Secondary ion mass spectrometry depth profiling of amorphous

Secondary ion mass spectrometry depth profiling of amorphous polymer
multilayers using O2+ and Cs+ ion bombardment with a magnetic
sector instrument
S. E. Harton
Department of Materials Science and Engineering, North Carolina State University, Raleigh, North
Carolina 27695
F. A. Stevie
Analytical Instrumentation Facility, North Carolina State University, Raleigh, North Carolina 27695
H. Adea兲
Department of Physics, North Carolina State University, Raleigh, North Carolina 27695
共Received 22 September 2005; accepted 17 January 2006; published 27 February 2006兲
Thin planar polymer films are model systems in a number of fields, including nano- and
biotechnology. In contrast to reciprocal space techniques such as reflectivity or diffraction,
secondary ion mass spectrometry 共SIMS兲 can provide depth profiles of tracer labeled polymers in
real space directly with sufficient depth resolution to characterize many important aspects in these
systems. Yet, continued improvements in characterization methods are highly desirable in order to
optimize the trade-offs between depth resolution, mass resolution, detection sensitivity, data
acquisition time, and artifacts. In this context, the utility of a magnetic sector SIMS instrument for
amorphous polymer film analysis was evaluated using model polymer bilayer systems of
polystyrene 共PS兲 with poly共methyl methacrylate兲 共PMMA兲, PS with poly共2-vinylpyridine兲, and
poly共cyclohexyl methacrylate兲 共PCHMA兲 with PMMA. Deuterium-labeled polystyrene embedded
in PS or PCHMA at concentrations ranging from 5% to 20% 共v / v兲 was used as tracer polymer.
Analysis conditions for a magnetic sector SIMS instrument 共CAMECA IMS-6f兲 were varied to
achieve a depth resolution of ⬃10 nm, high signal/noise ratios, and high sensitivity, while
minimizing matrix effects and sample charging. Use of Cs+ and O+2 primary ions with detection of
negative and positive secondary ions, respectively, has been explored. Primary beam impact energy
and primary ion species have been shown to affect matrix secondary ion yields. Sputtering rates
have been determined for PS and PMMA using both primary ion species and referenced to values
for intrinsic 共100兲 silicon 共Si兲 under identical analysis conditions. © 2006 American Vacuum
Society. 关DOI: 10.1116/1.2172948兴
I. INTRODUCTION
Thin polymer films and multilayers are model systems for
the investigation of phenomena related to physical confinement of polymer chains near surfaces and interfaces,1–3 such
as reaction mechanisms of functional polymers,4,5 block copolymer segregation,6,7 and block copolymer ordering.8
These systems are ideally characterized using experimental
techniques such as neutron or x-ray scattering reflectometry
共NR or XR兲,9 scanning probe microscopy 共SPM兲,10 scanning
transmission x-ray microscopy 共STXM兲,11–13 forward recoil
spectrometry 共FRES兲,14 and secondary ion mass spectrometry 共SIMS兲,15 with the details depending on whether depth
or lateral characteristics need to be probed. As the applications of these systems continually increase in a number of
fields that include nano- and biotechnology, improvements in
characterization methods and spatial resolution 共depth and
lateral兲 are essential, as the size scale of interest has become
exceedingly small.3
a兲
Author to whom correspondence should be addressed; electronic mail:
[email protected]
362
J. Vac. Sci. Technol. A 24„2…, Mar/Apr 2006
Direct depth profiling techniques such as FRES and SIMS
have undergone significant advances in the recent years.14,16
The relatively low depth resolution of ⬃80 nm 共Ref. 17兲 of
conventional FRES 共⬇3.0 MeV 4He++兲18 has been improved
to a depth resolution of ⬃30 nm with the development of the
low-energy
FRES
共LE-FRES兲
共⬇1.3 MeV 4He++兲19 and time-of-flight FRES 共TOF-FRES兲
共⬇2.0 MeV 4He++兲20 techniques. Since the convolution resulting from secondary scattering below the surface is an
inherent limitation of FRES, further improvements are
unlikely.14
Reciprocal space techniques such as NR and XR can have
excellent depth resolution on the order of ⬃1 nm. In XR, the
contrast depends on the relative electron density of the layers
to be probed. In neutron scattering and NR depth profiling,
deuterium 共 2H兲 has a much higher scattering length density
than protium 共 1H兲 and is therefore frequently used as contrast agent for carbonaceous matter.21 Although NR has superior depth resolution to FRES and SIMS, the inversion of
the reciprocal space information obtained with NR to the
composition profiles in real-space can be model dependent,
requiring some a priori knowledge of the real-space
0734-2101/2006/24„2…/362/7/$23.00
©2006 American Vacuum Society
362
363
Harton, Stevie, and Ade: SIMS depth profiling of amorphous polymer multilayers
FIG. 1. Gaussian convolution of a Heaviside 共step兲 function ⌽共z兲, where
⌽共z兲 = 0 from −⬁ to 0 and ⌽共z兲 = 1 from 0 to ⬁. Typical depth resolutions
full width at half maximum are represented for NR 共solid line兲, SIMS
共dashed line兲, LE- or TOF-FRES 共bold line兲, and FRES 共dotted line兲.
