3638-3650.
(24) Hayward, M. J.; Park, F. D. S.; Phelan, L. M.; Bernasek, S. L.;
Somogyi, AÄ .; Wysocki, V. H. J. Am. Chem. Soc. 1996, 118, 8375-8380.
(25) Gu, C.; Wysocki, V. H. J. Am. Chem. Soc. 1997, 119, 12010-
12011.
(26) Phelan, L. M.; Hayward, M. J.; Flynn, J. C.; Bernasek, S. L. J.
Phys. Chem. B 1998, 102, 5667-5672.
(27) Pradeep, T.; Ast, T.; Cooks, R. G.; Feng, B. J. Phys. Chem. 1994,
98, 9301-9311.
(34) Pradeep, T.; Miller, S. A.; Cooks, R. G. J. Am. Soc. Mass Spectrom.
1993, 4, 769-773.
(35) Schey, K. L.; Durkin, D. A.; Thornburg, K. R. J. Am. Soc. Mass
Spectrom. 1995, 6, 257-263.
(36) Burroughs, J. A.; Wainhaus, S. B.; Hanley, L. J. Phys. Chem. 1994,
98, 10913-10919.
(37) Ast, T.; Riederer, D. E.; Miller, S. A.; Morris, M.; Cooks, R. G.
Org. Mass Spectrom. 1993, 28, 1021-1033.
(28) Pradeep, T.; Riederer, D. E.; Hoke, S. H.; Ast, T.; Cooks, R. G.;
Linford, M. R. J. Am. Chem. Soc. 1994, 116, 8658-8665.
(29) Pradeep, T.; Feng, B.; Ast, T.; Patrick, J. S.; Cooks, R. G.; Pachuta,
S. J. J. Am. Soc. Mass Spectrom. 1995, 6, 187-194.
(30) Ada, E. T.; Kornienko, O.; Hanley, L. J. Phys. Chem. B 1998, 102,
3959-3966.
(38) Koppers, W. R.; Beijersbergen, J. H. M.; Tsumori, K.; Weeding,
T. L.; Kistemaker, P. G.; Kleyn, A. W. Phys. ReV. B 1996, 53, 207-209.
(39) Saecker, M. E.; Govoni, S. T.; Kowalski, D. V.; King, M. E.;
Nathanson, G. M. Science 1991, 252, 1421-1424.
(40) (a) Ulman, A. Characterization of Organic Thin Films; Butterworth-
Heinemann: Boston, 1995. (b) Ulman, A. An introduction to ultrathin
organic films: from Langmuir-Blodgett to self-assembly; Academic
Press: Boston, 1991. (c) Petty, M. C. Langmuir-Blodgett films: an
introduction; Cambridge University Press: Cambridge, UK, 1996.
(41) (a) Zasadzinski, J. A.; Viswanathan, R.; Madsen, L.; Garnaes, J.;
Schwartz, D. K. Science 1994, 263, 1726-1733. (b) Peterson, I. R. J. Phys.
D: Appl. Phys. 1990, 23, 379-395.
(31) Winger, B. E.; Julian, R. K.; Cooks, R. G.; Chidsey, C. E. D. J.
Am. Chem. Soc. 1991, 113, 8967-8969.
(32) Wysocki, V. H.; Jones, J. L.; Ding, J. M. J. Am. Chem. Soc. 1991,
113, 8969-8970.
(33) Kane, T. E.; Wysocki, V. H. Int. J. Mass Spectom. Ion Process.
1994, 140, 177-184.

10556 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Gu et al.
ice, then 10% H2SO4 was added slowly until Al(OH)3 dissolved, and
extracted with Et2O. The Et2O layer was washed with saturated NaCl,
dried over MgSO4, and concentrated under vacuum. The residue was
purified by column chromatography (SiO2, CH2Cl2-MeOH, 99:1) to
give 12,12,12-trifluoro-1-dodecanol (4, 2.40 g, 99%). 4: Colorless
crystals. Mp 42.0 °C. MS m/z 222 (M+ - OH). HRMS Calcd for
C12H21F3 (M+ - OH) 222.160; found 222.160. IR (KBr) cm-1 3700-
3200 (O-H), 2928, 2860 (C-H). 1H NMR (CDCl3) δ 1.15-1.45 (15H,
b), 1.56 (4H, m), 2.05 (2H, qt, J ) 10.6, 7.5 Hz), 3.64 (2H, t, J ) 6.6
Hz). 19F NMR (CDCl3) ppm -3.42 (3F, t, J ) 10.6 Hz).
the outermost surface groups (e.g., terminal methyl group on
13CH3(CH2)16COOH or CD3(CH2)16COOH L-B film).25
A
reason for our choice of the L-B films for the present study,
quite similar to that in our previous Communication, is that the
material to make the desired surface (i.e., CF3(CH2)n-terminated
L-B film) is available, whereas the materials for making
equivalent SAM surfaces were not available when we began
this study.42 Recently, characterizations of SAM surfaces
prepared from CF3(CH2)nSH (n ) 9-15) have been re-
ported.43,44 Terminally fluorinated fatty acids, 18,18,18-trifluo-
rooctadecanoic acid and 12,12,12-trifluorododecanoic acid, were
synthesized initially for 19F NMR studies of the stabilizing ef-
fect of vitamin E on liposome membranes.45 The synthesis of
18,18,18-trifluorooctadecanoic acid, CF3(CH2)16COOH, and its
use in the investigation of ion-surface interactions are reported
here.
1-Bromo-12,12,12-trifluorododecane (5). Concentrated H2SO4 (7.60
mL) and 48% HBr (30.0 mL) were added to 12,12,12-trifluoro-1-
dodecanol (4, 9.77 g, 40.67 mmol) alternatingly under ice-cooling, and
the mixture was refluxed at 135 °C for 5 h. After cooling, the mixture
was extracted with CH2Cl2. The CH2Cl2 layer was washed with H2O
then with saturated NaHCO3, dried over MgSO4, and concentrated under
vacuum. The residue was purified by column chromatography (SiO2,
hexane) to give 1-bromo-12,12,12-trifluorododecane (5, 12.27 g, 99%).
5: Pail yellow oil. Bp 114.0 °C/5.00 mmHg. MS m/z 302 (M+), 304
(M + 2). HRMS Calcd for C12H22BrF3 (M+) 302.086; found 302.087.
