Skip to main content

PhD and Employment Opportunities

General enquiries related to reserach positions in the Electromagnetic and Acoustic Materials group should be directed to the group's Academic Lead, or directly to members of academic staff

Job vacancies

Current vacancies in academic and research positions are listed on the University job vacancy site.

PhD Opportunities

Potential applicants should read more about the academic training and research PhD programmes that can be hosted by the Electromagnetic and Acoustic Materials Group by visiting the Postgraduate Research in Physics or the EPSRC Centre of Doctoral Training in Metamaterials pages.

Funded PhD Studentships

We are currently recruiting to a number of Fully-funded PhD Studentships.  The funder of these projects (e.g. UKRI, university, industry) will normally pay tuition fees, and provide funds for a research, travel and training allowance, and a tax-free stipend to cover living expenses for the student.  Funder requirements and differing fee levels sometimes restrict these positions to UK or EU nationals.  Applicants can find funded studentships listed here, some of which are part of the EPSRC Centre of Doctoral Training in Metamaterials, which are separately listed here. Please see the indiviual adverts for details.

PhD Project Proposals

The list below describes potential reserach projects that are not directly associated with specific funding.  They provide a basis for applicants to write research proposals in any applicantion they might make (e.g. Exeter's Doctoral Training Partnershp or other university funding schemes, non-UK govenment or funding for international students etc).  Applicants are invited to discuss the project proposal with the named supervisors and together write a full studentship reserach proposal.  

Waves propagating in quasicrystals are at a curious point where neither Bloch’s theorem (applicable to periodic media), nor the diffusion approximation (applicable to random media) are appropriate. While quasicrystals can be constructed from a deterministic rule, they do not exhibit the translational invariance that allows propagation to be understood in terms of a single unit cell. However, it has been known for some time that diffraction from photonic quasicrystals exhibits similar sharp peaks as are observed for true periodic crystals, due to the presence of long range order.  This project explores the propagation of electromagnetic waves on the surface of 2D quasicrystals. Such lattices (e.g. those generated using the Fibonacci sequence in particular) have an interesting link to topology via Chern numbers, and fractals.  Although little experimental work has been done to explore these aperiodic structures, our fabrication and characterisation techniques for exploration in the microwave domain naturally lend themselves to this project, and to accompany this experimental strand, there will be a challenging programme of numerical and analytical modelling work for the student to undertake.

Biological tissues appear opaque because they scatter visible light, thus scrambling the spatial information needed to form an image. While this process is completely deterministic, reconstructing the position of the scattering centres from the scattered light is a very difficult inverse problem. It is important to realize that in most cases of interest one doesn't need a perfect reconstruction, but "just" a good guess. So we propose that a trained machine learning algorithm is ideally suited to estimate the most likely scatterers' distribution from the scattered light.

The crystal – a lattice made of atoms, plays the fundamental role in science. Its properties are defined by the geometry, interactions, and potentials, and are typically difficult to control. One possible way to change it is to consider artificial lattices made of optical resonators, where optical engineering and coupling to optically active medium can turn into versatile tool for simulating crystals. For instance, this research direction has brought two highly intriguing fields of physics. First, the precise control of coupling and effective spin orbit interaction allows to study topologically nontrivial lattices with emergent chiral states of polaritons – hybrid light-matter quasiparticles [S. Klembt et al., Nature 562, 552 (2018)].  Second, when light in the optical lattice is coupled to nonlinear medium, this potentially allows to reach the physics of strongly correlated polaritons [A. Greentree et al., Nature Physics 2, 856 (2006)]. Bringing the two directions together is an exciting possibility, which is yet to be explored, and serves the basis for the project. In particular, the unique combination of chiral edge states, multicavity system, and nonlinearity at the level of few quanta, can become a building block for spatially distributed quantum devices even in the presence of noise and dissipation.

Quantum thermodynamics is a new research field, where thermodynamics is studied in the small-system size limit where fluctuations become non-negligible, weak coupling between system and environment cannot be assumed, and a fully quantum mechanical description may be needed. This project is concerned with the application of theoretical tools from the fields of quantum thermodynamics and open quantum systems to the problem of modelling the dynamics, the equilibrium state, and the timescale of equilibration, of a collection of interacting spins that are strongly coupled to their environment. The aim of this project is to find analytical equations that describe the spin dynamics when they are driven by a changing magnetic field and exposed to a heated environment (with potentially multiple temperatures). These will then be used to quantify the importance of quantum mechanical and strong coupling effects in the magnetization dynamics and in the long-time steady state of the spins. 

