Quantum Computing
Research Topics
& Complete Guide
The definitive resource for students entering quantum computing research — covering the field’s foundational principles, 90+ rigorously framed research topics across eight sub-domains, a complete guide to quantum hardware platforms, methodology frameworks for different research types, and step-by-step guidance from identifying a research gap to producing a publishable contribution. Designed for students from advanced undergraduate through doctoral level.
⚛️ Need expert help with your quantum computing research paper or dissertation? Our specialists are ready.
Get Expert Help →What Is Quantum Computing Research — and Why Is It One of Science’s Greatest Frontiers?
Quantum computing research is the scientific investigation of computation that exploits quantum mechanical phenomena — superposition (the ability of quantum bits to represent 0 and 1 simultaneously), entanglement (the correlated quantum states of physically separated particles), and interference (the amplification of correct computational paths and cancellation of incorrect ones) — to solve certain categories of problems exponentially or polynomially faster than any known classical algorithm. As a research discipline, it spans the most abstract mathematical foundations of computation through the engineering frontier of qubit fabrication, encompassing quantum algorithm design, quantum error correction, post-quantum cryptography, quantum hardware physics, quantum simulation of chemistry and materials, quantum machine learning, quantum networking, and quantum information theory. This guide covers all eight sub-domains with 90+ specific, research-ready topic ideas and the analytical frameworks required to develop any of them into a genuine scholarly contribution.
In 2019, Google claimed “quantum supremacy” — demonstrating that their Sycamore processor completed in approximately 200 seconds a sampling task they estimated would take the world’s fastest classical supercomputer 10,000 years. This milestone, contested in some of its details and narrow in its practical scope, nonetheless announced an era. Quantum computing had moved from a theoretical curiosity into a genuine engineering competition with national security, pharmaceutical, financial, and materials-science implications that governments and corporations are investing billions to capture.
For student researchers, this moment creates extraordinary opportunity. Quantum computing is young enough that genuinely fundamental open problems exist at accessible levels — a determined graduate student with strong mathematical foundations can make real contributions to quantum error correction, quantum algorithm optimisation, post-quantum cryptographic analysis, or quantum machine learning benchmarking. The field is also, by necessity, interdisciplinary: the best quantum computing researchers combine expertise from computer science, physics, and mathematics in ways that distinguish their work from what any single discipline could produce alone.
Understanding what kind of quantum computing researcher you want to be — theorist, experimentalist, engineer, or algorithm designer — is the first step toward identifying a research direction. The eight domains covered in this guide require different foundational skills, different methodologies, and different publication venues. Choosing wisely, based on your genuine strengths and interests, is as important as the topic itself.
The Foundational Physics Every Researcher Must Understand
A qubit is the fundamental unit of quantum information. Unlike a classical bit, which is definitively 0 or 1, a qubit exists in a superposition of both states simultaneously — described mathematically as |ψ⟩ = α|0⟩ + β|1⟩, where α and β are complex amplitudes satisfying |α|² + |β|² = 1. The probabilities of measuring 0 or 1 are |α|² and |β|² respectively. Measurement collapses the superposition irreversibly.
Entanglement is the non-classical correlation between two or more qubits — when qubits are entangled, measuring one instantly determines information about the other(s) regardless of the physical distance between them. The maximally entangled Bell states, such as (|00⟩ + |11⟩)/√2, are the fundamental resource for quantum teleportation, superdense coding, and quantum key distribution. Interference is the mechanism through which quantum algorithms achieve their power: by constructing quantum circuits that constructively interfere the probability amplitudes of correct answers and destructively interfere incorrect ones, algorithms like Grover’s search and Shor’s factoring achieve their speedups. All three phenomena — superposition, entanglement, and interference — are non-classical resources that quantum algorithms must use deliberately; any quantum circuit that fails to exploit all three provides no speedup over classical computation.
Mathematical Prerequisites by Research Domain
- Quantum algorithms and complexity theory: Linear algebra (Hilbert spaces, unitary operators, tensor products), complex analysis, probability theory, computational complexity (P, NP, BPP, BQP), group theory, number theory (for cryptographic applications)
- Quantum error correction: Linear algebra, coding theory (Hamming codes, stabiliser formalism), abstract algebra (group theory, field theory), probability theory
- Quantum hardware physics: Undergraduate quantum mechanics, electromagnetism, solid-state physics or atomic physics depending on qubit modality, differential equations, signal processing
- Quantum machine learning: Linear algebra, probability and statistics, classical machine learning theory, optimisation (gradient methods, variational methods), quantum circuit theory
- Post-quantum cryptography: Classical cryptography, lattice theory, number theory, computational complexity, algebra (rings, modules, ideal lattices)
The Quantum Computing Research Landscape in 2025: Where We Are and What Remains Unsolved
The current era of quantum computing is defined by a single technological constraint: noise. Every physical qubit implementation — superconducting, trapped ion, photonic, spin, neutral atom, topological — suffers from decoherence, gate errors, and measurement errors at rates that make extended quantum computations unreliable without error correction. The NISQ (Noisy Intermediate-Scale Quantum) era, a term coined by physicist John Preskill in 2018, describes the present period in which quantum processors of 50–1000+ qubits are available but not fault-tolerant — capable of remarkable demonstrations but not yet of the deep circuits that theoretically promise exponential advantage over classical computers.