profile.15,22 This often necessitates a direct depth profiling
technique such as FRES or SIMS to be used in conjunction
with NR.18
Figure 1 shows a comparison of typical depth resolutions
encountered using NR 共1 nm兲, FRES 共80 nm兲, LE- or TOFFRES 共30 nm兲, and SIMS 共10 nm兲.18 Because of its sensitivity, excellent depth resolution 共⬃10 nm兲, and ability to provide real-space depth profiles directly, SIMS has become a
prominent method for the analysis of polymer thin films and
multilayers.4–6,23 There are three types of SIMS instruments
distinguished by the mass analyzer 共i.e., quadrupole, TOF,
and magnetic sector兲 that can be utilized.24,25 The quadrupole
instrument has been extensively used in all areas of surface
and interface science, including semiconductor26 and polymer film15 depth profiling. For depth profiling of deuteriumlabeled polymers, quadrupole SIMS instruments provide excellent depth resolution of ⬍10 nm 共Refs. 6, 27, and 28兲 due
to the low impact energies 共⬍2 keV兲 possible.25 This high
depth resolution comes at the cost of requiring low primary
ion currents to minimize charging of the insulating polymer
films, which can create long analysis times with somewhat
poor signal/noise 共S/N兲 ratios.15,18 Furthermore, the quadrupole mass analyzer 共mass resolution m / ⌬m typically about
300兲 cannot separate molecular hydrogen 1H2 from 2H 共required mass resolution of m / ⌬m ⬃ 1250兲,25 and as such,
negative secondary ions must be detected when depth profiling deuterium-labeled polymers 共negative 1H2 is essentially
nonexistent兲.15
TOF SIMS is a powerful technique for probing the surface composition of materials29 and it can be used for depth
profiling of various organic and inorganic films,30 including
polymer films and multilayers.31–34 The mass resolution and
mass range of these instruments are unmatched, making it
particularly valuable for depth profiling of relatively high
molecular weight 共⬃100 Da兲 secondary ions.33 Because of
its lateral resolution of ⬃1 ␮m it can also be used for twoand three-dimensional imaging while providing detailed
chemical information of the atomic and molecular species
present.16,35 Although its use for depth profiling of
JVST A - Vacuum, Surfaces, and Films
363
deuterium-labeled polymers has been limited,36 the application of TOF SIMS for depth profiling of polymer films is
currently an area of active investigation.31–34 TOF SIMS
technology is rapidly evolving with the availability of a
+
, Au+, Au+2 ,
larger number of primary ion species, such as C60
16,37
+
+
Bi , and Bi2 , constantly emerging.
Magnetic sector instruments have very high mass resolving capabilities 共maximum m / ⌬m ⬃ 2 ⫻ 104兲,25 which allows
for complete separation of many ions possessing equal nominal mass numbers,24 such as 2H and 1H2, 13C and 12C 1H, and
28
Si and 12C 16O. They can also provide improved signal/
noise ratios when compared to the quadrupole instruments
because of the high transmission of secondary ions into the
detector.25 However, charge buildup can be problematic for
the analysis of insulators 共e.g., polymers兲 with magnetic sector SIMS.24,38 These types of materials often require a conductive coating and/or other charge neutralization techniques, such as the addition of an electron flood gun, to help
reduce charge buildup.39 Recent magnetic sector SIMS instruments can operate at impact energies reduced to ⬍1 keV,
although initial investigations have shown significant charging during the analysis of polymer films at these low energies. Depth resolutions ⬇5–15 nm have been reported for
depth profiling of deuterium-labeled polymers in polymer
films using a magnetic sector spectrometer at the higher impact energies of ⬃5 keV.4,23,40
SIMS can be used to quantify an interfacial excess,4,6 surface excess,15 and diffusion gradient41 of a tracer polymer
within a polymer film. At strongly segregated polymer/
polymer 共heterogeneous兲 interfaces, the SIMS analysis can
become difficult due to problems associated with changing
matrix ion yields and varying sputtering rates for the two
immiscible polymers. These problems exist due to complex
molecular interactions during sputtering that are strongly dependent on the primary ion species and chemical
environment.42,43 By systematically investigating depth profiles through these polymer/polymer interfaces, one can better understand the parameters required to obtain accurate
SIMS analysis of polymer films and multilayers.