Experimental Sections
1
IR (neat) cm-1 2932, 2860 (C-H). H NMR (CDCl3) δ 1.15-1.49
(A) Synthesis of 18,18,18-Trifluorooctadecanoic Acid and 12,-
12,12-Trifluorododecanoic Acid. The experiments at Setsunan Uni-
versity, Japan, for synthesis and purification of these two acids are as
follows.
(14H, b), 1.55 (2H, t, J ) 7.0, 6.7 Hz), 1.85 (2H, tt, J ) 7.3, 7.2 Hz),
2.05 (2H, qt, J ) 10.6, 7.5 Hz), 3.40 (2H, t, J ) 7.0 Hz). 19F NMR
(CDCl3) ppm -3.42 (3F, t, J ) 10.6 Hz).
1-(12,12,12-Trifluorododecyl)cyclohexanol (6). In an atmosphere
of Ar, 1-bromo-12,12,12-trifluorododecane (5, 0.010 g, 0.033 mmol)
was added to a suspension of Li (0.054 g, 7.78 mmol) in Et2O (4.0
mL), and the mixture was immersed in an ultrasonic bath till the reaction
started (about 10 min). After the reaction started, a solution of 1-bromo-
12,12,12-trifluorododecane (5, 1.00 g, 3.32 mmol) in Et2O (10.0 mL)
was added dropwise in 3.5 h, then the mixture was stirred for an
additional 1 h. After cyclohexanone (0.70 mL, 6.75 mmol) was added
to this mixture, it was stirred at room temperature overnight. The
mixture was poured into ice-water, neutralized with 10% HCl, then
extracted with Et2O. The Et2O layer was washed with saturated NaCl,
dried over MgSO4, and concentrated under vacuum. The residue was
purified by column chromatography (SiO2, hexane-CH2Cl2, 7:3) to
give 1-(12,12,12-trifluorododecyl)cyclohexanol (6, 0.431 g, 40%). 6:
Yellow oil. MS m/z 322 (M+). HRMS Calcd for C18H33F3O (M+)
322.248; found 322.249. IR (neat) cm-1 3600-3200 (O-H), 2932,
2860 (C-H). 1H NMR (CDCl3) δ 1.15-1.75 (31H, b), 2.05 (2H, qt, J
) 10.3, 7.5 Hz). 19F NMR (CDCl3) ppm -3.42 (3F, t, J ) 10.3 Hz).
18,18,18-Trifluoro-6-oxooctadecanoic Acid (7). To a solution of
1-(12,12,12-trifluorododecyl)cyclohexanol (6, 1.52 g, 4.73 mmol) in
acetic acid (55.0 mL) was added CrO3 (0.379 g, 3.79 mmol) with
vigorous stirring for 10 min. Then CrO3 (4.13 g, 41.3 mmol) was added
slowly keeping the mixture under 30 °C by ice-cooling. After the
mixture was stirred for 3 h, H2O (50.0 mL) was added, and the mixture
was extracted with Et2O. The Et2O layer was washed by water, then
extracted with 5% NaOH. The aqueous layer was acidified with 36%
HCl and extracted with Et2O. The Et2O layer was dried over MgSO4
and concentrated under vacuum. The residue was recrystallized from
hexane to give 18,18,18-trifluoro-6-oxooctadecanoic acid (7, 0.773 g,
46.4%). 7: Colorless plates. Mp 52.0-53.0 °C. MS m/z 352 (M+). IR
(KBr) cm-1 3650-3200 (O-H), 2920, 2856 (C-H), 1734, 1710
Methyl 12,12,12-Trifluoro-10-iodododecanoate (2). Methyl 10-
undecenoate (1, 8.72 g, 44.0 mmol) was dissolved in MeOH (20.0 mL).
This solution and Raney Ni (W-2) (2.09 g) were then sealed in a
stainless steel tube. The tube was cooled to -78 °C, and CF3I (6.50
mL) was added using a vacuum line. The tube was sealed and stirred
at 80 °C for 20 h. After the catalyst was filtered off, the solvent was
evaporated under vacuum, and the residue was dissolved in CH2Cl2.
The CH2Cl2 layer was washed with saturated NaHCO3 and dried over
MgSO4. After evaporation of the solvent, the residue was distilled to
give methyl 12,12,12-trifluoro-10-iodododecanoate (2, 16.43 g, 95%).
2: Colorless oil. Bp 103.0 °C/0.0190 mmHg. MS m/z 394 (M+). HRMS
Calcd for C13H22F3IO2 (M+) 394.062; found 394.063. IR (neat) cm-1
2932, 2864 (C-H), 1742 (COOCH3). 1H NMR (CDCl3) δ 1.15-1.45
(10H, b), 1.62 (2H, quin, J ) 7.3 Hz), 1.75 (2H, m), 2.30 (2H, t, J )
7.6 Hz), 2.83 (2H, m), 3.67 (3H, s), 4.19 (1H, m). 19F NMR (CDCl3)
ppm (from C6H5CF3) -1.12 (3F, t, J ) 10.3 Hz).
Methyl 12,12,12-Trifluorododecanoate (3). Zn (1.97 g, 30.1 mmol)
NiCl2‚6H2O (0.36 g, 1.50 mmol), and H2O (5 dr) were stirred in THF
(25 mL) for 15 min in a stream of Ar, and methyl 12,12,12-trifluoro-
10-iodododecanoate (2, 5.85 g, 14.8 mmol) was added to this mixture.
After the mixture was stirred at room temperature for 3 h, it was poured
into saturated NaHCO3. After the mixture was stirred for 30min, it
was filtered through a Celite layer. The layer was washed with Et2O.
The filtrate and washings were combined and extracted with Et2O. The
Et2O layer was washed with H2O, dried over MgSO4, and concentrated
under vacuum. The residue was separated by column chromatography
(SiO2, hexane-Et2O, 9:1) to give methyl 12,12,12-trifluorododecanoate
(3, 3.75 g, 94%). 3: Colorless oil. MS m/z 268 (M+). HRMS Calcd
for C13H23F3O2 (M+) 268.165; found 268.166. IR (neat) cm-1 2936,
2864 (C-H), 1744 (COOCH3). 1H NMR (CDCl3) δ 1.15-1.45 (14H,
b), 1.58 (2H, m), 2.05 (2H, qt, J ) 10.6, 7.5 Hz), 2.30 (2H, t, J ) 7.6
Hz), 3.66 (3H, s). 19F NMR (CDCl3) ppm -3.42 (3F, t, J ) 10.6 Hz).