Where We Are Now (2025)
Quantum processors of 100–1000+ physical qubits from IBM, Google, IonQ, Quantinuum, and others. Quantum volume improving. First demonstrations of error-corrected logical qubits. Variational algorithms showing promise on specific problem classes. No definitive quantum advantage on practically important problems demonstrated yet.
The Fault-Tolerant Threshold
Achieving practical fault-tolerant quantum computation (FTQC) requires physical error rates below approximately 10⁻³ per gate and scalable qubit connectivity. Current best rates: superconducting ~10⁻³, trapped ion ~10⁻⁴. FTQC will require thousands to millions of physical qubits per logical qubit. First industry-scale demonstrations expected late this decade.
What Fault-Tolerant QC Enables
With fault-tolerant QC: Shor’s algorithm threatens RSA and ECC at scale; full quantum simulation of molecules enables drug and materials discovery at unprecedented resolution; quantum optimisation may transform logistics, finance, and energy; quantum networks enable provably secure communication across global distances.
The Quantum Advantage Gap
Despite theoretical proofs of quantum speedup for specific problems (Shor’s factoring, Grover’s search, HHL linear systems, quantum simulation), no practical demonstration of quantum advantage on a problem that classical computers cannot solve efficiently in practice has been definitively established. Proving quantum advantage on a problem of real-world relevance — not just a contrived sampling problem — is perhaps the field’s most pressing empirical challenge. Research into variational quantum eigensolvers (VQE), the quantum approximate optimisation algorithm (QAOA), and quantum chemistry simulation all aim at this target, with each sub-community developing increasingly sophisticated benchmarking methodologies to make honest comparisons with classical state-of-the-art.
Scalable Quantum Error Correction
Fault-tolerant quantum computation requires quantum error correction (QEC) — encoding each logical qubit in many physical qubits so that errors can be detected and corrected without disturbing the quantum information. The surface code is currently the leading QEC architecture: it requires approximately 1000 physical qubits per logical qubit at current error rates. Achieving practical FTQC therefore requires either dramatically reducing physical error rates or discovering QEC codes with far better thresholds. Both directions are active research frontiers with significant open problems at accessible levels for graduate students.
The BQP vs. Classical Complexity Landscape
BQP (Bounded-error Quantum Polynomial time) is the complexity class of decision problems efficiently solvable by a quantum computer. We know that P ⊆ BQP ⊆ PSPACE, and that BQP ⊄ BPP (quantum computers can solve problems classical probabilistic computers cannot solve efficiently) — but the proof relies on oracle separations rather than absolute complexity results. The most important open question in quantum complexity theory is whether BQP contains problems outside NP, or whether quantum speedups are restricted to problems within NP. Shor’s factoring algorithm operates on a problem in NP (but not known to be NP-complete), while the quantum sampling problems used to demonstrate supremacy are believed to be outside NP entirely. Understanding the precise relationship between BQP and the classical complexity hierarchy remains one of the deepest open problems in theoretical computer science.
We are at the very beginning of the quantum age. The machines we have now are like the vacuum tube computers of the 1950s — powerful demonstrations of a new computational paradigm, but nowhere near the devices that will change the world.
— Adapted from John Preskill, coiner of the term “quantum supremacy” and director of the Institute for Quantum Information and Matter, CaltechQuantum Algorithm Research Topics
Designing, analysing, and benchmarking algorithms that exploit quantum mechanical advantage
Quantum algorithm research sits at the intersection of physics, mathematics, and theoretical computer science — designing circuits that exploit superposition, entanglement, and interference to solve problems faster than any classical algorithm. The canonical quantum algorithms (Shor’s factoring, Grover’s unstructured search, the quantum Fourier transform, HHL for linear systems) were developed in the 1990s and 2000s, but the field remains wide open. The challenge of identifying new problems in BQP that are outside BPP — particularly problems with real-world relevance — is both theoretically deep and practically urgent.
Quantum Approximate Optimisation Algorithm (QAOA): Depth vs. Approximation Trade-offs
Analysing how QAOA circuit depth (p) affects approximation ratio for combinatorial optimisation problems including Max-Cut, graph colouring, and vehicle routing, and comparing against classical approximation algorithms at equivalent circuit depths.
Research Question: Does increasing QAOA depth beyond p = 10 provide statistically significant approximation ratio improvements for Max-3-SAT instances compared to the Goemans-Williamson algorithm on current NISQ hardware?Variational Quantum Eigensolver Convergence Under Shot Noise
Characterising how measurement noise (shot noise) from finite circuit repetitions affects VQE convergence to ground state energies for molecular Hamiltonians, and evaluating noise-mitigation strategies for NISQ implementations.
Research Question: What minimum shot count is required for VQE to converge to chemical accuracy (1 kcal/mol) for the hydrogen chain H₄ using the UCCSD ansatz, and how does this scale with system size?Grover’s Algorithm Extensions for Structured Search Problems
Extending Grover’s unstructured search to problems with partial structure — constraint satisfaction, database search with domain-specific oracles — and measuring the quantum speedup achievable over classical branch-and-bound algorithms for structured instances.
