Calibration of charge state conversion surfaces for neutral particle detectors

Abstract

Molecular oxygen and hydrogen ions were scattered off hydrogen terminated diamondlike carbon (DLC) charge state conversion surfaces at incident grazing angles. The energy range of the scattered particles was 390–1000 eV, and the surface roughness of the DLC surface was of the order of 1 Å rms. For all surfaces almost equal angular scattering and negative ion fractions were found within the uncertainties of the measurement. This result supports the fact that charge state conversion with DLC surfaces is a reliable technology for neutral particle sensing instruments. Furthermore, these instruments can work in the laboratory as well as in the harsh environment on board a satellite.The surfaces measured here are used in the IBEX-lo sensor, a neutral particle sensing instrument on the NASA IBEX mission, which is scheduled for launch into orbit around Earth in July 2008.

Full text

Page 1

Calibration of charge state conversion surfaces for neutral particle
detectors
P. Wahlström,1,a?J. A. Scheer,1P. Wurz,1E. Hertzberg,2and S. A. Fuselier2
1Physikalisches Institut, University of Bern, Sidlerstrasse 5, 3012 Bern, Switzerland
2Lockheed Martin Advanced Technology Center, 3251 Hanover St., Palo Alto, California 94304, USA
?Received 24 September 2007; accepted 16 May 2008; published online 1 August 2008?
Molecular oxygen and hydrogen ions were scattered off hydrogen terminated diamondlike carbon
?DLC? charge state conversion surfaces at incident grazing angles. The energy range of the scattered
particles was 390–1000 eV, and the surface roughness of the DLC surface was of the order of 1 Å
rms. For all surfaces almost equal angular scattering and negative ion fractions were found within
the uncertainties of the measurement. This result supports the fact that charge state conversion with
DLC surfaces is a reliable technology for neutral particle sensing instruments. Furthermore, these
instruments can work in the laboratory as well as in the harsh environment on board a satellite.The
surfaces measured here are used in the IBEX-lo sensor, a neutral particle sensing instrument on the
NASA IBEX mission, which is scheduled for launch into orbit around Earth in July 2008. © 2008
American Institute of Physics. ?DOI: 10.1063/1.2957064?
I. INTRODUCTION
The interaction of atomic and molecular particles with
insulating surfaces has been researched extensively in recent
years.1–10Reports of relatively high fractions of negative
ions that result from scattering of positive atomic and mo-
lecular ions off insulating surfaces have suggested possibili-
ties for several new applications. Currently, we use this pro-
cess for efficient detection of 10 eV–2 keV neutral particles
in interplanetary and interstellar space.11–13The proof of con-
cept for this detection technique has already been demon-
strated with the IMAGE mission.14The mass spectrograph
used there is designed to detect low energy neutral atoms. It
uses a conversion surface of volatile adsorbates on a highly
polished polycrystalline tungsten substrate to convert a frac-
tion of the incoming neutral atoms into negatively charged
ions.14,15For future missions, such as the NASA IBEX mis-
sion, diamondlike carbon surfaces will be used as conversion
surfaces because of their better long-term stability and higher
negative ion yield.9,10,16
Nine flight-spare and nonflight hydrogen terminated dia-
mondlike carbon surfaces have been tested for their negative
ionization efficiencies and their angular scattering properties.
The dependency of these two properties on the incident angle
of the particles was investigated. These surfaces are identical
to all 28 trapezoid-shaped ?2.5 cm and 2.9?6.2 cm2? neu-
tral to negative conversion surfaces, which will be used for
the IBEX mission. Obviously, surfaces with nearly identical
scattering properties, i.e., angular scattering and ion conver-
sion efficiency, are required.
Since interstellar gas is expected to consist mainly of H
and He with traces of O, N, C, and Ne,17we focused our
tests on these particles. Measurements of the fraction of
negative ions produced upon scattering were performed with
O and H. He and Ne cannot produce stable negative ions.
However, He can produce a metastable negative ion.18The
yield of negative He ions upon surface scattering is of the
order of 10−5,19which is several orders of magnitude lower
than the sputter yield of negative ions. Thus, negative ion
yields measured by scattering He and Ne on charge state
conversion surfaces are dominated by several orders of mag-
nitude by the sputtered yield in this experiment. Therefore
He and Ne were used to determine the sputtering yields.
In the case of hydrogen and oxygen, molecular ions were
used because they can be produced far more efficiently than
atomic ions in our test system. The impact of using posi-
tively charged molecular ions on the results is discussed in
detail below.
II. EXPERIMENT
The Si wafers used as substrates for the tetrahedral
amorphous carbon ?ta-C? films are 1 mm thick with the ?100?
face exposed and boron doped for conductivity in the
1–20 ? cm range. Cut and polished wafer surfaces were
measured to be within 0.13° of the actual ?100? lattice plane.
The ta-C thin films were prepared by pulsed laser deposition
using an excimer laser ?248 nm, Lambda Physik Compex®
150 configured in the power oscillator power amplifier
mode? at high fluence ??100 J/cm2? to ablate a pyrolytic
graphite target in vacuum ??1?10−6mbar?. Under these
conditions the ablated carbon flux is highly ionized with a
broad energy distribution ranging from 20 to 300 eV. The
carbon ions penetrate and implant in the near surface region
?called subplantation20? causing densification and build up of
stress, ultimately leading to high fourfold ?i.e., sp3-bonded?
carbon content measured at ?80% ?Ref. 21? of the amor-
phous films. The subplantation also suppresses surface diffu-
sion, which is thought to be one reason why these films are
so smooth. Electrical conductivity22,23depends sensitively on
the amount of threefold ?i.e., sp2-bonded? coordinated carbon
present, and the degree of clustering and/or chain formation.
The electron conduction is accomplished by hopping from
a?Author to whom correspondence should be addressed. FAX: ?41-31-631-
4405. Electronic mail: peter.wahlstroem@space.unibe.ch.
JOURNAL OF APPLIED PHYSICS 104, 034503 ?2008?
0021-8979/2008/104?3?/034503/6/$23.00© 2008 American Institute of Physics
104, 034503-1
Downloaded 11 Aug 2008 to 130.92.9.58. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

