C. W. Hagen

Delft University of Technology, Delft, South Holland, Netherlands

Are you C. W. Hagen?

Claim your profile

Publications (77)142.14 Total impact

  • Oddur Ingólfsson · Ragesh Kumar TP · Cornelis W Hagen · Sangeetha Hari ·

    Journal of Physics Conference Series 09/2015; 635(7):072088. DOI:10.1088/1742-6596/635/7/072088
  • T. Verduin · S.R. Lokhorst · P. Kruit · C.W. Hagen ·
    [Show abstract] [Hide abstract]
    ABSTRACT: We have investigated in a numerical study the determination of sidewall roughness (SWR) from top down scanning electron microscopy (SEM) images. In a typical metrology application, top-down SEM images are acquired in a (critical-dimension) SEM and the roughness is analyzed. However, the true size, shape and roughness characteristics of resist features are not fully investigated in the analysis of top-down SEM images. In reality, rough resist features are complex three-dimensional structures and the characterization naturally extends to the analysis of SWR. In this study we randomly generate images of rough lines and spaces, where the lines are made of PMMA on a silicon substrate. The lines that we study have a length of 2 μm, a width of 32nm and a height of 32 nm. The SWR is modeled by using the power spectral density (PSD) function of Palasantzas, which characterizes roughness by the standard deviation σ, correlation length ξ and roughness exponent α. The actual roughness is generated by application of the method of Thorsos in two dimensions. The images are constructed by using a home-built program for simulating electron-specimen interactions. The program that we have developed is optimized for complex arbitrary geometries and large number of incident low energy primary electrons by using multi-core CPUs and GPUs. The program uses the dielectric function model for inelastic scattering events and has an implementation specifically for low energy electrons. A satisfactory comparison is made between the secondary electron yields from the home-built program and another program found in literature. In order to reduce the risk of shrinkage, we use a beam energy of 300 eV and a spot size of 3 nm. Each pixel is exposed with 20 electrons on average ( 276 μC/cm2), following the Poisson distribution to account for illumination shot noise. We have assumed that the detection of electrons is perfect and does not introduce additional noise. We measure line edge roughness (LER) in simulated top-down SEM images of randomly generated rough lines by using PSD analysis. The measurements are then compared to the actual SWR that was used to generate the rough lines. We conclude that the bias in the determination of SWR is a non-linear function of the correlation length ξ3D of the actual SWR. The measured correlation length ξ 2D shows a linear trend with the correlation length ξ 3D of the SWR. From another simulation run, we conclude that the relation between measured LER in the top-down image and the standard deviation ξ 3D of the SWR is linearly biased. We see that the amount of bias relates to the correlation length ξ 3D of the SWR: The bias in the determination of SWR from top-down images increases for decreasing correlation length ξ 3D of the actual SWR. The results of this study, with respect to the metrology of rough resist features, touches upon the reliability and comparability of roughness characterization in top-down images.
    Proceedings of SPIE - The International Society for Optical Engineering 01/2015; 9424. DOI:10.1117/12.2085768 · 0.20 Impact Factor
  • C. W. Hagen ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A perspective is sketched for the field of focused electron beam-induced processing (FEBIP). The FEBIP lithography technique is compared to the very successful resist-based electron beam lithography (EBL) technique. The advantages of FEBIP over EBL are identified, the main advantage being its high spatial resolution. This will enable FEBIP to become an important lithography technique for the fabrication of devices with critical dimension in the range between 1 and 20 nm and serve as a complementary technique to EBL. It will be discussed what needs to be done to achieve this and what the potential applications are.