Research Question: Can quantum walk-based generalisations of Grover’s search achieve sub-quadratic speedup over classical branch-and-bound for structured constraint satisfaction problems with exploitable local structure?Quantum Algorithms for Graph Problems: Triangle Finding and Graph Isomorphism
Designing and analysing quantum algorithms for graph problems including triangle detection, shortest paths, and graph isomorphism, seeking polynomial quantum speedups over classical algorithms using quantum walks and phase estimation.
Research Question: Can quantum walk algorithms for triangle detection achieve a sub-cubic quantum query complexity advantage over classical algorithms, and does this advantage extend to graphs with bounded degree distributions?Quantum Linear Algebra: HHL Algorithm Utility and Practical Constraints
Critically evaluating the practical utility of the Harrow-Hassidim-Lloyd (HHL) algorithm for linear systems, examining the dequantisation challenges (classical algorithms achieving equivalent complexity) and identifying problem instances where genuine quantum advantage survives.
Research Question: For which categories of sparse linear systems with efficient quantum state preparation does the HHL algorithm provide polynomial quantum speedup that is not recoverable by Tang’s classical sampling-based dequantisation?Quantum Speedup for Monte Carlo Methods in Financial Risk Modelling
Applying quantum amplitude estimation (a generalisation of Grover’s algorithm) to accelerate Monte Carlo risk calculations in financial derivative pricing and portfolio risk assessment, evaluating the quantum advantage under realistic circuit depth constraints.
Research Question: Does quantum amplitude estimation provide a quadratic speedup over classical Monte Carlo for option pricing problems that remains viable at the circuit depths achievable on near-term hardware with error mitigation?Quantum Algorithms for Differential Equation Solving: Fluid Dynamics Applications
Investigating quantum algorithms for solving systems of differential equations arising in computational fluid dynamics and climate modelling, examining the state preparation bottleneck and the data extraction problem that affect practical quantum advantage claims.
Research Question: What is the practical quantum advantage of quantum linear differential equation solvers for fluid dynamics simulations when the costs of quantum state preparation and measurement sampling are fully accounted?Quantum Walks for Network Centrality and Community Detection
Applying continuous-time and discrete-time quantum walks on complex networks to compute centrality measures and detect community structure, comparing performance against classical spectral methods and graph neural networks.
Research Question: Do quantum walk-based centrality measures on protein interaction networks provide biologically meaningful rankings that differ from classical PageRank, and do these differences correlate with known functional annotations?Parameterised Quantum Circuits: Trainability and Barren Plateau Avoidance
Investigating the barren plateau problem in parameterised quantum circuits — where gradients vanish exponentially with circuit width, making training infeasible — and evaluating initialisation strategies, layer-by-layer training, and circuit architecture choices that mitigate this pathology.
Research Question: Do problem-inspired ansatz circuit architectures for quantum optimisation problems avoid barren plateaus at circuit widths relevant for near-term applications, and what structural circuit properties predict trainability?Quantum Algorithms for Drug Target Identification via Protein Folding
Applying variational quantum algorithms and quantum annealing to protein structure prediction problems (energy minimisation on lattice and off-lattice protein models), evaluating against AlphaFold2 baselines on benchmark protein datasets.
Research Question: Do quantum annealing and VQE approaches to protein lattice model energy minimisation scale more favourably than classical simulated annealing and MCMC methods as protein chain length increases beyond 50 residues?Quantum Algorithm Benchmarking: Beyond Quantum Volume
Developing comprehensive, application-relevant benchmarking methodologies for quantum processors that go beyond single-metric measures like quantum volume, evaluating algorithmic tasks across the spectrum from random circuit sampling to structured optimisation.
Research Question: Does the Q-Score benchmarking metric for quantum optimisation correlate more reliably with practical quantum processor capability for relevant industrial optimisation tasks than IBM’s Quantum Volume metric?Adiabatic Quantum Computing and the Quantum Adiabatic Theorem in Practice
Examining the gap between the theoretical quantum adiabatic theorem (which guarantees computation correctness for sufficiently slow evolution) and practical adiabatic quantum computing implementations, characterising the minimum spectral gap and evolution time requirements for realistic optimisation problems.
Research Question: For which classes of quadratic unconstrained binary optimisation (QUBO) problems does adiabatic quantum computing on D-Wave hardware find provably optimal solutions at problem sizes where classical branch-and-bound requires exponential time?Quantum Signal Processing and Quantum Singular Value Transformation Unification
Exploring the quantum signal processing (QSP) and quantum singular value transformation (QSVT) framework as a unified approach to quantum algorithm design, and applying it to derive new algorithms for matrix analysis and linear algebra problems.
Research Question: Can the QSVT framework be applied to derive quantum algorithms for spectral density estimation of sparse Hamiltonians that achieve better gate complexity than existing quantum phase estimation approaches?Quantum Error Correction Research Topics
The path to fault-tolerant quantum computation — protecting quantum information from noise
Quantum error correction (QEC) is arguably the most critical challenge on the path to practical quantum computing. Unlike classical bits, qubits cannot be copied (the no-cloning theorem forbids it), and measuring a qubit destroys its quantum state. QEC must therefore detect and correct errors — bit flips (X errors), phase flips (Z errors), and combinations thereof — using entanglement and indirect syndrome measurements that reveal the error type without collapsing the encoded quantum information. The stabiliser formalism, developed by Daniel Gottesman, provides the mathematical framework for most practical QEC codes, including the surface code, the colour code, and the Bacon-Shor code.