Page 2

threefold clusters mediated by the cluster size and spacing.
Typically, resistivities of 106? cm are found for high four-
fold content films, but resistivities have been measured down
to 10 ? cm in annealed films with slightly lower fourfold
content.
The surfaces have been hydrogen terminated because at
the surface each carbon atom has one free bonding site that is
very reactive. The hydrogen termination process saturates all
free bonding sites and thus creates a chemically very stable
surface. Furthermore, the negative ion fraction is known to
increase slightly after the hydrogen termination.16Due to the
rather high base pressure of 5?10−8mbar in the vacuum
chamber used in this experiment, differences between hydro-
gen terminated and nonhydrogen terminated surfaces are dif-
ficult to identify because a layer of adsorbed hydrogen on the
surfaces is almost inevitable.
Surface smoothness is critical for minimizing scattering
losses in the downstream particle collection and analysis sys-
tem of any test apparatus or instrument. The exceptional
smoothness of these ta-C samples, approaching perfect
smoothness ??1 Å rms? when processed optimally, was an
important criterion in selecting this material.
The measurements presented here were made with the
imager for low energetic neutral atom ?ILENA? apparatus at
the University of Bern, Switzerland. The setup is described
briefly. More details on the experimental setup can be found
elsewhere.11,24ILENA consists of an ion source, a magnetic
beam filter, an ion guiding system, a sample stage with hous-
ing, and a detection unit. All these units are contained in a
single vacuum chamber pumped by an ion getter pump. A
schematic diagram of the setup is shown in Fig. 1.
The reflected beam is recorded using a two-dimensional
position-sensitive microchannel plate ?MCP? detector with a
viewing angle of ?12.5° in both azimuthal and polar direc-
tions. A retarding potential analyzer ?RPA? consisting of
three grids is mounted in front of the MCP detector. The
detector unit, including the RPA, is shielded electrostatically
and can be rotated independently of the converter surface
around the same axis as the converter surface. The outer
grids of the RPA are grounded to shield the inner grid, which
can be biased to suppress positive ions. An additional grid in
front of the MCP detector at negative potential with respect
to the MCP detector serves to reject secondary electrons
originating from the preceding grids and the converter sur-
face. The MCP detector may be floated to a high negative
voltage with respect to the converter surface to vary the
transmission threshold for negative particles. After baking
out the vacuum chamber at 80 °C for one day, a residual gas
pressure of 5?10−8mbar is achieved. During operation the
pressure may rise into the low 10−7mbar range as a result of
test gas leaking into the ion source chamber. This pressure
mirrors the conditions within a typical particle sensing satel-
lite instrument shortly after launch. From the ROSETTAmis-
sion of the European Space Agency, it is known that several
weeks after launch, the pressure in the vicinity of the space-
craft drops to the low 10−9mbar range, and into the
10−11mbar range after one year.25The pressure inside space
instruments with small openings to vent to the outside, such
as most particle instruments, is expected to be at least an
order of magnitude higher than the pressure outside the
spacecraft. Because of internal outgassing, pressures in the
10−8mbar range are expected to persist in space particle
instruments more or less indefinitely.
The fraction of negative ions is determined by taking
measurements with and without an applied floating voltage
on the MCP. In the first case, only neutral particles are re-
corded, in the latter, neutral particles and negative ions. From
the difference the fraction of negative ions is evaluated.
Each data point results from a series of successive mea-
surements that allow the detection of possible ion beam in-
stabilities and surface charging during each measurement se-
ries. With the MCP detector, we cannot distinguish between
negatively ionized primary particles and sputtered negative
ions. Therefore, measurements with incident positive noble
gas ions of comparable mass were performed, e.g., Ne+for
O+, and the negative ion fractions recorded there were taken
as an indication of the sputtering background of the previous
measurements.
Although we eventually want to use neutral atoms to
study surface ionization, for the hydrogen and oxygen mea-
surements we used positive molecular ions because they can
be produced far more efficiently and with much better en-
ergy, intensity, and angle control in our system than neutrals.
However, the charge and mass of molecules must be justi-
fied.
From previous experiments with several other insulating
surfaces
?polycrystalline
diamond,26,27and MgO ?Ref. 12??, it has been established
that incident hydrogen and oxygen ions are effectively neu-
tralized upon scattering. These previous measurements were
done with both incident positive ions and incident neutral
particles and they revealed the same negative ion fractions in
both cases. As a result, we can assume complete memory
loss of the incident charge state after scattering.
The use of molecules instead of atoms is justified as
follows. A molecule has a lot more electronic states than an
diamond,11
single-crystal
FIG. 1. Schematic diagram of the ILENA experiment. A positively charged
ion beam is produced in a Nier-type ion source with energies of ?100 up to
1200 eV. The beam then passes through an electromagnet that allows the
selection of the desired species. The sample surface is mounted on a revolv-
able sample holder which allows angles of incidence of 0°–90°. The MCP
can be turned freely around the same axis.
034503-2 Wahlström et al. J. Appl. Phys. 104, 034503 ?2008?
Downloaded 11 Aug 2008 to 130.92.9.58. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