12,12,12-Trifluoro-1-dodecanol (4). A solution of methyl 12,12,-
12-trifluorododecanoate (3, 2.70 g, 10.1 mmol) in Et2O (10.0 mL) was
dropped into a suspension of LiAlH4 (0.458 g, 12.1 mmol) in Et2O
(20.0 mL) at room temperature in an atmosphere of Ar, and the mixture
was stirred at room temperature for 5 h. The mixture was treated with
1
(COOH, CdO), H NMR (CDCl3) δ 1.15-1.88 (14H, b), 2.07 (2H,
m), 2.37 (4H, m), 2.47 (2H, t, J ) 7.3 Hz).19F NMR (CDCl3) ppm
-3.42 (3F, t, J ) 10.1 Hz).
18,18,18-Trifluorooctadecanoic Acid (8). A suspension of Zn (1.40
g, 21.3 mmol) and HgCl2 (0.124 g, 0.457 mmol) in 36% HCl (0.60
mL) and H2O (15.0 mL) was stirred for 5 min. After the surface of Zn
was amalgamated, the aqueous phase was decanted off. To this Zn,
H2O (4.00 mL), 36% HCl (11.0 mL), toluene (6.40 mL), and 18,18,-
18-trifluoro-6-oxooctadecanoic acid (7, 0.154 g, 0.437 mmol) were
added in this order, and the mixture was refluxed for 30 h under
vigorous stirring. To keep the acidity of the medium, 36% HCl (0.50
mL) was added every 6 h. The mixture was diluted with H2O and
extracted with Et2O. The Et2O layer was washed with H2O, dried over
MgSO4, and concentrated under vacuum. The residue was recrystallized
from hexane to give 18,18,18-trifluorooctadecanoic acid (8, 0.042 g,
28%). 8: Colorless plates. Mp 68.7-69.0 °C. MS m/z 338 (M+). HRMS
(42) Gu, C.; Angelico, V. J.; Wysocki, V. H.; Harada, A.; Takaya, H.;
Kumadaki, I. Proceedings of The 46th ASMS Conference on Mass
Spectrometry and Applied Topics; Orlando, FL, 1998.
(43) Miura, Y. F.; Takenaga, M.; Koini, T.; Graupe, M.; Garg, N.;
Graham, R. L.; Lee, T. R. Langmuir 1998, 14, 5821-5825.
(44) Houssiau, L.; Graupe, M.; Colorado, R.; Kim, H. I.; Lee, T. R.;
Perry, S. S.; Rabalais, J. W. J. Chem. Phys. 1998, 109, 9134-9147.
(45) Nishimura, M.; Koyama, M.; Takagi, T.; Ando, A.; Kumadaki, I.;
Urano, S. Bitamin E kenkyu no shimpo (Progress in Studies on Vitamin E
(Japanese); Sankyo Shuppan: Tokyo, 1995; Vol. V.

Ion-Surface Collisions with Langmuir-Blodgett Films
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10557
surface collision experiments. A Y-type multilayer L-B film is that
with hydrophobic-hydrophobic or hydrophilic-hydrophilic interactions
between two adjacent layers (i.e., tail-to-tail, head-to-head), built when
a vertically oriented substrate surface is repeatedly pulled-up and
pushed-down through the monomolecular film on the air-water
interface.40 The L-B films were deposited in a NIMA (Type-611,
England) L-B trough. The amphiphilic carboxylic acid molecules were
spread onto the water-air interface in a 0.1-1 mM solution of
chloroform (HPLC grade). The surface pressure of the monolayer during
the deposition was maintained at 32 × 10-3, 27 × 10-3, and 23 ×
10-3 N‚m-1 for n-C17H35COOH, n-C13F27COOH, and CF3(CH2)16COOH
L-B films (hereafter referred to as stearate, perfluorinated, and CF3-
(CH2)n-terminated L-B films, respectiVely). The dipping speed of the
metal substrate was kept at 3 mm‚min-1. Stearic acid (99%) and
perfluoroteradecanoic acid (97+%) were purchased from Aldrich and
used without further purification. The high-purity water ([conductivity]
-1
g 18 MΩ‚cm) used for the subphase in the L-B trough was
generated by a Nanopure (Barnstead, Dubuque, IA) or a Milli-Q-plus
(Millipore, Bedford, MA) ultrapure water system. Mechanically polished
aluminum plates, cleaned by ultrasonicating in various organic solvents
after the plates were exposed to air for a few hours (the presence of
oxide is presumed and desirable), were used as the substrates of the
L-B films. The L-B films prepared (the film covering area on the
substrate was as large as 20 mm × 30 mm) were put in a surface holder
and stayed in the vacuum chamber (of the instrument described below)
overnight before ion-surface collision experiments.
(C) Ion-Surface Collision Experiments. Ion-surface collisions
were carried out on a SID tandem mass spectrometer that has been
described previously.12,48 Briefly, the mass spectrometer consists of two
Extrel (now ABB Extrel, Pittsburgh, PA) 4000 amu quadrupole mass
analyzers positioned at 90° relative to each other, with the surface
bisecting the two quadrupoles. The precursor ions used in this study
were all generated in an electron impact ionization (EI) source. All
chemicals used to generate projectile ions, except C60, were introduced
into the EI source region through a needle valve inlet attached to a
sample reservoir. C60 was introduced by thermal desorption at 400 °C
from a direct probe for solids. The atomic metal ions (e.g., Mo+, Cr+)
used in the experiments were generated from the metal hexacarbonyls
by EI. The beam of precursor ion that are mass selected by the first
quadrupole was incident onto the surface at roughly 45° to the surface
normal. Product ions resulting from collisions with surfaces were
collected by focusing electrostatic-lenses and analyzed by the second
quadrupole. The collision energy was determined by the potential
difference, between the EI source block and the target surface,
multiplied by the charge of the ion. The SID instrument has a surface
holder that can carry several surfaces, which makes it possible to obtain
results for different surfaces under the same instrumental conditions
Figure 1. Surface pressure as a function of surface area (i.e., Langmuir
isotherms), measured at room temperature in a L-B trough, with 5 ×
10-5 M Al3+ in water as the subphase for CF3(CH2)16COOH, CF3(CH2)10-
COOH, and n-C13F27COOH and with 5 × 10-4 M Cd2+ in water as the
subphase for stearic acid.