Surface Code Thresholds on Realistic Noise Models
Computing fault-tolerance thresholds of the surface code under realistic noise models that include correlated errors, leakage to non-computational states, and crosstalk — going beyond the independent depolarising noise assumption of theoretical analyses.
Research Question: How does the fault-tolerance threshold of the rotated surface code change when correlated two-qubit errors from ZZ coupling crosstalk are included in the noise model, and which decoding algorithms are most robust to this correlation structure?Machine Learning Decoders for Quantum Error Correction Codes
Training neural network and reinforcement learning decoders for the surface code and other topological codes, evaluating whether learned decoders can approach or exceed the performance of minimum-weight perfect matching (MWPM) decoders at lower computational cost.
Research Question: Do recurrent neural network decoders for the surface code achieve threshold-beating performance at code distances d ≥ 7 with lower online computational cost than MWPM, enabling real-time decoding at hardware clock speeds?LDPC Quantum Error Correcting Codes: Encoding Rate vs. Threshold
Designing and analysing quantum low-density parity-check (LDPC) codes — which offer better encoding rates than the surface code — evaluating their fault-tolerance thresholds and decoder performance using belief propagation and other message-passing algorithms.
Research Question: Do good quantum LDPC codes (achieving constant encoding rate and linear distance) maintain fault-tolerance thresholds comparable to the surface code under circuit-level noise when decoded with iterative belief propagation?Leakage Reduction Units in Superconducting Qubit Systems
Designing and evaluating leakage reduction unit (LRU) circuits for superconducting transmon qubits, where leakage to non-computational states (|2⟩, |3⟩) bypasses standard error correction protocols and must be actively suppressed.
Research Question: What is the minimum overhead in physical qubits and gate cycles of leakage reduction units that maintain surface code performance on superconducting processors with measured leakage rates of 10⁻³ per gate?Fault-Tolerant Magic State Distillation Efficiency
Analysing and optimising magic state distillation protocols — which are required to implement non-Clifford gates (T gates) fault-tolerantly — evaluating overhead reduction strategies and comparing distillation factory architectures.
Research Question: What is the minimum spacetime overhead of magic state distillation required to implement a T gate with logical error rate below 10⁻¹⁰ using the 15-to-1 distillation protocol on a surface code processor at the achievable physical error rates of current hardware?Autonomous Quantum Error Correction Without Classical Measurement
Designing passive error correction schemes using engineered dissipation and feedback without explicit classical syndrome measurement, evaluating their robustness and scalability for superconducting and photonic qubit platforms.
Research Question: Can autonomous quantum error correction using engineered reservoir coupling correct both bit-flip and phase-flip errors simultaneously on a two-qubit encoding with correction timescales faster than the natural decoherence time?Concatenated vs. Topological Codes: Resources for Fault-Tolerant Arithmetic
Comparing resource requirements (physical qubit count, gate count, circuit depth) of concatenated code architectures versus topological (surface code) architectures for implementing fault-tolerant quantum arithmetic circuits needed by Shor’s algorithm.
Research Question: For implementing Shor’s factoring algorithm against 2048-bit RSA keys, do surface code or concatenated Steane code architectures require fewer physical qubits when both are optimised for the same logical error rate at current physical gate error rates?Quantum Error Mitigation for NISQ Algorithms: Probabilistic Error Cancellation
Implementing and evaluating quantum error mitigation techniques (zero-noise extrapolation, probabilistic error cancellation, symmetry verification) for variational quantum algorithms on NISQ hardware, measuring the sampling overhead and accuracy improvement.
Research Question: Does probabilistic error cancellation for VQE reduce systematic energy estimation bias to chemical accuracy levels on a 20-qubit trapped-ion processor, and what sampling overhead does this require relative to unmitigated measurements?Topological Quantum Codes: Colour Codes and Their Transversal Gate Sets
Investigating the transversal gate sets achievable by topological colour codes and their implications for fault-tolerant gate compilation, examining whether colour codes reduce magic state distillation overhead compared to surface codes.
Research Question: Do 3D topological colour codes that admit transversal T gates achieve lower total resource overhead for fault-tolerant universal quantum computation than surface code plus magic state distillation at equivalent logical error rates?Real-Time Syndrome Decoding Latency for Fault-Tolerant Quantum Computation
Designing real-time classical decoding hardware (FPGA, ASIC) for surface code syndrome decoding that operates within the measurement cycle time of superconducting qubit hardware (~1 microsecond), enabling genuinely real-time error correction feedback.
Research Question: Can an FPGA implementation of the union-find decoder for the surface code decode distance-7 code syndromes within 1 microsecond using hardware-parallel implementations, and what code distance is achievable within this latency budget?Bosonic Qubit Codes: Cat Qubits and GKP Encoding for Photonic Systems
Investigating bosonic QEC codes — including cat qubit encoding in superconducting cavities and Gottesman-Kitaev-Preskill (GKP) codes in optical systems — which encode a qubit in the state space of a harmonic oscillator to exploit the large Hilbert space of bosonic modes.