The lowest temperatures in the universe—of just a few nanokelvins—are nowadays routinely achieved in many labs. Ultracold atomic gases loaded in optical lattices stand out as an ideal platform for quantum simulation, which could help to crack longstanding condensed-matter physics problems, and most notably assist in the design of new metamaterials or drugs. Precise temperature control is indispensable to run reliable simulations and yet, measuring extreme temperatures with high accuracy still remains a formidable open challenge.

In this project we will combine the theory of open quantum systems with quantum metrology to address the problem of temperature sensing deep in the quantum regime. In particular, we will new develop methods to treat strongly coupled open quantum systems under non-linear dissipation, and a novel theory of non-Gaussian quantum metrology. We will then apply these tools to improve the desing, measurement, and data analysis in current thermometry experiments exploiting atomic impurities in Bose–Einstein condensates. This could enable reliable quantum simulation and pave the way towards next-generation material design.

With the advent of Big Data and the Internet of Things, demand for data storage is growing exponentially, placing additional strain on the world’s energy resources.  As scaling of existing technology, such as the hard disk drive, reaches its physical limits, there is an urgent need to develop new recording paradigms that are ultra-dense, ultrafast, and energy efficient.  Switching of magnetic “bits” by ultrafast laser pulses is highly attractive for its potential speed, but the physical mechanisms needed to deliver the required density and energy efficiency are as yet unclear.  All-optical “toggle switching” of ferrimagnetic materials has recently been demonstrated [1] but requires heating, full demagnetization, and a significant time for cooling.  It has been proposed that switching may also be induced by non-thermal mechanisms that are potentially faster and more energy efficient [2].  Optically induced non-thermal switching will be explored in artificial ferrimagnetic trilayer materials that allow the magnetic parameters of the material to be carefully tuned.  The underlying aim is to promote transfer of angular momentum to the electron spins within the material without significant absorption of energy.  The project will make extensive use of the EPSRC funded EXTREMAG facility ( that recently opened within Physics.

[1] T. A. Ostler et al., Nat. Commun3, 666 (2012); [2] S. Mangin et al., Nat. Mater. 13, 291 (2014).

Information is processed within the human brain through the mutual synchronization of a vast 3 dimensional network of neurons that oscillate at a frequency of order 40 Hz.   In contrast, spin transfer oscillators (STOs) are nanoscale non-linear oscillators with frequencies that can be tuned within the GHz frequency range. Synchronization of a chain of 4 STOs has been used for recognition of human vowel sounds [1], while mutual synchronization of 8 x 8 arrays of STOs has recently been demonstrated [2].  Exploiting the high frequency and small footprint of STOs, arrays of STOs can be connected together to form powerful devices for high-speed pattern recognition.  This project addresses a critical gap in present understanding of STO synchronization, namely the relative phase of oscillation in adjacent STOs.  Time resolved scanning Kerr microscopy will be used to directly resolve the GHz oscillations of each STO device so that its relative phase can be determined.  Spatial resolution of down to 50 nm will be achieved through the optimization of a plasmonic antenna mounted on the tip of an atomic force microscope [3]. The project will make extensive use of the EPSRC funded EXTREMAG facility ( that recently opened within Physics.

[1] M. Romera et al., Nature 563, 230 (2018); [2] M. Zahedinejad et al., arXiv:1812.09630v1 (2018); [3] P. S. Keatley et al., Rev. Sci. Instrum. 88, 123708 (2017).

Recently, a new branch of optics has developed, coined Complex Photonics, which in broad terms involves the study of light in multiply scattering environments. When light propagates through a scattering or opaque medium such as frosted glass, a sugar cube, or even living tissue, the spatial information it carries becomes scrambled. This corrupts the formation of images of objects behind or inside scattering environments. The rapid growth of complex photonics has been prompted by the realisation that many of these extremely complicated scattering systems are linear (in electric field) and deterministic in nature. Therefore, we can use digital light shaping technology to measure and reverse their scattering effects – unscrambling the light back to the state it was in before it entered the medium. At the heart of this capability lies the transmission matrix (TM) concept, which describes the scattering as a linear operation relating a set of input spatial light modes incident on one side of the scatterer, to a new set of output modes leaving on the opposite side of the scatterer. Once characterised, seemingly random materials are transformed into exquisitely complicated optical systems that can be used in a variety of previously unexpected scenarios. For example, these developments promise a new generation of photonic devices capable of unscrambling light to peer deep inside living tissue. They can also be used to create a new forms of classical and quantum optical computer, that use the high-dimensional nature of the scattering transform itself to perform calculations at the speed of light.

