Dynamic Secondary Ion Mass Spectrometry Research Articles

Secondary Ion Mass Spectrometry and Time-of-Flight Secondary Ion Mass Spectrometry Study of Impurity Measurements in HgCdTe

In this study, time-of-flight (TOF) secondary ion mass spectrometry (SIMS) was compared against dynamic SIMS to determine detection limits and background levels for nine impurities: Li, Na, K, Al, Ni, As, In, Fe, and Cu. Statistics were gathered by measuring six material test structure samples from six different liquid phase epitaxy (LPE) HgCdTe double layer heterojunction (DLHJ) wafers. Also included is a comparison between dynamic SIMS and TOF-SIMS capabilities.

Surface analysis of thermally annealed porous silicon

Quasi-monocrystalline porous silicon (QMPS) has high potential for photovoltaic application for its enhanced optical absorption compared to bulk silicon in the visible range of solar spectrum. In this study, QMPS was formed from low porosity (∼20–30%) porous silicon (PS) produced by electrochemical anodization, and thermal annealing in the temperature range 1050–1100 °C under pure hydrogen ambient for a duration of 30 min. We analyzed the material surface by grazing incidence X-ray diffraction (GIXRD), field emission scanning electron microscopy (FESEM), atomic force microscopy (AFM) and dynamic secondary ion mass spectroscopy (SIMS) study. The crystallinity was confirmed by GIXRD while FESEM studies revealed that the surface layer is pore free with voids embedded inside the body. AFM studies indicated relatively smooth and uniform surface and the dynamic SIMS study showed the depth profiles of impurities present in the material.

Anomalous Mn depth profiles for GaMnAs∕GaAs(001) thin films grown by molecular beam epitaxy

Mn concentration depth profiles in Mn-doped GaAs thin films grown at substrate temperatures of 580 and 250°C using various Mn cell temperatures have been investigated with dynamic secondary ion mass spectrometry and Auger electron spectroscopy. When the samples are grown at a low substrate temperature of 250°C, the Mn distributes uniformly. For the samples grown at a high substrate temperature of 580°C, the concentration depth profiles are easily fitted with a temperature-dependent Fermi function only if the Mn concentration is above the solubility limit. However, when the Mn concentration is below the solubility limit, unexpected peaks are observed in the concentration depth profiles.

Quantification of Liquid Crystal Concentrations in Periodically Stratified Polymer-Dispersed Liquid Crystal Films by Dynamic Secondary Ion Mass Spectrometry and Multivariate Statistical Analysis

Liquid crystal (LC) concentrations in stratified holographic polymer-dispersed liquid crystal (PDLC) films have been quantified from depth profiles obtained by time-of-flight secondary ion mass spectrometry (SIMS). The volatile nature of the LCs was hindered during SIMS analysis by capping the PDLC samples with poly(vinyl alcohol) and cooling to cryogenic temperatures. It remains difficult to gain quantitative results from SIMS analysis due to matrix effects yielding complex SIMS fingerprint spectra. With the multivariate statistical method discriminant function analysis the LC contents in the stratified holographic PDLC films were quantified. The concentration of the LC-rich layers was determined to be 32.9 ± 3.4 wt % LC, and for the polymer-rich layers it was 28.8 ± 2.7 wt % LC. The low concentration difference was supported by imaging analysis and modeling results.

Investigation of Potassium Contamination in SOI Wafer Using Dynamic SIMS

Mobile ions may cause critical failures in integrated circuits. For silicon-on-insulator (SOI) wafers, the mobile ions affect not only the reliability of back end of the line but also the performance of the transistors. This paper presents a case study of potassium (K) contamination in the SOI wafer using dynamic secondary ion mass spectrometry. The results showed that the presence of K in chemical and mechanical polish slurry contaminated the porous interlayer dielectrics (ILD), penetrating below the surface due to their porous structure. The K contamination has been greatly reduced by capping the porous ILD with a high-density-oxide layer.

