ACS Publications. Most Trusted. Most Cited. Most Read
The Role of 11B4C Interlayers in Enhancing Fe/Si Multilayer Performance for Polarized Neutron Mirrors
More

OR SEARCH CITATIONS

My Activity
Publications

Figure 1Loading Img
Download Hi-Res ImageDownload to MS-PowerPointCite This:J. Phys. Chem. C 2025, 129, 16, 7921-7930
  • Open Access
C: Physical Properties of Materials and Interfaces

The Role of 11B4C Interlayers in Enhancing Fe/Si Multilayer Performance for Polarized Neutron Mirrors
Click to copy article linkArticle link copied!

  • Anton Zubayer*
    Anton Zubayer
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    *Email: [email protected]
    More by Anton Zubayer
    Orcidhttps://orcid.org/0000-0002-5327-609X
  • Fredrik Eriksson
    Fredrik Eriksson
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Fredrik Eriksson
    Orcidhttps://orcid.org/0000-0003-2982-5914
  • Martin Falk
    Martin Falk
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Martin Falk
  • Marcus Lorentzon
    Marcus Lorentzon
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Marcus Lorentzon
    Orcidhttps://orcid.org/0000-0002-3428-5847
  • Justinas Palisaitis
    Justinas Palisaitis
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Justinas Palisaitis
  • Christine Klauser
    Christine Klauser
    PSI Center for Neutron and Muon Sciences, Villigen PSI 5232, Switzerland
    More by Christine Klauser
  • Gyula Nagy
    Gyula Nagy
    Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    More by Gyula Nagy
    Orcidhttps://orcid.org/0000-0003-3172-5736
  • Philipp M. Wolf
    Philipp M. Wolf
    Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    More by Philipp M. Wolf
    Orcidhttps://orcid.org/0000-0002-3555-5503
  • Eduardo Pitthan
    Eduardo Pitthan
    Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    More by Eduardo Pitthan
  • Radek Holeňák
    Radek Holeňák
    Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    More by Radek Holeňák
  • Daniel Primetzhofer
    Daniel Primetzhofer
    Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    More by Daniel Primetzhofer
    Orcidhttps://orcid.org/0000-0002-5815-3742
  • Gavin B.G. Stenning
    Gavin B.G. Stenning
    Rutherford Appleton Laboratory, ISIS Neutron and Muon Source, Didcot OX11 0QX, United Kingdom
    More by Gavin B.G. Stenning
  • Artur Glavic
    Artur Glavic
    PSI Center for Neutron and Muon Sciences, Villigen PSI 5232, Switzerland
    More by Artur Glavic
    Orcidhttps://orcid.org/0000-0003-4951-235X
  • Jochen Stahn
    Jochen Stahn
    PSI Center for Neutron and Muon Sciences, Villigen PSI 5232, Switzerland
    More by Jochen Stahn
  • Samira Dorri
    Samira Dorri
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Samira Dorri
    Orcidhttps://orcid.org/0000-0002-3630-8176
  • Per Eklund
    Per Eklund
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    More by Per Eklund
    Orcidhttps://orcid.org/0000-0003-1785-0864
  • Jens Birch
    Jens Birch
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Jens Birch
    Orcidhttps://orcid.org/0000-0002-8469-5983
  • Naureen Ghafoor
    Naureen Ghafoor
    Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    More by Naureen Ghafoor
Open PDFSupporting Information (1)

The Journal of Physical Chemistry C

Cite this: J. Phys. Chem. C 2025, 129, 16, 7921–7930
Click to copy citationCitation copied!
https://doi.org/10.1021/acs.jpcc.5c00068
Published April 15, 2025

Copyright © 2025 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY 4.0 .

Abstract

Click to copy section linkSection link copied!
High Resolution Image
Download MS PowerPoint Slide

This study investigates the effects of incorporating 11B4C interlayers into Fe/Si multilayers, with a focus on interface quality, reflectivity, polarization, and magnetic properties for polarizing neutron optics. It is found that the introduction of 1–2 Å 11B4C interlayers significantly improves the interface sharpness, reducing interface width and preventing excessive Si diffusion into the Fe layers. X-ray reflectivity and polarized neutron reflectivity measurements show enhanced reflectivity and polarization, with a notable increase in polarization for 30 Å period multilayers. The inclusion of interlayers also helps prevent the formation of iron-silicides, improving both the magnetic properties and neutron optical performance. However, the impact of interlayers is less pronounced in thicker-period multilayers (100 Å), primarily due to the ratio between layer and interface widths. These results suggest that 11B4C interlayers offer a promising route for optimizing Fe/Si multilayer performance in polarizing neutron mirrors.

This publication is licensed under

CC-BY 4.0 .
  • cc licence
  • by licence
Copyright © 2025 The Authors. Published by American Chemical Society

Subjects

what are subjects

1. Introduction

Click to copy section linkSection link copied!
Neutron scattering is a powerful technique for studying atomic and molecular structures in advanced materials research, (1) and polarized neutron scattering has the capability to investigate structural and magnetic properties through neutron spin polarization. (2,3) Enhanced neutron optics, capable of increasing the neutron flux at higher angles of reflection with thinner bilayer periods, enable novel types of optics. (4−7) A highly polarized neutron flux is preferable for any neutron scattering instrument, to decrease the measurement time and/or increase the statistics. The Fe/Si multilayer is the state-of-the-art materials system for neutron polarizer and analyzer optics used at neutron scattering beamlines.
Multilayer mirror optics consist of alternating layers of two materials, designed to produce constructive interference in accordance with Bragg’s modified law. The reflected intensity depends on several factors, including the contrast in scattering length density (SLD) between the layers, the number of periods/repetitions (N), and the interface width. Short interface widths, high SLD contrast and high number of periods are three factors yielding higher reflectivity. For multilayers with shorter period thicknesses, required for high-angle neutron reflection, maintaining excellent interface quality becomes increasingly critical. In this regard, the Fe/Si multilayer system faces specific challenges due to interface broadening caused by a solid-state reaction during growth. This reaction leads to the formation of iron silicides at the Fe–Si interfaces, resulting in interface widths as large as 8–10 Å. (8) Such broadening reduces the system’s reflectivity by creating a gradual rather than sharp SLD transition between layers. Although increasing the number of periods (N) can partially compensate for the reduced reflectivity, it does not mitigate the detrimental effect of magnetically ″dead″ silicide regions on neutron polarization. These nonmagnetic zones at the interfaces reduce the polarizing performance of the multilayer and can not be overcome by simply adding more periods.
In neutron optics, the m-value characterizes the performance of a supermirror, representing the ratio between its maximum reflection angle and the critical angle of total reflection for nickel. Higher m-values correspond to broader angular ranges for neutron reflection. Reduced interface widths have been achieved using ion beam sputtering (IBS), enabling the fabrication of supermirrors up to m = 3.9. (4) However, IBS is considerably more expensive than DC magnetron sputtering (DCMS), which remains the standard for producing state-of-the-art Fe/Si multilayers with m-values up to 5.5. (9) Reactive sputtering of Si in a nitrogen atmosphere has also been explored as a strategy for interface control, but it requires careful optimization depending on the supermirror’s layer geometry. (10)
In our previous work, we addressed these challenges by cosputtering 11B4C into Fe and Si layers, effectively preventing silicide-nanocrystal formation. (11) This resulted in enhanced reflectivity and polarization, magnetically soft Fe layers, and reduced diffuse scattering, even for bilayer periods as thin as 15 Å. (11) While incorporation of 11B4C can solve many critical issues with Fe/Si multilayers, (11) the cosputtering method has significant challenges in terms of uniformity over a large area, particularly when applied on large scale industrial deposition systems. These systems in most cases require modifications to sputtering system geometry or sputtering targets, making cosputtering less feasible for large-scale manufacturing with current industrial setups. Therefore, a more feasible alternative solution is to add the 11B4C as interlayers or barrier layers. Natural B4C interlayers have previously been used for improved X-ray optics. (12) The choice of 11B4C is based on its ability to form an amorphous alloy at the interfaces due to the B-metal bonding, in contrast to the amorphization of entire layers observed in cosputtering. Further, if prevention of iron-silicide is successful, the SLD profile will avoid the almost nanometer-sized intermediary nuclear and magnetic step. Unlike cosputtering, which dilutes Fe and Si and reduces SLD contrast, the use of 11B4C interlayers preserves high SLD contrast and should maintain a high reflectivity.
Here, we investigate the effect of varying 11B4C interlayer thicknesses on the performance of Fe/Si multilayers. Since the interlayer itself can be seen as part of the interface width, the introduced interlayer’s thickness should not exceed the interface width for state-of-the-art Fe/Si, which can be up to 8.6 Å. (5) In the other extreme, the minimum interlayer thickness is constrained by the size of the B atom to about 1 Å. To account for this, we have chosen to investigate interlayer thicknesses of 2 Å for thicker periods (100 Å), 1 Å for the thinnest periods (15 Å), and both 1 and 2 Å for intermediate period thicknesses (30 Å). Our analyses focus on reflectivity, polarization, and the impact on magnetization from the addition of interlayers. This approach offers a practical comparison between traditional Fe/Si multilayers and those utilizing interlayers, particularly for industrial sputtering systems lacking cosputtering capabilities. The findings aim to show that interlayer-based solutions are an accessible and effective alternative to cosputtering.

2. Experimental Details

Click to copy section linkSection link copied!
Ion-assisted magnetron sputter deposition in a high vacuum environment (approximately 5.6·10–5 Pa or 4.2·10–7 Torr) was utilized to deposit Fe/Si and Fe/11B4C/Si/11B4C multilayer thin films. The deposition system is described in detail elsewhere. (13) The multilayers were grown onto 001-oriented single-crystalline Si substrates measuring 10 × 10 × 1 mm3 in size, with a native surface oxide layer. Sputtering targets were Fe (99.95% pure, 75 mm diameter), Si (99.95% pure, 75 mm diameter), and 99.8% chemically pure, > 90% isotopic purity 11B4C target with 50 mm diameter. Continuous operation of magnetrons during deposition, along with computer-controlled shutters for each target material, allowed precise control of atom fluxes, enabling multilayer deposition. The deposition rates for Fe and Si were both around 0.4 Å/s, while the rate for 11B4C was 0.08 Å/s. The 11B4C rate was calculated from the period thicknesses of multilayers with and without interlayers, as determined from X-ray reflectivity fitting. The substrates were kept at ambient temperature during deposition and rotated at 8 rpm to ensure even thickness distribution. For the first 3 Å of each layer, the substrate was kept at a floating potential followed by a −30 V substrate bias for the remaining thickness. The plasma was condensed toward the substrate by a magnetic field aligned with the substrate normal by a coil, enhancing the Ar-ion flux to the growing film. (13) The design parameters for all the deposited films are listed in Table 1.
Table 1. Summary of Parameters for All Multilayers Investigated in This Studya
Multilayers11B4C layer thickness [Å]Period thickness (Λ) [Å]Fe layer thickness/Si layer thickness [Å]Number of periods (N)
Multilayers with varying 11B4C interlayer thicknesses, periods, and total number of periods0 and 210050/50, 48/4820
0, 1 and 23015/15, 14/14, 13/1320
0 and 1157.5/7.5, 6.5/6.520
0 and 1157.5/7.5, 6.5/6.540
0 and 1157.5/7.5, 6.5/6.580
0 and 11511.25/3.75, 10.25/2.7520
a

The listed layer thicknesses represent nominal values.

X-ray reflectivity experiments were performed using a Malvern Panalytical Empyrean diffractometer equipped with Cu-Ka radiation and a PIXcel detector. A Göbel mirror, along with a 0.5° divergence slit, was implemented on the incident beam side, while a parallel beam collimator and a 0.27° collimator slit were utilized on the diffracted beam side. Reflectivity data from X-rays were analyzed to calculate the multilayer period, the thickness of individual layers, and interface roughness, using the GenX3 software. (14)
Additionally, X-ray diffraction measurements were carried out with a Panalytical X’Pert diffractometer, using Bragg–Brentano geometry, scanning over a 2θ range from 20° to 90°. For enhanced clarity, the analysis primarily focused on the 39.5° to 52° region.
The magnetic properties of the samples were assessed at room temperature using vibrating sample magnetometry (VSM) in a longitudinal setup, with measurements performed across a magnetic field range of −25 mT to 25 mT. Polarized neutron reflectometry (PNR) experiments were performed using the Morpheus instrument at the Swiss Spallation and Neutron Source (SINQ) located at the Paul Scherrer Institute (PSI) in Switzerland. PNR is sensitive to the spin-dependent scattering length density (SLD) of the sample, thus providing insights into the magnetization profile. The two spin states produce distinct reflectivity curves, with Bragg peaks occurring due to constructive interference. In these experiments, a polarized beam of neutrons was directed at small incidence angles (θ) toward the sample, reflecting off of the sample before detection by a He-3 detector. Measurements were conducted with the samples in an external magnetic field of approximately 20 mT, covering a 2θ range of 0° to 15°, using neutrons with a wavelength of 4.825 Å.
Transmission electron microscopy (TEM) cross-sectional samples were prepared using conventional mechanical polishing followed by argon ion etching at 5 keV, with a final etching step at 2 keV to remove surface damage. High-angle annular dark-field scanning transmission electron microscopy (HAADF STEM) at atomic resolution was performed using a double Cs-corrected Titan (3) 60–300 microscope in Linköping, operated at 300 keV. TEM images and selected area electron diffraction (SAED) patterns were acquired using a FEI Tecnai G2 microscope operated at 200 keV.
Ion beam analysis was carried out at the Tandem Laboratory at Uppsala University. (15) Time-of-Flight Elastic Recoil Detection Analysis (ToF-ERDA) measurements was performed at the 5 MV Pelletron accelerator utilizing with a 36 MeV I8+ beam at an incident angle of 67.5° and a recoil detection angle of 45.0°. The energy of the recoiled particles was measured using a gas ionization chamber, while C foils were employed to record the time-of-flight. Nuclear Resonance Analysis (NRA) was performed using the 350 keV ion implanter, employing the 11B(p,α)8Be resonance at 163 keV to investigate the amount of 11B in the samples. (16) All NRA measurements were performed at an incident angle of 60°. Additionally, Time-of-Flight Medium Energy Ion Scattering (ToF-MEIS) measurements were carried out with a 120 keV He+ beam, recorded at a scattering angle of 160° and an incidence angle of 0°.

3. Results

Click to copy section linkSection link copied!

3.1. X-Ray Reflectivity, XRR

X-ray reflectivity (XRR) measurements were performed on all multilayers investigated in this study, and the resulting profiles were fitted using the GenX software. Figure 1(a) compares the XRR profiles of multilayers with a period thickness of 100 Å, both without and with 2 Å 11B4C interlayers. Without interlayers, the Fe/Si multilayer exhibits 10 Bragg peaks, whereas the addition of interlayers increases the number to 13, extending to a scattering angle of 2θ = 10°. Notably, the odd-order Bragg peaks are more intense in the interlayered sample. The inset in Figure 1(a) presents the corresponding fitted SLD depth profiles. These reveal that the Si-on-Fe interface width increases slightly from 5 Å to 6 Å with the addition of interlayers, while the Fe-on-Si interface width is significantly reduced from 13.3 Å to 7.6 Å. This results in a more homogeneous and symmetric SLD profile for the interlayered multilayer.

Figure 1

Figure 1. X-ray reflectivity (XRR) data normalized to the respective critical edges, shown alongside simulated interface widths (σ) for the Si-on-Fe and Fe-on-Si interfaces. Data are presented for multilayers without 11B4C interlayers (black), with 1 Å 11B4C interlayers (red), and with 2 Å 11B4C interlayers (blue), for the following cases:

(a) Λ = 100 Å multilayers with N = 20. The inset shows the corresponding scattering length density (SLD) depth profiles. (b) Λ = 30 Å multilayers with N = 20. The inset shows the first Bragg peak region on a linear scale.

(c) Λ = 15 Å multilayers with N = 20, 40, and 80, vertically offset for clarity.

High Resolution Image
Download MS PowerPoint Slide
Figure 1(b) presents the XRR comparison for multilayers with a nominal period thickness of Λ = 30 Å. The inset shows the first Bragg peak plotted on a linear scale. Higher intensities are observed for the multilayers with 1 Å and 2 Å interlayers, even when accounting for scattering attenuation at higher angles due to the Debye–Waller factor. Simulations indicate that the Si-on-Fe interface widths are similar across all samples, ranging from σ = 6.0–6.2 Å. This is consistent with the findings for the Λ = 100 Å multilayers in Figure 1(a), where the addition of interlayers has only a minor effect on this interface. However, a significant reduction is observed for the Fe-on-Si interface width: from 8.2 Å (no interlayer) to 6.1 Å (1 Å interlayer), and further down to 5.2 Å (2 Å interlayer). Since the Si-on-Fe interface width slightly increases for the 2 Å interlayer sample, we consider the optimum interlayer thickness to lie between 1 Å and 2 Å.
For Λ = 15 Å multilayers, where the individual layer thicknesses (∼7.5 Å) are comparable to the expected interface widths of Fe/Si, samples without interlayers and with 1 Å interlayers were investigated. Different numbers of periods (N = 20, 40, 80) were used to enhance overall reflectivity and evaluate the effects of accumulated roughness in these ultrathin stacks. Figure 1(c) shows the XRR profiles for these samples, vertically shifted for clarity. The first-order Bragg peaks for interlayered samples appear at higher angles than those for pure Fe/Si, indicating slightly thinner actual periods than nominally intended. This also confirms that, despite the Debye–Waller attenuation with increasing angle, the reflected intensity remains higher for samples containing 11B4C interlayers than for the pure Fe/Si multilayers.
Additionally, pronounced Kiessig fringes are visible in the interlayered samples, highlighting smoother and more well-defined interfaces. Simulations for the N = 80 samples support this, yielding average interface widths of 9.5 Å for Fe/Si and 5.2 Å for the interlayered sample. Since in the Fe/Si multilayers the interface width of at least one interface often exceeds the nominal layer thickness, it is likely that no distinct Si layers are present. Instead, the entire nonmagnetic layer appears intermixed with Fe. As such, the reported interface widths represent averages between the Si-on-Fe and Fe-on-Si interfaces.

3.2. TEM, STEM, and SAED

To further investigate the distribution and effect of 11B4C interlayers, large-period multilayers with a period of Λ = 100 Å (N = 20) were analyzed using analytical transmission electron microscopy (TEM). Cross-sectional samples from two multilayers, Fe/Si and Fe/Si with 2 Å 11B4C interlayers, were examined, as shown in Figure 2. High-resolution TEM images [Figures 2(a) and 2(b)] reveal a more distinct alternating structure in the pure Fe/Si multilayer, with Fe-rich layers appearing darker and Si-rich layers brighter. In contrast, the interlayered sample exhibits more diffuse interfaces and thinner bright regions. Additionally, the Fe-rich layers in the Fe/Si sample appear more textured, whereas those in the interlayered sample show reduced texture. The Si-rich layers in both multilayers appear amorphous. Selected area electron diffraction (SAED) patterns confirm the textured nature of the Fe-rich layers in the Fe/Si multilayer. However, no clear distinction is observed between reflections from Fe and Fe-silicides, likely due to the overlap of two body-centered cubic (bcc) phases. These results are consistent with X-ray diffraction (XRD) data (see Supporting Information), which also indicate the presence of superlattice fringes─particularly around the 110 reflections, suggesting a crystalline structure in the Fe-rich layers. High-angle annular dark-field scanning TEM (HAADF-STEM) images support the TEM findings, showing sharper interfaces in the Fe/Si sample and more diffuse interfaces in the interlayered sample. In both samples, the Fe-containing layers appear bright in HAADF-STEM, with a noticeable bright core and relatively flat interfaces. Although the interlayered sample shows diffuse contrast at the interfaces in HAADF-STEM, elemental mapping provides a more nuanced picture. Energy-dispersive X-ray spectroscopy (EDX, not shown) and electron energy loss spectroscopy (EELS) confirm that the Fe and Si interfaces remain chemically well-defined. EELS maps, shown in Figures 2(g) and 2(h), clearly indicate a central Fe-rich core in each layer and demonstrate reduced Si diffusion into Fe in the interlayered sample compared to the pure Fe/Si multilayer. EDX analysis further confirms these findings, showing compositional modulations of Fe, Si, and Ar in both samples. Since EDX results are consistent with the EELS data, they are not presented in the figure.

Figure 2

Figure 2. TEM/STEM characterization of Λ = 100 Å Fe/Si and Fe/Si + 2 Å 11B4C interlayers, respectively: HR-TEM images (a and b), HAADF-STEM images (c and d), SAED patterns of the films obtained along the <011> zone axis of the Si(001) substrate (e and f), EELS elemental maps of Fe, Si, and Ar and corresponding HAADF-STEM images (g and h).

High Resolution Image
Download MS PowerPoint Slide
An additional observation from the EDX and EELS maps reveal an unexpected compositional modulation of argon, the sputtering gas used during deposition. Argon atoms appear to be trapped within the Si layers, with a higher concentration localized toward the center of the Si layers in the interlayered sample. This may be attributed to Si self-confinement resulting from reduced diffusion. This finding is reported here for the first time, and further investigation will be conducted in future studies, as it could potentially influence the scattering length density (SLD) contrast in mirror applications.
No signals for the lighter elements, boron and carbon, were detected in the EELS and EDX maps, likely due to their low concentration. However, EELS analysis indicates that the interlayers reduce Si diffusion into the Fe layers, resulting in relatively sharper and more abrupt interfaces.

3.3. Ion Beam Analysis, IBA

XRR analysis revealed that the introduction of interlayers effectively reduces interface width in the multilayers. However, analytical TEM did not provide any direct evidence of the presence of 11B4C layers or their role in achieving smooth or abrupt interfaces. On the contrary, interfaces appear rougher in the TEM images, but with a sharper contrast between Fe-rich and Si-rich layers, suggesting that 11B4C hinders the intermixing to some extent. To investigate the presence of 11B4C and further study the silicide formation, we investigated multilayers with a period of Λ = 100 Å (N = 20) using multiple ion beam analysis methods. As shown in Figure 3 (a-b), ToF-ERDA determined the average film composition, revealing that 2 at. % of the Fe/Si + 2 Å 11B4C multilayer consisted of 11B + C. Due to limited depth resolution, individual layers could not be resolved in the profiles. Additionally, about 2 at. % of Ar was detected in the films, supporting the TEM results and indicating that Ar is trapped during the deposition process rather than being introduced during the TEM sample preparation.

Figure 3

Figure 3. Ion beam analysis of Λ = 100 Å Fe/Si multilayers without and with 2 Å 11B4C interlayers. ERDA results of multilayers with (a) and without (b) interlayers. (c) Depth profile of 11B obtained via NRA. (d-e) ToF-MEIS data along with corresponding simulations of multilayers with and without interlayers, respectively.