The aim of this project is to design and build scattering systems with pre-determined scattering properties, and apply them to a range of imaging and optical computing applications.

Quantum computing represents a paradigm-shifting approach to perform calculations, where a genuine quantum setup is used as a processor. From the theoretical perspective, foundations of quantum information science were laid down by proposals for quantum algorithms which can run exponentially faster than classical versions (e.g. Shor’s algorithm). More recently, the problems of quantum chemistry and simulation of materials were considered as the most prominent application for quantum computing, leading to a huge economic impact in the future. This serves as a great motivation for the field. However, to implement them in practise two key ingredients are needed: 1) quantum hardware of sufficient size and quality, where noise level is reduced; 2) quantum software suitable to be run on existing hardware and solve industrially relevant problems. For the first point, the important step was made by achieving quantum supremacy [Google AI group, Nature 574, 505 (2019)] and increasing size of processors. To improve on the second point, novel strategies for quantum simulation of molecules shall be proposed, and shall represent the core goal of the project.

Within the doctoral studies, we will use the best state-of-the-art quantum computing techniques, and merge easy-to-implement variational approaches with novel ways to represent a system Hamiltonian. This shall allow to create efficient algorithms for tackling wide range of problems, and advance the field of quantum chemistry, materials, and machine learning. The project requires both the background in theoretical physics and mathematics, and the desire to challenge existing state of the art in computational science.

Our civilization is defined by the functionalities delivered by devices, ranging from smartphones to airplanes. These functionalities are however constrained by materials available to engineers, when constructing and indeed even conceiving a device. Radically new dynamical properties and advanced functionalities can be created by tailor-tuning the spectra of wave excitations in structured media – so-called metamaterials. Among other waves, ‘surface acoustic waves’ have been investigated for over one hundred years and currently used e.g. for a wide and diverse range of functions, e.g. analogue signal processing in mobile phones. Recently, the field of metamaterials research has expanded to acoustic waves. To date, however, there have been very few suggested ways of designing acoustic metamaterials that can be dynamically reconfigured and tuned. Integration with magnetic materials, well known for their ability to store information e.g. in magnetic hard disk drives, offers an exciting route for achieving non-volatile tuning of acoustic metamaterials. We strive to develop a new class of magneto-acoustic metamaterials in which the role of their building blocks (“meta-atoms”) is played by magneto-acoustic resonators [1,2]. Such metamaterials will add magnetic field tunability to structures aimed to control the propagation of surface acoustic waves, opening intriguing opportunities both in fundamental science and technology. The memory phenomenon inherent to magnetism will enable significant energy savings in non-volatile magneto-acoustic data and signal processing devices. For instance, they would be instantly bootable and could be more easily integrated with the existing magnetic data storage devices. From the point of view of fundamental science, the magneto-acoustic metamaterials will serve as an excellent test bed for studying the physics of wave propagation in non-uniform and non-stationary media.

View the abstract for Controlling acoustic metamaterials with magnetic resonances

  • Please contact Prof Volodymyr Kruglyak, Prof Geoff Nash, Dr. A. V. Shytov, and / or Dr. S. A. R. Horsley (depending on whether an experimental, theoretical or computational project is preferred) via for more information.


[1] O. S. Latcham, et al, “Controlling acoustic waves using magneto-elastic Fano resonances”, Appl. Phys. Lett. 115, 082403 (2019). 

[2] O. S. Latcham, et al, “Hybrid magnetoacoustic metamaterials for ultrasound control”, Appl. Phys. Lett. 117, 102402 (2020).