Page 3

atom, so we cannot expect the charge exchange process
while scattering to be identical. In separate studies,10,11it has
been demonstrated that more than 80% of molecules with
energies 300–1000 eV dissociate shortly before reaching the
surface when scattered off a polycrystalline diamond surface.
As a consequence, the final charge state fraction is deter-
mined mainly by charge exchange processes between the
surface and dissociated atoms. Therefore, we conclude that
the use of molecules causes a negligible change to the charge
state fractions measured in this study. In addition, various
measurements with different surfaces have shown that the
results gained here with energies of 390–1000 eV are com-
parable to the results which are achieved with incident hy-
drogen and oxygen atoms with energies of 195–500 eV.
The detection efficiency of the MCP is taken from a
previous study28,29in which an identical detector was used.
III. RESULTS AND DISCUSSION
The two key performance requirements for the applica-
tion of conversion surfaces in neutral particle sensing instru-
ments are high ionization yield and low angular scattering,
the latter to minimize scattered particle loss in the detection
systems of the sensor. In general, neutral particle sensing
instruments use more than one conversion surface. As a con-
sequence, it is essential for accurate measurements that all
surfaces used in the instrument show equal performance in
the two key requirements.
In the IBEX-lo sensor, 28 circularly arranged charge
state conversion surfaces are used. More details on the
IBEX-lo sensor can be found elsewhere.30,31
A. Angular scattering
The angular scattering properties of the surfaces are veri-
fied by measuring the angular full width at half maximum
?FWHM? of the scattered particle beam. The component of
angular deviation from specular reflection that resides in a
plane containing the incoming trajectory and is normal to the
surface is defined as polar scattering. The scattering compo-
nent normal to the polar angle plane is defined as the azi-
muthal scattering with zero indicating a true specular reflec-
tion.
The measured angular scattering dependence on surface
roughness for different IBEX conversion surfaces is shown
in Fig. 2. The rms roughness of the surfaces has been mea-
sured with an atomic force microscope ?AFM?.32
The scattering angles in azimuthal direction are gener-
ally larger than the scattering angles in polar direction and
the FWHM increases significantly with increasing energy.
However, a dependency on the surface roughness cannot be
seen here. The variations in the angular FWHM measured for
surfaces with different ?rms? surface roughness are almost
within the measurement uncertainty of ?1° for ILENA, as
can be seen in Fig. 2, and there is no noticeable trend. With
the ILENA apparatus it is not possible to find a dependency
of the scattering properties on the very small variations in the
surface roughness measured with the AFM.
B. Angular dependency
For most of the measurements presented in this paper,
the angle of incidence of the ion beam is 8° with respect to
the surface plane. This was chosen because all previous mea-
surements during the selection process of the charge state
conversion surfaces were done with this setting, even before
the angle of incidence in the IBEX-lo sensor was known. To
obtain comparable results, the angle of incidence was kept at
the same value. In the IBEX-lo instrument the angle of inci-
dence is 15° with respect to the surface plane and the field of
view is ?7°. Therefore, the dependence of the angular scat-
tering and the fraction of negative ions on the angle of inci-
dence were tested.
The influence of the incidence angle on the scattered
beam can be seen in the pictures taken with the imaging
detector, as shown in Fig. 3. The three pictures show a
500 eV O2beam scattered off an IBEX-lo charge state con-
version surface with different angles of incidence. The beam
profile broadens with increasing angle of incidence. The base
height of the profile also increases with increasing angle of
incidence. At an angle of incidence of 2° relative to the sur-
face plane, the plateau of almost isotropic scattering can be
observed at about 5%. At 8° the flanks of the scattered peak
nearly reach the limit of the field of view of the detector;
thus the plateau can only be estimated to be between 10%
and 15%. The last profile shown in Fig. 3 shows an even
broader distribution where the plateau is completely masked
by the flanks of the scattered peak. It is therefore only pos-
sible to give an upper limit of the base height at 12° of about
30%.
In Figs. 4 and 5 the FWHM dependency on the angle of
incidence is shown. The FWHM increases considerably with
increasing angle of incidence relative to the surface plane
and also with increasing energy of the scattered particles.
FIG. 2. Surface roughness ?rms? measured with an AFM for the IBEX-lo
conversion surfaces vs angular FWHM in polar and azimuthal directions of
a scattered O2beam for energies of incident molecular ions of 390, 500, and
780 eV. The FWHM decreases with decreasing energy of the incident par-
ticle and is narrower in polar direction than in azimuthal direction. No
dependence on the surface roughness can be seen.
034503-3Wahlström et al. J. Appl. Phys. 104, 034503 ?2008?
Downloaded 11 Aug 2008 to 130.92.9.58. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