Analysis of Reaction Kinetics of End-Functionalized Polymers at a PS/P2VP Interface by DSIMS

Reactions to produce graft copolymers at polymer interfaces during extruder mixing are important for controlling dispersed phase size by retarding droplet coalescence and reducing interfacial tension while reinforcing the polymer interface. Such reactions are investigated at various temperatures in a model bilayer film system consisting of amine end-functional deuterated polystyrene (dPS-NH 2 ) in PS and anhydride end-functional poly(2-vinylpyridine) (P2VP-ah) in P2VP. The interfacial excess (z*) of block copolymer formed by the reaction of dPS-NH 2 and P2VP-ah at the interface is determined by detecting the 2 H - ion using dynamic secondary ion mass spectrometry (DSIMS). The reaction kinetics is analyzed using a model based on a reaction-controlled mechanism for various reaction temperatures and molecular weights (M n ) of the end-functional chains. At low initial volume fractions Φ 0 (∼0.01) of dPS-NH 2 and P2VP-ah, such that the normalized interface excess (z*/R g ) < 1 and the blocks are unstretched, the forward reaction rate constant (k + ) decreases as M n -0.70 , in rough agreement with predictions of the model (k + ∼ M n -0.55 ) for this regime. The reaction is thermally activated with an activation enthalpy (∼160 kJ/mol) that is independent of M n . The interfacial reaction, which is carried out under high-vacuum conditions so that any water produced by reaction would be efficiently removed, is reversible with an equilibrium constant K rxn that has an enthalpy of reaction of -55 kJ/mol. This result means that the interface reaction does not proceed to form the cyclic imide, but likely stops at the amic acid.

Chain Architecture Effects on the Diffusion of Cylinder-Forming Block Copolymers

We investigate the effect of chain architecture on the self-diffusion of poly(styrene)−poly(ethylene) triblock (PS−PE−PS or SES) and pentablock (PS−PE−PS−PE−PS or SESES) copolymers with a cylindrical domain structure (E cylinders). Transmission electron microscopy (TEM) and grazing-incidence small-angle X-ray scattering (GISAXS) are employed to reveal the microstructure of the block copolymer films, and dynamic secondary ion mass spectroscopy (DSIMS) is used to obtain diffusion coefficients. The triblock copolymer (SES) chains diffuse through a randomly oriented cylindrical structure via a “hopping” mechanism where the normalized diffusion coefficient D/D0 (D is the diffusion coefficient of the block copolymer and D0 is the diffusion coefficient of a disordered (hypothetical) block copolymer of similar Mw) decreases exponentially with the thermodynamic barrier χNPE, where χ is the Flory interaction parameter between styrene and ethylene and NPE is the degree of polymerization of the poly(ethylene) block. However, the pentablock copolymer (SESES) diffuses via a “walking” mechanism through an ordered cylindrical structure which involves the activation of a single E block at a time to overcome the thermodynamic barrier. Both mechanisms have been reported previously for diblock and triblock copolymers, respectively. Hence, for a similar overall Mw the pentablock copolymer diffuses faster than the corresponding triblock copolymer. Moreover, a comparison is made between perpendicular and parallel diffusion in the pentablock copolymer as both mechanisms are observed due to a change in the cylinder orientation (from parallel to perpendicular cylinders) as a function of depth. The relative independence of the parallel diffusion on χNPE suggests that the thermodynamic barrier for perpendicular diffusion does not affect parallel diffusion coefficients. However, the normalized parallel and perpendicular diffusion coefficients of the SESES pentablock copolymer as well as the normalized perpendicular diffusion coefficient of the SES triblock copolymer are substantially less than the corresponding diffusion coefficients of a cylinder-forming diblock copolymer investigated by Cavicchi and Lodge at the same values of χNcyl, where Ncyl is the degree of polymerization of the cylinder block. This comparison suggests that entangled loops in the PE midblock cylinders impose additional entropic barriers to diffusion that are absent for diblock copolymers.

Corrosive attack of glass by a pharmaceutical compound

Glass delamination, or the generation of glass flakes, continues to be an unwanted occurrence in the manufacture of parenteral (injectable) solutions and suspensions. In this root cause analysis study, advanced analytical tools including atomic force microscopy, environmental scanning electron microscopy, quantitative image analysis, and dynamic secondary ion mass spectroscopy (D-SIMS) showed significant differences in glass characteristics and performance. By observing the size and spatial arrangement of defects found on the interior surface of vials used as primary packaging for these products, in conjunction with the chemical changes that can occur to the glass because of product contact, a considerable amount of insight can be obtained into this phenomenon. Elemental depth profiling obtained by D-SIMS revealed that the interior vial surface was significantly altered by the presence of the parenteral solution, while another vial (manufactured by another vendor) was not. Although significant chemical changes can occur to the glass, the surface defect structure appears to be the dominant factor controlling the generation of glass flakes.