Antiferromagnetic materials have generated great excitement due to their ability to conduct pure spin currents over micron scale distances.  In materials such as NiO, that have biaxial magnetic anisotropy, it has been proposed that spin amplification may also be possible, paving the way to more energy efficient operation of magnetic random access memory (MRAM) devices.  In this project, the relationship between the propagation of spin current and the underlying antiferromagnetic spin wave excitations will be explored so that the optimum conditions for spin amplification can be determined and realised.  Ultimately, it may be possible to realise multi-stage spin amplification through the creation of multi-layered antiferromagnetic metamaterials. Antiferromagnetic spin waves will be detected by means of ultrafast magneto-optical measurements, while spin current propagation will be detected by x-ray detected ferromagnetic resonance measurements at a synchrotron source as shown schematically within the figure.

View the abstract for Spin current propagation through antiferromagnetic thin film metamaterials

Spin waves (elementary excitations of magnetically ordered materials) boast extreme nonlinearity and modest loss while having micrometre to nanometre wavelengths at GHz frequencies. This presents a unique path towards miniature and powerful yet energy efficient devices for unconventional computing. In this project, you will seek to combine two inherently energy-efficient technology paradigms: (i) magnonics (using spin waves to process signals and data) and (ii) neuromorphic computing (using large-scale integrated systems and analog circuits to solve data-driven problems in a brain-like manner). Going well beyond existing paradigms, we will use nanoscale chiral magnonic resonators [1] as building blocks of artificial neural networks. The power of the networks will be demonstrated by creating magnonics versions of field programmable gate arrays, reservoir computers, and recurrent neural networks. The ultimate efficiency of the devices will be achieved by (a) maximising their magnetic nonlinearity (via spin wave power focusing within chiral magnonic resonators of minimal intrinsic loss); (b) using epitaxial yttrium iron garnet (YIG), which has the lowest known magnetic damping allowed by physics, for thin film magnonic media and resonators; and (c) using wireless delivery of power (minimising Ohmic loss in interconnects). Sensitive to the resonators’ micromagnetic states, such artificial neural networks will be conveniently programmable and trainable within existing paradigms of magnetic data storage. The latter includes magnetic random-access memory (MRAM), which is already compatible with CMOS, while compatibility with other technology paradigms of spintronics will also be sought, explored, and exploited.

View the abstract for Magnonics for Unconventional Computing Devices


[1] V. V. Kruglyak, “Chiral magnonic resonators: Rediscovering the basic magnetic chirality in magnonics”, Appl. Phys. Lett. 119, 200502 (2021). 

Wave excitations of the magnetic order (so called spin waves, quanta of which are called magnons) have a dispersion relation like no other kind of waves: it is nonlinear, anisotropic and often non-reciprocal. In addition, the spin wave dispersion is very sensitive to the sample’s magnetic properties and micromagnetic configuration, so that spin waves are rarely observed to propagate in uniform media. Inspired by and feeding from other, more traditional areas of wave physics, we explore the excitation, propagation and control of spin waves in media with continuously varying properties, i.e. media with a ‘graded magnonic index’. Not only this research is full of exciting fundamental physics questions, but also a new, spin wave based technology for data and signal processing is out there to be grabbed by those who dare to face the challenges!

1. N. J. Whitehead, et al, “A Luneburg lens for spin waves”, Appl. Phys. Lett. 113, 212404 (2018).

2. N. J. Whitehead, et al, “Graded index lenses for spin wave steering”, Phys. Rev. B 100, 094404 (2019).

A stone thrown into water creates waves that propagate in form of circles. An intense and tightly focused ultrashort optical pulse incident on a thin solid film of a magnetic material also launches waves – so called spin waves as well as more familiar acoustic waves. By studying the shape of the excited wave fronts, one can obtain vital information about the waves’ properties, which are in turn inherently related to the interatomic and magnetic interactions in the studied films.  These phenomena can be studied using the time-resolved optically pumped scanning optical microscopy (TROPSOM), a unique technique pioneered in Exeter [1] and now set up within the EPSRC funded Exeter Time-Resolved Magnetism Facility (ExTReMag,  Starting from continuous thin films, we will proceed to magnonic and acoustic metasurfaces. Disentangling optically excited dynamics of their electron, lattice and spin sub-systems is a challenging and exciting task. Yet, a bunch of radically novel functionalities arising from interplay between surface acoustic and spin waves is a prize for those who solve it!  When the primary goal of in-depth understanding of the observed dynamics is achieved, we will proceed to exploration of lithographically shaped, optically driven magnonic and acoustic sources that will enable control of the excited waves.  