Calcd for C18H33F3O2 (M+) 338.243; found 338.242. IR (KBr) cm-1
3700-3300 (O-H), 2920, 2852 (C-H), 1706 (COOH). 1H NMR
(CDCl3) δ 1.15-1.45 (24H, b), 1.55 (2H, quin, J ) 7.8 Hz), 1.64 (2H,
quin, J ) 7.2 Hz), 2.07 (2H, qt, J ) 10.1, 7.5 Hz). 19F NMR (CDCl3)
ppm -3.42 (3F, t, J ) 10.1 Hz).
12,12,12-Trifluorododecanoic acid (9) was converted from methyl
12,12,12-trifluorooctadecanoate (3) via NaOH catalyzed hydrolysis
followed by acidification. 9: Colorless wool. Mp 49.5-50.0 °C. MS
m/z 254 (M+). HRMS Calcd. for C12H21F3O2 (M+) 254.149; found
254.149. IR (KBr) cm-1 3700-2200 (O-H), 2928, 2860 (C-H), 1702
(COOH). 1H NMR (CDCl3) δ 1.15-1.45 (14H, b), 1.58 (2H, m), 2.05
(2H, qt, J ) 11.0, 7.50 Hz), 2.30 (2H, t, J ) 7.63 Hz). 19F NMR
(CDCl3) ppm -3.42 (3F, t, J ) 11.0 Hz).
(B) Preparation of L-B Films. The preparation of the L-B film
from octadecanoic acid (n-C17H35COOH, stearic acid) follows a well-
established “textbook” method with Cd2+ as the counterion in the
aqueous subphase (5 × 10-4 M CdCl2).40 The preparation of the L-B
film from perfluorotetradecanoic acid (n-C13F27COOH) follows the
method of Nakahama et al.,46 who first successfully deposited the L-B
film of a perfluorinated fatty acid (e.g., n-C10F21COOH) by using a
subphase containing Al3+. We have found that a well-packed monolayer
of 18,18,18-trifluorooctadecanoic acid forms on 5 × 10-5 M Al3+
(AlCl3) aqueous subphase, the same subphase used by Nakahama et
al. for a perfluorinated fatty acid.46 It is noteworthy that we found no
packed monolayer of 18,18,18-trifluorooctadecanoic acid forms on the
subphase of either Cd2+ (0.5 mM CdCl2) or 0.01 M HCl aqueous
solution which are two typical subphases40 used for making L-B films
with “normal” fatty acids. Figure 1 gives the surface pressure as a
function of area per molecule (i.e., π ∼ A, Langmuir isothermal curve),
measured in a L-B trough with 18,18,18-trifluorooctadecanoic acid
molecules on the air-water interface. The measured isothermal curves
for stearic acid and perfluorotetradecanoic acid are provided as dashed
lines in Figure 1 for comparison. The isothermal curve of 12,12,12-
trifluorododecanoic acid, which has a shorter chain length than 18,-
18,18-trifluorooctadecanoic acid, is also provided in Figure 1. Note
this shorter chain length ω,ω,ω-trifluoroalkanoic acid does not form a
close-packed (“solid phase”) monolayer on the air-water interface.47
by moving each surface into the ion path. The ion beam density is less
49
than 1 nA in a 0.5 cm2 area
of the ion-surface impact. In the
experiments of the present study, each film was used for 10-30 min
in ion-surface collisions. Note that the “static” limit of ∼1012 ions/
cm2 in SIMS1 may not be applied here, because the collision energies
in our study are much lower than those in SIMS.
Results and Discussion
Different chemistry occurs when organic projectile ions
collide with flurocarbon vs hydrocarbon terminated organic thin
films (L-B films in the present study). Several sets of
experimental data are presented and discussed below that are
aimed at determining the effect of different terminal surface
groups (namely CF3 vs CH3) on the different chemistry of
hyperthermal ion-surface interactions. The experimental data
include those for internal energy deposition, electron transfer,
and ion-surface reaction. Also provided are comparisons of
(47) This is not surprising. The longer the amphiphilic molecule, the
better the quality of the L-B film. For example, to prepare a vinyl-
terminated L-B film, naturally occurring CH2dCH(CH2)8COOH had to
be expanded to CH2dCH(CH2)20COOH by chemical synthesis, see:
Barraud, A.; Rosilio, C.; Ruaudel-Teixier, A. J. Colloid Interface Sci. 1977,
62, 509-523.
Three-layer Y-type L-B films, with the hydrophobic end as the
terminal group at the vacuum-film interface, were used in our ion-
(46) Nakahama, H.; Miyata, S.; Wang, T. T.; Tasaka, S. Thin Solid Films
1986, 141, 165-169.

10558 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Gu et al.
Figure 2. The product ion spectra acquired when benzene molecular
ion (m/z 78) collides at 30 eV kinetic energy with Y-type three-layer
L-B films prepared from (a) n-C13F27COOH (Al3+ salt), (b) CF3(CH2)16-
COOH (Al3+ salt), and (c) n-C17H35COOH (Cd2+ salt).
Figure 4. Energy-resolved mass spectra for surface-induced dissocia-
tion (SID) of W(CO)6+ with (a) n-C13F27COOH, (b) CF3(CH2)16COOH,
and (c) n-C17H35COOH L-B films. The products of “chemical side
+
reactions” other than the unimolecular dissociation of W(CO)6 are
not included (e.g., ion-surface reactions and chemical sputtering of
surface species during SID). The total abundance of those “byproduct”
ions increases in the raw spectra as collision energy increases (e.g.,
approximately 10% of the total scattered ions at 60 eV collisions with
perfluorinated and CF3(CH2)n-terminated L-B films, a little more than
20% at 70 collisions with the same two films).
to those for the perfluorinated film, but quite different from
those for the stearate film. This is evident by the following three
observations: (i) overall shape (product ion distribution) of the
spectra, (ii) ratios of low mass fragments, and (iii) ion-surface
reaction products.
It is seen in Figures 2 and 3 that the CF3(CH2)n-terminated
L-B film provides a percent fragmentation of the projectile
ions similar to that given by the perfluorinated L-B film
(Figures 2a vs 2b and 3a vs 3b), but different from that produced
by collision with the stearate L-B film (Figures 2c and 3c).