Research Question: Do GKP-encoded photonic qubits with homodyne syndrome measurement achieve lower logical error rates than dual-rail photon-number-resolving encoded qubits at equivalent photon loss rates in waveguide quantum electrodynamics platforms?Quantum Error Correction in the Presence of Temporal Correlations
Characterising and mitigating temporally correlated noise (1/f noise, two-level system fluctuators) in superconducting qubits that violates the Markovian error model assumed by standard QEC analyses, developing adaptive decoding strategies.
Research Question: How does temporally correlated 1/f charge noise in superconducting transmon qubits affect surface code threshold estimates, and can dynamical decoupling sequences interleaved with syndrome measurements restore Markovian-equivalent performance?Post-Quantum Cryptography Research Topics
Securing digital infrastructure against the quantum threat — algorithms, protocols, and migration
The cryptographic implications of quantum computing are profound and immediate. Shor’s algorithm, running on a sufficiently large fault-tolerant quantum computer, breaks RSA, DSA, and elliptic curve cryptography — the foundation of almost all currently deployed public-key cryptography — in polynomial time. The timeline for “cryptographically relevant” quantum computers is uncertain but plausible within 10–20 years, driving an urgent global effort to develop, standardise, and deploy post-quantum cryptography (PQC). NIST completed its initial PQC standardisation process in 2024, selecting CRYSTALS-Kyber (now ML-KEM), CRYSTALS-Dilithium (ML-DSA), and FALCON for standardisation — but significant research questions remain open around their security, performance, and deployment.
Lattice-Based Cryptography Security: Hardness of LWE Under Quantum Attack
Analysing the quantum security of the Learning With Errors (LWE) problem — the computational hardness assumption underlying ML-KEM and ML-DSA — evaluating whether quantum algorithms provide any sub-exponential advantage over the best classical lattice reduction algorithms.
Research Question: Do quantum lattice sieving algorithms (quantum-accelerated BKZ) provide asymptotic security reductions against LWE that require meaningfully larger lattice parameters than classical sieving-based attacks for equivalent security levels at NIST security levels 1, 3, and 5?Implementation Attacks on NIST PQC Standards: Side-Channel Analysis
Investigating power and electromagnetic side-channel vulnerabilities in hardware and software implementations of ML-KEM and ML-DSA, developing and evaluating countermeasures including masked implementations and constant-time algorithms.
Research Question: Are reference implementations of ML-KEM vulnerable to power analysis side-channel attacks that recover secret key material within a feasible measurement budget, and does the NIST recommended constant-time implementation guidance fully mitigate this risk?Hybrid Classical-PQC TLS Protocol Design and Performance Overhead
Designing and evaluating hybrid TLS handshake protocols that combine classical ECDH and post-quantum ML-KEM for key exchange, measuring handshake latency, certificate size, and bandwidth overhead in realistic network conditions.
Research Question: What is the practical TLS handshake latency and bandwidth overhead of hybrid ECDH + ML-KEM key exchange compared to ECDH-only, and at what network latency and bandwidth constraints does the overhead become operationally significant?“Harvest Now, Decrypt Later” Attacks: Assessing the Real Timeline
Modelling the threat posed by adversaries currently harvesting encrypted traffic for decryption once cryptographically relevant quantum computers become available, and evaluating which categories of data are most at risk given realistic quantum development timelines.
Research Question: Given the most recent engineering progress data on superconducting qubit development trajectories, what probability distribution over quantum computer capability timelines is most consistent with the evidence, and which classified data categories face the highest expected harvest-now-decrypt-later risk?Code-Based Cryptography: McEliece Variants for Compact Public Keys
Designing and analysing variants of the McEliece cryptosystem that reduce public key size — the system’s main practical limitation — using quasi-cyclic codes, moderate-density parity-check codes, and rank-metric codes while preserving security against known attacks.
Research Question: Do quasi-cyclic moderate-density parity-check (QC-MDPC) code-based encryption schemes achieve public key sizes competitive with ML-KEM while maintaining equivalent security against information-set decoding attacks?Quantum Key Distribution Protocol Security Under Device Imperfections
Rigorously proving the composable security of practical QKD protocols (BB84, E91, measurement-device-independent QKD) under realistic device imperfections including detector efficiency mismatch, source photon number fluctuations, and channel loss.
Research Question: What is the asymptotic secret key rate achievable by measurement-device-independent QKD over 200 km of optical fibre with realistic single-photon detector dark count rates, and how does this compare to twin-field QKD under identical assumptions?Hash-Based Digital Signatures: SPHINCS+ Performance Optimisation
Optimising the performance of stateless hash-based signature schemes (SPHINCS+, a NIST PQC standard) through algorithmic improvements in the hypertree construction and hardware acceleration strategies for constrained IoT environments.
Research Question: Can hardware-specific optimisation of SPHINCS+ signing on ARM Cortex-M4 microcontrollers achieve signature generation times below 100 milliseconds at NIST security level 1 without compromising security properties?Quantum-Safe Blockchain: Migrating Proof-of-Work and Signature Schemes
Designing and evaluating quantum-resistant blockchain architectures that replace ECDSA signatures and SHA-256 proof-of-work with post-quantum alternatives, assessing the performance and security implications at network scale.