Approach for detecting localization of inkjet ink components using dynamic-SIMS analysis

Organic compounds having high boiling points included in inkjet inks remain in the printed media after printing. It has been said that these residual organic compounds influenced the fastness of pigment ink prints, but details of the analytical methods have not been reported, yet. In this paper, a new analytical method was applied for detecting localization of those organic compounds in inkjet prints after printing. Glycerin was chosen as a typical example of those organic compounds. To analyze the position within the printed paper, deuterated glycerin was formulated in the ink as a probe and dynamic Secondary Ion Mass Spectroscopy was applied to detect the deuterium peak in the whole printed paper thickness. It was found that glycerin remained in the range from the bottom of the pigment ink layer to the pigment coating layer of the printed media. This peak was sufficiently detectable for at least 7 days after printing. The maximum peak of residual glycerin was located at the surface of the coating layer of paper.

State-of-the-Art Characterization for 65 nm CMOS Processes and Beyond

Characterization of advanced CMOS devices and control of front- end of line (FEOL) processes require state-of-the-art analytical techniques and specific measurement protocols. Dynamic Secondary Ion Mass Spectrometry (D-SIMS) is well suited for in-depth quantitative analysis of both bulk- and dopant concentrations in GexSi1-x binary alloys. A similar D-SIMS protocol -in combination with Auger Electron Spectroscopy (AES) - allows monitoring nickel-silicide phase transitions upon silicidation of (doped) Ni Totally Silicided (TOSI) gates. Combination of Low-Energy Electron X-ray Emission Spectroscopy (LEXES) and high-precision DSIMS provides accurate control of dose, junction depth, and wafer uniformity for Ultra-Shallow Junctions (USJ) at high analytical throughput, opening the way to USJ metrology. Similar techniques (LEXES, SIMS) are employed to control wafer uniformity of deposited NiPt metal films as Source/Drain (S/D) contact metal, to monitor NixSi1-x phase transformations and Pt-redistribution upon silicidation, and to determine root causes of defectivity at the device level (micro- AES). The given examples demonstrate the importance of having available (novel) state-of-the-art characterization techniques - pushing the analytical limits in terms of resolution, sensitivity, and accuracy while simultaneously increasing throughput and reproducibility- to support process development, improve yield, and to speed up yield learning of advanced CMOS processes.

Nitrogen doping and multiplicity of stacking faults in SiC

This paper reports on the strong enhancement of stacking fault (SF) formation in 4H–SiC by heavy nitrogen doping. The paper consists of two separate observations. The first part reports on localized but severe deformation bands observed in certain regions of 4H–SiC wafers that had undergone high temperature processing during device fabrication. Using a combination of dynamic secondary ion mass spectroscopy (SIMS) and conventional, weak-beam (WB) and high-resolution (HR) transmission electron microscopy (TEM), the affected regions of the wafers were found to have a much higher concentration of nitrogen and to contain a high density of stacking faults. In contrast, in the non-affected regions of the wafers, the nitrogen concentration was lower and no lattice defects could be observed by TEM, indicating that the severely deformed morphology of the affected regions was due to the high stacking fault content. Moreover, the stacking faults in the affected regions were found to be invariably double and not single-layered, formed by the glide of two leading partial dislocations on adjacent (0001) planes. The second part of the paper reports on the occurrence of stacking faults during deformation tests on heavily nitrogen-doped 4H–SiC. Combining optical microscopy, HR and weak-beam (WB) TEM, the generated faults were found to be double-layered as well. It is interesting that in neither type of experiment, trailing partials were observed: it appears that the SFs were not in the form of ribbons bound by leading and trailing partials but rather in the form of faulted loops on two adjacent planes, each loop bound by a leading Shockley partial of the same Burgers vector. The results of the two observations are explained by the stabilization of double-layer stacking faults (DSFs) when the Fermi level of the faulted crystal is pushed up by nitrogen doping to above the stacking fault energy level.

Diblock Copolymer Surfactant Transport across the Interface between Two Homopolymers

Dynamics of adsorption and desorption of a diblock copolymer to an interface between two homopolymers was measured using dynamic secondary-ion mass spectrometry (SIMS). Thin films were constructed consisting of a layer of saturated polybutadiene with 90% 1,2-addition (sPB90), followed by a layer of saturated polybutadiene with 63% 1,2-addition (sPB63), and finally by another layer of the sPB90 homopolymer. A sPB90-sPB63 diblock copolymer was initially included only in the top sPB90 layer of the film at a volume fraction of 0.05. The thin films were annealed at ambient temperature for times ranging between 0.2 and 108 h, and the concentration profiles of the diblock copolymer through the films were measured using SIMS. The dynamics of adsorption and desorption of the diblock copolymer at the two sPB90-sPB63 interfaces was gauged by comparing the different transient concentration profiles. The sorption process was modeled as diffusion in an external field, generated from self-consistent field theory (SCFT). All parameters for the model were determined independently. Although the model neglects the dynamics of conformational change, experimental results matched theory very well.

Vertical phase separation in spin-coated films of a low bandgap polyfluorene/PCBM blend—Effects of specific substrate interaction

We report on the effect of the substrate on the vertical phase separation in spin-coated thin films of poly[(9,9-dioctylfluorenyl-2,7-diyl)-co-5,5-4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiazole] (APFO-3) blended with [6,6]-phenyl-C 61-butyric acid methyl ester (PCBM). Compositional depth profiles of the films are measured by dynamic secondary ion mass spectrometry (SIMS). We found that changing the substrate from silicon to gold affects the composition profile near the substrate interface. This is caused by a specific interaction between the polymer (APFO-3) and the gold surface, as confirmed by X-ray photoelectron spectroscopy (XPS). The composition profile in the area away from the substrate interface, as well as the enrichment of the free surface with APFO-3, remain however unaffected by the choice of substrate. The vertical composition was also analysed for APFO-3:PCBM films spin-coated on indium tin oxide (ITO) coated with a thin layer of (3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS).

Growth and characterization of chromium oxide thin films prepared by reactive ac magnetron sputtering

Cr O x thin films were prepared on single crystal silicon (111 orientation) and glass substrates by midfrequency (41kHz) ac sputtering technique in an Isoflux ICM-10 sputter deposition system consisting of two hollow cylindrical targets of Cr in an argon-oxygen plasma at a discharge power of 5kW and without any deliberate substrate heating. X-ray diffraction studies showed that the coatings were of hcp rhombohedral α-phase chromium oxide. CrOx coating samples were found to grow preferentially towards the (110) crystal orientation of α-Cr2O3. Texturing was found to depend on the orientation of the substrates relative to the targets and also on the nature of substrates. One coating sample grown on Si substrate was annealed in air up to 1173K. Whereas heat treatment increased the crystallite size, CrOx coating was thermally stable and did not show any structural transformations. Scanning electron microscopy studies showed differences in the surface morphology of the coatings grown on glass and silicon substrates. Dynamic secondary ion mass spectrometry measurements performed on one CrOx coating deposited on silicon showed that the O∕Cr ratio in the films was 1.38. Ar and H impurity concentrations were also measured in this coating as a function of film thickness. Alumina coatings were prepared on Si substrates by reactive sputtering technique using CrOx coating as template layers. X-ray diffraction studies showed that CrOx template layers improved the crystallinity of alumina coatings grown on top of it and also facilitated the formation of the thermodynamically stable α-alumina phase.

Smooth Nb surfaces fabricated by buffered electropolishing

It was demonstrated that smooth Nb surfaces could be obtained through buffered electropolishing (BEP) employing an electrolyte consisting of lactic, sulfuric, and hydrofluoric acids. Parameters that control the polishing process were optimized to achieve a smooth surface finish. The polishing rate of BEP was determined to be 0.646 μm/min which was much higher than 0.381 μm/min achieved by the conventional electropolishing (EP) process widely used in the superconducting radio frequency (SRF) community. Root mean square measurements using a 3D profilometer revealed that Nb surfaces treated by BEP were an order of magnitude smoother than those treated by the optimized EP process. The chemical composition of the Nb surfaces after BEP was analyzed by static and dynamic secondary ion mass spectrometry (SIMS) systems. SIMS results implied that the surface oxide structure of Nb might be more complicated than what usually believed and could be inhomogeneous. Preliminary results of BEP on Nb SRF single cell cavities and half-cells were reported. It was shown that smooth and bright surfaces could be obtained in 1800 s when the electric field inside a SRF cavity was uniform during a BEP process. This study showed that BEP is a promising technique for surface treatment on Nb SRF cavities to be used in particle accelerators.

Characterization of high explosive particles using cluster secondary ion mass spectrometry

The use of secondary ion mass spectrometry (SIMS) for the detection and spatially resolved analysis of individual high explosive particles is described. A C(8) (-) carbon cluster primary ion beam was used in a commercial SIMS instrument to analyze samples of high explosives dispersed as particles on silicon substrates. In comparison with monatomic primary ion bombardment, the carbon cluster primary ion beam was found to greatly enhance characteristic secondary ion signals from the explosive compounds while causing minimal beam-induced degradation. The resistance of these compounds to degradation under ion bombardment allows explosive particles to be analyzed under high primary ion dose bombardment (dynamic SIMS) conditions, facilitating the rapid acquisition of spatially resolved molecular information. The use of cluster SIMS combined with computer control of the sample stage position allows for the automated identification and counting of explosive particle distributions on silicon surfaces. This will be useful for characterizing the efficiency of transfer of particulates in trace explosive detection portal collectors and/or swipes utilized for ion mobility spectrometry applications.

Dynamic SIMS analysis of cryo-prepared biological and geological specimens

The modification of a dynamic magnetic sector secondary ion mass spectrometry (SIMS) instrument to permit the analysis of frozen biological and geological samples is described. The SIMS instrument used for this study combines SIMS analysis with the generation of ion-induced secondary electron images, allowing direct superposition of the SIMS image onto the image of cellular structures. Secondary ion maps have been acquired with sub-micron resolution, permitting the characterisation of sub-cellular elemental distributions in plant cells and human fibroblast cells, as well as the distribution of chemical impurities at grain boundaries in polar ice samples. This cryo-preparation technique clearly extends the applicability of SIMS analysis to a wide range of samples.

Molecular Recognition in Structured Matrixes: Control of Guest Localization in Block Copolymer Films

We demonstrate the use of molecular recognition to control the spatial distribution of guest molecules within block copolymer films. Block copolymers bearing recognition units were combined with complementary and noncomplementary molecules, and the extent of segregation of these molecules into the different domain types within microphase-separated thin films was quantitatively analyzed using dynamic secondary ion mass spectrometry (SIMS). Complementarity between the guest molecules and the polymer functionalities proved to be a key factor and an efficient tool for directing the segregation preference of the molecules to the different domain types. The effect of segregation preference on the glass transition temperature was studied using differential scanning calorimetry (DSC), and the results corroborate the SIMS findings. In a complementary study, guests with tunable sizes (via dendron substituents) were used to control block copolymer morphology. Morphological characterization using transmission electron microscopy (TEM) and X-ray diffraction reveal that selectivity differences can be directly translated into the ability to obtain different morphologies from recognition unit-functionalized block copolymer scaffolds.

Recovery of the Si–SiO2 interface studied by self-diffusion after high fluence ion implantation of 28Si

In this work, we discuss the recovery of the Si–SiO2 interface after high fluence ion implantation of isotope pure 28Si fluences of 1×1016–5×1016at./cm2 into 50nm thick thermally grown SiO2 on (100)-silicon and subsequent thermal treatment. The implantation energy was chosen to be 32keV in order to place the maximum of the implanted concentration profile into the interface. The ion implanted Si provokes ion induced atomic mixing and over-stoichiometric 28Si in the SiO2 layer.The depth profiles of 28Si, 29Si and 30Si isotopes, the trimer 28Si3 and oxygen were measured by dynamic secondary ion mass spectrometry (SIMS) using low energy Cs+ sputter ions with 1.5keV before and after annealing of the samples. Matrix effects, often encountered in dynamic SIMS depth profiling, obscuring the measured secondary ion yield depth profiles in close vicinity to the SiO2–Si interface, were eliminated by analyzing the Si isotope composition. By this technique material transport processes involved in the recovery of the interface by self-diffusion was studies. It was found, that transport of oxygen dominates and leads to oxidation of excess Si in close proximity to the interface.

Thermally cured low surface-tension epoxy films

A simple method is described on the preparation of low surface-tension epoxy films via thermal curing. First a partially fluorinated diamine was synthesized by reacting perfluoroalkyl epoxide with a known excess of diamine. Then a mixture of the fluorinated diamine and diglycidyl ether of bisphenol-A was thermally cured, resulting in films with surface tensions as low as 15 mN/m by using less than 2 wt% of perfluoroalkyl epoxide. The fluorinated species segregated preferentially at the film surface to minimize the coating-air interfacial energy. The surface tension of the epoxy films can be adjusted by varying the type of the fluorinated epoxides and their amounts. The surface enrichment of fluorine was verified by angle-resolved X-ray photoelectron spectroscopy (XPS), showing a more than 80-fold excess in the outermost ∼9 nm compared to the theoretical bulk concentration. Deeper depth profiles on the fluorine content were qualitatively revealed by dynamic secondary-ion mass spectrometry (DSIMS) measurements.