- Email: [email protected]
Characterization of surface structure by cluster coincidental ion mass spectrometry
Applied Surface Science 231–232 (2004) 106–112
Characterization of surface structure by cluster coincidental ion mass spectrometry R.D. Rickman, S.V. Verkhoturov, S. Balderas, N. Bestaoui, A. Clearfield, E.A. Schweikert* Department of Chemistry, Texas A&M University, College Station, TX 77843-3144, USA Available online 30 April 2004
Abstract Coincidental ion mass spectrometry, CIMS, is a technique in which spatial and/or chemical relationships are investigated by collection and analysis of coincidental ion emission events. Practical application of CIMS is dependant upon two factors. One is that the primary ion be selected so as to maximize coincidental ion, CI, yields. Polyatomic projectiles are best suited for this type of analysis because, in certain cases, they not only enhance conventional secondary ion yields but CI yields as well compared to that of equal velocity atomic projectiles. The second factor is that data collection and analysis be as efficient as possible. Experiments were run in an event-by-event bombardment/detection mode. All secondary ions from each impact were recorded with subsequent offline analysis of coincidental ion emission. We report here the combination of cluster SIMS (22 keV Au3 þ projectiles) with a fresh approach to data analysis for evaluation of a-zirconium phosphate in both gel and crystalline states. Results presented here show that it is possible to distinguish between these two phases although the stoichiometry of the compound is the same. # 2004 Elsevier B.V. All rights reserved. Keywords: Cluster SIMS; Nanostructural characterization; Zirconium phosphate; Coincidental ion mass spectrometry; Phase change
1. Introduction Secondary ion mass spectrometry, SIMS, is well documented as a sensitive and versatile surface analytical technique [1]. An issue which has received little attention is the question to what extent the SIMS signal can reveal differences in surface structure. It has been shown earlier that a compound probed successively in a gel and crystalline form can show significant differences in the corresponding SIMS spectra albeit there is * Corresponding author. Tel.: þ1-979-845-2344; fax: þ1-979-845-1655. E-mail address: [email protected] (E.A. Schweikert).
no change in the compounds stoichiometry [2]. We report here on the further development of a SIMS methodology pertinent for probing surface structure. If one divides a SIMS measurement into a sequence of single temporally and spatially resolved bombardment and detection events, one probes with each impact a nanometric volume [3]. Indeed the impact of a single projectile addresses areas as small as 10 nm in diameter [4] implying that any secondary ions desorbed from the impact of a single projectile originate from areas on this order. It has been shown experimentally that negatively charged polyatomic secondary ions are ejected by the direct emission processes and are fragments representative of the surface structure [2]. Thus the combination of sensitivity and ability to detect
0169-4332/$ – see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2004.03.079
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
prehensive data acquisition scheme, dubbed Total Matrix of Events#, TME, in cluster SIMS experiments [5]. Polyatomic projectiles are known to generate enhanced secondary ion emission, including coincidental ion emission, thus making this bombardment mode particularly appropriate for data analysis in the TME# mode. The surface structure studies were carried out on specimens of a-zirconium bis-(monohydrogen orthophosphate) monohydrate, a-ZrP, samples in the gel and crystalline states. a-ZrP, a clay-like material is used as an ion exchanger, consists of Zr, P, O, and H atoms arranged in a structure that can range from an amorphous gel to a well ordered crystalline state making this compound well suited for this type of analysis.