    Applied Physics A 12/2014; 117(4). DOI:10.1007/s00339-014-8847-8 · 1.70 Impact Factor
  • A.C. Zonnevylle · C. Th. H. Heerkens · C. W. Hagen · P. Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In single electron beam columns the alignment of the beam is obtained either by mechanical shifting the lenses or by x/y alignment deflectors. We are developing multi electron beam columns for electron microscopy and lithography. The problem in multi-beam systems is that mechanical alignment and simple x/y deflectors can only align the total array of beams in the x/y direction and not correct the position of individual beamlets. We present here a simple design, fabrication, electron-optical analysis and experimental results of a multi-beam x/y deflector array that can deflect each beamlet separately. The array is fabricated with micro-fabrication technology with in-plane deflection plates made of molybdenum. The electron optical properties of this in-plane deflector are simulated and compared to a traditional deflector. The experimental measurements are compared with the simulations and are in agreement.
    Microelectronic Engineering 07/2014; 123:140-148. DOI:10.1016/j.mee.2014.06.014 · 1.20 Impact Factor
  • S. Hari · A. M. Goossens · L.M.K. Vandersypen · C. W. Hagen ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Graphene, a monolayer of carbon atoms arranged in a hexagonal lattice, is an interesting material because of its unique electronic properties. Electron Beam Induced Deposition (EBID) is an attractive direct-write technique to fabricate graphene devices for electronic applications, due to the combination of the high resolution and the versatility of the technique. We study electron induced damage to graphene at the typical EBID energy and dose range. In particular, we investigate the role of the substrate by carrying out electron exposure studies of graphene on two different substrates - silicon oxide and hexagonal boron nitride. Raman measurements reveal the emergence of a disorder D peak upon irradiation of graphene on silicon oxide, whereas graphene on boron nitride does not show this damage. Further, we pattern structures on graphene using EBID from the platinum precursor MeCpPtMe3 and analyze the changes in the Raman spectrum of graphene coming about due to electron stimulated effects in the presence of the precursor.
    Microelectronic Engineering 06/2014; 121:122-126. DOI:10.1016/j.mee.2014.04.037 · 1.20 Impact Factor
  • Thomas Verduin · Pieter Kruit · Cornelis W. Hagen ·
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigated the off-line metrology for line edge roughness (LER) determination by using the discrete power spectral density (PSD). The study specifically addresses low-dose scanning electron microscopy (SEM) images in order to reduce the acquisition time and the risk of resist shrinkage. The first attempts are based on optimized elliptic filtering of noisy experimental SEM images, where we use threshold-based peak detection to determine the edge displacements. The effect of transversal and longitudinal filterings cannot be ignored, even when considering an optimized filter strength. We subsequently developed a method to detect the edge displacements without the use of a filter and thus avoiding biasing. This makes it possible to study how much image noise is acceptable and still determine the LER. The idea is to generate random images of line edges using the model of Palasantzas and the algorithm of Thorsos. We study the simulated PSDs as a function of the number of line edges and report on the convergence of the parameters (LER, correlation length, and roughness exponent) by fitting the Palasantzas model extended with a white noise term. This study demonstrates that a very noisy image with 12 line edges and about 2 electrons per pixel on average (charge density approximate to 10 mu C) already produces an estimation for LER with a relative error (one-sigma) of about 10%. Furthermore, increasing the dose beyond 20 electrons per pixel does not significantly improve the LER determination. (C) 2014 Society of Photo-Optical Instrumentation Engineers (SPIE)
    SPIE Advanced Lithography; 04/2014
  • Sangeetha Hari · Cornelis W. Hagen · Thomas Verduin · Pieter Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In a first study to analyze the feasibility of Electron Beam Induced Deposition (EBID) for creating certain patterns in advanced lithography, line patterns were fabricated on silicon wafers using EBID. The growth conditions were such that the growth rate is fully determined by the electron flux (the current limited growth regime). It is experimentally verified that different patterning strategies, such as serial versus parallel patterning and single pass patterning versus multiple pass patterning, all lead to the same result in this growth regime. Images of EBID lines, imaged in a scanning electron microscope, were analyzed to determine the position of the lines, the width of the lines and the line edge roughness (LER). The results are that the lines have an average width of 13.7 nm, an average standard deviation of 1.6 nm in the center position of the lines, and an average LER of 4.5 nm (1σ value). As an example of the capabilities of EBID a logicresembling lithography pattern was fabricated.
    SPIE Advanced Lithography; 03/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, the authors discuss a new electron optical component: an electrostatic lens in which one electrode is intentionally shifted laterally, breaking the rotational symmetry. This lens is called a “shift lens.” Usually, a shifted electrode is undesired, and the resulting aberrations are calculated only for the purpose of setting manufacturing requirements. However, the shift lens can be applied as a deflector. Thus, in multibeam systems with an individual microlens for each beam, all beams can be deflected with a single voltage. By giving a different shift to each lens, the deflection can be different for each beam. This allows the creation of a multibeam rotation error corrector. The optical properties of an electrostatic five-electrode lens with a shifted middle electrode are analyzed in this paper. For describing the optical properties of the shift lens, a simple mirror symmetric model in combination with Taylor polynomials is used. This model is then verified with a newly developed ray-tracing program, and the obtained aberrations are discussed. The middle electrode is shifted over a range of 1%–20% of the diameter of the lens. The authors have found dependences of deflection, defocus, astigmatism, and second order on shift distance and excitation. The authors expect the shift lens to be a useful new optical component, especially in multibeam systems.
    Journal of vacuum science & technology. B, Microelectronics and nanometer structures: processing, measurement, and phenomena: an official journal of the American Vacuum Society 11/2013; 31(6):06F702-06F702-7. DOI:10.1116/1.4826250 · 1.46 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The current understanding in the study of focused electron beam induced processing (FEBIP) is that the growth of a deposit is mainly the result of secondary electrons (SEs). This suggests that the growth rate for FEBIP is affected by the SE emission from the support. Our experiments, with membranes thinner than the SE escape depth, confirm this hypothesis. We used membranes of 1.4 and 4.3 nm amorphous carbon as supports. At the very early stage, the growth is support-dominated and the growth rate on a 4.3 nm thick membrane is three times higher than on a 1.4 nm thick membrane. This is consistent with Monte Carlo simulations for SE emission. The results suggest that SEs are dominant in the dissociation of W(CO)6 on thin membranes. The best agreement between simulations and experiment is obtained for SEs with energies between 3 and 6 eV.With this work we revisit earlier experiments, working at a precursor pressure 20 times lower than previously. Then, despite using membranes thinner than the SE escape depth, we did not see an effect on the experimental growth rate. We explain our current results by the fact that very early in the process, the growth becomes dominated by the growing deposit itself.
    Nanotechnology 07/2013; 24(34):345301. DOI:10.1088/0957-4484/24/34/345301 · 3.82 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Here we describe the production, using lithography and micro-engineering technologies, of patterned arrays of nanofabricated gold dots on a thin Si3N4 electron transparent layer, supported by silicon. We illustrate that the support with a patterned structure of nanosized gold can be exploited for (cryo) electron tomography application as a specimen support with predefined alignment markers. This nanogold patterned support has several advantages. The Si3N4 window provides a 50nm thin, strong and flat support with a ∼0.7mm(2) large electron-beam transparent window. The nanogold pattern has a user-defined size and density, is highly regular and stable. This facilitates accurate tracking during tilt series acquisition, provides sufficient contrast for accurate alignment during the image reconstruction step and avoids an uneven lateral distribution and movement of individual fiducials. We showed that the support is suitable for electron tomography on plastic sections.
    Ultramicroscopy 07/2013; 135C:99-104. DOI:10.1016/j.ultramic.2013.06.015 · 2.44 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Using three different precursors [MeCpPtMe3, Pt(PF3)4, and W(CO)6], an ultra-high vacuum surface science approach has been used to identify and rationalize the effects of substrate temperature and electron fluence on the chemical composition and bonding in films created by electron beam induced deposition (EBID). X-ray photoelectron spectroscopy data indicate that the influence of these two processing variables on film properties is determined by the decomposition mechanism of the precursor. For precursors such as MeCpPtMe3 that decompose during EBID without forming a stable intermediate, the film's chemical composition is independent of substrate temperature or electron fluence. In contrast, for Pt(PF3)4 and W(CO)6, the initial electron stimulated deposition event in EBID creates surface bound intermediates Pt(PF3)3 and partially decarbonylated Wx(CO)y species, respectively. These intermediates can react subsequently by either thermal or electron stimulated processes. Consequently, the chemical composition of EBID films created from either Pt(PF3)4 or W(CO)6 is influenced by both the substrate temperature and the electron fluence. Higher substrate temperatures promote the ejection of intact PF3 and CO ligands from Pt(PF3)3 and Wx(CO)y intermediates, respectively, improving the film's metal content. However, reactions of Pt(PF3)3 and Wx(CO)y intermediates with electrons involve ligand decomposition, increasing the irreversibly bound phosphorous content in films created from Pt(PF3)4 and the degree of tungsten oxidation in films created from W(CO)6. Independent of temperature effects on chemical composition, elevated substrate temperatures (>25 °C) in- reased the degree of metallic character within EBID deposits created from MeCpPtMe3 and Pt(PF3)4.
    Journal of Vacuum Science & Technology B Microelectronics and Nanometer Structures 09/2012; 30(5):1805-. DOI:10.1116/1.4751281 · 1.36 Impact Factor
  • Source
    J. C. van Oven · F. Berwald · K. K. Berggren · P. Kruit · C. W. Hagen ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper demonstrates electron-beam-induced deposition of few-nm-width dense features on bulk samples by using a scanning electron-beam lithography system. To optimize the resultant features, three steps were taken: (1) features were exposed in a repetitive sequence, so as to build up the deposited features gradually across the entire pattern, and thus avoid proximity effects; (2) an additional delay was added between exposures to permit diffusion of reactants into the exposed area; and (3) the exposures were phase-synchronized to the dominant noise source (the 50-Hz line voltage) to minimize the effect of noise. The reasons these steps led to significant improvements in patterning resolution are discussed.
    Journal of Vacuum Science & Technology B Microelectronics and Nanometer Structures 11/2011; 29(6). DOI:10.1116/1.3640743 · 1.36 Impact Factor
  • P. C. Post · A. Mohammadi-Gheidari · C. W. Hagen · P. Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Lithography techniques based on electron-beam-induced processes are inherently slow compared to light lithography techniques. The authors demonstrate here that the throughput can be enhanced by a factor of 196 by using a scanning electron microscope equipped with a multibeam electron source. Using electron-beam induced deposition with MeCpPtMe3 as a precursor gas, 14 × 14 arrays of Pt-containing dots were deposited on a W/Si3N4/W membrane, with each array of 196 dots deposited in a single exposure. The authors demonstrate that by shifting the array of beams over distances of several times the beam pitch, one can deposit rows of closely spaced dots that, although originating from different beams within the array, are positioned within 5 nm of a straight line.
    Journal of Vacuum Science & Technology B Microelectronics and Nanometer Structures 11/2011; 29(6). DOI:10.1116/1.3656027 · 1.36 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The effect of 500 eV electrons on nanometer-thick films of the platinum precursor tetrakis(trifluorophosphine)platinum [Pt(PF3)4] has been studied in situ under ultra-high-vacuum (UHV) conditions using a combination of X-ray photoelectron spectroscopy (XPS), mass spectrometry (MS), and high-resolution electron energy loss spectroscopy (HREELS). Electron irradiation of adsorbed Pt(PF3)4 molecules initially proceeds through a single Pt—P bond-cleavage event and the ejection of one PF3 ligand, analogous to the electron-stimulated reactions of Pt(PF3)4 in the gas phase. The electron-stimulated deposition cross section of Pt(PF3)4, σPt(PF3)4, is governed by the rate of this initial Pt—PF3 cleavage event, which is calculated to be 2.5 × 10–15 cm2 at an incident electron energy of 500 eV. In contrast to the initial deposition step, subsequent electron-stimulated reactions of the surface-bound Pt(PF3)3 intermediate occur exclusively through P—F bond cleavage and the release of fluorine into the gas phase. In this second phase of the decomposition process, oxygen uptake into the film is observed because of reactions between water vapor and the coordinatively unsaturated phosphorus atoms formed by P—F bond cleavage. Electron-beam-induced deposition (EBID) of Pt(PF3)4 was also performed by electron irradiating a substrate at room temperature and at higher electron fluxes, in the presence of a constant partial pressure of Pt(PF3)4. The absence of fluorine in these films underscores the role of electron-stimulated P—F bond cleavage, whereas the absence of oxygen highlights the important role that deposition conditions (e.g., substrate temperature and background gas composition) play in determining the ultimate composition of typical EBID films.