Page 4

C. Ionization yield
The negative ion yields measured for scattering of mo-
lecular hydrogen and oxygen ions off the IBEX-lo charge
state conversion surfaces are shown in Fig. 6. We found in-
creasing negative charge state fractions with increasing en-
ergy of the incident ions. Similar findings have been reported
for other insulating surfaces like MgO,12LiF,8and BaZrO3.33
For hydrogen we recorded negative ion yields of about
2%–4% and for oxygen about 11%–19%. Projectiles incident
on different IBEX charge state conversion surfaces, with a
given energy, give rise to equally high negative ion yields for
all surfaces within the experimental uncertainties.
D. Angular dependency
The dependency of the negative ion yield on the angle of
incidence has also been investigated. The fraction of nega-
tive ions versus the velocity component normal to the surface
is shown in Fig. 7. As a first approximation, the negative ion
yield depends only on the velocity normal to the surface and
thus no difference between the particles with different pri-
mary energies should be seen. The time during which elec-
FIG. 3. Scattering images and profiles of a 500 eV O2beam scattered under
various angles of incidence ?2°, 8°, and 12° with respect to the surface
plane?.
FIG. 4. FWHM in azimuthal direction of the distribution of particles scat-
tered from an IBEX charge state conversion surface for various angles of
incidence.
FIG. 5. FWHM in polar direction of the distribution of particles scattered
from an IBEX charge state conversion surface for various angles of
incidence.
FIG. 6. Ionization efficiencies of all measured IBEX charge state conversion
surfaces indicated by the different symbols. The error bars for all measure-
ments include statistical and experimental uncertainties.
034503-4Wahlström et al. J. Appl. Phys. 104, 034503 ?2008?
Downloaded 11 Aug 2008 to 130.92.9.58. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

Page 5

tronic processes can occur depends strongly on the velocity
normal to the surface. This is reflected in the measurements.
The ionization yield also depends on the velocity parallel to
the surface at high parallel velocities, which was explained
via a kinetic resonance ?Winter7?. This effect is less impor-
tant for our experimental setup. The data show a clear trend
of increasing fractions of negative ions up to a level of
roughly 23%, which is reached at a normal velocity of
20 km/s. When increasing the normal velocity further, the
yield remains stable within the measurement uncertainty. No
trend for further increasing or decreasing negative ion yield
has been found here. However the error bars allow different
interpretations of the measurements.
E. Summary
As expected the distribution of scattered particles broad-
ens with increasing angles of incidence relative to the surface
plane. This can be understood due to the increase in kinetic
energy perpendicular to the surface with increasing angle of
incidence. Particles with higher kinetic energies normal to
the surface penetrate into deeper levels of the surface poten-
tial where the corrugation of the potential is larger, resulting
in larger angular scattering distributions ?Figs. 3–5?.
The same effect can be seen for particles with different
kinetic energies for a constant angle of incidence. The distri-
bution broadens with increasing kinetic energies as this also
includes an increase in the kinetic energy normal to the sur-
face.
The fraction of negative ions produced by scattering
neutrals or positive ions from a charge state conversion sur-
face depends on the kinetic energy of the particle and the