co-located surface molecule via single impact event SIMS analysis provides an approach for nano-structure characterization [3]. Earlier work which demonstrated that surface structure can affect the appearance of a mass spectrum relied on coincidence counting mass spectrometry, CCMS [2]. In this methodology, one registers secondary ions emitted in coincidence with a selected secondary ion. However recording secondary ions only when a pre-selected secondary ion is registered entails a measurement time dictated by the frequency of detecting the specific secondary ion. A more efficient approach is to record all events, i.e. all secondary ions detected from each impact. One can then in a subsequent offline data analysis seek coincidental ion emission. In this study we apply a com-
185
O
O
P
20 1 O
O
O P
O
281 O
O
P
O
O
O
Zr
Zr
Zr
O
O O
343
P
O
O O
Zr
O
OH
O
465
O
O P
O
P
O O
107
O
P
O Zr
O
O O
P
O
O
589
O
O
P
O
Zr
O
Zr
O O
P
O
O O
HO O
P
O
O O
O
Zr
O
P O
O
Zr
O
P
O
O
Fig. 1. Structure-specific fragments that can be traced back to the crystal structure of alpha-zirconium phosphate.
108
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
2. Experimental Gold polyatomic projectiles were generated using a liquid metal ion source described elsewhere [6]. The impact energy studied was 22 keV with the primary ion kinetic energy at þ14 keV and the target potential set at 8 kV. Secondary electrons from the impact of the primary ion are steered towards a microchannel
plate, MCP, detector and serve as the start signal for the time-of-flight mass analysis of any secondary ions ejected by that projectile. The secondary ions strike the MCP stop detector where their arrival times are recorded and used to identify that particular secondary ion. The signal from each detector is processed through a constant fraction discriminator providing the logic pulse required by the high resolution
Fig. 2. (A) Mass spectrum of a ZrP gel target from the impact of 22 keV Au3 þ projectiles. Inset shows an expansion of the 100–600 amu region. (B) Powder diffraction spectrum of the same sample.
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
(400 ps/channel) time-to-digital multistop (>2000 stops/start) converter. Detected secondary ions from 3 106 events were collected and stored for offline analysis by our custom designed TME# software. This program allows for multidimensional mass spectral analysis based on user defined rules. For instance, one can organize data according to the number of secondary ions ejected per event, events in which one
109
or more secondary ions were detected in the same event, or any combination of these two conditions. For each case the experiments were run in the eventby-event bombardment/detection mode, i.e. secondary ion pulses from single primary ion impact events were recorded. The base pressure in the vacuum chamber is 105 Pa. Targets were prepared by mixing 50 (mg) of the ZrP into 1 ml ethanol and
Fig. 3. (A) Mass spectrum of a ZrP crystalline target from the impact of 22 keV Au3 þ projectiles. Inset shows an expansion of the 100– 600 amu region. (B) Powder diffraction spectrum of the same sample.
110
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112 -
PO3 (m/e = 79) 0.24
ZrP Gel ZrP Crystal
0.22
YSI(n) / YTotal(n)
0.20 0.18 0.16 0.14 0.12 0.10 0.08 1
2
(A)
3
4
n -
0.0050
(ZrO)(PO3)2OH (m/e = 281)
0.0045 0.0040
YSI(n) / YTotal(n)
0.0035 0.0030
ZrP Gel ZrP Crystal
0.0025 0.0020 0.0015 0.0010 0.0005 0.0000 1
2
3
4
n
(B)
Fig. 4. (A) The yield of m=z ¼ 79 (PO3 ) as a function of the total number of secondary ions ejected per single impact event. (B) The yield of m=z ¼ 281 as a function of the total number of secondary ions ejected per single impact event. Lines are guides for the eye.
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
depositing 50 ml of the slurry onto a stainless steel target.