Furthermore, some low mass fragments that require relatively
higher energies, such as C2H2+ (m/z 26) from benzene (Figure
Figure 3. The product ion spectra acquired when dimethyl sulfoxide-
d6 (DMSO-d6) molecular ion (m/z 84) collides at 20 eV kinetic energy
with (a) n-C13F27COOH, (b) CF3(CH2)16COOH, and (c) n-C17H35 COOH
L-B films.
+
2a,b) and CD3 (m/z 18) from DMSO-d6 (Figure 3a,b), are
present in SID spectra for both perfluorinated and CF3(CH2)n-
terminated L-B films, but are absent in the spectra for the
stearate L-B film (Figures 2c and 3c). All of these results
indicate that similar internal energy is deposited into the
projectile ions with CF3(CH2)n-terminated and perfluorinated
L-B films.
data for polyatomic ions vs atomic ions, and lower collision
energies (e.g., <100 eV) vs a slightly higher energy (e.g., 220
eV).
Transfer of Kinetic Energy into Internal Modes for
Polyatomic Projectile Ions. The energy deposition at collision
energies <100 eV was investigated by using several different
projectile ion probes: e.g., benzene molecular ion, dimethyl
sulfoxide-d6 (DMSO-d6) molecular ion, and tungsten hexacar-
bonyl molecular ion. The product ion spectra recorded when
benzene molecular ion (m/z 78) collides at 30 eV with three
different L-B films are shown in Figure 2. The spectra of
DMSO-d6 molecular ion (m/z 84) at 20 eV collisions with the
same L-B films are given in Figure 3. As shown by these
spectra, the CF3(CH2)n-terminated L-B film gives results similar
+
The fragmentation patterns produced when the W(CO)6
projectile ion collides with the films were used to estimate the
percentage of kinetic energy transferred to internal energy (T
+
f V). W(CO)6 is an established “thermometer” ion (with a
known unimolecular dissociation pattern and established dis-
sociation energetics) that has been used for this purpose
previously.6,18,34,35,48,50 The energy-resolved SID mass spectra
+
of W(CO)6 (up to 70 eV) obtained with the three L-B film
targets are illustrated in Figure 4. The trend observed with
+
W(CO)6 (Figure 4) for these three L-B films is the same as

Ion-Surface Collisions with Langmuir-Blodgett Films
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10559
Figure 6. Total scattered-ion current (TIC) recorded by the detector
of the SID tandem mass spectrometer when benzene molecular ions
collide at 30 eV kinetic energy with three different L-B films.
+
Figure 5. Estimated average internal energy deposited into W(CO)6
during SID with three different L-B films as a function of collision
energies. Two data points at 60 and 70 eV collisions with perfluorinated
and CF3(CH2)n-terminated L-B films are not included in the linear fit
(see discussion in text).
Another point that must be made clear is that we do not intend
to allege that the deposited internal energy as a function of the
collision energy should be linear over a large range of energy,
although a good linear relationship is shown in Figure 5 within
a range of collision energies. However, we also cannot conclude
that the ratio of Tf V decreases at the collision energies above
60 eV with perfluorinated and CF3(CH2)n-terminated L-B films,
just based on the deviation of two data points at 60 and 70 eV
for those films. There are limitations of using a “thermometer”
ion for the estimation of Tf V during SID: (i) “Chemical side
reactions”, such as ion-surface reactions and chemical sput-
tering during SID, occur and are not included in the estimation
of internal energy. Significant errors could be caused if such
“chemical side reactions” are not negligible. (ii)The maximum
internal energy that can be estimated using a “thermometer”
ion is limited by the molecular structures, dissociation pathways,
and activation energies for dissociation. For example, the energy
that observed with the two small organic molecular ions (Figure
2 and 3). For example, the maximum abundance of the fragment
W(CO)4+, formed by the loss of two CO molecules, occurs at
approximately 20 eV collision energy with both perfluorinated
and CF3(CH2)n-terminated L-B films (Figure 4a,b), but at
approximately 40 eV collision energy with the stearate film
(Figure 4c). Using an algorithm previously published by Cooks
and co-workers,50,51 the distribution of internal energy deposited
+
into W(CO)6 at each collision energy was estimated. The
weighted average of each estimated internal energy distribution
is plotted against collision energy in Figure 5.
The estimated ratio of T f V for W(CO)6+ is determined as
18.1% ((1.5), 17.4% ((0.8), and 11.9% ((0.6) for the
perfluorinated, CF3(CH2)n-terminated, and stearate L-B films,
respectively, assuming a linear fit is applicable (Figure 5). The
errors listed in parentheses above represent twice the standard
deviation of the least-squares linear fit shown in Figure 5, but
do not represent the genuine error that could be caused by the
algorithm for the estimation of the internal energies. The
estimated T f V ratios for perfluorinated and CF3(CH2)n-
terminated L-B films are in good agreement with the previously
reported value of 19-20% for the fluoroalkanethiolate SAM
surface on gold prepared from CF3(CF2)11(CH2)2SH,18 and the
reported value of 18% for liquid perfluoropolyether (PFPE) thin
films on a metal.34 Those cited values were obtained at a
collision angle similar to that in our SID instrument and using
the same algorithm as that employed for L-B films in this study.
The estimated value for the stearate L-B film is comparable
to the value of 11-13% for SID on stainless steel covered with
adventitious hydrocarbons.18,48 Different methods for the internal
energy estimation other than the algorithm used in this study
have also been published.23,36,52 However, this study concerns
mainly the trend in internal energy deposition among three L-B
films and other estimation methods will not be discussed here.
+
range that can be estimated with W(CO)6 is about 13.0 eV
(i.e., W(CO)6+ f W+, E0 ) 13.0 eV). As a consequence, when
a “thermometer” ion with a small maximum limit range such
as Cr(CO)6 is used (Cr(CO)6 f Cr+, E0 ) 5.9 eV), the
observed deviation from linearity starts at even lower collision
energies with perfluorinated and CF3(CH2)n-terminated L-B
films where no ion-surface reactions and chemical sputtering
were observed (data not shown here).
+
+
Electron Transfer from Surface to Polyatomic Projectile
Ions. Neutralization of projectile ions occurs by electron transfer
from target surfaces during ion-surface collisions.6,53 The
relative tendency of three L-B films to cause neutralization of
benzene molecular ions incident onto the films at 30 eV is
presented in Figure 6. The figure is a plot of total scattered-ion
currents (TIC) recorded when the precursor ions collide with
the L-B films, measured with the electron multiplier detector
of the SID tandem mass spectrometer. The total scattered-ions
include intact precursor ions, ionic fragments, and ion-surface
reaction product ions, in short, all the ions appearing in the
spectra shown in Figure 2. The baseline in Figure 6 was recorded
when the two filaments in the EI source (Extrel, Pittsburgh,
PA) were turned off. Higher total scattered-ion currents (TIC)
were recorded for perfluorinated and CF3(CH2)n-terminated
L-B films compared with that for the stearate film (Figure 6).