1. Y. Au, et al, “Direct excitation of propagating spin waves by focused ultrashort optical pulses”, Phys. Rev. Lett. 110, 097201 (2013). 

Biosensors are vital across a wealth of industries, from healthcare diagnostics, pharmaceuticals to food safety. Multiple complex techniques are currently needed in order to and monitor, and differentiate, specific analytes in real-time. A label-free, compact and highly sensitive biosensing platform capable of detecting major application-specific analytes in real-time would be highly desirable. For example, a dynamic signaling biosensor functionalized to detect major human diseases would be invaluable as a point-of-care diagnostic tool in 21st century patient-focused healthcare.

In this project, we will develop a new class of biosensor based on space-folded all-dielectric metaphotonics (Faraji-Dana, M., Arbabi, E., Arbabi, A. et al. Nat Commun 9, 4196 (2018)) integrated atop a CMOS image sensor (CIS). Leveraging design concepts in planar integrated optics,  metasurface design and spectral sensing, the proposed platform will consist of a multitude of metasurface optical elements (MOEs) lithographically patterned on a thin glass substrate which in-turn is built upon a CIS. MOEs can be designed to engineer light-matter interactions in order to locally control the spectral, amplitude, polarization and phase of light—along with enhancing fluorescence and Raman signals. Top-down lithographic patterning means all MOEs can be fabricated in a single / double-step (two sides of substrate), reducing cost while permitting design complexity. Integrating all optics either side of a single glass substrate folds the optical path, greatly reducing size, and the CIS provides cost effective (<£25) electro-optic readout capability.  

Read the full project description Space-folded metaphotonics for multiplexed biosensing on a single chip.

Please contact the supervisor, Dr Calum Williams for more information. 

Colour cameras utilize absorptive filter arrays atop the image sensor to spectrally discriminate light into red, green and blue (RGB) bands. These colour filter arrays (CFAs) are typically arranged in 2x2 unit cells (RGGB Bayer kernel) and tessellated across the image sensor (Fig.1A). Albeit providing spectral sensitivity, this spatial arrangement means only 50% of the total incident light reaches the green pixels, 25% the blue pixels and 25% the green pixels. Further, as the absorptive dyes themselves only transmit ~40% light, it means in combination ~70% of incident light upon the sensor is lost.

In recent years, nanophotonic colour routers (light sorters) have been proposed as an alternative filtering approach to absorptive CFAs.1,3 Colour routers split the incident light into separate colours (wavelengths) and route the energy to specific pixels (Fig.1B). This meta-optic is composed of many sub-wavelength scatterers, with a designed distribution such that light is routed to different output positions depending on its wavelength. 4 Theoretical efficiencies as high as ~95% have been reported3 yet experimental realization is challenging.

In this project, we will develop generalized light sorters based on inverse-designed meta-optics to efficiently route different wavelengths to different spatial positions. Our approach will employ 2D and ~2.5D meta-optic approaches (Fig.1C) in order to increase manufacturability while maintaining high optical performance (transmission efficiency, spectral sensitivity, angular sensitivity). We will extend our light sorter design scheme to: (1) longer waveband imaging (i.e. short-wave and mid-wave infrared); (2) polarimetric imaging, and (3) plenoptic (depth) imaging.

The research spans fundamental optical physics through to applications, and the student will develop a diverse skillset during the PhD project, including: computational optics, electromagnetic simulation (incl. Lumerical FDTD and COMSOL), nanofabrication within a state-of-the-art cleanroom (incl. e-beam lithography, physical vapour deposition etc.), electro-optic systems characterisation, validation of performance, and advanced data analysis.

Read the full project description Abstract for Inverse-designed meta-optics for light sorting‌.

Please contact the supervisor, Dr Calum Williams for more information.

Photocatalysis plays a pivotal role in addressing environmental concerns and advancing the quest for achieving a net-zero future. However, one of the major bottlenecks in photocatalytic processes is the limited optical activity of the catalytic metals employed. This project aims to revolutionize photocatalysis by leveraging the power of Artificial Intelligence (AI) to design novel catalytic metal-based nanostructures that significantly enhance the interaction between light and matter. By doing so, we strive to elevate energy efficiency in photocatalytic reactions, with a specific focus on CO2 reduction and hydrogen production.  


Read the full project description Two-photon holographic interference lithography

Please contact the supervisor, Dr Changxu Liu for more information. 