3. Results and discussion Fig. 1 lists several fragment ions that can be traced back to the crystal structure of a-ZrP [2]. Negatively charged cluster secondary ions are the result of direct emission processes which suggests that they exist intact on the surface or result from the fragmentation of larger assemblies [2]. Thus, their intensities in the mass spectrum of the crystalline material will be higher than in the amorphous material. Fig. 2a shows a mass spectrum of a ZrP gel sample with the high mass region expanded in the inset. The two predominant peaks resulting from the ZrP sample are 63 (PO2 ) and 79 amu (PO3 ). Most of the higher mass structure-specific fragments are in low abundance. This has been attributed to lack of long range ordering, on the scale of the emission area [7]. Fig. 2b shows a powder X-ray diffraction (XRD) spectrum of the same sample. The peaks are broad suggesting some evidence of structure in the gel. This may explain why some structure-specific fragments are present in the mass spectrum of the gel sample (Fig. 2a). Fig. 3a shows a mass spectrum of a crystalline ZrP sample. Again, the two predominant peaks are 63 and 79 amu. However, the structure-specific fragments are clearly present (inset). The presence of these peaks can be attributed to the fact that the surface is ordered to a greater degree compared to the previous sample. A powder XRD spectrum, Fig. 3b, further supports this observation. The TME# program allows for comparison of the gel and crystalline samples in a unique manner. One method to compare samples is to plot the ratio of yields, Y, for a particular secondary ion as a function of the number of secondary ions detected per event. The ratio of yields, Y, is defined as: Y¼
YSI ðnÞ YTotal ðnÞ
where YSI ðnÞ is the yield of a particular secondary ion from the sum of all n-ion detection events and YTotal ðnÞ the yield of all secondary ions from the sum of all n-ion detection events. A plot of this ratio for m=z ¼ 79 as a function of the number of second-
111
ary ions detected per single impact event is given in Fig. 4a. The yield of PO3 for the gel sample is relatively constant however, the yield of PO3 decreases as the number of secondary ions per event increases in the crystalline sample. In the case of the crystalline sample the ratio of yields for m=z ¼ 281 (Fig. 4b) increases as the number of secondary ions ejected per event increases. This same trend is also present when looking at the fragment corresponding to m=z ¼ 343. One explanation that can account for these trends is that secondary ions from multiple ion ejection events are less excited vibrationally. The vibrational energy deposited by the projectile may be more efficiently dissipated through a larger crystal lattice network resulting in a fragment ion that is more stable. Indeed as the number of secondary ions ejected per event increases there is an increase in the yield of higher mass fragments, Fig. 4b, along with a concomitant decrease in the yield of PO3 , Fig. 4a, for the crystalline material. In the amorphous sample the vibrational energy is imparted to a disordered system with less energy dissipation resulting in more fragmentation. This is apparent when considering the yields of m=z ¼ 281, Fig. 4b, and the relatively constant yield of PO3 , Fig. 4a, for the amorphous material.
4. Conclusions We have demonstrated, from a qualitative standpoint, that cluster SIMS can reveal changes in surface structure in the absence of any change in the analyte stoichiometry. Furthermore these results demonstrate that monitoring selected emission events, e.g. events when three or four secondary ion are ejected, can be very sensitive to changes in the surface structure. Further work is needed to investigate possible correlations between the low and high mass fragments in events where multiple secondary ions are detected.
Acknowledgements This study was supported by grants from the National Science Foundation (CHE-0135888) and the Robert A. Welch Foundation (A-1482).
112
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
References [1] A. Benninghoven, P. Bertrand, H.N. Migeon, H.W. Werner (Eds.), Secondary Ion Mass Spectrometry—SIMS XII, Elsevier, Amsterdam, 2000. [2] M. VanStipdonk, M.A. Park, E.A. Schweikert, P. Sylvester, A. Clearfield, Int. J. Mass Spectrom. Ion Process. 128 (1993) 133– 141. [3] S.V. Verkhoturov, E.A. Schweikert, N.M. Rizkalla, Langmuir 18 (2002) 8836.
[4] K. Wien, Lect. Notes Phys. 269 (1986) 1. [5] R.D. Rickman, S.V. Verkhoturov, E.A. Schweikert, Phys. Rev. Lett. 92 (2004) 047601. [6] M. Benguerba, A. Brunelle, S. Della-Negra, J. Depauw, H. Joret, Y. Le Beyec, M.G. Blain, E.A. Schweikert, G. Ben Assayag, P. Sudreau, Nucl. Instrum. Meth. B 62 (1991) 8. [7] M.J. VanStipdonk, J.B. Shapiro, E.A. Schweikert, Vacuum 46 (8–10) (1995) 1227.
Characterization of surface structure by cluster coincidental ion mass spectrometry R.D. Rickman, S.V. Verkhoturov, S. Balderas, N. Bestaoui, A. Clearfield, E.A. Schweikert* Department of Chemistry, Texas A&M University, College Station, TX 77843-3144, USA Available online 30 April 2004
Abstract Coincidental ion mass spectrometry, CIMS, is a technique in which spatial and/or chemical relationships are investigated by collection and analysis of coincidental ion emission events. Practical application of CIMS is dependant upon two factors. One is that the primary ion be selected so as to maximize coincidental ion, CI, yields. Polyatomic projectiles are best suited for this type of analysis because, in certain cases, they not only enhance conventional secondary ion yields but CI yields as well compared to that of equal velocity atomic projectiles. The second factor is that data collection and analysis be as efficient as possible. Experiments were run in an event-by-event bombardment/detection mode. All secondary ions from each impact were recorded with subsequent offline analysis of coincidental ion emission. We report here the combination of cluster SIMS (22 keV Au3 þ projectiles) with a fresh approach to data analysis for evaluation of a-zirconium phosphate in both gel and crystalline states. Results presented here show that it is possible to distinguish between these two phases although the stoichiometry of the compound is the same. # 2004 Elsevier B.V. All rights reserved. Keywords: Cluster SIMS; Nanostructural characterization; Zirconium phosphate; Coincidental ion mass spectrometry; Phase change
1. Introduction Secondary ion mass spectrometry, SIMS, is well documented as a sensitive and versatile surface analytical technique [1]. An issue which has received little attention is the question to what extent the SIMS signal can reveal differences in surface structure. It has been shown earlier that a compound probed successively in a gel and crystalline form can show significant differences in the corresponding SIMS spectra albeit there is * Corresponding author. Tel.: þ1-979-845-2344; fax: þ1-979-845-1655. E-mail address: [email protected] (E.A. Schweikert).
no change in the compounds stoichiometry [2]. We report here on the further development of a SIMS methodology pertinent for probing surface structure. If one divides a SIMS measurement into a sequence of single temporally and spatially resolved bombardment and detection events, one probes with each impact a nanometric volume [3]. Indeed the impact of a single projectile addresses areas as small as 10 nm in diameter [4] implying that any secondary ions desorbed from the impact of a single projectile originate from areas on this order. It has been shown experimentally that negatively charged polyatomic secondary ions are ejected by the direct emission processes and are fragments representative of the surface structure [2]. Thus the combination of sensitivity and ability to detect
0169-4332/$ – see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2004.03.079
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
prehensive data acquisition scheme, dubbed Total Matrix of Events#, TME, in cluster SIMS experiments [5]. Polyatomic projectiles are known to generate enhanced secondary ion emission, including coincidental ion emission, thus making this bombardment mode particularly appropriate for data analysis in the TME# mode. The surface structure studies were carried out on specimens of a-zirconium bis-(monohydrogen orthophosphate) monohydrate, a-ZrP, samples in the gel and crystalline states. a-ZrP, a clay-like material is used as an ion exchanger, consists of Zr, P, O, and H atoms arranged in a structure that can range from an amorphous gel to a well ordered crystalline state making this compound well suited for this type of analysis.
co-located surface molecule via single impact event SIMS analysis provides an approach for nano-structure characterization [3]. Earlier work which demonstrated that surface structure can affect the appearance of a mass spectrum relied on coincidence counting mass spectrometry, CCMS [2]. In this methodology, one registers secondary ions emitted in coincidence with a selected secondary ion. However recording secondary ions only when a pre-selected secondary ion is registered entails a measurement time dictated by the frequency of detecting the specific secondary ion. A more efficient approach is to record all events, i.e. all secondary ions detected from each impact. One can then in a subsequent offline data analysis seek coincidental ion emission. In this study we apply a com-
185
O
O
P
20 1 O
O
O P
O
281 O
O
P
O
O
O
Zr
Zr
Zr
O
O O
343
P
O
O O
Zr
O
OH
O
465
O
O P
O
P
O O
107
O
P
O Zr
O
O O
P
O
O
589
O
O
P
O
Zr
O
Zr
O O
P
O
O O
HO O
P
O
O O
O
Zr
O
P O
O
Zr
O
P
O
O
Fig. 1. Structure-specific fragments that can be traced back to the crystal structure of alpha-zirconium phosphate.
108
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
2. Experimental Gold polyatomic projectiles were generated using a liquid metal ion source described elsewhere [6]. The impact energy studied was 22 keV with the primary ion kinetic energy at þ14 keV and the target potential set at 8 kV. Secondary electrons from the impact of the primary ion are steered towards a microchannel
plate, MCP, detector and serve as the start signal for the time-of-flight mass analysis of any secondary ions ejected by that projectile. The secondary ions strike the MCP stop detector where their arrival times are recorded and used to identify that particular secondary ion. The signal from each detector is processed through a constant fraction discriminator providing the logic pulse required by the high resolution
Fig. 2. (A) Mass spectrum of a ZrP gel target from the impact of 22 keV Au3 þ projectiles. Inset shows an expansion of the 100–600 amu region. (B) Powder diffraction spectrum of the same sample.
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
(400 ps/channel) time-to-digital multistop (>2000 stops/start) converter. Detected secondary ions from 3 106 events were collected and stored for offline analysis by our custom designed TME# software. This program allows for multidimensional mass spectral analysis based on user defined rules. For instance, one can organize data according to the number of secondary ions ejected per event, events in which one
109
or more secondary ions were detected in the same event, or any combination of these two conditions. For each case the experiments were run in the eventby-event bombardment/detection mode, i.e. secondary ion pulses from single primary ion impact events were recorded. The base pressure in the vacuum chamber is 105 Pa. Targets were prepared by mixing 50 (mg) of the ZrP into 1 ml ethanol and
Fig. 3. (A) Mass spectrum of a ZrP crystalline target from the impact of 22 keV Au3 þ projectiles. Inset shows an expansion of the 100– 600 amu region. (B) Powder diffraction spectrum of the same sample.
110
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112 -
PO3 (m/e = 79) 0.24
ZrP Gel ZrP Crystal
0.22
YSI(n) / YTotal(n)
0.20 0.18 0.16 0.14 0.12 0.10 0.08 1
2
(A)
3
4
n -
0.0050
(ZrO)(PO3)2OH (m/e = 281)
0.0045 0.0040
YSI(n) / YTotal(n)
0.0035 0.0030
ZrP Gel ZrP Crystal
0.0025 0.0020 0.0015 0.0010 0.0005 0.0000 1
2
3
4
n
(B)
Fig. 4. (A) The yield of m=z ¼ 79 (PO3 ) as a function of the total number of secondary ions ejected per single impact event. (B) The yield of m=z ¼ 281 as a function of the total number of secondary ions ejected per single impact event. Lines are guides for the eye.
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
depositing 50 ml of the slurry onto a stainless steel target.