angle of incidence. The negative ion yield increases with
increasing kinetic energy normal to the surface, thus with
increasing angle of incidence with respect to the surface
plane and with increasing overall kinetic energy of the par-
ticle, as seen in Figs. 6 and 7. It has been suggested8,12,33that
higher energies cause smaller distances of closest approach
between scattered ions and surface atoms, which results in
higher probabilities for charge exchange processes and thus
higher fractions of negative ions.
In Table I the arithmetic mean and the standard deviation
is shown for the angular scattering and the fraction of nega-
tive ions produced. The angle of incidence was 8° with re-
spect to the surface plane. The similarity of the results for all
surface samples and the low angular scattering demonstrates
that the key requirement of a similar and smooth ??1 Å rms
as measured in Ref. 32? surface for all conversion facets with
good angular scattering properties is satisfied.
As seen in Fig. 6 and Table I the negative ion yields are
equal within the uncertainties of the experiment and the val-
ues are within the limits posed by the IBEX team for the
IBEX-lo sensor. Thus these charge state conversion surfaces
satisfy the second key requirement, which also proves the
usability of this technology for further space applications.
IV. CONCLUSIONS
We have tested nine charge state conversion surfaces of
nonflight and flight-spare quality of the IBEX-lo sensor for
their negative ion yield and their angular scattering distribu-
tions. The influence of the angle of incidence on these two
parameters was investigated. All surfaces measured met the
requirements concerning surface roughness ??1 Å rms?,
FWHM, and negative ion yield defined in the instrument
specifications. The surfaces are indistinguishable as seen
from the ILENA point of view.
As expected, both FWHM and negative ion yield in-
crease with increasing projectile energy. The angular depen-
dency shows a clear trend of decreasing FWHM and thus
better transmission through the instrument with decreasing
angle of incidence with respect to the surface plane. On the
other hand, the negative ion yield decreases with decreasing
angle of incidence. Furthermore the charge state conversion
surfaces would have to be longer for smaller angles of inci-
dence, which increase the size and the weight of the detector.
Therefore, the incident angle for a flight instrument must be
a compromise between high transmission and high negative
ion yield and has to be within the limits for weight and size
of the detector. For the IBEX-lo sensor an angle of incidence
of 15° relative to the surface plane was chosen as the best
compromise.
The negative ion yield increases with increasing velocity
normal to the surface up to about 2.5?104m/s. At higher
FIG. 7. Fraction of negative ions produced by an O2beam scattered from
IB0064 is shown as dependent on the velocity normal to the surface.
TABLE I. Arithmetic mean and standard deviation for different energies for
an angle of incidence of 8°. The negative ion fraction is given for hydrogen
and oxygen molecules, whereas the FWHM is only shown for oxygen mol-
ecules in polar and azimuthal directions. All values are given with statistical
error.
Energy ?eV?
Neg. ion fraction ?%?
H2
FWHM O2?deg?
PolarO2
Azimuth
1000
780
500
390
3.9?0.4
3.1?0.2
2.8?0.2
2.3?0.2
18.6?0.7
17.1?0.7
14.1?0.8
12.3?0.9
11.1?0.7
9.7?1.1
7.7?1.2
6.7?1.3
18.8?1.3
16.9?1.5
13.1?1.9
11.2?2.2
034503-5Wahlström et al. J. Appl. Phys. 104, 034503 ?2008?
Downloaded 11 Aug 2008 to 130.92.9.58. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