    The Journal of Physical Chemistry C 08/2011; 115(35). DOI:10.1021/jp204189k · 4.77 Impact Factor
  • V Castaldo · C W Hagen · P Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Scanning ion microscopy has received a boost in the last decade, thanks to the development of novel ion sources employing light ions, like He(+), or ions from inert gases, like Ne(+) and Ar(+). Scanning ion images, however, might not be as easy to interpret as SEM micrographs. The contrast mechanisms are different, and there is always a certain degree of sample sputtering. The latter effect, on the one hand, prevents assessing the resolution on the basis of a single image, and, on the other hand, limits the probing time and thus the signal-to-noise ratio that can be obtained. In order to fully simulate what happens when energetic ions impact on a sample, a Monte Carlo approach is often used. In this paper, a different approach is proposed. The contrast is simulated using curves of secondary electron yields versus the incidence angle of the beam, while the surface modification prediction is based on similar curves for the sputtering yield. Finally, Poisson noise from primary ions and secondary electrons is added to the image. It is shown that the evaluation of an ion imaging tool cannot be condensed in a single number, like the spot size or the edge steepness, but must be based on a more complex analysis taking into account at least three parameters: sputtering, contrast and signal-to-noise ratio. It is also pointed out that noise contributions from the detector cannot be neglected for they can actually be the limiting factor in imaging with focused ion beams. While providing already good agreement with experimental data in some imaging aspects, the proposed approach is highly modular. Further effects, like edge enhancement and detection, can be added separately.
    Ultramicroscopy 04/2011; 111(8):982-94. DOI:10.1016/j.ultramic.2011.03.019 · 2.44 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: It is often suggested that the growth in focused electron beam induced processing (FEBIP) is caused not only by primary electrons, but also (and even predominantly) by secondary electrons (SEs). If that is true, the growth rate for FEBIP can be changed by modifying the SE yield. Results from our Monte Carlo simulations show that the SE yield changes strongly with substrate thickness for thicknesses below the SE escape depth. However, our experimental results show that the growth rate is independent of the substrate thickness. Deposits with an average size of about 3 nm were written on 1 and 9 nm thick carbon substrates. The apparent contradiction between simulation and experiment is explained by simulating the SE emission from a carbon substrate with platinum deposits on the surface. It appears that the SE emission is dominated by the deposits rather than the carbon substrate, even for deposits as small as 0.32 nm(3).