Colour cameras utilize absorptive filter arrays atop the image sensor to spectrally discriminate light into red, green and blue (RGB) bands. These colour filter arrays (CFAs) are typically arranged in 2x2 unit cells and tessellated across the image sensor.  Albeit providing spectral sensitivity, this spatial arrangement means only 50% of the total incident light reaches the green pixels, 25% the blue pixels and 25% the green pixels. Further, as the absorptive dyes themselves only transmit ~40% light, it means in combination ~70% of incident light upon the sensor is lost.

In recent years, nanophotonic colour routers (light sorters) have been proposed as an alternative filtering approach to absorptive CFAs. Rather than absorb, colour routers split the incident light into separate colours (wavelengths) and route the energy to specific pixels.  Optical efficiencies as high as ~95% have been reported.

In this project, we will develop generalized light sorters based on inverse-designed meta-optics to efficiently route different wavelengths to different spatial positions. Our approach will employ 2D and ~2.5D meta-optics in order to increase manufacturability while maintaining high optical performance We envisage this project may expand to other imaging and sensing architectures which could benefit from light sorting, including plenoptic imaging and chiral sensing.

Read the full project description Inverse-designed meta-optics for light sorting.

Please contact the supervisor, Dr Calum Williams for more information. 

Current computing systems are based on the manipulation and transfer of information using electric charge or voltage in semiconductor devices and circuits such as transistors and digital logic gates and circuits, with copper based interconnects.  With the increased integration of semiconductor devices and threats of reaching device limits and increased power densities, alternative computing technologies are in demand to address current challenges in semiconductor devices and memories.

Spin-waves are oscillatory excitations of magnetic moments in magnetic material with wave-like nature that can propagate along magnetic waveguides or interconnects with intrinsically low energy and at high speeds, thus providing an alternative to semiconductor based devices. Spin-wave magnetic devices have thus been successfully developed and demonstrated at laboratory level to realise Boolean logic and circuits such as NOT, NOR and NAND logic gates, and multiplexer and majority functions.  

Spin-wave computing technology, however, is still at its infancy and face a number of challenges to enable their practicable device realisation and commercialisation, such as scaling of device dimensions, fan-out and energy efficiency of the generation and detection of spin-waves.  This project investigates, using multi-physics modelling and simulation, the fundamentals of the spin-wave propagation in magnetic devices and the effects of the waveguide materials and transducer topology on device performance to address the above challenges.


Supervisors: Prof Mikhail Portnoi and Dr Eros Mariani



This theoretical project aims at unveiling the electronic and optical properties of novel Weyl metasurfaces hosting exotic quasiparticles with unusual “designer” dispersion, including tilted Dirac cones in the band structure.



Developing novel materials and metamaterials with “designer” spectra for electrons, photons, magnons and phonons is the growing trend in contemporary condensed matter physics. The tide of research is turning towards theoretical predictions followed by experimental observations of exotic quasiparticles in solid state systems such as Weyl and Majorana fermions. Since the 1980’s, two-dimensional (2D) systems provided a particularly rich playground for the discovery of new quasiparticles, including fractionally charged anyons in quantum Hall systems and massless Dirac fermions in graphene and topological insulators as the most spectacular examples. Simultaneously, the most recent advances in nanotechnology include the development of van der Waals heterostructures (artificial few-layer materials held by van der Waals forces) and finding a way of producing carbon nanotubes of selected chirality.

The proposed theoretical PhD research will start by addressing the electronic and optical properties of novel designer 2D van der Waals materials formed by a planar arrays of single-walled carbon nanotubes. We will then extend the core ideas from this setup to other 2D Weyl metasurfaces (two-dimensional metamaterials with energy spectrum described by Weyl equation), including Borophene and spatially modulated graphene layers.

Single-walled carbon nanotubes are long cylindrical molecules with electronic properties defined entirely by the way they are rolled from a graphene sheet. Namely, they can be semiconducting with a band gap up to several electron-volts, metallic without any band gap or quasi-metallic with a tiny band gap of a few meV induced by curvature effects. Combining the tubes in a regular planar array should result in a strongly anisotropic dispersion of the emerging 2D quasiparticles. The most interesting results are expected for arrays of metallic and quasi-metallic nanotubes for which the motion normal to the nanotube axis will result in the most dramatic changes in the electronic properties. In particular, a band gap may be opened in an array of metallic tubes or collapse for quasi-metallic nanotubes. The motion normal to the tube axis will also result in lifting the valley degeneracy in highly-symmetric zigzag nanotubes leading to tilted Dirac cones supporting a new type of 2D Weyl fermions with hyperbolic equipotential lines.