Page 6

energies the negative ion yield remains relatively constant
over the range of the energies measured here.
ACKNOWLEDGMENTS
This work is supported by the Swiss National Science
Foundation.
1C. Auth, A. G. Borisov, and H. Winter, Phys. Rev. Lett. 75, 2292 ?1995?.
2P. Stracke, F. Wiegershaus, S. Krischok, H. Müller, and V. Kempter, Nucl.
Instrum. Methods Phys. Res. B 125, 63 ?1997?.
3S. A. Deutscher, A. G. Borisov, and V. Sidis, Phys. Rev. A 59, 4446
?1999?.
4P. Roncin, J. Villette, J. P. Atanas, and H. Khemliche, Phys. Rev. Lett. 83,
864 ?1999?.
5A. G. Borisov and V. A. Esaulov, J. Phys.: Condens. Matter 12, R177
?2000?.
6R. Souda, Int. J. Mod. Phys. B 14, 1139 ?2000?.
7H. Winter, Phys. Rep. 367, 387 ?2002?.
8J. A. Scheer, P. Wurz, and W. Heiland, Nucl. Instrum. Methods Phys. Res.
B 212, 291 ?2003?.
9J. A. Scheer, M. Wieser, P. Wurz, P. Bochsler, E. Hertzberg, S. A. Fuselier,
F. A. Koeck, R. J. Nemanich, and M. Schleberger, Nucl. Instrum. Methods
Phys. Res. B 230, 330 ?2005?.
10J. A. Scheer, M. Wieser, P. Wurz, P. Bochsler, E. Hertzberg, S. A. Fuselier,
F. A. Koeck, R. J. Nemanich, and M. Schleberger, Adv. Space Res. 38,
664 ?2006?.
11P. Wurz, R. Schletti, and M. R. Aellig, Surf. Sci. 373, 56 ?1997?.
12M. Wieser, P. Wurz, K. Brüning, and W. Heiland, Nucl. Instrum. Methods
Phys. Res. B 192, 370 ?2002?.
13P. Wurz, The Outer Heliosphere: Beyond the Planets ?Copernikus Gesell-
schaft e. V., Katlenburg-Lindau, 2000?, p. 251.
14T. E. Moore, D. J. Chornay, M. R. Collier, F. A. Herrero, J. Johnson, M.
A. Johnson, J. W. Keller, J. F. Laudadio, J. F. Lobell, K. W. Ogilvie, P.
Rozmarynowski, S. A. Fuselier, A. G. Ghielmetti, E. Hertzberg, D. C.
Hamilton, R. Lundgren, P. Wilson, P. Walpole, T. M. Stephen, B. L. Peko,
B. van Zyl, P. Wurz, J. M. Quinn, and G. R. Wilson, Space Sci. Rev. 91,
155 ?2000?.
15P. Wurz, M. R. Aellig, P. Bochsler, A. G. Ghielmetti, E. G. Shelley, S.
Fuselier, F. Herrero, M. F. Smith, and T. Stephen, Opt. Eng. 34, 2365
?1995?.
16J. A. Scheer, Surface Ionization at Insulating Surfaces. A Novel Concept to
Detect Neutral Particles in Space ?University of Bern, Switzerland, 2005?.