    Nanotechnology 03/2011; 22(11):115303. DOI:10.1088/0957-4484/22/11/115303 · 3.82 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Electron beam induced deposition (EBID) is a direct-write lithographic technique that utilizes the dissociation of volatile precursors by a focused electron beam in a low vacuum environment to create nanostructures. Notable advantages of EBID over competing lithographic techniques are that it is a single step process that allows three-dimensional free-standing structures to be created, including features with single-nanometer scale dimensions. However, despite the inherent advantages of EBID, scientific and technological issues are impeding its development as an industrial nanofabrication tool. Perhaps the greatest single limitation of EBID is that metal-containing nanostructures deposited from organometallic precursors typically possess unacceptable levels of organic contamination which adversely affects the material's properties. In addition to the issue of purity, there is also a lack of understanding and quantitative information on the fundamental surface reactions and reaction cross-sections that are responsible for EBID. In this prospective, we describe how surface analytical techniques have begun to provide mechanistic and kinetic insights into the molecular level processes associated with EBID. This has been achieved by observing the effect of electron irradiation on nanometer thick films of organometallic precursors adsorbed onto solid substrates at low temperatures (
    Surface Science 02/2011; 605(3):257-266. DOI:10.1016/j.susc.2010.10.035 · 1.93 Impact Factor
  • B. Cook · T. Verduin · C. W. Hagen · P. Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Emission theory predicts that high brightness cold field emitters can enhance imaging in the electron microscope. This (neglecting chromatic aberration) is because of the large (coherent) probe current available from a high brightness source and is based on theoretically determined values of reduced brightnesses up to 10<sup>14</sup> A /( m <sup>2</sup>  sr   V ) . However, in their analysis, the authors find that statistical Coulomb interactions limit the reduced brightness of even atomically sharp cold field emitters to 10<sup>11</sup> A /( m <sup>2</sup>  sr   V ) and regular tungsten cold field emitters to around 2×10<sup>8</sup> A /( m <sup>2</sup>  sr   V ) . The authors also find that for tip radii in the range from 5 nm to 1 μ m , cold field emitters do not outperform larger Schottky (thermal field) emitters. Although this is applied to only one geometry, they expect that similar results will occur for most other cases due to a distinct difference in the behavior of different beam regimes.
    Journal of vacuum science & technology. B, Microelectronics and nanometer structures: processing, measurement, and phenomena: an official journal of the American Vacuum Society 12/2010; 28(6-28):C6C74 - C6C79. DOI:10.1116/1.3502642 · 1.46 Impact Factor
  • A. Mohammadi-Gheidari · C.W. Hagen · P. Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The authors present the first results obtained with their multibeam scanning electron microscope. For the first time, they were able to image 196 (array of 14×14) focused beams of a multielectron beam source on a specimen using single beam scanning electron microscope (SEM) optics. The system consists of an FEI Novanano 200 SEM optics column equipped with a multielectron beam source module. The source module consists of the multibeam source and an accelerator lens. In the multibeam source, the wide angle beam of a high brightness Schottky source is divided into 196 beamlets and focused by an aperture lens array. The accelerator lens is positioned on the image plane of the multibeam source to direct the beams toward the SEM column. The array of source images is further imaged by the SEM magnetic lenses, and the beam opening angle is defined at the variable aperture of the SEM. The system is designed to deliver 14×14 arrays of beamlets with a minimum probe size of 1 nm. In this article, the performance of the system is examined for a fixed magnification case.
    Journal of vacuum science & technology. B, Microelectronics and nanometer structures: processing, measurement, and phenomena: an official journal of the American Vacuum Society 11/2010; 28(6):C6G5-C6G10. DOI:10.1116/1.3498749 · 1.46 Impact Factor
  • R.H. van Aken · D.J. Maas · C.W. Hagen · J.E. Barth · P Kruit ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The low-voltage foil corrector is a novel type of foil aberration corrector that can correct for both the spherical and chromatic aberration simultaneously. In order to give a realistic example of the capabilities of this corrector, a design for a low-voltage scanning electron microscope with the low-voltage foil corrector is presented. A fully electrostatic column has been designed and characterised by using aberration integrals and ray tracing calculations. The amount of aberration correction can be adjusted relatively easy. The third order spherical and the first order chromatic aberration can be completely cancelled. In the zero current limit, a FW50 probe size of 1.0 nm at 1 kV can be obtained. This probe size is mainly limited by diffraction and by the fifth order spherical aberration.
    Ultramicroscopy 10/2010; 110(11):1411-9. DOI:10.1016/j.ultramic.2010.07.012 · 2.44 Impact Factor