We present a compact in situ electromagnet with an active cooling system for use in ultrahigh vacuum environments. The active cooling enhances the thermal stability and increases the electric current that can be applied through the coil, promoting the generation of homogeneous magnetic fields, required for applications in real-time deposition experiments. The electromagnet has been integrated into a reflectance difference magneto-optic Kerr effect (RD-MOKE) spectroscopy system that allows the synchronous measurement of the optical anisotropy and the magneto-optic response in polar MOKE geometry. Proof of principle studies have been performed in real time during the deposition of ultra-thin Ni films on Cu(110)-(2 × 1)O surfaces, corroborating the extremely sharp spin reorientation transition above a critical coverage of 9 monolayers and demonstrating the potential of the applied setup for real-time and in situ investigations of magnetic thin films and interfaces.

The constantly developing field of magnetic thin films and interfaces requires reliable and sensitive methods to address and correlate structural and magnetic properties with high precision. Since the mid-1980s, the magneto-optic Kerr effect (MOKE) has been employed to investigate the magnetic properties of novel thin films and interfaces.1,2 In the case of surfaces and thin films, the magnetic properties are strongly related to the structure and to the morphology and can be tuned by the choice of substrate and adsorbate materials, the film thickness, or the growth conditions. A widely investigated class of materials are inorganic/organic interfaces, where magnetic molecules are adsorbed on ferromagnetic transition metal substrates. The “spinterface” between both layers often reveals a ferro- or antiferromagnetic coupling.3 Therefore, the methods required to experimentally investigate such systems need to simultaneously address and correlate the magnetism with the underlying morphology.

An optical technique, which elegantly incorporates the above requirements, is the combination of surface sensitive reflectance difference spectroscopy (RDS) with the MOKE.4–6 A further improvement of this technique is the magnetic field modulated RD-MOKE spectroscopy that allows examining morphological and magnetic changes in real time during the deposition of magnetic molecules with high sensitivity.6,7 While existing spectroscopic RD-MOKE systems often employ ex situ generated magnetic fields,4,5 it is desired to apply the magnetic field in situ directly at the sample position and during deposition. This requires an ultrahigh vacuum (UHV) compatible design that does not obstruct the optical path to and from the sample surface.1,2,8

The state-of-the-art in the field of in situ electromagnets is mainly based on devices for very specific applications, rather than on general UHV-compatible solutions. In the context of the implementation of a transversal MOKE setup in Ref. 9, the authors presented a Helmholtz-type double-coil electromagnet with a field of 5 mT, for which a large current of 45 A is required. A substantially improved current-to-magnetic field ratio is achieved in the yoke design electromagnet reported in Ref. 10, which allows the generation of a magnetic field of 170 mT with a current of only 4 A. This setup, however, is not equipped with a cooling system, and so its full potential can be exploited over very limited timescales or in pulse mode. An alternative solution for continuous operation is a superconducting magnet. However, superconducting electromagnets are significantly space-consuming and costly and are usually integrated in a specific measuring system, such as a scanning tunneling microscope.11 Until now, no universal solution for UHV chambers has been proposed.

In this work, we present a compact in situ UHV-compatible electromagnet with active cooling that provides a magnetic field located directly at the position of the investigated sample. A characterization of the magnetic field and of the thermal behavior of the electromagnet is provided in Sec. II. As a proof of principle, the implementation of the device into a UHV chamber in combination with real-time modulated RD-MOKE spectroscopy is described in Sec. III. The results show the advantages of employing the present design for real-time investigations of the magnetic properties of growing ferromagnetic thin films and its potential to elucidate the magnetic coupling at surfaces and interfaces between organic/inorganic systems.3,12–17

An electromagnet for in situ applications under UHV conditions has to fulfill two main requirements. (1) In order to ensure operation in real time during sample manipulation or thin film deposition, the electromagnet has to be positioned as close to the sample as possible, while complying with the geometrical restrictions and dimensions of the chamber. (2) The electromagnet has to be manufactured from UHV-compatible low outgassing materials, and its temperature should be kept as low as possible during operation, since, otherwise, the pressure inside the chamber would exceed the acceptable range for surface science experiments. For this reason, an actively cooled electromagnet was conceived, which allows applying high currents to reach large magnetic fields at limited outgassing rates. In fact, the cooling not only reduces the thermal desorption rates, but the cooled wires also have a lower electrical resistance, which further reduces the Joule heating. All the materials employed in the assembly have been chosen carefully to comply with all UHV requirements, in particular, the ability to withstand recurring “bakeouts” of the chamber up to 150 °C.

The developed electromagnet, shown in Fig. 1(a), features a cylindrical and regular single pancake-coil design to enhance the magnetic field homogeneity beyond that of a standard electromagnetic coil.18 The coil, consisting of N = 254 windings, is wound on a copper coil holder with radially arranged cooling fins, in order to facilitate the heat flow away from the wires. The heat generation depends strongly on the electric current flowing through the coil.

FIG. 1.

(a) Image of the pancake-coil electromagnet. Schematic representations of (b) the electromagnet and the active cooling system and (c) the wire arrangement in a conventional coil and (d) in the present electromagnet.

FIG. 1.

(a) Image of the pancake-coil electromagnet. Schematic representations of (b) the electromagnet and the active cooling system and (c) the wire arrangement in a conventional coil and (d) in the present electromagnet.

Close modal

The active cooling of the electromagnet, which is the centerpiece of the development, is achieved by inlet and outlet pipes that are directly mounted into a copper heat exchanger. Cooling can be accomplished in two different ways: (i) employing an air or gaseous nitrogen (N2) flow at room temperature (RT) or using (ii) nitrogen gas pre-cooled by liquid nitrogen (LN2). For the second method, the inlet pipe is connected to a copper pipe array that is submerged in liquid nitrogen, as shown schematically in Fig. 1(b). The gaseous N2 is, thus, cooled close to its boiling temperature of 77.4 K,19 resulting in a (partial) transition into the liquid state. A Kapton coated wire of 1.1 mm in diameter was used for the coil. Kapton is a polyimide film and a widespread electrical insulator in UHV environments due to its low outgassing rate and its thermal stability in the low and medium temperature regime.20,21 An additional feature of Kapton is its low thermal conductivity in the range of 0.37 W/(m·K)–0.52 W/(m·K),22 while commercially available copper shows a thermal conductivity coefficient of 240 W/(m·K)–380 W/(m·K).23 Even for a Kapton film thickness of 0.05 mm, as in the case of the used wire, this results in a major challenge concerning the heat transport from the coil. To optimize cooling efficiency, the present design consists of a pancake-like coil system, featuring thin copper cooling fins where only two wires are wound next to each other within one layer of the coil. In such an arrangement, every winding is in direct contact with one of the cooling fins, as shown in Fig. 1(d), so that the heat can be more easily transferred to the coil holder. In contrast, for the heat transfer in a standard coil configuration, presented in Fig. 1(c), only the wires in direct contact with the coil frame are cooled. In such an arrangement, the inner wires of the coil tend to overheat, limiting considerably the electric current through the coil and, thus, the maximum achievable magnetic field.

Another important feature of the present electromagnet is its compact design and DN40 CF flange mounting, which makes it exchangeable and compatible with most UHV chambers. Due to the fact that the electromagnet is mounted on an x-manipulator, the positioning of the magnet inside the UHV chamber is adjustable. Furthermore, the present design also allows the incorporation of a rotational drive to the mounting, adding a further degree of freedom to the electromagnet inside the UHV chamber.

The homogeneity of the magnetic field of the electromagnet was checked by measuring the magnetic flux density B along the z-direction with a commercial Hall probe, as shown schematically in Fig. 2(a). The position of the Hall probe is controlled by optical x-y manipulator stages. In this way, the distribution of the magnetic flux density in the x-y plane of the magnet was determined, providing a full 2D-map of B, as shown in Fig. 2(b). Note that the distance in the z-direction is not measured from the center of the coil (z), but from the front edge of the magnet (Z = zL/2), as depicted in Fig. 2(a). The magnetic flux density was measured every 2.5 mm in x- and y-directions over the magnet area at Z = 1.8 mm and for I = 1 A.

FIG. 2.

(a) Measurement configuration to determine the magnetic flux density B(x, y, Z). (b) 2D-map of the magnetic flux density of the present electromagnet for a current of I = 1 A and Z = 1.8 mm.

FIG. 2.

(a) Measurement configuration to determine the magnetic flux density B(x, y, Z). (b) 2D-map of the magnetic flux density of the present electromagnet for a current of I = 1 A and Z = 1.8 mm.

Close modal

The measured magnetic flux density along the central axis normal to the sample surface as a function of the distance Z is illustrated in Fig. 3(a). The data are compared with a numerical calculation (blue solid line) based on the Biot-Savart law24 considering our electromagnet as a multi-layered assembly of conducting loops,

Baxial(Z)=B0g(Z,L,ro,ri),
(1)

with

g(Z,L,ro,ri)=(Z+L)lnro+ro2+(Z+L)2ri+ri2+(Z+L)2Zlnro+ro2+(Z)2ri+ri2+(Z)21rori,
(2)

where B0 = μ0 · (I · L)/N, L is the length of the coil, and ro and ri are the outer and inner radius of the coil, respectively. For the present electromagnet, N = 254, L = 3.6 cm, ro = 3.4 cm, ri = 2.1 cm, and B0 = 4.9 mT. The experimental results are in good agreement with the theoretical prediction. The magnetic flux density B as a function of the applied current is depicted in Fig. 3(b), showing a linear dependence in accordance with the theoretical expectations. These results have been obtained directly at the front edge of the magnet (Z = 0).

FIG. 3.

Magnetic flux density of the electromagnet, both experimental (orange filled dots) and calculated (blue solid line): (a) as a function of the distance Z from the magnet and (b) as a function of the applied electric current.

FIG. 3.

Magnetic flux density of the electromagnet, both experimental (orange filled dots) and calculated (blue solid line): (a) as a function of the distance Z from the magnet and (b) as a function of the applied electric current.

Close modal

In the present device, the heat transfer from the windings to the coil holder is considerably improved through the cooling fins. Moreover, the coil holder is attached to a heat exchanger with an inflow and an outflow for a cooling fluid as shown in Fig. 1(b). As a result, the heat flow is mainly convective, rather than just relying on conductive heat transfer. The thermal properties of the magnet were measured in air conditions using a FLUKE Ti400 infrared camera. This thermal imaging camera, with a resolution of 320 × 240 pixels, features a temperature measurement range between −20 °C and +1200 °C and a ±2 °C or ±2% accuracy, depending on which of the two values is larger. It offers qualitative temperature monitoring with an adjustable color scale over the whole image area and an option for the quantitative temperature measurement in up to 3 pixels (markers) within the image. Precise temperature acquisition using a thermal imaging camera relies on the manual setting of the emissivity (0–1.0). Considering the different materials on the present magnet, e.g., bare copper for the magnet holder and Kapton for the wire insulation, and different viewing angles, the temperature measurements performed using the FLUKE Ti400 are more suited to observe a general trend in thermal behavior and relative temperature differences, rather than absolute and reliable temperature values. However, the results provide a good estimate of the temperature distribution within the magnet. In order to obtain a quantitative value of the temperature, an additional K-type thermocouple was attached to the electromagnet. These thermocouples are considered to be very reliable in the examined temperature range, usually featuring an accuracy of ±2.2 °C or ±0.75%.25 

An infrared camera was used to monitor the heating process, by taking an image every 10 s, while the electromagnet was operated with a current of I = 4 A. As soon as the temperature of the electromagnet reached ∼70 °C, the cooling was turned on. The temperature of the magnet was measured quantitatively using two available markers (A and B) of the infrared camera, as displayed in Fig. 4(a). Additionally, the temperature of the magnet during the heating process was also estimated by measuring the change in electrical resistance of the whole assembly, consisting of the coil plus the wires connecting it to the electrical feedthrough. The resistivity ρ rises with increasing temperature, according to the Taylor expansion,

ρ=ρ20(1+α20(T20°C)+β20(T20°C)2+),
(3)

where ρ20 = 0.017 86 Ω·mm2/m is the resistivity of copper at T = 20 °C and α20 and β20 are the temperature coefficients of the copper wire at 20 °C.26 In the measured temperature range below T = 150 °C, it is sufficient to use only the first-order linear temperature coefficient α20 = 3.93 × 10−3 K−1. Considering the relation between resistance R and resistivity ρ, with being R = ρ l/A, where l is the length and A is the cross-sectional area of a wire, the above formula transforms into

R=R201+α20(T20°C).
(4)

In the following experiments, the values of the resistance were obtained by measuring the applied voltage U at a constant current of I = 4 A and using Ohm’s law, R = U/I. The evolution of the temperature and the resistance upon cooling with nitrogen (N2) gas at RT are depicted in Fig. 4(b). It can be noticed that the coil holder reacts upon cooling by keeping its temperature in a rather constant range, making the setup partly suitable for RT cooling fluids. As can be seen in Fig. 4(c), the resistance scales linearly with the temperature, which is an indication for a homogeneous temperature distribution.

FIG. 4.

(a) Infrared image of the electromagnet with the indicated positions of the markers A and B. The red circle above letter A is a hole that was drilled in the coil holder to outgas the inside of the coil holder. (b) Evolution of the temperature and the resistance as a function of time upon cooling with gaseous N2 at RT. (c) Resistance as a function of temperature for markers A and B.

FIG. 4.

(a) Infrared image of the electromagnet with the indicated positions of the markers A and B. The red circle above letter A is a hole that was drilled in the coil holder to outgas the inside of the coil holder. (b) Evolution of the temperature and the resistance as a function of time upon cooling with gaseous N2 at RT. (c) Resistance as a function of temperature for markers A and B.

Close modal

The behavior of the temperature and the resistance upon cooling with LN2 is shown in Fig. 5(a). It can be clearly stated that the coil is cooling down very homogeneously and the time needed to reach 0 °C is 15 min. The resistance as a function of the temperature for both markers, A and B, is plotted in Fig. 5(b). The resistance for marker A, which is not placed on the coil, deviates slightly from a linear behavior during cooling. However, during the heating process, the resistance for both markers is comparable and linear. This provides further evidence of a highly homogeneous temperature distribution of the pancake coil. The electromagnet was further tested under vacuum conditions while cooling with LN2. No current was applied during the experiments. The temperature was measured solely with the thermocouple on the coil holder since infrared cameras are not suitable for measurements through UHV chamber windows. The results are illustrated in Fig. 5(c), confirming that the employed cooling method is suitable for UHV.

FIG. 5.

(a) Evolution of the temperature and the resistance as a function of time while cooling with LN2. (b) Resistance as a function of temperature taken from the measured points in (a) during heating and cooling of the coil, as indicated by the arrows. (c) Temperature evolution of the electromagnet upon cooling with LN2 in UHV.

FIG. 5.

(a) Evolution of the temperature and the resistance as a function of time while cooling with LN2. (b) Resistance as a function of temperature taken from the measured points in (a) during heating and cooling of the coil, as indicated by the arrows. (c) Temperature evolution of the electromagnet upon cooling with LN2 in UHV.

Close modal

The proof of principle of the in situ electromagnet with active cooling has been carried out in a UHV chamber equipped with an RD-MOKE spectroscopy setup, as illustrated in Fig. 6. The electric current from the power source (1) is modulated by a function generator (2) into a sinusoidal waveform. The modulated current, which is also redirected as a reference signal (ref.) to the computer (PC), generates a modulated magnetic field H (light and dark green arrows) in the pancake-coil electromagnet (3). During the experiments, the electromagnet was cooled by LN2. A current of 1 A corresponds to a magnetic field of μ0H ≈ 3 mT at the sample position. During operation of the electromagnet with current amplitudes of up to 6 A (corresponding to μ0H = 18 mT at the sample surface), the base pressure in the vacuum chamber was less than 2 × 10−10 mbar and the temperature of the electromagnet during operation and cooling with LN2 stabilized at −20 °C. Operation at constant currents of up to 7 A (μ0H = 21 mT) is also possible, preferably for shorter periods of time, e.g., in pulse mode, or for recording of single hysteresis curves.

FIG. 6.

Schematic representation of the employed RD-MOKE setup with a modulated external magnetic field provided by the newly developed in situ UHV electromagnet with active cooling. The parts labeled (1)–(7) in the figure are described in the text.

FIG. 6.

Schematic representation of the employed RD-MOKE setup with a modulated external magnetic field provided by the newly developed in situ UHV electromagnet with active cooling. The parts labeled (1)–(7) in the figure are described in the text.

Close modal

In the chamber, either metals (4) or organic materials (5) can be deposited from the respective deposition sources onto the sample (6), in this case, consisting of a Cu(110) single crystal. In the geometry employed here, when the magnetic easy axis of the film switches from in-plane to out-of-plane as a result of the spin reorientation transition (SRT), the magnetic moment follows the modulated field of the magnet, yielding a sizable signal in the polar RD-MOKE setup.4,6 The phase modulated information from the light reflected from the sample surface at a small angle of about 2° measured with the RDS (7) is transformed into an intensity modulated signal with an analyzer.27 The resulting modulated RD signal and the reference signal (ref.) are recorded and processed in the PC. The front end displays the real-time view of the magnetization curve and a Fourier decomposition of the modulated RD signal, from which other measurement related parameters can be derived.7 

The oxygen pre-coverage on the Cu(110) surface is required to overcome the rather disfavored 3D-type growth mode of Ni on the pristine surface.4,28 The pre-adsorbed oxygen acts as a surfactant and induces a layer-by-layer growth mode over an extended coverage regime.6 The oxygen atoms float on top of the grown Ni layers and induce a (2 × 1) O reconstruction at the surface of the pseudomorphic Ni(110) film. The shape anisotropy initially favors an in-plane orientation of the magnetic easy axis, which switches to an out-of-plane orientation at a Ni film thickness of 9 monolayers (ML).4,6 This sharp SRT arises from the increasing magneto-elastic anisotropy as a consequence of the accumulating strain during the pseudomorphic growth of the Ni film.

The sudden onset of the out-of-plane magnetization is reproduced in Fig. 7(a), where the RD signal at a photon energy of E = 3.5 eV is monitored during the deposition of 12.3 ML of Ni on a Cu(110)-(2 × 1)O surface. Ni was deposited with a rate of 0.37 ML/min and with the sample held at RT. A sinusoidally modulated magnetic field with an amplitude of μ0H = 6 mT and a frequency of f = 0.02 Hz, as shown in the upper panel of Fig. 7(b), was generated by the in situ pancake-coil electromagnet applied normal to the surface during deposition. The onset of the sharp SRT at 9 ML is clearly identified since, at this point, the RD signal starts oscillating in response to the applied magnetic field. The oscillating RD-MOKE signal after the SRT, presented in detail in the lower panel of Fig. 7(b), quickly takes the shape of a square-wave signal, which is an indication for out-of-plane ferromagnetism.29,30 From this oscillating RD-MOKE signal, magnetization curves, more precisely, the Kerr rotation, which is proportional to the sample magnetization, as a function of the external magnetic field, can be extracted. As an example, the hysteresis curve, shown in Fig. 7(c), was obtained from the shadowed area in Fig. 7(b), corresponding to a full cycle of the external magnetic field.

FIG. 7.

(a) RD transient at a photon energy of E = 3.5 eV recorded as a function of the Ni film thickness dNi during the deposition of Ni while applying a modulated magnetic field of ±6 mT. The SRT at a coverage of 9 ML is marked with a red arrow. (b) Comparison of the applied external field μ0H and the Kerr rotation as a function of dNi after the SRT, showing the linear dependence of the Kerr rotation with the increasing layer thickness. (c) Hysteresis curve extracted from the shaded area in (b): data (symbols) and fit (solid line).

FIG. 7.

(a) RD transient at a photon energy of E = 3.5 eV recorded as a function of the Ni film thickness dNi during the deposition of Ni while applying a modulated magnetic field of ±6 mT. The SRT at a coverage of 9 ML is marked with a red arrow. (b) Comparison of the applied external field μ0H and the Kerr rotation as a function of dNi after the SRT, showing the linear dependence of the Kerr rotation with the increasing layer thickness. (c) Hysteresis curve extracted from the shaded area in (b): data (symbols) and fit (solid line).

Close modal

In this way, it is possible to distinguish between a paramagnetic and ferromagnetic response and to follow its evolution during growth. The magnetization curves can further be fitted to model functions, as shown by the solid blue line in Fig. 7(c). From these fits, important parameters such as the Kerr amplitude, the remanent Kerr rotation, and the coercive field can be determined in real time as a function of coverage and sample temperature.7 

A compact in situ pancake-coil electromagnet for operation in UHV environments has been developed with an active cooling system to reduce the challenges related to the Joule heating and the outgassing occurring in UHV. The functionality of the present electromagnet was successfully characterized and tested both ex situ and in situ. The advantages of the present electromagnet clearly stand out from comparable solutions for UHV environments. Due to its in situ nature, the magnet requires much lower currents for achieving the same magnetic field at the sample position than ex situ setups. Moreover, because of the proximity to the sample, the field is quite homogeneous and not distorted by any additional chamber components. As a result of the active cooling of the electromagnet, high currents of 7 A can be applied, yielding magnetic field strengths as high as 21 mT in continuous mode. In addition, its energy- and space-saving characteristics, the electromagnet is mounted on a DN40 CF flange and is compatible with any standard UHV chamber, which makes it a simple and powerful add-on to UHV systems where a magnetic field is required.

The here developed electromagnet was combined with an existing RD-MOKE setup, and the first in situ modulated RD-MOKE experiments have been realized. The modulation of the magnetic field allows monitoring the magnetic properties in real time as demonstrated here for the case of Ni thin films deposited on a Cu(110)-(2 × 1)O surface. The ease and beauty of the field modulation technique can be fully exploited using a suitable in situ electromagnet, which, thanks to its design, delivers highly homogeneous magnetic fields up to 21 mT. Such magnetic field strengths would also be sufficient for real-time investigations of hybrid organic/inorganic magnetic systems.13,15–17

The authors acknowledge the financial support from Austrian Science Fund (FWF) Project No. V-478-N36. For the purpose of open access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. For the purpose of open access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. Technical support of Robert Leimlehner, Ewald Fink, and Konrad Fragner is greatly acknowledged. The authors would also like to thank Margherita Matzer for her support with the Austrian Patent Office.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
S. D.
Bader
and
E. R.
Moog
, “
Magnetic properties of novel epitaxial films
,”
J. Appl. Phys.
61
,
3729
(
1987
).
2.
Z. Q.
Qiu
and
S. D.
Bader
, “
Surface magneto-optic Kerr effect
,”
Rev. Sci. Instrum.
71
,
1243
(
2000
).
3.
S.
Sanvito
, “
The rise of spinterface science
,”
Nat. Phys.
6
,
562
564
(
2010
).
4.
T.
Herrmann
,
K.
Lüdge
,
W.
Richter
,
K. G.
Georgarakis
,
P.
Poulopoulos
,
R.
Nünthel
,
J.
Lindner
,
M.
Wahl
, and
N.
Esser
, “
Optical anisotropy and magneto-optical properties of Ni on preoxidized Cu(110)
,”
Phys. Rev. B
73
,
134408
(
2006
).
5.
M.
Fronk
,
B.
Bräuer
,
J.
Kortus
,
O. G.
Schmidt
,
D. R. T.
Zahn
, and
G.
Salvan
, “
Determination of the Voigt constant of phtalocyanines by magneto-optical Kerr-effect spectroscopy
,”
Phys. Rev. B
79
,
235305
(
2009
).
6.
R.
Denk
,
M.
Hohage
, and
P.
Zeppenfeld
, “
Extremely sharp spin reorientation transition in ultrathin Ni films grown on Cu(110)-(2×1)O
,”
Phys. Rev. B
79
,
073407
(
2009
).
7.
G.
Mendirek
,
A.
Brozyniak
,
M.
Hohage
,
A.
Navarro-Quezada
, and
P.
Zeppenfeld
(unpublished) (
2021
).
8.
G.
Vinai
,
F.
Motti
,
A. Y.
Petrov
,
V.
Polewczyk
,
V.
Bonanni
,
R.
Edla
,
B.
Gobaut
,
J.
Fujii
,
F.
Suran
,
D.
Benedetti
,
F.
Salvador
,
A.
Fondacaro
,
G.
Rossi
,
G.
Panaccione
,
B. A.
Davidson
, and
P.
Torelli
, “
An integrated ultra-high vacuum apparatus for growth and in situ characterization of complex materials
,”
Rev. Sci. Instrum.
91
,
085109
(
2020
).
9.
P.
Froemel
, “
Implementation of a T-MOKE setup for transient studies of magnetic materials at the HELIOS XUV photon source
,” Technical Report,
Uppsala University
,
2013
, available at https://www.diva-portal.org/smash/get/diva2:716785/FULLTEXT01.pdf.
10.
F.
Heigl
,
O.
Krupin
,
G.
Kaindl
, and
K.
Starke
, “
Two-axis rotatable magnet in ultrahigh vacuum
,”
Rev. Sci. Instrum.
73
,
369
(
2002
).
11.
J.
Wiebe
,
A.
Wachowiak
,
F.
Meier
,
D.
Haude
,
T.
Foster
,
M.
Morgenstern
, and
R.
Wiesendanger
, “
A 300 mk ultra-high vacuum scanning tunneling microscope for spin-resolved spectroscopy at high energy resolution
,”
Rev. Sci. Instrum.
75
,
4871
(
2004
).
12.
M.
Gobbi
,
M. A.
Novak
, and
E.
Del Barco
, “
Molecular spintronics
,”
J. Appl. Phys.
125
,
240401
(
2019
).
13.
M.
Denk
,
D.
Queteschiner
,
M.
Hohage
,
A.
Navarro-Quezada
, and
P.
Zeppenfeld
, “
Magnetic switching in Ni/Cu(110)-(2×1)O induced by CoPc
,”
J. Appl. Phys.
125
,
142902
(
2019
).
14.
M.
Cinchetti
,
V. A.
Dediu
, and
L. E.
Hueso
, “
Activating the molecular spinterface
,”
Nat. Mater.
16
,
507
515
(
2017
).
15.
F. A.
Ma’Mari
,
T.
Moorsom
,
G.
Teobaldi
,
W.
Deacon
,
T.
Prokscha
,
H.
Luetkens
,
S.
Lee
,
G. E.
Sterbinsky
,
D. A.
Arena
,
D. A.
MacLaren
,
M.
Flokstra
,
M.
Ali
,
M. C.
Wheeler
,
G.
Burnell
,
B. J.
Hickey
, and
O.
Cespedes
, “
Beating the stoner criterion using molecular interfaces
,”
Nature
524
,
69
73
(
2015
).
16.
H.
Wende
,
M.
Bernien
,
J.
Luo
,
C.
Sorg
,
N.
Ponpandian
,
J.
Kurde
,
J.
Miguel
,
M.
Piantek
,
X.
Xu
,
P.
Eckhold
,
W.
Kuch
,
K.
Baberschke
,
P. M.
Panchmatia
,
B.
Sanyal
,
P. M.
Oppeneer
, and
O.
Eriksson
, “
Substrate-induced magnetic ordering and switching of iron porphyrin molecules
,”
Nat. Mater.
6
,
516
520
(
2007
).
17.
C.
Wäckerlin
,
F.
Donati
,
A.
Singha
,
R.
Baltic
,
A.-C.
Uldry
,
B.
Delley
,
S.
Rusponi
, and
J.
Dreiser
, “
Strong antiferromagnetic exchange between manganese phthalocyanine and ferromagnetic europium oxide
,”
Chem. Commun.
51
,
12958
(
2015
).
18.
A.
Brozyniak
, “
Real-time modulated magneto-optic Kerr effect of (ferro-)magnetic thin films
,” M.Sc. thesis,
Johannes Kepler University Linz
,
Austria
,
2020
.
19.
R. T.
Jacobsen
,
R. B.
Stewart
, and
M.
Jahangiri
, “
Thermodynamic properties of nitrogen from the freezing line to 2000 K at pressures to 1000 MPa
,”
J. Phys. Chem. Ref. Data
15
,
735
909
(
1986
).
20.
P.
Kittel
,
Advances in Cryogenic Engineering
(
Springer
,
1998
), Vol. 43, p.
1366
.
21.
See https://www.cmc.de/page/kapton-folie-klebeband for Kapton-Polyimidfolie von DuPont und Kapton Klebeband,
CMC Klebetechnik GmbH
,
2020
.
22.
H.
Czichos
, “
Wärmeleitfähigkeit von werkstoffen
,” in
Die Grundlagen der Ingenieurwissenschaften, D Werkstoffe
(
Springer
,
2000
), p.
D54
.
23.
H.-J.
Bargel
and
H.
Hilbranns
,
Werkstoffkunde
(
Springer
,
2008
), p.
275
.
24.
I. S.
Grant
and
W. R.
Phillips
,
Electromagnetism
, 2nd ed. (
John Wiley & Sons
,
2008
), p.
31
.
25.
See https://www.thermocoupleinfo.com/type-k-thermocouple.htm for type-K thermocouple,
REOTEMP Instrument Corporation
,
2020
.
26.
L.
Stiny
, “
Elektrische eigenschaften der metalle
,” in
Passive Elektronische Bauelemente
(
Springer
,
2015
), p.
9
.
27.
P.
Weightman
,
D. S.
Martin
,
R. J.
Cole
, and
T.
Farrell
, “
Reflection anisotropy spectroscopy
,”
Rep. Prog. Phys.
68
,
1251
(
2005
).
28.
M.
Wahl
,
T.
Herrmann
,
N.
Esser
, and
W.
Richter
, “
Structure and magneto optical properties of ferromagnetic Ni films grown on Cu(110)
,”
Phys. Status Solidi C
0
,
3002
3006
(
2003
).
29.
M.
Denk
, “
Optical and magnetic spectroscopy of Ni clusters embedded in organic molecular matrices
,” Ph.D. thesis,
Johannes Kepler University Linz
,
Austria
,
2019
.
30.
A.
Aspelmeier
,
M.
Tischer
,
M.
Farle
,
M.
Russo
,
K.
Baberschke
, and
D.
Arvanitis
, “
Ac susceptibility measurements of magnetic monolayers: MCXD, MOKE, and mutual inductance
,”
J. Magn. Magn. Mater.
146
,
256
266
(
1995
).