Research Question: Does replacing ECDSA with Dilithium signatures in Ethereum-compatible blockchain nodes produce acceptable transaction throughput and block validation latency at current network scale, and what hardware upgrades are required for full network transition?Isogeny-Based Cryptography After SIKE: Remaining Constructions
Investigating the remaining isogeny-based cryptographic constructions after the catastrophic classical break of SIKE in 2022 — including CSIDH and SQISign — evaluating their security against both classical and quantum attacks and their practical performance.
Research Question: Do the computational isogeny assumptions underlying SQISign remain hard against both classical and quantum algorithms after the techniques that broke SIKE are applied, and what security level does SQISign provide relative to NIST standards?Fully Homomorphic Encryption with Post-Quantum Security
Designing efficient lattice-based fully homomorphic encryption (FHE) schemes with post-quantum security guarantees, evaluating their performance for privacy-preserving machine learning inference and secure multi-party computation applications.
Research Question: What is the computational overhead of CKKS-scheme fully homomorphic encryption with post-quantum security parameters for neural network inference on medical imaging data, and is this overhead compatible with cloud deployment latency requirements?Crypto-Agility Frameworks for Enterprise PQC Migration
Designing software architecture frameworks for cryptographic agility — the ability to seamlessly replace cryptographic primitives without system-wide changes — that enable organisations to migrate from classical to post-quantum cryptography incrementally.
Research Question: Does a cryptographic agility layer implemented as an abstract cryptographic service interface in enterprise software systems introduce acceptable performance overhead while enabling seamless algorithm substitution for PQC migration?Quantum Hardware Research Topics
Qubit fabrication, control electronics, decoherence characterisation, and scalable integration
Quantum hardware research is the most physically demanding sub-field of quantum computing — requiring deep expertise in solid-state physics, atomic physics, optics, microwave engineering, or materials science depending on the qubit modality. The six major physical qubit implementations each have distinct advantages and limitations, and the race to build practical quantum computers is driving intense competition between them at the research level.
Superconducting
Josephson junction transmons (IBM, Google, Rigetti)
- Gate time: ~10–50 ns
- T₂ coherence: ~100–500 μs
- Gate fidelity: ~99.5% (2Q)
- Advantage: CMOS-compatible fabrication, fast gates
- Challenge: Cryogenic operation (15 mK), crosstalk
Trapped Ion
Ytterbium and barium ions (IonQ, Quantinuum, Oxford)
- Gate time: ~10–1000 μs
- T₂ coherence: ~seconds to minutes
- Gate fidelity: ~99.9% (2Q)
- Advantage: Long coherence, all-to-all connectivity
- Challenge: Slow gates, scaling beyond ~100 ions
Photonic
Linear optical qubits (PsiQuantum, Xanadu)
- Gate time: ~femtoseconds
- Coherence: Room temperature operation
- Advantage: No decoherence, integrated optics
- Challenge: Probabilistic gates, photon loss, detection
- Key: Requires massive resource state overhead for FTQC
Neutral Atom
Rydberg atom arrays (QuEra, Pasqal, Atom Computing)
- Gate time: ~μs (Rydberg gates)
- T₂ coherence: ~seconds
- Gate fidelity: ~99.5% (2Q)
- Advantage: Reconfigurable connectivity, 2D/3D arrays
- Challenge: Atom loading variability, gate speed
NV Centres
Nitrogen-vacancy defects in diamond (room temperature)
- Gate time: ~μs–ms
- T₂ coherence: ~ms (room temp)
- Advantage: Room temperature operation, optical interface
- Challenge: Scaling, fabrication reproducibility, gate speed
- Key application: Quantum sensing and quantum networking
Topological
Majorana fermion qubits (Microsoft)
- Status: Pre-commercial research phase
- T₂ coherence: Theoretically exponentially long
- Advantage: Topological protection from local noise
- Challenge: Majorana braiding not yet demonstrated reliably
- Potential: Intrinsically fault-tolerant if realised
Crosstalk Characterisation and Mitigation in Superconducting Qubit Arrays
Systematic characterisation of ZZ-coupling crosstalk between neighbouring transmon qubits in superconducting quantum processors, and designing dynamically decoupled gate sequences that suppress crosstalk below error correction threshold levels.
Research Question: What is the dominant mechanism of residual ZZ coupling in fixed-frequency superconducting transmon arrays, and can pulse-level gate optimisation reduce its contribution to two-qubit gate error rates below 10⁻³?Motional Mode Engineering for Scalable Trapped Ion Quantum Processors
Designing and characterising ion trap architectures that control the density of motional modes — the phonon buses used for two-qubit gates in trapped ion systems — to enable high-fidelity gates in large ion chains beyond 50 ions.
Research Question: Does segmented ion trap architecture with active zone separation maintain two-qubit Mølmer-Sørensen gate fidelity above 99% for chains of 50+ ytterbium ions without significant motional mode heating?Two-Dimensional Neutral Atom Arrays: Rydberg Gate Fidelity at Scale
Characterising and improving two-qubit Rydberg gate fidelity in 2D neutral atom arrays as array size increases, investigating the role of atomic position disorder, Rydberg blockade radius variation, and laser phase noise.