This indicates less neutralization on both the perfluorinated film
and the CF3(CH2)n-terminated films than that on the stearate
L-B film. The measurements reflected by the TIC curve are
(48) Wysocki, V. H.; Ding, J. M.; Jones, J. L.; Callahan, J. H.; King, F.
L. J. Am. Soc. Mass Spectrom. 1992, 3, 27-32.
(49) Kane, T. E.; Angelico, V. J.; Wysocki, V. H. Anal. Chem. 1994,
66, 3733-3736, 4564-4564.
(50) Wysocki, V. H.; Kenttamaa, H. I.; Cooks, R. G. Int. J. Mass
Spectrom. Ion Process. 1987, 75, 181-208.
(51) Dekrey, M. J.; Kenttamaa, H. I.; Wysocki, V. H.; Cooks, R. G.
Org. Mass Spectrom. 1986, 21, 193-195.
(53) Cooks, R. G.; Ast, T.; Pradeep, T.; Wysocki, V. Acc. Chem. Res.
1994, 27, 316-323.
(52) Vekey, K. J. Mass Spectrom. 1996, 31, 445-463.

10560 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Gu et al.
Figure 8. Spectra resulting from 220 eV collisions of C602+ (m/z 360)
with the same films as those for Figure 7, i.e., (a) n-C13F27COOH, (b)
CF3(CH2)16COOH, and (c) n-C17H35COOH L-B films.
2+
Figure 7. Spectra resulting from 80 eV collisions of C60 (m/z 360)
with L-B films prepared from (a) n-C13F27COOH, (b) CF3(CH2)16-
COOH, and (c) n-C17H35COOH.
by five different methods). Almost all measured IE[C60] are
consistent at 7.6 eV, whereas eleven reported IE[C60•+] values
range from approximately 9 to 12 eV with the arithmetic mean
at ∼11.0 eV.56 Regardless of the exact value, the second
ionization energy of C60 is higher than, or at least comparable
to, the expected ionization energy of the CH3(CH2)n-terminated
surface (e.g., IE[C10H22] ) 9.7 eV). Thus, electron transfer from
an alkyl-terminated surface to C602+ is expected to be exother-
mic. This is in agreement with the fact that more than 99% of
in agreement with the current measured at the surface by a
digital picoamperemeter connected to the target surface during
the collisions. The higher the electric current at the target
surface, the lower the ion current at the detector of the tandem
mass spectrometer. For example, one set of measurement
averages of surface current is recorded as 0.37 nA:0.39 nA:
0.50 nA for perfluorinated:CF3(CH2)n-terminated:stearate L-B
films, respectively. In summary, the efficiency of electron
transfer from the films to benzene molecular ion follows the
order perfluorinated ≈ CF3(CH2)n-terminated < stearate L-B
films.
•+
2+
the scattered ions recorded are C60 ions when C60 collides
with the stearate L-B film (Figure 7c). In contrast, the expected
higher ionization energy for the fluorocarbon surface (e.g., IE-
[C3F8] ) 13.4 eV) reduces the likelihood that electron transfer
The relative electron transfer from three L-B films was also
investigated in a different way by using the C60 (buckminster-
2+
will occur when C60 is incident on a fluorocarbon surface.
2+
fullerene, “bucky ball”) doubly charged molecular ion (C60
,
As a result of the higher IE of the fluorocarbons, a large portion
m/z 360). C60 was chosen because of its remarkable stability in
2+
of the C60 ions scatter from the surface without charge
the gas phase. One noteworthy example in the literature is that,
reduction (Figure 7a). Finally, more evidence for control of the
electron transfer by the outermost group on the surfaces is that
•+
in keV collisions with helium as target gas (keV CID), C60
can encage a helium atom instead of fragmenting.54,55 Due to
the stability of C60, almost no fragments are observed when 80
eV kinetic energy is applied in the collision of C60 with
2+
the CF3(CH2)n-terminated L-B film gives a ratio of C60•+:C60
that is quite similar to that for the perfluorinated surface (Figure
7b vs 7a).
2+
the L-B films (Figure 7). The electron-transfer reaction
measured is
2+
220 eV Collisions of C60 with L-B Films. SID spectra
of C602+ at 220 eV are given in Figure 8. At this SID collision
energy, a number of fragment ion peaks at 12 or 24 u apart are
present in the spectrum with the perfluorinated L-B film (Figure
8a). These peaks correspond to doubly or singly charged
fragments that are produced either by loss of a single entity
C2n57 or by the sequential loss of nC258 from C60 ions. In contrast,
the major fragment formed when C602+ collides with the stearate
L-B film is seen as one peak of [C60-C2]+ at a very low
abundance (Figure 8c). The differences between the spectra
acquired with the perfluorinated vs the stearate L-B films are
similar to that reported previously using SAM surfaces of
fluoroalkanethiolate vs alkanethiolate.13 Even at this relatively
higher collision energy, the CF3(CH2)n-terminated L-B film
C602+ + e [from surface] f C60•+, RE (≈-IE[C60•+])
where RE is the recombination energy and IE the ionization
energy. Many researchers have measured the IE of C60 by using
various methods. For example, Scheier et al.56 compiled 8
reported values for the first ionization energy of C60 (i.e., IE-
[C60], measured by six different methods) and 11 reported values
for the second ionization energy of C60 (i.e., IE[C60•+], measured
(54) (a) Weiske, T.; Bohme, D. K.; Hrusak, J.; Kratschmer, W.; Schwarz,
H. Angew. Chem., Int. Ed. Engl. 1991, 30, 884-886. (b) Weiske, T.; Bohme,
D. K.; Schwarz, H. J. Phys. Chem. 1991, 95, 8451-8452. (c) Weiske, T.;
Hrusak, J.; Bohme, D. K.; Schwarz, H. Chem. Phys. Lett. 1991, 186, 459-
462.
(55) Ross, M. M.; Callahan, J. H. J. Phys. Chem. 1991, 95, 5720-5723.