Several model polymer bilayer systems have been characterized in order to delineate the utility of a magnetic sector
SIMS instrument for the soft matter depth analysis. The
polymer systems included atactic polystyrene 共PS兲 and syndiotactic poly共methyl methacrylate兲 共PMMA兲, PS and atactic
poly共2-vinylpyridine兲 共P2VP兲, and atactic poly共cyclohexyl
methacrylate兲 共PCHMA兲 and PMMA. In all three cases, atactic deuterium-labeled polystyrene 共dPS兲 was embedded in PS
or PCHMA at concentrations ranging from 5% to 20% 共v / v兲
as the tracer polymer. Using a CAMECA IMS-6f magnetic
sector spectrometer, SIMS depth profiles have been obtained
using Cs+ and O+2 primary ion bombardment. Analysis conditions were varied to optimize S/N ratios and sensitivity,
while preserving a high depth resolution of ⬃10 nm. Matrix
12
C ion yields 共Y M 兲, which are the detected intensities
共counts/ s兲 of atomic 12C, were measured for all the three
systems, with particular attention paid to the heterogeneous
interface. Sputtering rates 共SR兲 for the PS and PMMA films
364
Harton, Stevie, and Ade: SIMS depth profiling of amorphous polymer multilayers
364
have been quantified for both Cs+ and O+2 bombardments and
referenced to SR for intrinsic 共100兲 silicon under identical
conditions. Deuterium depth profiles are shown for PCHMA/
PMMA bilayers along with a comparison of the depth resolving capabilities for O+2 and Cs+ bombardment.
II. EXPERIMENT
FIG. 2. Schematic of thin film assembly used for the SIMS analysis. From
the bottom up: Si substrate 共with SiOx or SiH at the surface兲, first polymer
layer 共⬇100–200 nm PMMA or P2VP兲, second polymer layer
共⬇100–200 nm PS or PCHMA兲 “doped” with dPS, 50 nm PS sacrificial
layer, and 20 nm Au coating.
A. Material and sample preparation
The polymers used in this investigation were purchased
from Polymer Source and Scientific Polymer Products. The
PS, PMMA, and dPS were low polydispersity polymers
共M w / M n ⬍ 1.1兲 with molecular weights ⬇70–100 kDa, while
PCHMA 共M w ⬇ 70 kDa兲 and P2VP 共M w ⬇ 100 kDa兲 had
higher polydispersities 共M w / M n ⬎ 1.5兲. Silicon 共100兲 wafers
共Wafer World兲 were cleaned using the previously outlined
procedures.41 They were cut into 2 ⫻ 2 cm2 squares and
soaked in a hydrogen peroxide/sulfuric acid solution at approximately 100 °C for 30 min and subsequently washed
with de-ionized 共DI兲 water. They were then soaked in
10% 共v / v兲 hydrofluoric acid for 1 min and again washed
with DI water. Some of the wafers were placed in an UVozone environment to produce a thin 共⬇2 nm兲 native oxide
layer 共SiOx兲, while others were used with the hydrogen passivation treatment 共SiH兲.
All polymer solutions were allowed to fully dissolve for
approximately 12 h and were subsequently filtered with
0.45 ␮m pore polytetrafluoroethylene 共PTFE兲 syringe filters
prior to spin casting. For bilayer film preparation, PMMA
共Tg ⬇ 125 ° C兲 or P2VP 共Tg ⬇ 100 ° C兲 was spun cast onto the
substrate using toluene or chlorobenzene, respectively, and
annealed at 150 °C for approximately 30 min to remove residual solvent, allow the polymer chains to attain equilibrium
conformations 共relax兲, and reduce the surface roughness. The
second layer, PS 共Tg ⬇ 100 ° C兲 or PCHMA 共Tg ⬇ 100 ° C兲,
was cast from a selective solvent directly onto the first layer.
It has been found that 1-chloropentane serves as a selective
solvent for spin casting of PS or PCHMA directly onto
PMMA or P2VP, thereby providing sharp, uniform interfaces
and smooth surfaces.4 This is extremely important for measuring high-resolution depth profiles. Before the SIMS
analysis was performed, a 50 nm PS sacrificial layer was
added to the sample surface to ensure that primary beam
concentration uniformity was reached before the start of the
second layer. This was done by casting PS onto SiOx, scoring
the film with a sharp tip, and floating it onto water. The
floating film was then picked up with the bilayer. A 20 nm
Au coating was sputtered onto the sacrificial layer to minimize charge buildup. The analyzed film assembly is shown
in Fig. 2. For the sputtering rate 共SR兲 measurements,
⬇100 kDa PS and PMMA single layer films were cast onto
SiH from toluene with a nominal thickness of 150 nm as
determined by ellipsometery. The PMMA films were annealed for approximately 30 min at 150 °C, which is a sufficient time for the chain relaxation to occur,44 and both anJ. Vac. Sci. Technol. A, Vol. 24, No. 2, Mar/Apr 2006
nealed 共150 °C兲 and as-cast PS films were analyzed for
comparison. No sacrificial layer or Au coating was used with
these samples.