An analytical description of the peculiar low-energy electronic dispersion in this system will be at the core of the proposed research. We will study a plethora of unique physical effects stemming from this dispersion including unusual magneto-transport phenomena and inter-band dipole transitions, which are expected to be in the highly sought-after terahertz frequency range.


Please contact Prof Mikhail Portnoi ( or Dr Eros Mariani ( for additional information.


There is a mismatch between astrophysical observations and theory. Observations of galaxy dynamics and gravitational lensing are consistent with there being considerably more mass in galaxies than we can see. The prevailing theory postulates new species of “dark matter” particles, which have significant mass but insignificant interaction with the electromagnetic field [1], thus being invisible to close to earth observations of astrophysical objects. This project builds on recent proposals to use resonant wire-medium based metamaterials [2,3,4] as the basis for detectors that enhance the interaction between candidate dark matter particles and the electromagnetic field.

One candidate dark matter particle is the axion, a field introduced in 1977 to explain the observation of zero (or at least very weak) CP violation by the strong force [5]. This theory modifies the Lagrangian density for the standard model to contain both our usual massless electromagnetic field, and an axion field with non-zero mass. The interaction between the two is parametrized by an interaction constant χ.

In infinite free space, conservation laws prevent photon creation from the axion field, hence this matter being “dark”. However, within a plasma subject to a strong magnetic field, the interaction can be greatly increased [2-4], becoming resonant when the axion rest mass frequency equals the plasma frequency. The theoretically expected axion mass puts this resonant frequency in the terahertz or lower. In this frequency range, artificial plasma wire-based metamaterials have been developed [3], with a tunable plasma frequency. This makes it feasible to develop metamaterial based detectors for axionic dark matter particles. This theoretical project will investigate the design and feasibility of these detectors.


In this project we will:

(1) Examine the theory of electromagnetism coupled to axions and understand the general conditions for enhancing their interaction with materials, as well as how to numerically simulate these effects through e.g. modifying COMSOL multiphysics. We will consider a broader class of materials than the wire media in [2], including both dielectrics and metasurfaces.

(2) Search for candidate metamaterials for enhancing the rate of axion to photon conversion. This will be done using both analytic theory and numerical simulations using e.g. adjoint optimization to develop inhomogeneous tunable structures where the coupling to axions is maximized within some region, analogous to increasing the local density of states for an antenna [6].

(3) Compare metamaterial-based axion detectors to existing cavity-based designs.

(4) More broadly explore the topic of axion electrodynamics and the similarity between bianisotropic metamaterials and coupling to the proposed axion field. This may allow us to also use the designs developed in (1-2) to enhance the usually small resonant bianisotropic response of metamaterials.

View the abstract for Tunable metamaterials and the search for dark matter


[1] J. L. Feng “Dark Matter Candidates from Particle Physics and Methods of Detection”, Annu. Rev. Astron. Astrophys. 48, 495 (2010).
[2] M. Lawson, A. J. Millar, M. Pancaldi, E. Vitagliano, and F. Wilczek, “Tunable Axion Plasma Haloscopes”, Phys. Rev. Lett. 123, 141802 (2019)
[3] R. Balafendie, C. Simovski, A. Millar, P. Belov, “Wire metamaterial use for dark matter detection”, 2022 Sixteenth International Congress on Artificial Materials for Novel Wave Phenomena (Metamaterials), 1-3 (2022).
[4] R. Cervantes et al/. “Search for 70 μeV Dark Photon Dark Matter with a Dielectrically Loaded Multiwavelength Microwave Cavity” Phys. Rev. Lett. 129, 201301 (2022).
[5] R. D. Peccei and H. R. Quinn, “CP Conservation in the Presence of Instantons”, Phys. Rev. Lett. 38, 1440 (1977).
[6] S. Mignuzzi, S. Vezzoli, S. A. R. Horsley, W. L. Barnes, S. A. Maier, and R. Sapienza “Nanoscale Design of the Local Density of Optical States”, Nano Lett. 19, 113-117 (2019).