(56) Scheier, P.; Dunser, B.; Worgotter, R.; Lezius, M.; Robl, R.; Mark,
T. D. Int. J. Mass Spectrom. Ion Process. 1994, 138, 77-93.
(57) McHale, K. J.; Polce, M. J.; Wesdemiotis, C. J. Mass Spectrom.
1995, 30, 33-38.
(58) Scheier, P.; Dunser, B.; Worgotter, R.; Muigg, D.; Matt, S.; Echt,
O.; Foltin, M.; Mark, T. D. Phys. ReV. Lett. 1996, 77, 2654-2657.

Ion-Surface Collisions with Langmuir-Blodgett Films
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10561
still gives a SID spectrum more like that with the perfluorinated
film (Figure 8b vs 8a), other than that with the stearate L-B
film (Figure 8c), in terms of fragmentation of C60 and charge
2+
reduction from C60 to C60•+. Nonetheless, at this higher
collision energy, a little more charge reduction and a little less
fragmentation is provided by the CF3(CH2)n-terminated film in
comparison with the perfluorinated L-B film (Figure 8b vs 8a).
The results of a molecular dynamics (MD) simulation done by
Mowrey et al. show that the carbon lattice on a hydrogen-
terminated diamond surface {111} is crumpled three layers deep
during the impact of a neutral C60 at 200 or 250 eV.59,60 In
other words, the differences as shown with CF3(CH2)n-
terminated vs perfluorinated L-B films (Figure 8b vs 8a)
indicate some significant contributions from the underlying
(CH2)n vs (CF2)n for C60 collisions with L-B films at 220 eV.
Ion-Surface Reactions with L-B Films. DMSO-d6 mo-
lecular ion (m/z 84) was chosen as a probe for hydrogen
abstraction from the films (e.g., Figure 3c). Benzene molecular
ion (m/z 78) is a good probe to investigate fluorine addition
versus the addition of a hydrocarbon unit from the films, e.g.,
the product ions C6H4F+ and C6H5F+ (m/z 95 and 96, respec-
tively; Figure 2a) versus the product ion C7H7+ (m/z 91; Figure
2c). For the perfluorinated L-B film, a small amount of m/z
91 (not labeled, Figure 2a) and a small amount of [DMSO-d6
+ H]+ (not labeled, Figure 3a) indicate that the film is slightly
contaminated by hydrocarbon, either during the preparation or
in the vacuum chamber (10-7 to 10-6 Torr pressure). The ratio
of m/z 91 to m/z 95 and 96 in benzene spectra with the
perfluorinated L-B film (Figure 2a) is similar to that with a
fluoroalkanethiolate SAM on gold prepared by 12 h immersion
of gold substrate in the fluoroalkanethiol solution (unpublished
data in the Wysocki research group), whereas the same SAM
surface prepared by 72 h immersion gives almost no m/z 91.21
Note that hydrocarbon is much more reactive for the CHx
transfer to the projectile ions than fluorocarbon for the F
addition. Therefore, the relative abundance of m/z 91 vs m/z 95
and& 96 (Figure 2a) does not represent the relative composition
of two different species on the surface. With the CF3(CH2)n-
terminated L-B film, the reaction products are similar to
(although not exactly the same as) those formed with the
perfluorinated L-B films (Figures 2b vs 2a and 3b vs 3a). This
suggests that the terminal group is also the main determinant
of these atomic/group transfers to the polyatomic projectile ions,
in addition to being the determinant of the energy transfer and
electron transfer for polyatomic projectile ions as discussed in
previous sections. A slightly higher CHx/H adduct ion intensity
for the CF3(CH2)n-terminated L-B film relative to that for the
perfluorinated film (Figures 2b vs 2a and 3b vs 3a) could be
the result of several possible factors: First, the packing quality
of the L-B film for 18,18,18-trifluorooctadecanoic acid and
perfluoroteradecanoic acid could be slightly different and thus
the films could have different amounts of defect sites, and/or,
different tendencies for contamination after formation (e.g., in
the vacuum chamber). Second, exposed methylenes of CF3-
(CH2)n- at surface defect sites or some disruption sites caused
by ion impact could contribute to a slightly higher CHx/H adduct
ion intensity. Future investigation of these ion-surface reactions
with the SAM surface of CF3(CH2)nSH43 on gold will be of
interest. It is worth noting that a L-B film made from 12,12,-
12-trifluorododecanoic acid [CF3(CH2)10COOH] that does not
form a “solid phase” on the air-water interface in the L-B
Figure 9. The metal atomic ion and its fluorine adduct ion region of
the product ion spectra of Mo+ incident at 60 eV kinetic energy on
(a) n-C13F27COOH and (b) CF3(CH2)16COOH L-B films. Solid
line: isotope 98Mo+ selected as the precursor ion. Dashed line: isotope
92Mo+ selected as the precursor ion.
trough (refer to Figure 1) gives a much higher intensity of m/z
91 vs m/z 95 and 96 for the product ion spectrum of benzene
molecular ion (spectra not shown here). The percentage
fragmentation is accordingly lower with the CF3(CH2)10COOH
L-B film (spectra not shown here) than that obtained with the
CF3(CH2)16COOH L-B film.
Fluorine addition to an atomic ion with the L-B films was
investigated by using Mo+ or Cr+ ions as the projectile probes.
Figure 9 shows the precursor ion and adduct ion regions of the
product ion spectra for a 98Mo+ (solid line) or 92Mo+ (dashed
line) incident at 60 eV onto the perfluorinated L-B film (part
a) and CF3(CH2)n-terminated L-B film (part b). Additions of
multiple fluorine atoms to a Mo+ projectile are seen with the
perfluorinated L-B film (Figure 9a). The spectra obtained with
perfluorinated L-B films are consistent with that previously
reported by Cooks and co-workers using a fluoroalkanethiolate
SAM surface of CF3(CF2)11(CH2)2SH on gold.28 Cooks and co-
workers suggested that some of the fluorine atoms in the
additions come from the underlying -CF2- on the fluoroal-
kanethiol SAM surface (moreover, two or more of the multiple
fluorine atoms could be from the same surface fluoroalkyl
chain).28 Arguments in their proposed mechanism include the
following: (i) Some of the required energy for the endothermic
processes of M+ f MFn could be provided by the energy
+
released in the concomitant CdC bond formation on the surface
(e.g., a double bond formed between two carbon atoms of
terminal CF3CF2 by abstraction of one fluorine atom from each
carbon) and (ii) the number of fluorine atoms in the additions
and the relative intensities of the adduct ion peaks are dependent
on the incident angle of the primary ion beam, with the
maximum at the angle of the surface normal.28 In other words,
the multiple number of fluorine additions could be dependent
on the penetration of the atomic metal ion into a depth of the
surface to reach at least the first CF2 unit underneath the terminal
CF3 group. According to their proposed mechanism,28 the
probability of fluorine additions would be predicted to decrease
for the CF3(CH2)n-terminated film. This is in agreement with
the experimental observation that the adduct ion peaks are of
lower intensities with CF3(CH2)n-terminated vs perfluorinated
L-B films (Figure 9b vs 9a). Similar results were also observed
(59) Mowrey, R. C.; Brener, D. W.; Dunlap, B. I.; Mintmire, J. W.;
White, C. T. J. Phys. Chem. 1991, 95, 7138-7142.