B. Secondary ion mass spectrometry
Depth profiles were measured using a CAMECA IMS-6f
magnetic sector secondary ion mass spectrometer. Typical
analysis conditions for O+2 primary ion bombardment included a 25 nA primary current rastered over a 180
⫻ 180 ␮m2 area at a 41° angle of incidence from normal,
with 5.5 keV impact energy and a mass resolution m / ⌬m of
1250. Positive secondary ions were detected from a 60 ␮m
diam optically gated area positioned in the center of the raster. For Cs+, typical analysis conditions included a 15 nA
primary current rastered over a 200⫻ 200 ␮m2 area at a 27°
angle of incidence, with 6.0 keV impact energy and similar
mass resolution as with the O+2 bombardment. Negative secondary ions were detected from a 60 ␮m diam optically
gated area positioned in the center of the raster.
III. RESULTS AND DISCUSSION
Most of the physically relevant information obtainable
from a depth profile of a polymer film structure is gathered at
or near the surface or a heterogeneous interface. These are
the most important regions that require a detailed understanding of conditions for the optimal SIMS analysis. The
chemical structures of the polymers used in the presented
investigation are shown in Fig. 3. Both Cs+ and O+2 were
used to evaluate Y M for the three types of bilayers, as shown
in Fig. 4. For the profiles determined using Cs+, a nonmonotonic change in ion yield is observed through the heterogeneous interface for PS/PMMA and PCHMA/PMMA 关Figs.
4共a兲 and 4共b兲兴. The decrease in Y M from PS to PMMA away
from the interface 关see Fig. 4共a兲兴 can be attributed partly to
the decrease in number density of carbon atoms. The anomalous, nonmonotonic yield through the PS/PMMA interface,
however, has been found to be nearly independent of primary
ion current, as shown in Fig. 5, which implies that it cannot
be explained with sample charging. This same type of behavior was also observed through the PS/PMMA interface for
the ion yields of other detected species, as shown in Fig. 6.
This nonmonotonic change in ion yield was not observed
with the Cs+ analysis of the PS/P2VP films 关see Fig. 4共c兲兴,
nor was it observable with any of the O+2 profiles 关see Figs.
365
Harton, Stevie, and Ade: SIMS depth profiling of amorphous polymer multilayers
FIG. 3. Chemical structures of polystyrene 共PS兲, poly共2-vinylpyridine兲
共P2VP兲, poly共methyl methacrylate兲 共PMMA兲, and poly共cyclohexyl methacrylate兲 共PCHMA兲. Deuterium-labeled polystyrene 共dPS兲 is identical to PS,
except a percentage of hydrogen sites are occupied with deuterium. For the
dPS used here, ⬇100% of the hydrogen sites are occupied with deuterium.
4共d兲–4共f兲兴. The underlying mechanism behind this behavior
is unknown at this time and merits further investigation.
The constant Y M through the PS/PMMA and PCHMA/
PMMA interfaces using the O+2 bombardment 关see Figs. 4共d兲
and 4共e兲兴 with the detection of positive secondary ions is an
ideal situation for the analysis of these bilayer systems. Because the separation of positive secondary ions of H2 and 2H
FIG. 4. Matrix 12C ion yields 共Y M 兲, which are the detected intensities
共counts/ s兲 of atomic 12C, normalized to the nominal value for the second
layer 共PS or PCHMA兲 for Cs+ with 共a兲 PS/PMMA, 共b兲 PCHMA/PMMA,
and 共c兲 PS/P2VP and O+2 with 共d兲 PS/PMMA, 共e兲 PCHMA/PMMA, and 共f兲
PS/P2VP. The x axis is the sputtering time 共t兲 normalized to the maximum
time of analysis 共tmax兲. The location of the heterogeneous polymer-polymer
interface is indicated by the vertical dashed line, as determined by the deuterium profiles 共not shown兲. All bilayers had a 50 nm PS sacrificial layer and
a 20 nm Au coating at the surface 共see Fig. 2兲.