3. Results and discussion Fig. 1 lists several fragment ions that can be traced back to the crystal structure of a-ZrP [2]. Negatively charged cluster secondary ions are the result of direct emission processes which suggests that they exist intact on the surface or result from the fragmentation of larger assemblies [2]. Thus, their intensities in the mass spectrum of the crystalline material will be higher than in the amorphous material. Fig. 2a shows a mass spectrum of a ZrP gel sample with the high mass region expanded in the inset. The two predominant peaks resulting from the ZrP sample are 63 (PO2 ) and 79 amu (PO3 ). Most of the higher mass structure-specific fragments are in low abundance. This has been attributed to lack of long range ordering, on the scale of the emission area [7]. Fig. 2b shows a powder X-ray diffraction (XRD) spectrum of the same sample. The peaks are broad suggesting some evidence of structure in the gel. This may explain why some structure-specific fragments are present in the mass spectrum of the gel sample (Fig. 2a). Fig. 3a shows a mass spectrum of a crystalline ZrP sample. Again, the two predominant peaks are 63 and 79 amu. However, the structure-specific fragments are clearly present (inset). The presence of these peaks can be attributed to the fact that the surface is ordered to a greater degree compared to the previous sample. A powder XRD spectrum, Fig. 3b, further supports this observation. The TME# program allows for comparison of the gel and crystalline samples in a unique manner. One method to compare samples is to plot the ratio of yields, Y, for a particular secondary ion as a function of the number of secondary ions detected per event. The ratio of yields, Y, is defined as: Y¼
YSI ðnÞ YTotal ðnÞ
where YSI ðnÞ is the yield of a particular secondary ion from the sum of all n-ion detection events and YTotal ðnÞ the yield of all secondary ions from the sum of all n-ion detection events. A plot of this ratio for m=z ¼ 79 as a function of the number of second-
111
ary ions detected per single impact event is given in Fig. 4a. The yield of PO3 for the gel sample is relatively constant however, the yield of PO3 decreases as the number of secondary ions per event increases in the crystalline sample. In the case of the crystalline sample the ratio of yields for m=z ¼ 281 (Fig. 4b) increases as the number of secondary ions ejected per event increases. This same trend is also present when looking at the fragment corresponding to m=z ¼ 343. One explanation that can account for these trends is that secondary ions from multiple ion ejection events are less excited vibrationally. The vibrational energy deposited by the projectile may be more efficiently dissipated through a larger crystal lattice network resulting in a fragment ion that is more stable. Indeed as the number of secondary ions ejected per event increases there is an increase in the yield of higher mass fragments, Fig. 4b, along with a concomitant decrease in the yield of PO3 , Fig. 4a, for the crystalline material. In the amorphous sample the vibrational energy is imparted to a disordered system with less energy dissipation resulting in more fragmentation. This is apparent when considering the yields of m=z ¼ 281, Fig. 4b, and the relatively constant yield of PO3 , Fig. 4a, for the amorphous material.
4. Conclusions We have demonstrated, from a qualitative standpoint, that cluster SIMS can reveal changes in surface structure in the absence of any change in the analyte stoichiometry. Furthermore these results demonstrate that monitoring selected emission events, e.g. events when three or four secondary ion are ejected, can be very sensitive to changes in the surface structure. Further work is needed to investigate possible correlations between the low and high mass fragments in events where multiple secondary ions are detected.
Acknowledgements This study was supported by grants from the National Science Foundation (CHE-0135888) and the Robert A. Welch Foundation (A-1482).
112
R.D. Rickman et al. / Applied Surface Science 231–232 (2004) 106–112
References [1] A. Benninghoven, P. Bertrand, H.N. Migeon, H.W. Werner (Eds.), Secondary Ion Mass Spectrometry—SIMS XII, Elsevier, Amsterdam, 2000. [2] M. VanStipdonk, M.A. Park, E.A. Schweikert, P. Sylvester, A. Clearfield, Int. J. Mass Spectrom. Ion Process. 128 (1993) 133– 141. [3] S.V. Verkhoturov, E.A. Schweikert, N.M. Rizkalla, Langmuir 18 (2002) 8836.
[4] K. Wien, Lect. Notes Phys. 269 (1986) 1. [5] R.D. Rickman, S.V. Verkhoturov, E.A. Schweikert, Phys. Rev. Lett. 92 (2004) 047601. [6] M. Benguerba, A. Brunelle, S. Della-Negra, J. Depauw, H. Joret, Y. Le Beyec, M.G. Blain, E.A. Schweikert, G. Ben Assayag, P. Sudreau, Nucl. Instrum. Meth. B 62 (1991) 8. [7] M.J. VanStipdonk, J.B. Shapiro, E.A. Schweikert, Vacuum 46 (8–10) (1995) 1227.