Research Question: How does two-qubit Rydberg CZ gate fidelity degrade as a function of array size in reconfigurable 2D rubidium atom arrays, and does site-to-site laser intensity variation or atom position disorder dominate this degradation?Cryogenic Control Electronics: Scalability Bottleneck for Superconducting Qubits
Designing and evaluating cryogenic classical control electronics (cryo-CMOS) that operate at millikelvin temperatures inside the dilution refrigerator, enabling the control signal multiplexing required to scale beyond hundreds of qubits without a classical electronics bandwidth bottleneck.
Research Question: Does cryo-CMOS control circuitry operating at 4K achieve sufficient signal-to-noise and bandwidth for single-qubit gate control of transmon qubits with fidelity compatible with surface code fault-tolerance thresholds?Quantum Processor Benchmarking: Randomised Benchmarking Extensions
Developing extensions of randomised benchmarking (RB) — including interleaved RB, character RB, and gate set tomography — that provide more informative error characterisation of quantum processors beyond average gate fidelity.
Research Question: Does gate set tomography applied to a 10-qubit superconducting quantum processor identify error model structure (coherent vs. incoherent, correlated vs. local) that standard randomised benchmarking cannot distinguish, and does this additional information enable more effective error mitigation?Quantum Dot Spin Qubits: Singlet-Triplet Splitting and Gate Fidelity
Characterising charge noise, nuclear spin noise, and spin-orbit coupling effects on the coherence and gate fidelity of silicon quantum dot spin qubits, and designing dynamically corrected gates that are robust to these dominant noise sources.
Research Question: What is the dominant decoherence mechanism for singlet-triplet spin qubits in isotopically purified silicon-28 quantum dots at sub-kelvin temperatures, and do Carr-Purcell-Meiboom-Gill dynamical decoupling sequences extend T₂ coherence times to millisecond scales?Interconnects for Modular Quantum Computing: Microwave Photon Transduction
Designing and characterising microwave-to-optical quantum transducers that enable entanglement distribution between superconducting quantum processor modules via optical fibre, addressing the fundamental challenge of connecting cryogenic quantum processors.
Research Question: Can electro-optomechanical transducers achieve microwave-to-optical quantum state conversion fidelity sufficient for Bell state measurement at rates enabling entanglement generation between superconducting qubit modules at a rate of 1 kHz?Flux-Tunable Transmon Qubit Calibration Automation
Developing automated calibration pipelines for large-scale superconducting quantum processors that maintain qubit frequency, gate fidelity, and cross-resonance gate performance over extended operating periods without manual recalibration.
Research Question: Can Bayesian optimisation-based automated calibration maintain single-qubit and two-qubit gate fidelities within 0.1% of optimal values over 24-hour operating periods on a 50-qubit superconducting processor without human intervention?Photonic Integrated Circuits for Linear Optical Quantum Computing
Designing and characterising silicon-photonics integrated circuit platforms for linear optical quantum computing, evaluating photon loss, mode overlap, and beamsplitter fidelity in integrated waveguide architectures for boson sampling and photonic cluster state generation.
Research Question: What photon loss per component budget is required for silicon-photonics integrated optical quantum circuits to achieve boson sampling with 50+ photons at a sampling rate that surpasses classical simulation capacity, and is this achievable with current fabrication technology?NV Centre Spin-Photon Entanglement for Quantum Repeater Nodes
Engineering the spin-photon entanglement generation rate and fidelity at nitrogen-vacancy centre in diamond quantum memory nodes, and demonstrating entanglement swapping between adjacent nodes for quantum repeater demonstration.
Research Question: Does entanglement purification applied to NV centre spin-photon Bell pairs generated at room temperature achieve fidelities sufficient for quantum teleportation-based quantum repeater protocols over 50 km fibre segments?Materials Characterisation for Reducing Superconducting Qubit Surface Loss
Investigating materials science approaches to reducing two-level-system (TLS) loss at metal-substrate interfaces in superconducting transmon qubits — including surface treatments, substrate choices, and junction fabrication protocols — to extend T₁ relaxation times.
Research Question: Which combination of substrate cleaning procedure, superconducting metal deposition technique, and Josephson junction oxidation protocol produces the lowest TLS-limited T₁ relaxation times in transmon qubits fabricated on silicon versus sapphire substrates?Quantum Processor Architecture Compilation: Qubit Mapping and Routing Optimisation
Designing and evaluating algorithms for compiling quantum circuits to specific quantum processor architectures — performing qubit mapping, SWAP insertion, and gate commutation to minimise circuit depth and two-qubit gate count for a given hardware connectivity graph.
Research Question: Do reinforcement learning-based qubit routing algorithms achieve lower two-qubit gate overhead than SABRE and lookahead-based routing algorithms for compilation of quantum chemistry circuits to IBM heavy-hex processor graphs?Quantum Simulation Research Topics
Using quantum computers to simulate nature — chemistry, materials, and condensed matter physics
Richard Feynman’s original 1982 proposal for quantum computers was specifically motivated by simulation: classical computers cannot efficiently simulate quantum systems because the Hilbert space grows exponentially with system size. Quantum simulation — using controlled quantum systems to simulate other quantum systems — is widely regarded as the most near-term path to practical quantum advantage, because the resource requirements for useful quantum chemistry and materials simulations appear lower than for the deep algorithmic circuits needed by Shor’s algorithm.
