Open Access
Add to BibSonomyAbstract
When cobalt thin films are illuminated with femtosecond laser pulses, we observe the emission of terahertz pulses. For a cobalt film thickness less than about 40 nm, the THz electric field direction rotates when the sample is rotated about the surface normal. This azimuthal angle-dependent emission is consistent with the assumption that laser-induced changes in an in-plane magnetization are responsible for the emission. For thicker layers, however, we observe the development of an azimuthal angle-independent contribution to the THz emission which we attribute to laser-induced changes in an out-of-plane magnetization component. We show that the relative contribution of this component grows when the cobalt film thickness increases. Our observations are supported by magnetic force microscopy measurements which show that for film thicknesses below 40 nm, the magnetization is predominantly in-plane whereas for thicknesses larger than 40 nm, an out-of-plane magnetization component develops.
© 2015 Optical Society of America
1. Introduction
The most common way of generating terahertz (THz) radiation is by illuminating non-linear optical materials and semiconductors with ultrashort laser pulses. The emission of terahertz radiation from non-linear crystals and semiconductors is due to laser-induced changes in current and/or a polarization in the sample [1]. THz radiation can be also generated by thin metal films after illumination with femtosecond laser pulses [2]. For example, in 2004 Kadlec et al. showed emission of THz pulses from gold thin films after illumination with femtosecond laser pulses [3]. Subsequently, Welsh et al. and Ramakrishnan et al. showed generation of THz radiation from metallic nanostructured surfaces [4, 5]. Both second-order and higher-order nonlinear processes were responsible for the emission. More recently, there has been a lot of interest in THz generation from ferromagnetic metal thin films. The interest in ferromagnetic thin films is motivated by the fact that such films have potential applications in magnetic data storage. In the past few years THz radiation has been used as an interesting tool for studying the magnetization dynamics of such ferromagnetic thin films. Beaurepaire et al. were the first to show that laser-induced ultrafast demagnetization of ferromagnetic films results in the emission of THz electromagnetic pulses [6]. Far away from the sample, the radiated electric field E(t) was assumed to be proportional to the second time derivative of the magnetization (d2M/dt2). In their research, the authors suggested that thermal effects are responsible for the ultrafast demagnetization. When laser pulses are incident on the sample, the temperature of the pump spot increases and as soon as it reaches the Curie temperature, demagnetization starts immediately. In 2004, Hilton et al. showed terahertz emission from iron thin films due to a second order magnetic nonlinearity. Without external magnetic field present, the ferromagnetic sample is multi-domain. However, when averaged over these multiple magnetic domains it still has a non-negligible net magnetization [7]. In 2012, Shen et al. also reported on terahertz emission from Ni-Fe alloy thin films through ultrafast demagnetization [8]. Recently Battiato et al. presented a microscopic model for femtosecond laser-induced demagnetization [9]. They argue that in ferromagnetic metals, where the number of spin-up and spin-down electrons is different, majority spin electrons (conventionally the majority spin electrons are referred to as ”spin up electrons” and the minority spin electrons as ”spin down electrons”) have a longer mean free path. This may lead to a depletion of majority carriers in the magnetic film. As a result, a spin current is established leading to the transfer of magnetization away from the surface. Subsequently, Eschenlohr et al. also supported the idea of superdiffusive spin transport as the mechanism responsible for the ultrafast demagnetization [10]. In this work the authors deposited a thin layer of gold on top of a ferromagnetic thin film in such a way that only a small part of the incident laser energy is able to reach the ferromagnetic film. Surprisingly, they observe that the efficiency of the ultrafast demagnetization remains the same. This indicates that for ultrafast demagnetization it is not necessary to directly illuminate the ferromagnetic metal but that it can also be achieved by the transport of electrons excited by the laser elsewhere. Yet again Schellekens et al. showed that for ultrafast demagnetization, superdiffusive spin transport may have an effect but it is not the dominant one [11].
Clearly, the microscopic mechanism behind the generation of THz light in these ferromagnetic films is not yet completely understood. A complicating and, perhaps, underestimated factor is that non-magnetic metal films are known to emit THz radiation upon illumination with a femtosecond laser too (as mentioned above), and that ”non-magnetic” contributions to the THz emission from ferromagnetic films cannot a priori be excluded. More experiments are thus essential to provide further information on the origin of THz emission from these materials.
Here, we show measurements of THz emission from cobalt (Co) thin films, illuminated with femtosecond laser pulses from a Ti:sapphire oscillator. We find that for Co thicknesses smaller than about 40 nm, the THz electric field polarization rotates when the sample is rotated around the surface normal. As a result, when only the p- or s-polarized field component is detected, the THz field changes sign every 180 degrees when the sample is rotated. Such behavior is typically absent in experiments on femtosecond laser induced THz emission from non-magnetic metals, and suggests ultrafast changes in an in-plane magnetization as the source of the THz emission for these thicknesses. For increasing thicknesses, however, an additional, azimuthal angle-independent, contribution to the signal is observed which grows in size with respect to the angle-dependent contribution. This angle-independent contribution is attributed to ultrafast changes in an out-of-plane magnetization which emits only p-polarized THz light and which dominates the emission for thicknesses larger than about 175 nm. This is consistent with magnetic force microscopy (MFM) measurements on these samples that show that for low thicknesses, the magnetization is predominantly in-plane, whereas for larger thicknesses the magnetization acquires a strong out-of-plane character. Our results show that for cobalt thin films, the emission of THz light is strongly correlated with the magnetization dynamics.
2. Experimental
2.1. Sample fabrication
Cobalt thin films of different thicknesses (2 nm – 250 nm) were deposited on glass substrates by electron beam evaporation at a rate of 1 Å/s. The thickness of the evaporated thin films was measured using a quartz crystal resonator positioned inside the evaporation chamber. The crystallinity of the deposited cobalt thin films was investigated by standard X-ray diffraction (XRD) measurements. Figure 1 shows the experimental XRD data for a 100 nm thick cobalt film deposited on the glass substrate. The large bump in the XRD data around 2θ = 30° is due to the amorphous nature of the glass substrate while the presence of several sharp peaks shows that the cobalt film is polycrystalline. The origin of the small peak near 2θ = 49° is ambiguous. It is either due to the neighboring hexagonal (100) plane of Co or due to the presence of cobalt oxide (CoO).
2.2. THz generation and detection setup
The experimental setup used for our measurements is schematically shown in Fig. 2(a). We have used a Ti: sapphire oscillator which generates p-polarized light pulses of 50 fs duration. These pulses are centered at a wavelength of 800 nm with an average power of 800 mW. The laser beam is split into a pump beam and a probe beam by a 80:20 beam splitter. When the pump beam is incident on the sample at a 45° angle of incidence, THz radiation is generated. The generated THz pulses are collected and focused using parabolic mirrors onto a 0.5 mm thick zinc telluride (ZnTe) (110) detection crystal. The probe beam is also focused on the detection crystal. The instantaneous electric field of the THz radiation induces birefringence in the detection crystal. When the probe beam passes through the detection crystal, it is modified from a linearly polarized to an elliptically polarized beam. The amount of ellipticity is proportional to the instantaneous THz electric field. The probe beam then passes through a quarter waveplate and a Wollaston prism and is finally focused on the photo-diodes of a differential detector. By varying the delay between the pump and the probe pulse we obtain a full 20 ps long THz electric field time-trace. In Fig. 2(b) we show the setup used for THz emission and detection at a zero degree angle of incidence. In this case the pump beam passes through a hole in the parabolic mirror and is incident on the sample at a zero degree angle of incidence. The THz radiation emitted from the cobalt thin films is collected in the back-propagating direction.
Fig. 2 Experimental setup for the generation and detection of THz pulses. The pump beam is incident on the sample at (a) a 45° degree angle of incidence and (b) a 0° angle of incidence. In (b), the THz emission is detected in the back-reflected direction.
2.3. Magnetic force microscopy
Magnetic force microscopy (MFM) is used to determine the size of magnetic domains having an out-of-plane magnetization. In MFM there is a sharp tip, which is coated with a ferromagnetic thin film and magnetized along the axis of the tip. During MFM measurements, this tip first scans the surface and gathers the topographical information over a sample area. During a second scan, in which the tip is slightly raised, this information is used to maintain a locally constant separation between the tip and the sample throughout the image. This second scan then only measures the long-range magnetic interactions [12]. MFM, as used by us, is not sensitive to the in-plane magnetization but only to a magnetization perpendicular to the plane. MFM can be used for investigating the domains and domain walls of magnetic thin films, nanoparticles and nanostructures [13].
3. Result and discussions
3.1. THz emission from cobalt thin film
In Fig. 3(a) we plot the THz electric field as a function of time, emitted from a 100 nm thick cobalt film deposited on a glass substrate, illuminated with femtosecond laser pulses at a 45° angle of incidence (see Fig. 2(a)). The amplitude of the emitted THz radiation is fairly weak and is roughly 0.4% of the THz emission from a conventional, semi-insulating GaAs (100) surface depletion field emitter. The emitted THz amplitude increases linearly with the laser power incident on the sample as shown in Fig. 3(b). This suggests that a second-order nonlinear process is responsible for the THz emission. The red solid line is a guide to the eye. In Fig. 3(c) we plot the emitted THz amplitude and the optical absorption as a function of pump-beam polarization angle for a 40 nm thick cobalt film. A 0° angle corresponds to a p-polarized pump beam, while a 90° angle corresponds to an s-polarized beam. We attribute the dependence on the polarization angle to changes in the efficiency with which the pump light is coupled into the film.
Fig. 3 (a) Measured THz electric field vs. time, emitted from a 100 nm thick cobalt film deposited on the glass substrate at 45° angle of incidence (b) Pump power dependence of THz emission from cobalt thin films. (c) The measured percentage of absorbed pump power (blue) and the electric field amplitude of the THz pulses emitted (red) from the cobalt thin films, as a function of pump beam polarization.
To investigate the relation between the THz emission and the magnetic order of the thin film, we applied an external magnetic field using permanent magnets, as shown in Fig. 4(a). We observe that upon reversal of the applied magnetic field direction, the THz electric field is also reversed, as shown in Fig. 4(b), where we plot the THz electric field amplitude as a function of time for a film thickness of 40 nm. Since the sample magnetization is reversed when we flip the direction of the applied magnetic field. This suggests that there is a strong connection between the magnetic order and the polarity of the THz pulse.
Fig. 4 (a) Schematic detail of setup used to apply an external magnetic field to cobalt thin films. (b) Measured THz emission from a 40 nm thick Co film on glass, as a function of time. Black and red traces indicate THz emission with magnetic fields applied in opposite directions.
3.2. Azimuthal angle dependence
In Fig. 5(a) we plot the p-polarized THz amplitude emitted by a 40 nm thick Co film, as a function of sample azimuthal angle. The p-polarized pump beam is incident at an angle of 45° and there is no externally applied magnetic field. The figure shows that the amplitude of the THz signal shows a sinusoidal-like dependence on azimuthal angle. The generated THz electric field flips sign when the sample is rotated by 180°. This kind of dependence is typically not observed in experiments on THz emission from non-ferromagnetic metals. This suggests that an in-plane magnetization change is responsible for the THz emission. We performed the same experiment with cobalt films of different thicknesses, in the range of 10 nm − 250 nm. For cobalt films with thicknesses less than 90 nm, the azimuthal angle dependence is similar to that of the 40 nm thick sample, meaning that the polarity of the THz signal changes when the sample is rotated by 180 degrees, as shown in Figs. 5(b) and 5(c). For 90–150 nm thick cobalt films, upon rotation, the THz signal changes polarity but an apparent positive offset reduces the angular range in which the signal changes sign, as shown in Figs. 5(d) and 5(e). When the thickness of the cobalt thin film is greater than 175 nm, an azimuthal angle dependence is still observed but there is no sign change anymore. An example of this is shown in Fig. 5(f) where we plot the change in THz amplitude as a function of azimuthal angle for a 250 nm thick cobalt film. As we will argue below, this suggests that for thick cobalt films, along with an in-plane magnetization, an additional perpendicular magnetization component is also present. Since the in-plane magnetization is no longer the dominant component, when we rotate the sample, the transient does not change sign and remains positive.
Note that for the thinnest sample, 40 nm, it seems that the range of azimuthal angles where the amplitude is negative, is smaller than for the next, 60 nm thick, sample. We are not really sure about the reason for this, but perhaps this may be explained if we can assume that, in the 40 nm case, the location where the laser beam hits the sample has a small offset with respect to the rotation axis. If this is true, then upon rotation of the sample, the laser would not hit exactly the same spot all the time. Instead it would, to a certain extent, also hit different parts of the sample which may give rise to somewhat different THz emission amplitudes and, thus, somewhat different looking azimuthal angle dependence.
Fig. 5 The azimuthal angle dependence of the measured THz electric-field amplitude emitted by a (a) 40 nm (b) 60 nm (c) 80 nm (d) 125 nm (e) 150 nm and (f) 250 nm thick cobalt film. Incident light is p-polarized. Error bars indicate the RMS uncertainty in the measured THz amplitude.
3.3. Relation between the components of the THz electric field
If changes in the in-plane magnetization are responsible for the THz emission, then rotating the sample around the surface normal should also rotate the THz field polarization. In general, when the sample magnetization is in-plane, both, s- and p-polarized THz light should be present. At the same time, for the samples with the magnetization perpendicular to the plane, only p-polarized THz emission should be observed. In Fig. 6 we plot the measured azimuthal angle dependence for both p-polarized and s-polarized components in the generated THz pulse for a 30 nm thick Co film. The pump beam is p-polarized and is incident at a 45° angle. The figure shows that when we rotate the sample, both components change in such a way that when the amplitude of the p-polarized THz emission is maximum, the amplitude of the s-polarized THz emission is zero and vice-versa. This supports the assumption that changes in the in-plane magnetization are responsible for the THz emission from this sample. For the 250 nm thick Co film, the results are different, as already shown in Fig. 5(f). Here only p-polarized THz light is observed and this shows only a weak azimuthal angle dependence. No measurable s-polarized THz emission was observed from the 250 nm thick cobalt film. Note that the weak azimuthal angle dependence observed in Fig. 5(f) suggests that a weak contribution from changes in an in-plane magnetization is present, and that a weak s-polarized component should be present also. This component is presumably too weak to measure in view of the small signal to noise ratio observed for this film thickness.
Fig. 6 Measured p-polarized and s-polarized component of the emitted THz pulse for 30 nm thick cobalt film. Error bars indicate the RMS uncertainty in the measured THz amplitude.
3.4. THz emission in back reflection
In Fig. 7 we plot the p-polarized THz electric field measured in the back-reflected direction as a function of time from cobalt thin films, detected at a 0° angle of incidence for two different film thicknesses, 40 nm and 250 nm. We observe that for the 40 nm thick cobalt film, THz emission is detected in the back-reflected direction, but when the 250 nm thick cobalt film is illuminated, we do not detect any THz emission. For the 40 nm thick cobalt film the magnetization is assumed to be in plane and so THz emission in the back-reflected direction would be allowed. However, for the 250 nm thick cobalt film, the magnetization of the film is predominantly in the perpendicular direction and since an oscillating magnetic dipole oriented in the normal direction cannot emit an electric field in the same direction, we don’t detect any THz emission in the back-reflected direction. However, we note that in the zero degree angle configuration, a part of the pump beam is blocked while passing through the hole in the parabola, which results in smaller THz emission signals. This makes it more difficult to detect weak backreflected THz emission from a small in-plane magnetization, if present.
Fig. 7 Measured p-polarized THz emission from 40 nm (black) and 250 nm (red) thick cobalt films at a 0° angle of incidence.
3.5. Thickness dependent THz emission
In Fig. 8(a) we plot the reflected, transmitted and absorbed pump power as a function of the thickness of the cobalt thin films deposited on glass substrates. The absorbed pump power is obtained by measuring the incident, reflected and transmitted pump power from the samples. We see that the absorption of the cobalt thin films increases as we increase the thickness but becomes eventually constant for thicknesses larger than about 20 nm. In Fig. 8(b), we show the amplitude of the THz radiation emitted at a 45° angle of incidence as a function of the thickness of the cobalt thin films. For each film, the sample is rotated to find the maximum THz amplitude emitted by the film. We observe that there is no direct correlation between the absorbed pump power and the emitted THz radiation. When we increase the thickness of the cobalt thin film, THz emission initially increases with increasing thickness, peaks around 40 nm and then decreases. We propose that the in-plane magnetization component increases as we increase the thickness of the cobalt film, but around 40 nm a perpendicular magnetization component develops. As we further increase the thickness of the cobalt film, this perpendicular component of the magnetization grows bigger and more dominant. If we can assume that THz radiation emitted due to changes in the perpendicular magnetization component and that emitted by changes in the in-plane magnetization component are opposite in phase, then these contributions partially cancel each other. Consequently, with increasing thickness, as the out-of-plane magnetization component becomes stronger, the emitted THz amplitude becomes smaller.
Fig. 8 (a) Percentage reflection, transmission and absorption of the pump laser pulses by different thicknesses of cobalt thin films. (b) p-polarized THz emission as a function of thickness of cobalt film deposited on the glass substrate.
In order to confirm our assumption that changes in the orientation of the magnetic domains are responsible for the observed changes in THz emission as we increase the sample thickness, cobalt thin films were studied using magnetic force microscopy (MFM). Polycrystalline cobalt thin films typically possess an in-plane magnetization, but above a critical thickness a perpendicular magnetization component is expected [14]. This cross-over thickness for cobalt thin films is around 40 nm [15]. For thicker samples, the perpendicular magnetization component gives rise to a stripe domain pattern with alternating dark and bright contrast indicating domains pointing up and down. Saito et al. have demonstrated such domains for thin Ni-Fe films [16]. In Fig. 9 we show the MFM measurements of cobalt thin films with different thicknesses deposited on the glass substrate. We see that for thin cobalt films, with thicknesses less than 40 nm, there are no stripe domains present but the stripe pattern appears for thicker cobalt films, showing the presence of a perpendicular magnetization component. We also observe that the width of the stripes is a function of film thickness and for much thicker films (>200 nm) they grow wider. Based on these measurements we can confirm that the contrasting azimuthal angle dependence behavior shown by films with different thicknesses most probably arises due to a change from a predominantly in-plane magnetization to a predominantly out-of-plane magnetization for increasing thickness. Note that, although we cannot fully exclude the possibility that a non-magnetic contribution to the THz emission from cobalt thin films is present, we would expect this to be present also in the range of thicknesses below 90 nm, in contrast with what we observe.
Fig. 9 Magnetic force microscope images of cobalt thin films on glass with different thicknesses. No domains are observed for thin cobalt films; domains start appearing when the thickness of the film crosses the critical thickness (40 nm). For thicker cobalt films, the width of the domains increases as we increase the thickness of the film.
It is actually surprising that a net magnetization is present in our samples. If the orientation of many domains within the spotsize of the laser were truly random, no net magnetization would be present. In principle, a net magnetization can be induced by the atomic order of the underlying substrate but that is less likely in our case since all substrates are made of glass. However, a net magnetization can also result from a nonzero net angle of incidence of cobalt atoms impinging on the substrate during deposition [7]. A net magnetization can also result from stray magnetic fields, for example, when the cobalt film is deposited on the substrate using electron beam deposition in which a magnetic field is used to direct an electron beam to the cobalt target to melt it. We note that the MFM measurements for cobalt films with thicknesses greater than 40 nm give the impression of some order in the way the out-of-plane magnetic domains are organized. It is unclear whether this also implies that there is a net magnetization. Although the exact origin of the net magnetization is currently not well understood, the purpose of the current work is to discover whether changes in this magnetization are responsible for the THz emission. Similar conclusions were drawn by Hilton et al. who studied iron thin films [7].
Our results can be summarized by the four figures in Fig. 10. We show that when the femtosecond laser pulses are incident on the thin cobalt films with thicknesses less than 40 nm, i.e. when the magnetization is in-plane, THz emission is observed at both a 45° angle of incidence and a 0° angle of incidence, as shown in Figs. 10(a) and 10(b) respectively. For thicknesses in the range of 40 nm – 175 nm, in addition to an in-plane magnetization component, a perpendicular magnetization component is also present. The perpendicular magnetization component for these samples is relatively weak and the net magnetization mostly lies in the plane of the sample. Hence, the THz emission from these cobalt films is observed at both angles of incidence. When femtosecond laser pulses are incident at a 45° angle of incidence on the cobalt films thicker than 175 nm, with the sample magnetization predominantly perpendicular to the plane, a relatively weak THz emission is observed compared to the emission from a 40 nm thick cobalt film, as shown in Fig. 10(c). However, when the laser pulses are incident on these films at 0° angle of incidence, no THz emission is detected in the back-reflected direction (Fig. 10(d)).
Fig. 10 Schematic overview of our results. (a) THz generation from thin cobalt films at a 45° angle of incidence (b) THz generation from thin cobalt films at a 0° angle of incidence (c) THz generation from thick cobalt films at a 45° angle of incidence (d) Absence of THz emission from thick cobalt films at a 0° angle of incidence. Incident light is p-polarized
4. Conclusion
We have demonstrated that the THz radiation emitted by cobalt thin films upon illumination with ultrashort laser pulses shows a different azimuthal angle dependent behavior depending on the sample thickness. For cobalt thin films with thicknesses less than 175 nm, the THz signal changes sign when the sample is rotated around the surface normal whereas no sign change is observed for thicker cobalt films. This behavior is attributed to the change in the direction of magnetization from in-plane to out of plane as the film thickness increases. THz emission at zero degree angle of incidence is consistent with this change in orientation of the magnetization. The maximum amplitude of the THz radiation emitted from cobalt thin films at a 45° angle of incidence depends on film thickness. The THz amplitude increases with increasing thickness, peaks around 40 nm and then decreases. The measurements are consistent with MFM measurements of cobalt films showing the increasing presence of out-of-plane magnetization for increasing film thicknesses. Our results provide strong evidence that THz emission from cobalt thin films after illumination with femtosecond laser pulses, is the result of rapid change in the magnetization.
Acknowledgments
PCMP and AJLA acknowledge the financial support from the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO) and the Stichting voor Technische Wetenschappen (STW) in the form of VICI and VENI grants, respectively. NK gratefully acknowledges financial support from the European commission through a Marie Curie fellowship.
References and links
1. K. Sakai, Terahertz Optoelectronics (Springer, 2005). [CrossRef]
2. G. K. P. Ramanandan, G. Ramakrishnan, N. Kumar, A. J. L. Adam, and P. C. M. Planken, “Emission of terahertz pulses from nanostructured metal surfaces,” J. Phys. D: Appl. Phys. 47, 374003 (2014). [CrossRef]
3. F. Kadlec, P. Kuel, and J.-L. Coutaz, “Optical rectification at metal surfaces,” Opt. Lett. 29(22), 2674–2676 (2004). [CrossRef] [PubMed]
4. G. H. Welsh and K. Wynne, “Generation of ultrafast terahertz radiation pulses on metallic nanostructured surfaces,” Opt. Express 17(4), 2470–2480 (2009). [CrossRef] [PubMed]
5. G. Ramakrishnan and P. C. M. Planken, “Percolation-enhanced generation of terahertz pulses by optical rectification on ultrathin gold films,” Opt. Lett. 36(13), 2572–2574 (2011). [CrossRef] [PubMed]
6. E. Beaurepaire, G. M. Turner, S. M. Harrel, M. C. Beard, J.-Y. Bigot, and C. A. Schmuttenmaer, “Coherent terahertz emission from ferromagnetic films excited by femtosecond laser pulses,” Appl. Phys. Lett. 84, 3465 (2004). [CrossRef]
7. D. J. Hilton, R. D. Averitt, C. A. Meserole, G. L. Fisher, D. J. Funk, J. D. Thompson, and A. J. Taylor, “Terahertz emission via ultrashort-pulse excitation of magnetic metal films,” Opt. Lett. 29(15), 1805–1807 (2004). [CrossRef] [PubMed]
8. J. Shen, X. Fan, Z. Chen, M. F. DeCamp, H. Zhang, and J. Q. Xiao, “Damping modulated terahertz emission of ferromagnetic films excited by ultrafast laser pulses,” Appl. Phys. Lett. 101, 072401 (2012). [CrossRef]
9. M. Battiato, K Carva, and P. M. Oppeneer, “Superdiffusive spin transport as a mechanism of ultrafast demagnetization,” Phys. Rev. Lett. 105, 027203 (2010). [CrossRef] [PubMed]
10. A. Eschenlohr, M. Battiato, P. Maldonado, N. Pontius, T. Kachel, K. Holldack, R. Mitzner, A. Fhlisch, P. M. Oppeneer, and C. Stamm, “Ultrafast spin transport as key to femtosecond demagnetization,” Nat. Mater. 12, 332–336 (2013). [CrossRef] [PubMed]
11. A. J. Schellekens, W. Verhoeven, T. N. Vader, and B. Koopmans, “Investigating the contribution of superdiffusive transport to ultrafast demagnetization of ferromagnetic thin films,” App. Phys. Lett. 102, 252408 (2013). [CrossRef]
12. Y. Martin and H. K. Wickramasinghe, “Magnetic imaging by force microscopy with 1000 A° resolution,” Appl. Phys. Lett. 50, 1455 (1987). [CrossRef]
13. C. T. Hsieh, J. Q. Liu, and J. T. Lue, “Magnetic force microscopy studies of domain walls in nickel and cobalt films,” Appl. Surf. Sci. 252, 1899–1909 (2005) [CrossRef]
14. M. Hehn, S. Padovani, K. Ounadjela, and J. P. Bucher, “Nanoscale magnetic domain structures in epitaxial cobalt films,” Phys. Rev. B 54(5), 3428–3433 (1996). [CrossRef]
15. J. Brandenburg, R. Huhne, L. Schultz, and V. Neu, “Domain structure of epitaxial Co films with perpendicular anisotropy,” Phys. Rev. B 79, 054429 (2009). [CrossRef]
16. N. Saito, H. Fujiwara, and Y. Sugita, “A new type of magnetic domain structure in negative magnetostriction Ni-Fe films,” J. Phys. Soc. Jpn. 19(7), 1116 (1964). [CrossRef]