High Resolution Image
Download MS PowerPoint Slide
For the Λ = 100 Å period multilayer, the 2 Å thick 11B4C interlayers correspond to 11B + C is 2 at. % of the multilayer. In comparison, for Λ = 30 Å, the same 11B4C interlayers correspond to ∼ 7 at. % of the multilayer, which is roughly half of the ∼ 15 at. % of 11B + C observed for cosputtered Fe/Si + 11B4C multilayers in our previous work. A clear advantage of 11B4C interlayers over cosputtered 11B4C multilayers is therefore the lower amount of 11B4C, thus maintaining the high SLD contrast for spin-up neutrons.
Because of the sensitivity and selectivity to 11B of the Nuclear Reaction Analysis (NRA), we performed a 11B depth profile to investigate the boron composition modulation in the multilayers, as shown in Figure 3(c). Using the 163 keV energy of resonance (width = 6 keV) and considering the stopping power of protons in Si and Fe at 170 keV with an incident angle of 60°, the depth resolution was estimated to be in the range 13–30 nm. Although boron was detected, the limited depth resolution prevented the identification of clear depth features.
In addition to boron detection, we also attempted to investigate potential differences in silicide thickness between samples with and without interlayers using TOF-MEIS. (16,17) The measurements were performed at a 0° incidence angle with a depth resolution of approximately 7 nm for areal density determination, and at a 70° incidence angle, providing higher depth resolution of roughly 1.7 nm, however above the expected silicide layers. ToF-MEIS measurements revealed consistent layer thicknesses across the ∼ 2 mm2 measurement area. Since ToF-MEIS is sensitive to differences in the thickness of single layers, we can conclude that the consistent layer thickness also applies to the topmost layer. The areal densities of Fe and Si from ToF-MEIS are presented in Table 2. For Si, the uncertainty in areal density is influenced by the stopping cross section uncertainties of both Si and Fe, the scattering potential, statistical variations from total counts, and the nominal 11B4C values from deposition. In total the Si areal density uncertainty is estimated to 5% for the areal density for the Fe/Si sample and 7% for the Fe/Si + 2 Å 11B4C sample. For Fe areal densities uncertainties are lower, about 4%, since the signal of He ions scattered on the first Fe layer is directly visible in the spectrum. The main sources of uncertainties are the scattering potential and the measurement statistics. The measurements indicate a decrease in Fe areal density with increasing 11B4C, while the Si areal density remains unchanged within the error limits. However, due to limited depth resolution, it was not possible to estimate the silicide thickness for either samples.
Table 2. Areal Densities Based on ToF-MEIS Experiments Performed with 120 keV He+, a Scattering Angle of 160° and an Incident Angle of 0° Degreea
 Fe/SiFe/Si + 2Å 11B4C
Si39.0 ± 2.038.5 ± 2.7
Fe46.5 ± 1.941.1 ± 1.6
11B4C02.6 ± 0.2
a

The areal densities of Si and Fe were extracted using SIMNRA simulations. The areal density of 11B4C was assumed based on the nominal values as an input parameter for the simulations. All values are in units of 1015 atoms/cm2.

3.4. Vibrating Sample Magnetometry, VSM

Figure 4(a) presents the magnetization as a function of the external magnetic field for Fe/Si multilayers with a thickness of Λ = 100 Å (N = 20). Data are presented for both Fe/Si (black) and Fe/Si with 2 Å 11B4C interlayers (blue). As expected, the magnetization amplitude for the interlayered sample is slightly reduced, due to the lower overall Fe content compared to the Fe/Si multilayer. However, the coercivity curve of the interlayered sample deviates slightly from the typical ferromagnetic response. Specifically, the presence of additional features, resembling shoulders, suggests that certain magnetic domains or layers exhibit increased resistance to magnetization before reaching saturation. While the origin of this enhanced resistance, whether due to magnetic domain behavior or interlayer coupling, can not be conclusively determined from the current data, however it does not affect the main objective of the study or its potential applications. Notably, the saturation field remains similar for both samples, approximately 5 mT. The saturated magnetization values are 0.0177 emu for the Fe/Si multilayer and 0.0170 emu for the Fe/Si + 2 Å 11B4C sample.

Figure 4

Figure 4. Vibrating sample magnetometry (VSM) of Fe/Si and Fe/Si + 2 Å 11B4C interlayers, (a) Λ = 100 Å, (b) of Λ = 15 Å and (c) also with Λ = 15 Å but with a Fe thickness being 3 times thicker than the Si thickness. All samples have N = 20 periods.

High Resolution Image
Download MS PowerPoint Slide
Figure 4(b) presents the VSM results for samples with a period thickness of Λ = 15 Å and N = 20, incorporating 0 and 1 Å interlayers. The observed low coercivity is typical for these thin, amorphous multilayers, where a significant portion of the bilayer consists of silicide. The magnetization values for Fe/Si and Fe/Si + 1 Å 11B4C are 0.0013 and 0.0007 emu, respectively. When considering the total volume of the Fe layers, the Fe atom count in the Λ = 15 Å samples is approximately 6.67 times lower for Fe/Si and 7.38 times lower for Fe/Si + 1 Å 11B4C compared to the Λ = 100 Å samples shown in Figure 4(a). Scaling the magnetization values from Figure 4(a) (0.0177 emu for Fe/Si and 0.0170 emu for Fe/Si + 1 Å 11B4C) by these factors predict magnetization values of 0.0026 and 0.0022 emu, respectively, for Λ = 15 Å. These predictions are significantly higher than the experimentally measured values, which can be attributed to the large silicide fraction in the Λ = 15 Å multilayers. When the Fe layer thickness was increased by a factor of 3 relative to Si in the 15 Å period multilayers, the corresponding VSM data, shown in Figure 4(c), indicate saturated magnetizations of approximately 0.0042 emu for Fe/Si and 0.0039 emu for Fe/Si + 1 Å 11B4C. These values are in close agreement with the expected values of 0.0040 and 0.0035 emu, based on the magnetization scaling in the 100 Å period multilayers. This confirms that the lower magnetization observed in Figure 4(b) is due to the large nonmagnetic silicide fraction, which remains largely unaffected by the 1 Å interlayer.

3.5. Polarized Neutron Reflectivity, PNR

Figure 5(a-c) shows the spin-up PNR and corresponding fitted curves of Fe/Si multilayers with 0, 1, and 2 Å interlayers of 11B4C, multilayers with Λ = 30 Å and N = 20. Measurements were focused on the critical angle region and the first order Bragg peak due to the limited beam time. The 133% increase in reflectivity from 0 to 2 Å interlayers, indicates a significant reduction in interface width with the addition of 11B4C interlayers. The polarization at the Bragg peak position was calculated to be 53%, 79%, and 89% for 0, 1, and 2 Å interlayered samples, respectively, which highlight the effectiveness of 11B4C interlayers.

Figure 5

Figure 5. Polarized neutron reflectivity (PNR) measurements (a-c) of Fe/Si multilayers with 0, 1, and 2 Å interlayers of 11B4C, respectively, where all multilayers had Λ = 30 Å and N = 20. (d) shows the corresponding nuclear SLD profiles stemming from the fits in a-c. (e) shows the Bragg peak intensities of the first, second, and third order Bragg peaks for both spin-up (solid) and spin-down (dashed) neutron reflectivity for Λ = 100 Å and N = 20.

High Resolution Image
Download MS PowerPoint Slide
To further illustrate the difference between the samples, Figure 5(d) shows the corresponding nuclear SLD depth profile from 300 to 400 Å above the substrate, based on the fits shown in Figure 5(a-c). The SLD depth profiles reveal that the interlayered samples have higher oscillation amplitudes. Figure 5(e) shows the reflectivity value for the first, second, and third Bragg peak order for spin-up and spin-down for the two samples Fe/Si and Fe/Si + 2 Å 11B4C where Λ = 100 Å and N = 20. It shows that the Bragg peak intensities do not differ significantly in reflectivity and only show a slight improvement in polarization for all peaks for the 11B4C sample when it comes to Λ = 100 Å.

4. Discussion

Click to copy section linkSection link copied!
This study evaluates the effect of incorporating 11B4C interlayers into Fe/Si multilayers, focusing on interface sharpness, silicide suppression, and performance in neutron reflectivity and polarization.

4.1. Reflectivity and Polarization Enhancement

XRR analysis revealed enhanced reflectivity for all multilayers containing 11B4C interlayers, especially those with thinner periods, such as the 15 Å multilayers with individual layer thicknesses of approximately 7 Å. Based on the analysis of interface widths from the reflectivity data, the optimal interlayer thickness was found to be 2 Å for all multilayers. The observed increase in reflectivity with the number of periods (N = 20, 40, and 80) for the thinner-period multilayers (Λ = 15 Å) with 1 Å 11B4C interlayers demonstrates that the reflectivity gain scales with the number of periods. This also suggests that introducing 11B4C interlayers does not lead to accumulated roughness, a key requirement for high-performance supermirror fabrication, since a supermirror require several hundreds to thousands of layers. The improvement in XRR for the Λ = 15 Å multilayers with 11B4C interlayers is comparable to that observed in 11B4C cosputtered multilayers [see Supporting Information].
Consistent with the XRR results, polarized neutron reflectivity (PNR) measurements for 30 Å period multilayers (Figure 5) showed a clear increase in both reflectivity and polarization upon the addition of 1 Å and 2 Å 11B4C interlayers. Specifically, the polarization at the Bragg peak positions increased by 67% for the 2 Å 11B4C interlayer (Λ = 30 Å). Combined XRR and PNR results indicate that the interlayer approach enables reflection and polarization at q = 0.44 Å–1 (m = 16.5 for 15 Å period), which corresponds to a vector reflection approximately 3.5 times higher than that achieved by state-of-the-art polarized neutron optics.
While reflectivity and polarization improvements were more pronounced for thinner-period multilayers, the 100 Å multilayers exhibited more modest enhancements compared to the 11B4C cosputtered multilayers from previous studies. (11) This suggests that the advantages of the 11B4C interlayers are more significant for thinner periods, where the interface width and magnetic variations play a more significant role. Therefore, the inclusion of interlayers has a more substantial impact on the performance of thinner-period multilayers. The neutron reflectivity and polarization improvement for 2 Å interlayer samples is summarized in Table 3, where the dramatic improvement is mainly evident for thinner periods.
Table 3. Neutron Reflectivity and Polarization Improvement Using 2 Å 11B4C Interlayers for Λ = 30 and 100 Åa
Period (Å)Reflectivity improvement (%)Polarization improvement (%)
30133 ± 467 ± 3
1002.3 ± 0.11.5 ± 0.1
a

Improvement is the relative improvement compared to the equivalent Fe/Si sample.

4.2. Interface Width, Amorphization and Silicide Formation

Although the increased reflectivity and polarization observed with the introduction of 11B4C interlayers is clear, the underlying mechanisms responsible for these improvements remain a critical focus of investigation. We hypothesize that the 11B4C interlayer function primarily as a barrier, inhibiting the bonding and/or mixing of Fe with Si, thereby preventing the formation of iron-silicide phases or heavily mixed regions at the interfaces. This would contribute to the improvement of interface quality and performance. Additionally, the strong affinity of B for metals, particularly iron, contributes to amorphization of Fe layers at the interfaces, and in doing so, it may hinder the formation of Fe–Fe bonds, thus reducing nanocrystallites causing roughness at the interfaces. In the context of X-ray optics, the amorphization of the metal layers at the interfaces is beneficial, as it can improve interface quality with the same mechanism as seen here. This reduction in interface roughness is crucial for optimizing the performance of multilayer mirrors and improving their reflectivity and polarization capabilities.
The Bragg reflections observed in XRR and PNR measurements, along with reflectivity fitting for multilayers with various periods, confirm a reduction in interface width for all interlayered samples, including those with the thinner 15 Å periods. Reflectivity intensity is influenced by the scattering length density (SLD) contrast between the layers and the interface width, but in this case, the reflectivity gain cannot be attributed to changes in the SLD contrast between Fe and Si, as this remains unaffected by the interlayer. Instead, the improvement in reflectivity can be attributed solely to reduced interface widths, as confirmed by the fits. These indicate that, without an interlayer, the interface width between Fe and Si is significantly larger compared to the interlayered samples. SLD profiles (Figures 1(a) and 5(d)) show more abrupt SLD contrasts with 11B4C interlayers, supporting the reduction in interfacial roughness.
The effect of interlayers on interface width is found to be asymmetric for the two interfaces, Si-on-Fe and Fe-on-Si. Simulations revealed that only minor differences in the interface widths of Si-on-Fe interfaces with and without interlayers, while a significant reduction in the Fe-on-Si interface width was observed. For example, for a period of Λ = 30 Å, the Fe-on-Si interface width decreased from 8.2 Å to 6.1 Å with a 1 Å interlayer and further to 5.2 Å with a 2 Å interlayer. The larger interface width for Fe-on-Si interfaces in the absence of interlayers is attributed to enhanced diffusion and intermixing, driven by a combination of Fe-backscattered neutrals, surface energies, and phase-formation thermodynamics during deposition. Studies by Romano et al. (18) have shown that asymmetric magnetic and nonmagnetic silicide layers form at interfaces, with their thickness depending on composition and deposition conditions. In our study, ion-assisted magnetron sputter deposition, combined with 11B4C interlayers, affects surface energies and diffusion processes, influencing silicide formation. This appears to be particularly pronounced at the Fe-on-Si interface, suggesting that the interlayers play a critical role in reducing interfacial roughness by altering the diffusion dynamics and silicide formation mechanisms. (19)
While analytical TEM could not directly resolve interface asymmetries, STEM/EELS and EDX analyses revealed that interlayers reduced Si diffusion into Fe layers, resulting in relatively sharp interfaces. No boron or carbon signals were detected in these analyses, but ERDA confirmed the presence of 11B4C, with a 2 Å interlayer in 15 Å period multilayers corresponding to approximately 7 at. % 11B4C. This finding highlights an advantage of 11B4C interlayers over cosputtered 11B4C multilayers, as the lower 11B4C content minimizes the scattering length density (SLD) contrast dilution, thereby preserving reflectivity and polarization performance.
Despite extensive ion beam analysis to determine silicide layer thicknesses, poor resolution limited definitive conclusions about interlayer effects. ToF-MEIS measurements, however, indicated consistent layer thicknesses across a ∼ 2 mm2 measurement area for all multilayers, reflecting the overall quality and high uniformity of the multilayers.
For thicker periods of 100 Å, XRR showed a reduction in interface width by adding 2 Å interlayers, especially on the Fe-on-Si interface. STEM and EELS analysis suggested reduced Si diffusion into Fe layers, however, PNR measurements did not detect a significant difference in polarization compared to pure Fe/Si multilayers. In comparison, 11B4C cosputtering was shown to be advantageous for thinner as well as thicker periods. The key difference in the two designs is the amorphization of Fe and Fe silicide layers, which becomes prominent for thicker periods. For thicker periods, 2 Å interlayers reduce silicide formation and only amorphize the interface region, as seen in the STEM contrast images, but cannot overcome the roughness generated by highly textured Fe and Fe silicide nanocrystallites, and the interface width remains high compared to thinner periods. In contrast, in 11B4C cosputtered amorphous multilayers, the interface width does not depend on the multilayer period.

4.3. Effect on Saturation Magnetization and Coercivity

When investigating an alternative design to 11B4C cosputtered multilayers, i.e. by interlayers, it is essential to understand how the presence of interlayers affects magnetization and coercivity. For Fe/Si multilayers with a period of Λ = 100 Å, the saturation magnetization was measured at 0.0177 emu, while the interlayered sample with a 2 Å interlayer showed a value of 0.017 emu, corresponding to a 5.5% decrease, which is consistent with the reduced Fe content in the interlayered sample. The external field required for saturation is similar in both samples, around 5 mT. However, the hysteresis loop for the interlayered sample is less steep during magnetization reversal, though this does not impact practical performance, as polarizers and analyzers operate under a saturated external field. For thinner-period multilayers (15 Å), both with and without interlayers, the low magnetization suggests significant silicide formation. The 1 Å interlayer is insufficient to suppress this effect. To mitigate silicide formation and improve magnetic properties, a 2 Å interlayer combined with increasing the Fe layer thickness would yield better results in terms of magnetic properties, thereby improving polarization.

5. Conclusion

Click to copy section linkSection link copied!
In conclusion, incorporating 11B4C interlayers into Fe/Si multilayers significantly enhances interface quality by reducing interface width and preventing excessive Si diffusion into Fe. This results in improved neutron reflectivity and polarization, particularly for thinner-period multilayers, enabling high-angle neutron reflectivity beyond the capabilities of commercial Fe/Si supermirrors. The optimal interlayer thickness of 2 Å for all multilayers improve interface sharpness and prevent iron-silicide formation, which would otherwise degrade both magnetic properties and neutron optical performance. However, characterizing the elemental distribution at the interfaces remains challenging due to the limitations of current analytical techniques. While the interlayer approach yields significant improvements in reflectivity and polarization for thinner-period multilayers, as seen in Table 3, the impact is less pronounced for thicker periods (100 Å) due to roughness caused by Fe and silicide nanocrystallites. The inclusion of interlayers also results in a slight decrease in saturation magnetization, consistent with the reduced Fe content. The introduction of 11B4C interlayers, however, does not affect coercivity, which limits their potential for applications requiring low coercivity for polarized neutron mirrors. Overall, 11B4C interlayers offer a promising strategy for enhancing the performance of Fe/Si multilayers, particularly for polarized neutron optics and supermirror designs grown with current industrial systems, though further advancements in interface characterization are needed to fully understand the impact of interlayers.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcc.5c00068.

  • X-ray diffraction (XRD), X-ray reflectivity (XRR), and off-specular scattering measurements for multilayers with and without 11B4C interlayers; complementary samples grown in a different deposition system are also included to demonstrate reproducibility and to assess performance under conditions more representative of industrial fabrication; figures illustrating interface structure, crystallinity, and fit quality across various period thicknesses (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!
  • Corresponding Author
  • Authors
    • Fredrik Eriksson - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden;  Orcidhttps://orcid.org/0000-0003-2982-5914
    • Martin Falk - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    • Marcus Lorentzon - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden;  Orcidhttps://orcid.org/0000-0002-3428-5847
    • Justinas Palisaitis - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
    • Christine Klauser - PSI Center for Neutron and Muon Sciences, Villigen PSI 5232, Switzerland
    • Gyula Nagy - Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden;  Orcidhttps://orcid.org/0000-0003-3172-5736
    • Philipp M. Wolf - Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden;  Orcidhttps://orcid.org/0000-0002-3555-5503
    • Eduardo Pitthan - Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    • Radek Holeňák - Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden
    • Daniel Primetzhofer - Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden;  Orcidhttps://orcid.org/0000-0002-5815-3742
    • Gavin B.G. Stenning - Rutherford Appleton Laboratory, ISIS Neutron and Muon Source, Didcot OX11 0QX, United Kingdom
    • Artur Glavic - PSI Center for Neutron and Muon Sciences, Villigen PSI 5232, Switzerland;  Orcidhttps://orcid.org/0000-0003-4951-235X
    • Jochen Stahn - PSI Center for Neutron and Muon Sciences, Villigen PSI 5232, Switzerland
    • Samira Dorri - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden;  Orcidhttps://orcid.org/0000-0002-3630-8176
    • Per Eklund - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden;  Ångström Laboratory, Uppsala University, Box 538, Uppsala SE-751 21, Sweden;  Orcidhttps://orcid.org/0000-0003-1785-0864
    • Jens Birch - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden;  Orcidhttps://orcid.org/0000-0002-8469-5983
    • Naureen Ghafoor - Thin Film Physics Division, Department of Physics, Chemistry, and Biology (IFM), Linköping University, Linköping SE-581 83, Sweden
  • Notes
    The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

The authors gratefully acknowledge funding from the Swedish Government Strategic Research Area in Materials Science on Functional Materials at Linköping University (Faculty Grant SFO-Mat-LiU No. 2009 00971). Financial support was also provided by the Swedish Research Council (VR) through project grants 2019-04837 (F.E.), 2018-05190 (N.G.), and 2021-03826 (P.E.), as well as by the Hans Werthén Foundation (grant 2022-D-03 to A.Z.), the Royal Academy of Sciences Physics Grant PH2022-0029 (A.Z.), the Lars Hiertas Minne Foundation (grant FO2022-0273 to A.Z.), the Längmanska Kulturfonden (grant BA23-1664 to A.Z.), and the SNSS travel grant (M.F.). Additional support was provided by the Knut and Alice Wallenberg Foundation through the Wallenberg Academy Fellows program (KAW-2020.0196 to P.E.). This work was partly conducted on the Morpheus neutron reflectometer at the SINQ spallation source, Paul Scherrer Institute, Switzerland. We thank Materials Characterisation Laboratory at ISIS for their support. We also thank ARTEMI, the Swedish National Infrastructure for Advanced Electron Microscopy, for access to the TEM and assistance with TEM analysis, and Ingemar Persson for help analyzing EELS data.

References

Click to copy section linkSection link copied!

This article references 19 other publications.

  1. 1
    Willis, B. T. M.; Carlile, C. J. Experimental Neutron Scattering; Oxford University Press: 2009, 325.
    Google Scholar
    There is no corresponding record for this reference.
  2. 2
    Toperverg, B. P. Polarized Neutron Reflectometry of Magnetic Nanostructures. Phys. Met. Metallogr. 2015, 116 (13), 1337– 1375,  DOI: 10.1134/S0031918X15130025
    Google Scholar
    There is no corresponding record for this reference.
  3. 3
    Zhao, J. K.; Robertson, L.; Herwig, K.; Crabb, D. Polarized Neutron in Structural Biology – Present and Future Outlook. Phys. Procedia 2013, 42, 39– 45,  DOI: 10.1016/j.phpro.2013.03.173
    Google Scholar
    There is no corresponding record for this reference.
  4. 4
    Hino, M.; Hayashida, H.; Kitaguchi, M.; Kawabata, Y.; Takeda, M.; Maruyama, R.; Ebisawa, T.; Torikai, N.; Kume, T.; Tasaki, S. Development of Large-m Polarizing Neutron Supermirror Fabricated by Using Ion Beam Sputtering Instrument at KURRI. Phys. B Condens. Matter 2006, 385–386, 1187– 1189,  DOI: 10.1016/j.physb.2006.05.404
    Google Scholar
    There is no corresponding record for this reference.
  5. 5
    Kovács-Mezei, R.; Krist, T.; Révay, Z. Non-Magnetic Supermirrors Produced at Mirrotron Ltd.. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc. Equip. 2008, 586 (1), 51– 54,  DOI: 10.1016/j.nima.2007.11.034
    Google Scholar
    There is no corresponding record for this reference.
  6. 6
    Maruyama, R.; Yamazaki, D.; Ebisawa, T.; Hino, M.; Soyama, K. Development of Neutron Supermirrors with Large Critical Angle. Thin Solid Films 2007, 515 (14), 5704– 5706,  DOI: 10.1016/j.tsf.2006.12.022
    Google Scholar
    There is no corresponding record for this reference.
  7. 7
    Padiyath, J.; Stahn, J.; Allenspach, P.; Horisberger, M.; Böni, P. Influence of Mo in the Ni Sublayers on the Magnetization of Ni/Ti Neutron Supermirrors. Phys. B 2004, 350 (1–3 Suppl. 1), E237– E240,  DOI: 10.1016/j.physb.2004.03.059
    Google Scholar
    There is no corresponding record for this reference.
  8. 8
    Amir, S. M.; Gupta, M.; Gupta, A.; Ambika, K.; Stahn, J. Silicide Layer Formation in Evaporated and Sputtered Fe/Si Multilayers: X-Ray and Neutron Reflectivity Study. Appl. Surf. Sci. 2013, 277, 182– 185,  DOI: 10.1016/j.apsusc.2013.04.021
    Google Scholar
    There is no corresponding record for this reference.
  9. 9
    SwissNeutronics Neutron Supermirrors. https://www.swissneutronics.ch/products/neutron-supermirrors/ (accessed 2025–04–01).
    Google Scholar
    There is no corresponding record for this reference.
  10. 10
    Høghøj, P.; Anderson, I.; Siebrecht, R.; Graf, W.; Ben-Saidane, K. Neutron Polarizing Fe/Si Mirrors at ILL. Phys. B 1999, 267–268, 355– 359,  DOI: 10.1016/S0921-4526(99)00038-1
    Google Scholar
    There is no corresponding record for this reference.
  11. 11
    Zubayer, A.; Ghafoor, N.; Thórarinsdóttir, K. A.; Stendahl, S.; Glavic, A.; Stahn, J.; Nagy, G.; Greczynski, G.; Schwartzkopf, M.; Febvrier, A. L.; Eklund, P. Reflective, Polarizing, and Magnetically Soft Amorphous Neutron Optics with 11B Enriched B4C. Sci. Adv. 2024, 10 (7), eadl0402  DOI: 10.1126/sciadv.adl0402
    Google Scholar
    There is no corresponding record for this reference.
  12. 12
    Sakhonenkov, S. S.; Filatova, E. O. Effect of B4C Barrier Layer on Layer Intermixing in Nanoscale W/Be Multilayers before and after Annealing. Appl. Surf. Sci. 2022, 606, 154961  DOI: 10.1016/j.apsusc.2022.154961
    Google Scholar
    There is no corresponding record for this reference.
  13. 13
    Engström, C.; Berlind, T.; Birch, J.; Hultman, L.; Ivanov, I. P.; Kirkpatrick, S. R.; Rohde, S. Design, Plasma Studies, and Ion Assisted Thin Film Growth in an Unbalanced Dual Target Magnetron Sputtering System with a Solenoid Coil. Vacuum 2000, 56 (2), 107– 113,  DOI: 10.1016/S0042-207X(99)00177-3
    Google Scholar
    There is no corresponding record for this reference.
  14. 14
    Glavic, A.; Björck, M. GenX 3: The Latest Generation of an Established Tool. J. Appl. Crystallogr. 2022, 55 (4), 1063– 1071,  DOI: 10.1107/S1600576722006653
    Google Scholar
    There is no corresponding record for this reference.
  15. 15
    Ström, P.; Primetzhofer, D. Ion Beam Tools for Nondestructive In-Situ and in-Operando Composition Analysis and Modification of Materials at the Tandem Laboratory in Uppsala. J. Instrum. 2022, 17 (4), P04011  DOI: 10.1088/1748-0221/17/04/P04011
    Google Scholar
    There is no corresponding record for this reference.
  16. 16
    Corrêa, S. A.; Pitthan, E.; Moro, M. V.; Primetzhofer, D. A Multipurpose Set-up Using KeV Ions for Nuclear Reaction Analysis, High-Resolution Backscattering Spectrometry, Low-Energy PIXE and in-Situ Irradiation Experiments. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 2020, 478 (June), 104– 110,  DOI: 10.1016/j.nimb.2020.05.023
    Google Scholar
    There is no corresponding record for this reference.
  17. 17
    Tran, T. T.; Lavoie, C.; Zhang, Z.; Primetzhofer, D. In-Situ Nanoscale Characterization of Composition and Structure during Formation of Ultrathin Nickel Silicide. Appl. Surf. Sci. 2021, 536, 147781  DOI: 10.1016/j.apsusc.2020.147781
    Google Scholar
    There is no corresponding record for this reference.
  18. 18
    Badía-Romano, L.; Rubín, J.; Bartolomé, F.; Magén, C.; Bartolomé, J.; Varnakov, S. N.; Ovchinnikov, S. G.; Rubio-Zuazo, J.; Castro, G. R. Morphology of the Asymmetric Iron–Silicon Interfaces. J. Alloys Compd. 2015, 627, 136– 145,  DOI: 10.1016/j.jallcom.2014.12.019
    Google Scholar
    There is no corresponding record for this reference.
  19. 19
    Gupta, A.; Kumar, D.; Phatak, V. Asymmetric Diffusion at the Interfaces in Fe/Si Multilayers. Phys. Rev. B 2010, 81 (15), 155402  DOI: 10.1103/PhysRevB.81.155402
    Google Scholar
    There is no corresponding record for this reference.

Cited By

Click to copy section linkSection link copied!

This article has not yet been cited by other publications.

Download PDF
Go to The Journal of Physical Chemistry C
Get e-Alerts
Get e-Alerts

The Journal of Physical Chemistry C

Cite this: J. Phys. Chem. C 2025, 129, 16, 7921–7930
Click to copy citationCitation copied!
https://doi.org/10.1021/acs.jpcc.5c00068
Published April 15, 2025

Copyright © 2025 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY 4.0 .

Article Views

277

Altmetric

-

Citations

-
Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 1

    Figure 1. X-ray reflectivity (XRR) data normalized to the respective critical edges, shown alongside simulated interface widths (σ) for the Si-on-Fe and Fe-on-Si interfaces. Data are presented for multilayers without 11B4C interlayers (black), with 1 Å 11B4C interlayers (red), and with 2 Å 11B4C interlayers (blue), for the following cases:

    (a) Λ = 100 Å multilayers with N = 20. The inset shows the corresponding scattering length density (SLD) depth profiles. (b) Λ = 30 Å multilayers with N = 20. The inset shows the first Bragg peak region on a linear scale.

    (c) Λ = 15 Å multilayers with N = 20, 40, and 80, vertically offset for clarity.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 2

    Figure 2. TEM/STEM characterization of Λ = 100 Å Fe/Si and Fe/Si + 2 Å 11B4C interlayers, respectively: HR-TEM images (a and b), HAADF-STEM images (c and d), SAED patterns of the films obtained along the <011> zone axis of the Si(001) substrate (e and f), EELS elemental maps of Fe, Si, and Ar and corresponding HAADF-STEM images (g and h).

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 3

    Figure 3. Ion beam analysis of Λ = 100 Å Fe/Si multilayers without and with 2 Å 11B4C interlayers. ERDA results of multilayers with (a) and without (b) interlayers. (c) Depth profile of 11B obtained via NRA. (d-e) ToF-MEIS data along with corresponding simulations of multilayers with and without interlayers, respectively.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 4

    Figure 4. Vibrating sample magnetometry (VSM) of Fe/Si and Fe/Si + 2 Å 11B4C interlayers, (a) Λ = 100 Å, (b) of Λ = 15 Å and (c) also with Λ = 15 Å but with a Fe thickness being 3 times thicker than the Si thickness. All samples have N = 20 periods.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 5

    Figure 5. Polarized neutron reflectivity (PNR) measurements (a-c) of Fe/Si multilayers with 0, 1, and 2 Å interlayers of 11B4C, respectively, where all multilayers had Λ = 30 Å and N = 20. (d) shows the corresponding nuclear SLD profiles stemming from the fits in a-c. (e) shows the Bragg peak intensities of the first, second, and third order Bragg peaks for both spin-up (solid) and spin-down (dashed) neutron reflectivity for Λ = 100 Å and N = 20.

    High Resolution Image
    Download MS PowerPoint Slide

  • References

    This article references 19 other publications.

    1. 1
      Willis, B. T. M.; Carlile, C. J. Experimental Neutron Scattering; Oxford University Press: 2009, 325.
      There is no corresponding record for this reference.
    2. 2
      Toperverg, B. P. Polarized Neutron Reflectometry of Magnetic Nanostructures. Phys. Met. Metallogr. 2015, 116 (13), 1337– 1375,  DOI: 10.1134/S0031918X15130025
      There is no corresponding record for this reference.
    3. 3
      Zhao, J. K.; Robertson, L.; Herwig, K.; Crabb, D. Polarized Neutron in Structural Biology – Present and Future Outlook. Phys. Procedia 2013, 42, 39– 45,  DOI: 10.1016/j.phpro.2013.03.173
      There is no corresponding record for this reference.
    4. 4
      Hino, M.; Hayashida, H.; Kitaguchi, M.; Kawabata, Y.; Takeda, M.; Maruyama, R.; Ebisawa, T.; Torikai, N.; Kume, T.; Tasaki, S. Development of Large-m Polarizing Neutron Supermirror Fabricated by Using Ion Beam Sputtering Instrument at KURRI. Phys. B Condens. Matter 2006, 385–386, 1187– 1189,  DOI: 10.1016/j.physb.2006.05.404
      There is no corresponding record for this reference.
    5. 5
      Kovács-Mezei, R.; Krist, T.; Révay, Z. Non-Magnetic Supermirrors Produced at Mirrotron Ltd.. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc. Equip. 2008, 586 (1), 51– 54,  DOI: 10.1016/j.nima.2007.11.034
      There is no corresponding record for this reference.
    6. 6
      Maruyama, R.; Yamazaki, D.; Ebisawa, T.; Hino, M.; Soyama, K. Development of Neutron Supermirrors with Large Critical Angle. Thin Solid Films 2007, 515 (14), 5704– 5706,  DOI: 10.1016/j.tsf.2006.12.022
      There is no corresponding record for this reference.
    7. 7
      Padiyath, J.; Stahn, J.; Allenspach, P.; Horisberger, M.; Böni, P. Influence of Mo in the Ni Sublayers on the Magnetization of Ni/Ti Neutron Supermirrors. Phys. B 2004, 350 (1–3 Suppl. 1), E237– E240,  DOI: 10.1016/j.physb.2004.03.059
      There is no corresponding record for this reference.
    8. 8
      Amir, S. M.; Gupta, M.; Gupta, A.; Ambika, K.; Stahn, J. Silicide Layer Formation in Evaporated and Sputtered Fe/Si Multilayers: X-Ray and Neutron Reflectivity Study. Appl. Surf. Sci. 2013, 277, 182– 185,  DOI: 10.1016/j.apsusc.2013.04.021
      There is no corresponding record for this reference.
    9. 9
      SwissNeutronics Neutron Supermirrors. https://www.swissneutronics.ch/products/neutron-supermirrors/ (accessed 2025–04–01).
      There is no corresponding record for this reference.
    10. 10
      Høghøj, P.; Anderson, I.; Siebrecht, R.; Graf, W.; Ben-Saidane, K. Neutron Polarizing Fe/Si Mirrors at ILL. Phys. B 1999, 267–268, 355– 359,  DOI: 10.1016/S0921-4526(99)00038-1
      There is no corresponding record for this reference.
    11. 11
      Zubayer, A.; Ghafoor, N.; Thórarinsdóttir, K. A.; Stendahl, S.; Glavic, A.; Stahn, J.; Nagy, G.; Greczynski, G.; Schwartzkopf, M.; Febvrier, A. L.; Eklund, P. Reflective, Polarizing, and Magnetically Soft Amorphous Neutron Optics with 11B Enriched B4C. Sci. Adv. 2024, 10 (7), eadl0402  DOI: 10.1126/sciadv.adl0402
      There is no corresponding record for this reference.
    12. 12
      Sakhonenkov, S. S.; Filatova, E. O. Effect of B4C Barrier Layer on Layer Intermixing in Nanoscale W/Be Multilayers before and after Annealing. Appl. Surf. Sci. 2022, 606, 154961  DOI: 10.1016/j.apsusc.2022.154961
      There is no corresponding record for this reference.
    13. 13
      Engström, C.; Berlind, T.; Birch, J.; Hultman, L.; Ivanov, I. P.; Kirkpatrick, S. R.; Rohde, S. Design, Plasma Studies, and Ion Assisted Thin Film Growth in an Unbalanced Dual Target Magnetron Sputtering System with a Solenoid Coil. Vacuum 2000, 56 (2), 107– 113,  DOI: 10.1016/S0042-207X(99)00177-3
      There is no corresponding record for this reference.
    14. 14
      Glavic, A.; Björck, M. GenX 3: The Latest Generation of an Established Tool. J. Appl. Crystallogr. 2022, 55 (4), 1063– 1071,  DOI: 10.1107/S1600576722006653
      There is no corresponding record for this reference.
    15. 15
      Ström, P.; Primetzhofer, D. Ion Beam Tools for Nondestructive In-Situ and in-Operando Composition Analysis and Modification of Materials at the Tandem Laboratory in Uppsala. J. Instrum. 2022, 17 (4), P04011  DOI: 10.1088/1748-0221/17/04/P04011
      There is no corresponding record for this reference.
    16. 16
      Corrêa, S. A.; Pitthan, E.; Moro, M. V.; Primetzhofer, D. A Multipurpose Set-up Using KeV Ions for Nuclear Reaction Analysis, High-Resolution Backscattering Spectrometry, Low-Energy PIXE and in-Situ Irradiation Experiments. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 2020, 478 (June), 104– 110,  DOI: 10.1016/j.nimb.2020.05.023
      There is no corresponding record for this reference.
    17. 17
      Tran, T. T.; Lavoie, C.; Zhang, Z.; Primetzhofer, D. In-Situ Nanoscale Characterization of Composition and Structure during Formation of Ultrathin Nickel Silicide. Appl. Surf. Sci. 2021, 536, 147781  DOI: 10.1016/j.apsusc.2020.147781
      There is no corresponding record for this reference.
    18. 18
      Badía-Romano, L.; Rubín, J.; Bartolomé, F.; Magén, C.; Bartolomé, J.; Varnakov, S. N.; Ovchinnikov, S. G.; Rubio-Zuazo, J.; Castro, G. R. Morphology of the Asymmetric Iron–Silicon Interfaces. J. Alloys Compd. 2015, 627, 136– 145,  DOI: 10.1016/j.jallcom.2014.12.019
      There is no corresponding record for this reference.
    19. 19
      Gupta, A.; Kumar, D.; Phatak, V. Asymmetric Diffusion at the Interfaces in Fe/Si Multilayers. Phys. Rev. B 2010, 81 (15), 155402  DOI: 10.1103/PhysRevB.81.155402
      There is no corresponding record for this reference.

  • Supporting Information

    Supporting Information

    The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcc.5c00068.

    • X-ray diffraction (XRD), X-ray reflectivity (XRR), and off-specular scattering measurements for multilayers with and without 11B4C interlayers; complementary samples grown in a different deposition system are also included to demonstrate reproducibility and to assess performance under conditions more representative of industrial fabrication; figures illustrating interface structure, crystallinity, and fit quality across various period thicknesses (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.