JVST A - Vacuum, Surfaces, and Films
365
FIG. 5. Matrix 12C ion yields 共Y M 兲 for PS/PMMA normalized to the nominal
value for the PS layer for 6.0 keV 共impact energy兲 Cs+ with 5 共bottom兲, 10
共middle兲, and 15 共top兲 nA primary ion currents. The 10 and 15 nA profiles
have been offset by an arbitrary amount for clarity. The x axis is the sputtering time 共t兲 normalized to the maximum time of analysis 共tmax兲. The
location of the PS/PMMA interface 共circled兲 is obvious from the nonmonotonic change in Y M . This change in Y M at the PS/PMMA interface is nearly
independent of the primary ion current.
is not possible with quadrupole SIMS,15 these conditions
could not be implemented with that type of instrument. The
change in Y M from PS to P2VP for the Cs+ bombardment
关see Fig. 4共c兲兴 approximately scales with the number density
of carbon atoms, as P2VP has a similar mass density to PS,
yet there are 7 / 8 as many carbon atoms in a polymer segment 共see Fig. 3兲. For the O+2 bombardment, however, there
is a considerable increase in Y M going from PS to P2VP 关see
Fig. 4共d兲兴. A similar increase in Y M was also observed with
the O+2 bombardment at an interface between PS and a random copolymer of styrene and acrylonitrile 共SAN兲, with
25% 共w / w兲 acrylonitrile segments 共C3H3N兲. Nitrogen in the
sample thus appears to have a positive secondary ion yield
enhancing effect for the O+2 bombardment. At the interface
between the substrate and the first polymer layer, there is an
obvious increase in Y M for all polymers with both Cs+ and
FIG. 6. Secondary ion yields for 12C 共bold line兲, 12C 1H 共dashed line兲, and
16
O 共solid line兲 for 5 nA 共6.0 keV impact energy兲 Cs+ primary ion bombardment of a PS/PMMA bilayer as a function of sputtering time 共t兲.
366
Harton, Stevie, and Ade: SIMS depth profiling of amorphous polymer multilayers
366
TABLE I. Sputtering rates 共SR兲 in nm/ s for 25 nA, 5.5 keV O+2 and 10 nA, 6.0
keV Cs+ SIMSs, respectively. The values are normalized to SR for intrinsic
Si 共100兲 共SR / SR,Si兲. No sacrificial layer or Au coating was used with these
measurements.
Layer
O+2 SR共nm/ s兲
O+2 SR / SR,Si
Cs+SR共nm/ s兲
Cs+SR / SR,Si
共100兲 Si
PS annealed
PS as cast
PMMA annealed
0.16
0.28
0.25
0.57
1.00
1.75
1.56
3.56
0.19
0.22
0.21
0.62
1.00
1.16
1.11
3.26
O+2 bombardments 共see Figs. 4–6兲. This is most likely due to
the secondary ion enhancement resulting from a change in
matrix species 共i.e., going from a 12C-based matrix to a
28
Si-based matrix兲.24,45,46 The silicon matrix is oxidized more
readily than the carbon-based matrix and secondary ion yield
for carbon is enhanced as we start to reach the silicon matrix.
A change in sputtering rates from one polymer to the next
could have significant consequences for the interpretation of
the depth profile near the heterogeneous interface. Hence, SR
was determined for PS and PMMA for 10 nA 共6.0 keV impact energy兲 Cs+ and 25 nA 共5.5 keV impact energy兲 O+2
primary ion bombardments. Table I lists both the sputtering
rates and the rates relative to intrinsic 共100兲 Si 共SR / SR,Si兲.
There is a marked increase in the sputtering rate from PS to
PMMA for both Cs+ and O+2 bombardments, with a twofold
to threefold increase in SR for PMMA relative to PS. As-cast
PS was also analyzed to compare the effects of the sample
history on the sputtering rates. Previous O+2 SIMS analyses4
have shown that the sputtering rate is much greater 共nearly
double兲 for PS as cast from 1-chloropentane when compared
to identical samples of annealed PS. Here, SR for the PS as
cast from toluene is nominally identical to annealed PS, in
stark contrast to the increase observed after casting from
1-chloropentane. It has also been found that both types of
as-cast samples are far more susceptible to charging with
both Cs+ and O+2 bombardments than annealed PS. This
means that the factors affecting the SIMS analysis, including
sputtering rate, secondary ion yield, and sample charging, are
strongly dependent on the sample preparation and history as
well as chemical composition of the polymer constituents.
These effects would be most significant for investigations
involving as-cast films or measurements after relatively short
annealing times 共less than the terminal relaxation time兲.44
The ability of magnetic sector SIMS to readily and
quickly provide quantitative information from deuterium
depth profiles is demonstrated with the PCHMA/PMMA system, where dPS is embedded in the PCHMA. From the phase
behavior of the three polymer pairs,47,48 it is known that PS
and PCHMA are completely miscible with each other, yet PS
and PCHMA are completely immiscible with PMMA under
the conditions implemented here. Taking advantage of these
asymmetries in the thermodynamic interactions between
dPS/PMMA and PCHMA/PMMA, dPS has been driven to
the interface by annealing bilayer films at 150 °C for 42 h.