(60) Beck, R. D.; John, P.; Alvarez, M. M.; Diederich, F.; Whetten, R.
L. J. Phys. Chem. 1991, 95, 8402-8409.

10562 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Gu et al.
with a 52Cr+ ion (spectra not shown here). For example, the
modes of a polyatomic impinging ion. Moreover, our results
suggest that the outermost surface group (e.g., CF3 vs CH3) is
the main determinant of this energy transfer with fluorinated
vs hydrocarbon-terminated surfaces for a low-energy (e.g., 10-
100 eV) polyatomic projectile ion. On the other hand, it is still
not clear whether the high electronegativity and low polariz-
ability of fluorine plays a role in energy transfer. This question
could be addressed in the future using a series of model surfaces
terminated by fluorine vs chlorine vs bromine. At this point,
our ability to understand the roles of underlying methylene/
fluoromethylene chains or an assembly of the surface backbone
chains in the conversion of translational energy into internal
modes of a polyatomic ion is still limited. First, the physical
processes of polyatomic ion-surface collisions can be very
complicated. For example, multibody collision or collective
collisions of a polyatomic ion with a number of atoms or groups
on the surface can occur during the ion-surface impact. The
orientation or average orientation of a polyatomic projectile ion
during the impact can influence the energy transfer. Moreover,
the time frames of the collisions (e.g., a low-energy polyatomic
ion collision with organic thin films) are not yet well understood.
Our experimental observations reported in this paper may
encourage further experimental and theoretical investigations
to better understand the processes.
+
intensity ratio of Cr+:CrF+:CrF2 is detected as 100:95:8.5 at
40 eV collision with the perfluorinated L-B, whereas no trace
+
of CrF2 is detected for 40 eV collision with the CF3(CH2)n-
terminated film and the ratio of Cr+:CrF+ is 100:19. The
penetration of the atomic projectile ion into a depth of the CF3-
(CH2)n-terminated film is confirmed by our observation of the
extent of neutralization of Mo+ and Cr+ incident on the film.
By comparison of total scattered-ion currents (TIC) measured
for Mo+ incident at 60 eV on the L-B films (TIC not shown
here), the neutralization for 60 eV Mo+ collisions with the films
has the following order: perfluorinated L-B < CF3(CH2)n-
terminated L-B ≈ stearate L-B. This is also illustrated by
different signal-to-noise ratios in Figure 9. Relative to that for
the perfluorinated film, the signal-to-noise for the CF3(CH2)n-
terminated L-B is worse (Figure 9b vs 9a) due to the weaker
signals (i.e., as a result of higher neutralization), although the
resolution of the quadrupole mass analyzer was actually opened
a little in the acquisition of the spectra for the CF3(CH2)n-
terminated L-B film. Based on the results discussed above,
our experiments provide further evidence that could support the
mechanisms proposed by Cooks and co-workers.28 The similar-
ity between CF3(CH2)n- and CH3(CH2)n-terminated films for
electron transfer to the atomic ion Mo+ at 60 eV collisions is
in contrast to what is observed for polyatomic ions such as
The results of electron transfer (neutralization of a projectile
ion) and atom/group transfer (ion-surface reaction) to an
impinging ion are distinctive for polyatomic ions vs atomic ions.
For polyatomic ions, the outermost surface group seems to be
the main determinant for the efficiency of these transfers. The
trends for different L-B films toward the electron transfer to a
polyatomic ion, investigated in the absence of solvent and
electric double-layer effects, might provide information relevant
to studies of electron transfer at electrode-solution interfaces
and increase our knowledge of electron transfer through organic
thin films. Our results for neutralization and ion-surface
reaction of an atomic ion, together with ion scattering and ion-
surface reaction results reported by other researchers, suggest
the penetration of an atomic impinging ion into a depth of the
films (e.g., the methylene portion of a CF3(CH2)n-terminated
film).
2+
benzene molecular ion (Figure 6) and C60 (Figure 7), and is
consistent with penetration of atomic metal ions into the
hydrocarbon portion of the CF3(CH2)n- film. The observations
discussed above suggest that there is a projectile-size dependence
for the low-energy ion-surface interactions. The results of Ar+
scattering from a SAM surface of CF3(CH2)15S-Au, recently
reported by Rabalais and co-workers,44 also suggest the penetra-
tion of atomic Ar+ ion into a few atomic layers of the surface.
Conclusions
The use of model surfaces, e.g., labeled by fluorine (this
study) or by isotopes (our previous Communication to this
journal) at the outermost carbon, is an effective way to
investigate low-energy ion-surface interactions. Recent devel-
opments in preparation of new thiolate SAMs and new L-B
films will lead to future research in this direction that will better
characterize the ion-surface collision processes. In turn, the
knowledge that is being and will be gained in the investigations
of the ion-surface interactions with various model surfaces
might lead to a surface characterization technique based on the
interactions, e.g., to probe the outermost exposed surface groups
or atomic layers.
Acknowledgment. Financial support of this work from the
National Science Foundation (CHE-9224719) is acknowledged.
The use of the L-B trough in the laboratory of Dr. S. Scott
Saavedra at the University of Arizona and discussions with him
regarding the preparation of L-B films are gratefully appreci-
ated. Vincent J. Angelico is acknowledged for the assistance
with reproducing TIC measurements of benzene SID.
Our results from this study support the previous speculation
by many researchers that the effective mass of the surface
controls the conversion of translational energy into internal
JA990719A