17J. Geiss and M. Witte, Space Sci. Rev. 78, 229 ?1996?.
18B. Brehm, M. A. Gusinov, and J. L. Hall, Phys. Rev. Lett. 19, 737 ?1967?.
19P. Wurz, L. Saul, J. Scheer, E. Möbius, H. Kucharek, and S. Fuselier, J.
Appl. Phys. 103, 054904 ?2008?.
20Y. Lifshitz, S. R. Kasi, and J. W. Rabalais, Phys. Rev. Lett. 62, 1290
?1989?.
21T. M. Alam, T. A. Friedmann, P. A. Schultz, and D. Sebastiani, Phys. Rev.
B 67, 245309 ?2003?.
22J. P. Sullivan and T. A. Friedmann, Proceedings of the 1st International
Spezialist Meeting on Amorphous Carbon ?World Scientific, Singapore,
1998?.
23J. P. Sullivan, T. A. Friedmann, R. G. Dunn, E. B. Stechel, P. A. Schultz,
and N. Missert, The Electron Transport Mechanism in Amorphous
Tetrahedrally-Coordinated Carbon Films, MRS Symposia Proceedings
No. 498 ?Materials Research Society, Pittsburgh, 1998?.
24S. Jans, P. Wurz, R. Schletti, T. Fröhlich, E. Hertzberg, and S. Fuselier, J.
Appl. Phys. 87, 2587 ?2000?.
25S. Graf, K. Altwegg, H. Balsiger, P. Bochsler, B. Fiethe, and E. Montag-
non, J. Spacecr. Rockets 45, 57 ?2008?.
26J. A. Scheer, K. Brüning, T. Fröhlich, P. Wurz, and W. Heiland, Nucl.
Instrum. Methods Phys. Res. B 157, 208 ?1999?.
27P. Wurz, T. Fröhlich, K. Brüning, J. A. Scheer, W. Heiland, E. Hertzberg,
and S. Fuselier, Prague Proceeding Week of Postdoctoral Students, 1998,
edited by J. Safrankov and A. Kanka ?unpublished?, p. 257.
28B. L. Peko and T. M. Stephen, Nucl. Instrum. Methods Phys. Res. B 171,
597 ?2000?.
29T. M. Stephen and B. L. Peko, Rev. Sci. Instrum. 71, 1355 ?2000?.
30M. Wieser, P. Wurz, E. Moebius, S. A. Fuselier, E. Hertzberg, and D. J.
McComas, Rev. Sci. Instrum. 78, 124502 ?2007?.
31N. Lorente, J. Merino, F. Flores, and M. Yu. Gusev, Nucl. Instrum. Meth-
ods Phys. Res. B 125, 277 ?1997?.
32Evans Analytical Group ?EAG?. Sunnyvale, California, USA.
33S. Jans, P. Wurz, R. Schletti, K. Brüning, K. Sekar, W. Heiland, J. Quinn,
and R. E. Leuchter, Nucl. Instrum. Methods Phys. Res. B 173, 503 ?2001?.
034503-6Wahlström et al.J. Appl. Phys. 104, 034503 ?2008?
Downloaded 11 Aug 2008 to 130.92.9.58. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp