Jobs Internships

Open positions

The digital data generated annually around the world is now counted in zettabytes, or trillions of billions of bytes. This is equivalent to delivering hundreds of millions of books of data every second. The amount of data generated continues to grow because of the developpement of “Internet of Things”, “Autonomous Driving” , “ Artificial Intelligence” etc … . Its growth is so strong that if the same technologies continued to be used, by 2040 all of the current global electricity consumption would be devoted to data storage.

Spin-based rather than charge-based logic is in principle faster and more energy efficient, which might offer a solution to the handling of the exponential increase in data traffic. Superconductivity, on the other hand, allows zero-dissipation charge transport and the introduction of superconducting circuitry might considerably reduce the energy load associated to data traffic. Recent theoretical and experimental works have shown that the introduction of superconducting elements in spin-based devices can significantly
boost their performance for logic operation. One example of how superconducting order can enhance conventional spintronics is in the superconducting analogue of a giant magnetoresistance junction. The substitution of the metallic spacer between the two ferromagnets with a superconductor has been shown to lead to potentially much higher resistance changes. Moreover, quasiparticles in the superconducting state have been shown to have spin relaxation times that exceed those of normal metals by several orders of magnitude, in principle allowing for much more effective spin transport.

This project aims at joining the expertise of two groups at Cavendish laboratory in Cambridge University, and one group at institut Jean Lamour in Université de Lorraine to investigate how spin transport in superconducting spintronic heterostructures scale in the picosecond regime, comparable to the relaxation times of quasiparticles. Previous experiments have broadly focused on the DC limit whereas the experiments that we propose here are very timely and essential for accessing the potential for the
implementation of ultra-fast device functionalities.

Contacts:

Stéphane Mangin (stephane.mangin@univ-lorraine.fr)

Thomas Hauet (thomas.hauet@univ-lorraine.fr)

Chiara Ciccarelli (https://www.ciccarelli.phy.cam.ac.uk)

Rare Earth Nickelates constitute a class of materials with very rich properties, of strong interest for both fundamental investigation and potential applications. One can first mention the metallic-insulator transition (MIT) occurring in RNiO3 compounds with a perovskite structure. This phenomenon, often accompanied by drastic changes in optical properties, is especially interesting when the MIT temperature is close to room temperature, letting envision applications in the fields of solar panels and microelectronic
devices. A second noticeable example is the occurrence of superconducting properties, recently discovered in Sr-doped RNiO2 compounds. This new phenomenon rose a considerable enthusiasm in the scientific community that launched many studies to analyze and investigate the possible underlying physical mechanisms in comparison with high Tc cuprates materials.

A state of the art instrument for the synthesis and design of complex oxide based nanosystems is installed at the Institut Jean Lamour, as part of the Daυm equipment (Ultra High Vacuum interconnection of synthesis and in-situ analysis tools). This Molecular Beam Epitaxy chamber (DCA), equipped with an ozone generator, enables us to achieve the epitaxial growth with an ultimate control of the atomic arrangement and to design complex and/or artificial architectures at the nanometric scale. These are particularly relevant for nickelates-based systems since observed MIT and superconducting phenomena are extremely sensitive to fine details of structural/interface properties. Such a level of control/design is thus mandatory for the understanding and the control of these properties.

The M2 internship will take place at the Institut Jean Lamour (IJL, ijl.univ-lorraine.fr) in the SPIN team (spin.ijl.cnrs.fr). This team, composed of 15 permanent staff and about 20 PhD and post-doctoral students, has an international recognition in materials design, nanomagnetism and spintronics. The student will work at the design and elaboration of the nickelates systems (films, ordered heterostructures combining several compounds…) using the oxide MBE instrument and benefiting from other in-situ available analysis
(RHEED, XPS). He/she will be also in charge of the full structural characterization by X-ray diffraction and of the investigation of magnetic/electronic transport properties in using dedicated instruments of the Xgamma and magnetometry Competence Centers of IJL.

The candidate must have a Master 1 degree.
Good communication skills, curiosity and a taste for experimental work will be highly appreciated.  

Contacts:

Karine Dumesnil (karine.dumesnil@univ-lorraine.fr)

Nanoscale magnetic objects are already used in current technologies (hard disks, MRAM and magnetic sensors) and are expected to become a key component of several future technologies. For these reasons, they have been the object of particular attention for many years, their size and diversity following the most recent developments in materials science and solid state physics. In 2022, the IJL acquired the few NV center-scanning microscopes available in the world. This new breed of near-field microscope allows the measurement of extremely low stray field patterns (<1.3 µT/Hz-1/2) with nanometer resolution (30nm). During this internship, we will deploy this state-of-the-art microscopy technique to explore magnetic textures of interest for future spintronics development. Magnetic field maps obtained from ultrathin ferromagnetic films or patterned magnetic elements will be compared to magnetic force microscopy (MFM) images and micromagnetic simulations performed on GPU. 

The M2 internship will take place at the Institut Jean Lamour (IJL, ijl.univlorraine.fr) in the SPIN team (spin.ijl.cnrs.fr). This team, composed of 15 permanent
staff and about 20 PhD and post-doctoral students, has an internationally
recognized expertise in nanomagnetism and spintronics.

The successful candidate must have a Master 1 degree (knowledge in magnetism will be valued). Communication skills, rigor, curiosity, creativity, autonomy and a taste for teamwork will be highly appreciated. 

Contacts:

D. Lacour (daniel.lacour@univ-lorraine.fr)

M. Hehn (michel.hehn@univ-lorraine.fr)

Spintronics mixes conventional electronics with magnetism to offer new ways to manipulate devices and to study matter. In recent years, magnetic structures have been used to generate intense THz radiation, by exploiting spin transport and ultrafast optics [1]. These newly developed systems and methods allow us to access magnetic material properties, in a non-contact manner, with access to extremely short timescales [2]. Moreover, the interplay between very high frequency electromagnetic waves (THz) and magnetism is a topic of growing interest [3], with the potential to drastically speed up
spintronic devices. During this internship we will study experimentally the interaction between magnetism and THz in a number of different samples by using intense femtosecond lasers. The student will be trained on a laser bench, and learn extensively about aspects such as spintronics, magnetism and ultrafast optics.

Knowledge of electromagnetism and solid state physics, including optics, magnetism and electronic transport properties is essential.

Knowledge of English (oral and written) is important and knowledge of French would be an advantage.

 [1] T. Seifert, S. Jaiswal, U. Martens, et al., Nat. Photonics 10, 483 (2016). 

[2] J. Gorchon, S. Mangin, M. Hehn, et al., Appl. Phys. Lett. 121, 012402 (2022). 

[3] J. Walowski and M. Münzenberg, J. Appl. Phys. 120, 140901 (2016).

Contacts:

J. Gorchon (jon.gorchon@univ-lorraine.fr)

G. Malinowski (gregory.malinowski@univ-lorraine.fr)

Today, the study of the Spin Hall Effect (SHE) and its reciprocal effect, the Inverse Spin Hall Effect (ISHE), combined with the Spin-Orbit Torque (SOT), represents a major challenge in a sub-field of solidstate physics called spintronic or spin-orbitronic from the point of view of fundamental research and for future applications such as magnetic memories (MRAM) [1,2].
SOT-SHE consists in generating a pure spin current Js, i.e. without any net displacement of electrical charges, in a heavy non-magnetic metal (HM), such as platinum, and injecting it into a ferromagnetic material (FM) to control its magnetization. To create this spin current Js, we look for materials possessing a high Spin-Orbit Coupling (SOC) coefficient. The SOC will deflect injected electrons in opposite directions depending on their spin, which corresponds to the Spin Hall Effect (SHE). It is therefore possible to electrically control the magnetization dynamics of a ferromagnetic material in a structure consisting of the stacking of a thin HM layer on or under a thin FM layer (HM/FM structure). Conversely, in this same type of structure, if the magnetization of the ferromagnetic material FM is set in motion (more precisely in precession), it will be possible to generate in the metal layer HM thanks to the ISHE an electrically measurable charge current Jc [3].
Other spin currents sources have been shown, for example, in interfaces such as Rashba interfaces, FM/topological insulators [4] or simple NM/FM interfaces where the Heavy Metal (HM) is replaced by a so-called “normal” metal (NM) [5].

For a while we have overlooked that some magnetic materials also have strong SOC, and,
consequently, they could generated efficient spin current and SOT [6,7]. We have reported two different symmetries for the self-production of spin current in rare-earth transition metal (RE-TM) ferrimagnetic alloy [8, 9] : SHE-like and SAHE-like (spin anomalous Hall effect). The spin currents created by SAHE are polarized along the magnetization direction and depends on the angle between magnetization and current.
The objective of the M2, then the PhD, we propose is twofold: i) optimize systems based on RE-TM/spin sink towards a robust and efficient self-torque, ii) look for zero-field switching in the developed nanostructures.
There is a possibility of a PhD the thesis is funded by PERP Electronics and that PhD project will be carried out in collaboration with Spintec, in particular with Dr Kevin Garello, specialist in advanced spintronics components. We will study different RE-TM materials such as GdFeCo, CoTb and so on. For the spin-sink layer we look for materials without heavy metal.
The selected PhD student will investigate and evaluate the efficiency of the generation of the pure spin current Js in this type of system in order to get a better understanding of the underlying physics as well as the to optimize the system to perform the electrical switching of magnetization at different time scales ranging from the millisecond to the nano or sub-nanosecond [1]. She/he will work on an experimental setup for electronic transport characterization at the state of the art [8-11]. 

The successful candidate will perform the lithography process to patter the devices. Moreover, she/he will have to take control of the use of the setup, contribute to its improvement, participate to the interpretation of the results obtained. And for the PhD thesis, propose new experiments.

[1] K. Garello et al., Ultrafast magnetization switching by spin-orbit torques, Appl. Phys. Lett. 105, 212402 (2014). 

[2] K. Garello et al., Manufacturable 300mm platform solution for Field-Free Switching SOT-MRAM, 2019
Symposium on VLSI Circuits, T194-T195. 

[3] J.-C. Rojas-Sánchez, et al. Spin pumping and inverse spin Hall effect in platinum: the essential role of spin
memory loss at metallic interfaces, Phys. Rev. Lett. 112, 106602 (2014)

[4] J.-C. Rojas-Sánchez and Albert Fert, Compared efficiencies of conversions between charge and spin current
by spin-orbit interactions in two-and three-dimensional systems, Phys. Rev. Appl. 11, 054049 (2019) 

[5] S. C. Baek, V. P. Amin, et al., Spin currents and spin–orbit torques in ferromagnetic trilayers, Nat. Matt. 17, 509
(2018) 

[6] T. Taniguchi, J. Grollier and M. Stiles, Spin-Transfer Torques Generated by the Anomalous Hall Effect and
Anisotropic Magnetoresistance, Phys. Rev. Appl. 3, 044001 (2015) 

[7] S. Ihama et al., Spin-transfer torque induced by the spin anomalous Hall effect, Nature Electron. 1, 120 (2018) 

[8] D. Céspedes-Berrocal, H. Damas, et al. Current-induced spin torques on single GdFeCo magnetic layers, Adv.
Materials 33, 2007047 (2021) 

[9] H. Damas et al., Ferrimagnet GdFeCo characterization for spin-orbitronics: large field-like and damping-like
torques. Phys. Status Solidi RLL (invited), 2200035 (2022) 

[10] C. Guillemard, et al., Charge-spin current conversion in high quality epitaxial Fe/Pt systems: Isotropic spin Hall
angle along different in-plane crystalline directions, Appl. Phys. Lett. 113, 262404 (2018) 

[11] Er Liu et al. Strain-Enhanced Charge-to-Spin Conversion in Multilayers Grown on Flexible Mica Substrate,
Phys. Rev. Appl. 12, 044074 (2019)

Contacts:

J. Carlos Rojas-Sánchez (juan-carlos.rojas-sanchez@univ-lorraine.fr)

Sébastien Petit-Watelot (sebastien.petit@univ-lorraine.fr)

The race to save space, time and energy in technology products is driving a heavy research on solid state physics. One the one hand, it requires the reduction of the lateral sizes of electronic and mechanical systems to nanometric sizes where the electromagnetic effects of matter combine with the quantum effects. On the other hand, new bulk and interface phenomena have been discovered and implemented in natural or artificial nanomaterials.  In the field of magnetism, the magnetic memory (hard disk drive, MRAM), magnetic sensor and more generally Information Technology industries are particularly at the forefront in the development of atomically designed systems where information can be addressed at time scale down to femtosecond scale and always seek for new functionalities.   

               Asymmetric control of the atomic magnetic moment when applying a mechanical strain, so called piezomagnetism effect, has been predicted theoretically both for anti-ferromagnetic and ferromagnetic materials [1]. This effect is the counterpart of piezoelectricity in the theory of electrostatics.

Nevertheless, so far, it has been only observed in non-usual anti-ferromagnetic materials, like UO2 [2]. Very recently, our “SPIN” research group (https://spin.ijl.cnrs.fr), in collaboration with colleagues from MIT in Boston, made a fundamental discovery in this field. We have observed that the passage of an acoustic wave through a single crystal ferromagnetic thin film generates a variation in magnetic moment that does depend on the strain sign but does not depend on the direction of the magnetic moment, i.e. a piezomagnetic effect [3]          

                To date, we have observed this phenomenon on Co/Ni multilayer thin films where the thickness of Co and Ni are of the order or one to few atomic layers. During the internship then the thesis, we wish to carry out the same measurements for new samples by varying the thicknesses of the Cobalt and Nickel layers, in order to build a theoretical model for the appearance of piezomagnetism, to disentangle interface and bulk effects, and to be able to predict and test other piezomagnets. We also wish to carry out other complementary experiments (in our lab and in synchrotron) in order to characterize and understand the phenomenon on the timescale of the femtosecond up to the quasi-static state.

               The internship and phD thesis will be held at Institut Jean Lamour (IJL, https://ijl.univ-lorraine.fr), one of the largest European laboratories in the field of materials. It is located on the ARTEM site in the city center of Nancy. At IJL, around 50 researchers and students work in the field of magnetism from the atomic scale to applications.

              

[1] A.S. Borovik-Romanov et al. Magnetic properties, International Tables for Crystallography Vol. D 1.5, 105-149 (2006).

[2] M. Jaime et al. Nature communication 8, 99 (2017)

[3] T. Pezeril, T Hauet et al., submitted to Nature Materials (2020)

 

 

Contacts:

Thomas Hauet (thomas.hauet@univ-lorraine.fr)

Karine Dumesnil (karine.Dumesnil@univ-lorraine.fr)

Today, the study of the Spin Hall Effect (SHE) and its reciprocal effect, the Inverse Spin Hall Effect (ISHE), combined with the Spin-Orbit Torque (SOT), represents a major challenge in solid-state physics from the point of view of fundamental research and for future applications such as magnetic memories (MRAM)[1].

SOT-SHE consists in generating a pure spin current Js, i. e. without any net displacement of electrical charges, in a heavy non-magnetic metal (HM), such as Platinum, and injecting it into a ferromagnetic material (FM) to control its magnetization. To create this spin current Js, one method consists injecting an electric current Jc (charge current) in the heavy metal (HM) which must have a high Spin-Orbit Coupling coefficient. The Spin-Orbit Coupling will deflect electrons in opposite directions depending on their Spin, which corresponds to the Spin Hall Effect (SHE). For a given sign of Jc, it therefore results in a transverse pure spin current Js with an associated spin direction. By depositing a thin layer of this HM material on top of a ferromagnetic material (FM), part of the spin current Js is injected into the latter, causing a torque on its magnetization. If the sign of Jc injected into HM is reversed, the torque produced on the FM magnetization will be opposite. It is therefore possible to electrically control the magnetization dynamics of a ferromagnetic material in a structure consisting of the stacking of a thin HM layer on or under a thin FM layer (HM/FM structure).

So far, switching the magnetization of a system via Spin-Orbit Torque has been demonstrated with pulses down to 0.2 nanoseconds in duration [2]. This was considered as “ultrafast” at the time, and switching with shorter pulses was considered impractical. However, recently, we were able to demonstrate switching via Spin-Orbit Torques, using a mere 6 picoseconds electrical current pulse (10-12 seconds) [3]. To this aim, instead of using common commercial current pulse generators that are limited in bandwidth, we built a unique femtosecond laser platform with ultrafast optoelectronic switches, that allow for high intensity few picosecond electrical pulse generation. This achievement poses many new questions and provide us with a new method of study of spin-orbit torques close to various fundamental timescales.

The objective of the internship we propose is to study a number of different samples with our ultrafast current pulse platform, and to benchmark the results and compare them with slow switching experiments.

During the internship, the student will learn how to work around a femtosecond laser setup and generate ultrashort pulses and will be taught the basics of high-frequency electronics and spintronics. The student will need to systematically test different samples under different experimental conditions for switching via spin-orbit torques. The student will likely perform low frequency studies to characterize the spin-orbit torques, which will be useful to compare with the high speed results.

This work will likely result in a PhD. Therefore, working on the internship will be a strong plus for applying to the PhD position. We note that we have a strong collaboration with the Universities of California Berkeley and Riverside on this topic, and the student might have a chance to travel there during the PhD.

References

[1] Y. Niimi and Y. Otani, Reciprocal spin Hall effects in conductors with strong spin-orbit coupling: a review, Rep. Prog. Phys. 78, 124501 (2015).

[2] Garello, K. et al. Ultrafast magnetization switching by spin-orbit torques. Appl. Phys. Lett. 105, 212402 (2014).

[3] Jhuria K. et al. Picosecond spin-orbit torque switching, Accepted in Nature Electronics. arXiv:1912.01377

Contacts:

Jon Gorchon

Spin Orbit Toque (SOT) is a promising technique to induce magnetization switching with high switching speed and low energy. However, there is a serious limitations for SOT switching driven by the spin Hall effect. In usual geometry, an in-plane magnetic field needs to be applied to induce some symmetry-breaking and obtain full magnetization reversal.

In the past we have studied SOT switching in ferrimagnetic materials like TbCo as sketched in the figure below. In this project we wish to study other ferrimagnetic material such as GdCo, DyCo and heterostructures as GdCo/TbCo or GdCo/DyCo

During the internship the candidate will grow the ferrimagnetic materials. The structural and magnetic properties of the multilayers will be characterized. Following previous study (1) we will study the possibility to manipulate magnetization using SOT without any applied magnetic field

This project aims at joining the expertise of one group at the Center for Quantum Phenomena; New York University and one group at Institut Jean Lamour in Université de Lorraine. During the internship the candidate will work at both institutions.

  • Thermal contribution to the Spin-Orbit torque in Metallic Ferrimagnetic systems, Thai Ha Pham, Soong-Geun Je, Pierre Vallobra, T. Fache D. Lacour,  Marie-Claire Cyrille, Gilles Gaudin, Olivier Boulle, Michel Hehn, J-C Rojas-Sánchez, Stéphane Mangin, Rev. Appl. 9, 034028 (2018)

 

 

Contacts:

Pr. Andrew Kent: Center for Quantum Phenomena; New York University https://as.nyu.edu/content/nyu-as/as/faculty/andrew-d-kent.html

Pr. Stéphane Mangin stephane.mangin@univ-lorraine.fr : Institut Jean Lamour, Université de Lorraine, https://spin.ijl.cnrs.fr

In recent years, there has been significant interest in exploiting the unique properties of magnetic domain walls and skyrmions in next-generation magnetic data storage, nonvolatile memories, and computation schemes. Skyrmions are particularly attractive, given their efficient motion in response to the spin-transfer and spin-orbit torques that can arise when an electrical current is passed through a sample. Skyrmions may form in ferromagnetic materials that feature competition between the noncollinear Dzyaloshinskii- Moriya interaction (DMI) and conventional ferromagnetic exchange. One approach to achieving this competition is by sandwiching thin ferromagnetic (FM) layers between heavy metal layers with large spin-orbit coupling (SOC), where an interfacial DMI can develop when there is an asymmetry between the sign or strength of DMI at the top and bottom interfaces of the magnetic layers. As the DMI strength is sensitive to the degree of SOC present at each FM interface, the DMI can be tuned by modifying the deposition conditions or chemical identity of the high-SOC layers

During the internship the candidate will grow magnetic multilayers where skyrmion can be created. The structural and magnetic properties of the multilayers will be characterized. Following previous studies (1,2) we will study nucleating skyrmions using short current pulse and ultra-short laser pulse

This project aims at joining the expertise of one group at Center for Memory and Recording Research in University California San Diego, and one group at Institut Jean Lamour in Université de Lorraine. During the internship the candidate will work at both institutions.

(1) Current-induced generation of skyrmions in Pt/Co/Os/Pt thin films, J. A. Brock, P. Vallobra, R.D. Tolley, S.A. Montoya, S. Mangin, E.E. Fullerton, Phys. Rev. B 102 2 024443 (2020)

(2) Creation of magnetic skyrmion Buble Lattices by Ultrafast laser in Ultrathin Films, S. -G. Je, P. Vallobra, T. Srivastava, J. -C. Rojas-Sánchez, T. H. Pham, M. Hehn, G. Malinowski, C. Baraduc, S. Auffret, G. Gaudin, S. Mangin, H. Béa, and O. Boulle, Nano Letters 18, 7362  (2018)

 

Contacts:

Pr Eric E Fullerton, efullerton@ucsd.edu Center for Memory and Recording Research; University California San Diego https://cmrr.ucsd.edu/research/faculty-profiles/fullerton.html

Pr. Stéphane Mangin stephane.mangin@univ-lorraine.fr : Institut Jean Lamour, Université de Lorraine, https://spin.ijl.cnrs.fr

Ultrafast spin electronics is an emerging field of research that combines the ideas and concepts of magneto-optics and opto-magnetism with spin transport phenomena, supplemented with the possibilities offered by photonics for ultrafast low-dissipative manipulation and transport of information. Both light and spin currents can control magnetic order, though the mechanisms as well as the corresponding time scales and energy dissipations differ. The project aims at the best of both worlds, combining short time scales and non-dissipative propagation of light pulses with nanoscale selectivity and strong interactions of spin currents, to create novel concepts for data technology. The ultimate goals are the creation and implementation of non-volatile, low-dissipative, and ultrafast functional elements.

During the internship the candidate will grow Magnetic Tunnel Junction (MTJ) and characterize their structural and magnetic properties. The magnetization of the magnetic layers of the MTJ will then be manipulated using magnetic field or/and femto-second laser pulses

This project aims at joining the expertise of one at the Research Institut of Electrical Communication in Tohoku University, and one group at institut Jean Lamour in Université de Lorraine. During the internship the candidate will work at both institutions.

Contacts:

Pr. Stéphane Mangin stephane.mangin@univ-lorraine.fr

Institut Jean Lamour,  Université
de Lorraine,
https://spin.ijl.cnrs.fr

Pr Shunsuke Fukami

Research Institut of Electrical Communication,
Tohoku University http://www.spin.riec.tohoku.ac.jp/en/

The digital data generated annually around the world is now counted in zettabytes, or trillions of billions of bytes. This is equivalent to delivering hundreds of millions of books of data every second. The amount of data generated continues to grow because of the developpement of “Internet of Things”, “Autonomous Driving” , “ Artificial Intelligence” etc … . Its growth is so strong that if the same technologies continued to be used, by 2040 all of the current global electricity consumption would be devoted to data storage.

Spin-based rather than charge-based logic is in principle faster and more energy efficient, which might offer a solution to the handling of the exponential increase in data traffic. Superconductivity, on the other hand, allows zero-dissipation charge transport and the introduction of superconducting circuitry might considerably reduce the energy load associated to data traffic.  Recent theoretical and experimental works have shown that the introduction of superconducting elements in spin-based devices can significantly boost their performance for logic operation. One example of how superconducting order can enhance conventional spintronics is in the superconducting analogue of a giant magnetoresistance junction. The substitution of the metallic spacer between the two ferromagnets with a superconductor has been shown to lead to potentially much higher resistance changes. Moreover, quasiparticles in the superconducting state have been shown to have spin relaxation times that exceed those of normal metals by several orders of magnitude, in principle allowing for much more effective spin transport.

This project aims at joining the expertise of two groups at Cavendish laboratory  in Cambridge University, and one group at institut Jean Lamour in Université de Lorraine to investigate how spin transport in superconducting spintronic heterostructures scale in the picosecond regime, comparable to the relaxation times of quasiparticles. Previous experiments have broadly focused on the DC limit whereas the experiments that we propose here are very timely and essential for accessing the potential for the implementation of ultra-fast device functionalities.

Contacts:

Pr. Stéphane Mangin stephane.mangin@univ-lorraine.fr :

Institut Jean Lamour,  Université de Lorraine, https://spin.ijl.cnrs.fr

Pr Jason Robinson

Cavendish Laboratory, Cambridge University https://www.robinson.msm.cam.ac.uk

Dr Chiara Ciccarelli

Cavendish Laboratory, Cambridge University  https://www.ciccarelli.phy.cam.ac.uk