Using SIMS, the interfacial excess of dPS at the PCHMA/
PMMA interface was determined for initial dPS concentraJ. Vac. Sci. Technol. A, Vol. 24, No. 2, Mar/Apr 2006
FIG. 7. SIMS profiles for bilayers of dPS in PCHMA on PMMA with 共a兲
5%, 共b兲 10%, and 共c兲 20% 共v / v兲 initial concentrations of dPS. After depletion of dPS due to segregation to the interface during the 42 h anneal at
150 °C, the bulk concentration away from the interface was reduced to 共a兲
4.3%, 共b兲 9.2%, and 共c兲 19.2%. The interfacial excess 共Z*兲 was determined
from Eq. 共1兲 after fitting the SIMS profiles to a Gaussian error function
共dotted line兲 and a Gaussian peak 共bold line兲, which superimpose to represent the convoluted dPS profile 共solid line兲. The location of the bilayer
surface 共depth= 0兲 was determined from the inflection point in the Gaussian
error function.
tions of 5%, 10%, and 20% 共v / v兲 in PCHMA, as shown in
Fig. 7. Data acquisition times for these profiles were
⬃20 min.
The measured dPS profiles were fit to a Gaussian error
function for the equilibrium dPS in the PCHMA layer and a
Gaussian peak for the segregated dPS at the interface.4 The
interfacial excess was determined from the expression
Z* ⬅
冉 冊
␲
4 ln 2
1/2
␸ p⌬,
共1兲
where ␸ p is the height and ⌬ is the full width at half maximum of the Gaussian peak. The full width at half maximum
of the error function is used as a measure of the instrumental
resolution, which is approximately 8–10 nm for this system
under these analysis conditions. Using Eq. 共1兲, Z* / R, where
R is the rms end-to-end distance49 of a dPS chain 共R
⬇ 18 nm兲,50 was determined to be 0.053, 0.055, and 0.068
for initial concentrations of dPS of 5%, 10%, and 20%,
respectively.
The low excess of dPS at the interface 共Z* / R Ⰶ 1兲 is a
result of the highly favorable interaction between dPS and
PCHMA 共exothermic energy of mixing兲, which creates a
strong driving force to keep dPS within the PCHMA layer.51
Effects due to changes in depth resolution are shown in Fig.
8. The use of 6.0 keV impact energy Cs+ primary ions provides improved depth resolution 共⬇8–10 nm兲 when compared with bombardment using 5.5 keV impact energy O+2
primary ions 共⬇13–15 nm兲 for the PCHMA/PMMA system.
The analysis of the depth profiles in Fig. 8 results in nearly
367
Harton, Stevie, and Ade: SIMS depth profiling of amorphous polymer multilayers
FIG. 8. Comparison of the Cs+ 共䊊兲 and O+2 共쎲兲 primary ion bombardments
with 6.0 and 5.5 keV impact energies, respectively. For this system, Cs+
clearly provides improved depth resolution.
identical values for Z*, indicating that the interfacial excess
is faithfully convoluted with the instrument resolution but
not distorted. Primary ion bombardment with Cs+ was also
observed to provide improved depth resolution for the PS/
PMMA 共Ref. 23兲 and PS/P2VP systems when compared to
the O+2 bombardment under the conditions described here.
This is due to Cs+ having a lower penetration depth, thereby
reducing ion-induced mixing and chemical degradation below the surface.52 A decrease in depth resolution to ⬇30 nm,
which would typically be encountered with the analysis using LE-FRES or TOF-FRES,14 would smear out the interfacial excess of dPS in the dPS doped PCHMA/PMMA system
to make it less noticeable. Quantitative analysis would be
significantly more difficult due its dependence on the accuracy of the base line concentration profile and the high S/N
ratio required. The PCHMA/PMMA system is a great example of the excellent capabilities of magnetic sector SIMS
to provide high depth resolution and high signal-to-noise ratio simultaneously, thereby allowing accurate physical information to be determined from the depth profiles.
IV. CONCLUSIONS
The utility of using magnetic sector SIMS for analyzing
three different polymer/polymer bilayer systems 共PS/PMMA,
PCHMA/PMMA, and PS/P2VP兲 has been demonstrated. Excellent throughput, mass resolution, and S/N ratio could be
achieved at high depth resolution of ⬃10 nm with a CAMECA IMS-6f. The sputtering rates for PS and PMMA using
25 nA, 5.5 keV O+2 and 10 nA, 6.0 keV Cs+ beams have been
measured. The sputtering rates increased significantly 共two to
three times兲 from PS to PMMA with the Cs+ and O+2 bombardments, and the Cs+ bombardment resulted in a nonmonotonic change in matrix ion yield through the interface, which
was shown to be nearly independent of primary ion current
共see Fig. 5兲. This nonmonotonic change was also observed at
the PCHMA/PMMA interface, but was not witnessed with
the PS/P2VP system or in any of the systems when characterized with the O+2 bombardment 共see Fig. 4兲. The use of the
O+2 bombardment with the analysis of positive secondary
JVST A - Vacuum, Surfaces, and Films
367
ions showed constant matrix 12C secondary ion yields
through the PS/PMMA and PCHMA/PMMA interfaces.
These ideal conditions could not be replicated with quadrupole SIMS because these types of instruments are not capable of mass resolving H2 from 2H. It has been established
that there are profound effects due to sample preparation and
history, chemical composition of the samples, and type of
primary ion bombardment, which result in changes in the
sputtering rates and secondary ion yields. The excellent
depth resolution attainable using SIMS was exemplified with
deuterium depth profiling of dPS in the PCHMA/PMMA system, where a low interfacial excess of dPS was observed at
the interface 共see Fig. 7兲. The results for the PCHMA/
PMMA system show that quantitative analysis of this system
would be extremely difficult with other depth profiling methods such as FRES, where poorer depth resolution and S/N
ratio are encountered.18
ACKNOWLEDGMENT
This work was supported by the U. S. Department of Energy 共DE-FG02-98ER45737兲.
R. A. L. Jones, Polymers at Surfaces and Interfaces 共Cambridge University Press, New York, 1999兲.
G. J. Fleer, M. A. C. Stuart, J. M. H. M. Scheutjens, T. Cosgrove, and B.
Vincent, Polymers at Interfaces 共Chapman and Hall, New York, 1993兲.
3
S. Granick et al., J. Polym. Sci., Part B: Polym. Phys. 41, 2755 共2003兲.
4
S. E. Harton, F. A. Stevie, and H. Ade, Macromolecules 38, 3543 共2005兲.
5
B. J. Kim, H. Kang, K. Char, K. Katsov, G. H. Fredrickson, and E. J.
Kramer, Macromolecules 38, 6106 共2005兲.
6
B. J. Reynolds, M. L. Ruegg, T. E. Mates, C. J. Radke, and N. P. Balsara,
Macromolecules 38, 3872 共2005兲.
7
S. Zhu, Y. Liu, M. H. Rafailovich, J. Sokolov, D. Gersappe, D. A.
Winesett, and H. Ade, Nature 共London兲 400, 49 共1999兲.
8
P. Mansky, Y. Liu, E. Huang, T. P. Russell, and C. Hawker, Science 275,
1458 共1997兲.
9
W. Zhao et al., Physica B 173, 43 共1991兲.
10
S. Ge, Y. Pu, W. Zhang, M. Rafailovich, J. Sokolov, C. Buenviaje, R.
Buckmaster, and R. M. Overney, Phys. Rev. Lett. 85, 2340 共2000兲.
11
H. Ade, X. Zhang, S. Cameron, C. Costello, J. Kirz, and S. Williams,
Science 258, 972 共1992兲.
12
H. Ade and B. Hsiao, Science 262, 1427 共1993兲.
13
A. L. D. Kilcoyne et al., J. Synchrotron Radiat. 10, 125 共2003兲.
14
R. J. Composto, R. M. Walters, and J. Genzer, Mater. Sci. Eng., R. 38,
107 共2002兲.
15
S. A. Schwarz et al., Mol. Phys. 76, 937 共1992兲.
16
N. Winograd, Anal. Chem. 77, 142A 共2005兲.
17
Depth resolution is defined as the full width at half maximum of a Gaussian convolution function.
18
E. J. Kramer, Physica B 173, 189 共1991兲.
19
J. Genzer, J. B. Rothman, and R. J. Composto, Nucl. Instrum. Methods
Phys. Res. B 86, 345 共1994兲.
20
J. Sokolov, M. H. Rafailovich, R. A. L. Jones, and E. J. Kramer, Appl.
Phys. Lett. 54, 590 共1989兲.
21
R.-J. Roe, Methods of X-Ray and Neutron Scattering in Polymer Science
共Oxford University Press, New York, 2000兲.
22
A. Hariharan, S. K. Kumar, M. H. Rafailovich, J. Sokolov, X. Zheng, D.
H. Duong, S. A. Schwarz, and T. P. Russell, J. Chem. Phys. 99, 656
共1993兲.
23
S. E. Harton, F. A. Stevie, R. J. Spontak, T. Koga, M. H. Rafailovich, J.
C. Sokolov, and H. Ade, Polymer 46, 10173 共2005兲.
24
R. G. Wilson, F. A. Stevie, and C. W. Magee, Secondary Ion Mass Spectrometry: A Practical Handbook for Depth Profiling and Bulk Impurity
Analysis 共Wiley, New York, 1989兲.
25
V. K. F. Chia, G. R. Mount, M. J. Edgell, and C. W. Magee, J. Vac. Sci.
Technol. B 17, 2345 共1999兲.
1
2
368
Harton, Stevie, and Ade: SIMS depth profiling of amorphous polymer multilayers
26
C. W. Magee, D. Jacobson, and H.-J. Gossmann, J. Vac. Sci. Technol. B
18, 489 共2000兲.
27
X. Zhao, W. Zhao, J. Sokolov, M. H. Rafailovich, S. A. Schwarz, B. J.
Wildens, R. A. L. Jones, and E. J. Kramer, Macromolecules 24, 5991
共1991兲.
28
Y. M. Strzhemechny, S. A. Schwarz, J. Schacter, M. H. Rafailovich,
and J. Sokolov, J. Vac. Sci. Technol. A 15, 894 共1997兲.
29
D. G. Castner, Nature 共London兲 422, 129 共2003兲.
30
A. G. Sostarecz, S. Sun, C. Szakal, A. Wucher, and N. Winograd, Appl.
Surf. Sci. 231–232, 179 共2004兲.
31
E. R. Fuoco, G. Gillen, M. B. J. Wijesundara, W. E. Wallace, and L.
Hanley, J. Phys. Chem. B 105, 3950 共2001兲.
32
M. S. Wagner, Anal. Chem. 77, 911 共2005兲.
33
C. Szakal, S. Sun, A. Wucher, and N. Winograd, Appl. Surf. Sci.
231–231, 183 共2004兲.
34
C. M. Mahoney, J. Yu, and J. A. Gardella, Anal. Chem. 77, 3570 共2005兲.
35
J. Xu, S. Ostrowski, C. Szakal, A. G. Ewing, and N. Winograd, Appl.
Surf. Sci. 231–232, 159 共2004兲.
36
X. Hu et al., Macromolecules 36, 823 共2003兲.
37
D. Weibel, S. Wong, N. Lockyer, P. Blenkinsopp, R. Hill, and J. C.
Vickerman, Anal. Chem. 75, 1754 共2003兲.
38
J. M. McKinley, F. A. Stevie, C. N. Granger, and D. Renard, J. Vac. Sci.
Technol. A 18, 273 共2000兲.
39
A. L. Pivovarov, F. A. Stevie, and D. P. Griffis, Appl. Surf. Sci. 231–232,
J. Vac. Sci. Technol. A, Vol. 24, No. 2, Mar/Apr 2006
368
786 共2004兲.
G. Coulon, T. P. Russell, V. R. Deline, and P. F. Green, Macromolecules
22, 2581 共1989兲.
41
K. Shin, X. Hu, X. Zheng, M. H. Rafailovich, J. Sokolov, V. Zaitsev,
and S. A. Schwarz, Macromolecules 34, 4993 共2001兲.
42
Z. Postawa, B. Czerwinski, N. Winograd, and B. J. Garrison, J. Phys.
Chem. B 109, 11973 共2005兲.
43
R. G. Wilson, G. E. Lux, and C. L. Kirschbaum, J. Appl. Phys. 73, 2524
共1993兲.
44
K. Fuchs, C. Friedrich, and J. Weese, Macromolecules 29, 5893 共1996兲.
45
V. R. Deline, C. A. Evans, and P. Williams, Appl. Phys. Lett. 33, 578
共1978兲.
46
V. R. Deline, W. Katz, C. A. Evans, and P. Williams, Appl. Phys. Lett.
33, 832 共1978兲.
47
J. A. Pomposo, A. Mugica, J. Areizaga, and M. Cortazar, Acta Polym.
49, 301 共1998兲.
48
T. P. Russell, Macromolecules 26, 5819 共1993兲.
49
P. J. Flory, Principles of Polymer Chemistry 共Cornell University Press,
Ithaca, NY, 1953兲.
50
R = aN1/2, where a is the statistical segment length 共0.67 nm for PS
or dPS兲, and N is the number of segments in the polymer chain.
51
E. Helfand, Macromolecules 25, 1676 共1992兲.
52
P. Williams, R. K. Lewis, C. A. Evans, and P. R. Hanley, Anal. Chem. 49,
1399 共1977兲.
40