Fermionic Hamiltonian Simulation: Jordan-Wigner vs. Bravyi-Kitaev Encoding
Comparing qubit encoding methods for fermionic systems (Jordan-Wigner, Bravyi-Kitaev, parity basis) in terms of circuit depth, gate count, and noise sensitivity when used with VQE for second-quantised molecular Hamiltonians.
Research Question: Does the Bravyi-Kitaev encoding of fermionic Hamiltonians produce shallower VQE circuits with lower cumulative gate error for H₂O and NH₃ molecules than Jordan-Wigner encoding at equivalent basis set size?Quantum Phase Estimation for Ground State Energy of Strongly Correlated Systems
Applying quantum phase estimation (QPE) to compute ground state energies of strongly correlated electronic systems (Hubbard model, molecular transition metal complexes) where classical methods (DFT, CCSD(T)) fail, estimating resource requirements on fault-tolerant hardware.
Research Question: What are the quantum circuit resources (logical qubit count, T gate depth) required for QPE to compute the ground state energy of FeMoco (the active site of nitrogenase) to chemical accuracy, and how do these compare to DMRG-based classical estimates?Quantum Simulation of the Ising Model: Phase Transitions and Critical Phenomena
Using programmable quantum processors to simulate quantum phase transitions in transverse-field Ising models and characterise critical exponents, comparing quantum simulation data against exact diagonalisation and quantum Monte Carlo benchmarks.
Research Question: Can a 100+ qubit neutral atom quantum processor simulate the quantum phase transition of the 2D transverse-field Ising model and extract critical exponents in agreement with conformal field theory predictions beyond the scale accessible to classical simulation?Quantum Simulation of Battery Electrolyte Materials for Energy Storage
Applying quantum chemistry simulation algorithms to model lithium-ion battery electrolyte decomposition reactions and electrode-electrolyte interfaces at quantum mechanical accuracy, identifying materials for next-generation solid-state batteries.
Research Question: Can VQE-based quantum simulation of lithium-ion migration barriers in solid electrolyte materials achieve chemical accuracy predictions at a computational scale feasible on near-term quantum processors for systems relevant to solid-state battery design?Open Quantum Systems Simulation: Lindblad Dynamics on Quantum Hardware
Developing algorithms for simulating open quantum systems governed by Lindblad master equations on quantum hardware, enabling simulation of dissipative processes in photosynthetic light harvesting, superconducting circuits, and biological environments.
Research Question: Can Lindblad dynamics simulation using dilated unitary circuits on NISQ hardware reproduce experimentally measured energy transfer rates in the Fenna-Matthews-Olson (FMO) photosynthetic complex with accuracy sufficient to distinguish competing theories of quantum coherence in photosynthesis?Tensor Network Classical Simulation: When Can Classical Methods Compete?
Characterising the boundary between quantum circuits that can be efficiently classically simulated using tensor network methods (MPS, MERA, PEPS) and those that genuinely require quantum hardware, informing claims of quantum advantage.
Research Question: What is the entanglement entropy growth rate threshold beyond which random quantum circuits of practical depth resist efficient classical tensor network simulation, and does this threshold depend on circuit architecture (brickwork vs. random connectivity)?Drug-Receptor Binding Affinity via Quantum Computing: Benchmark Study
Benchmarking quantum simulation algorithms (VQE, QITE) for computing protein-ligand binding free energies against industry-standard classical free energy perturbation methods on a curated set of drug-receptor pairs.
Research Question: Can VQE-based quantum simulation of drug-receptor binding energetics for a set of HIV protease inhibitors achieve accuracy competitive with FEP+ classical calculations, and what system size threshold triggers a crossover in computational cost?Quantum Simulation of High-Temperature Superconductivity: Hubbard Model
Using quantum processors to simulate the 2D Hubbard model at half-filling — a central model for high-temperature superconductivity that classical methods cannot simulate in the strongly correlated regime — and characterising its phase diagram.
Research Question: Can a 50+ qubit quantum processor simulate the 2D Fermi-Hubbard model at intermediate coupling (U/t = 8) and half-filling sufficiently to verify or refute d-wave superconducting order in the ground state phase diagram?Quantum-Classical Hybrid Simulation of Reaction Dynamics
Developing hybrid workflows where quantum processors simulate the core quantum region of a chemical reaction (the reactive centre) while classical processors handle the surrounding environment, enabling simulation of enzyme-catalysed reactions.
Research Question: Does a QM/MM quantum-classical hybrid simulation workflow — using VQE for the reactive quantum region and molecular mechanics for the protein environment — predict reaction activation energies for dihydrofolate reductase catalysis within 1 kcal/mol of experimental values?Quantum Simulation for Materials Discovery: Topological Insulator Characterisation
Applying quantum simulation algorithms to characterise the electronic band structure and topological invariants of candidate topological insulator materials, potentially accelerating experimental discovery of new topological phases.
Research Question: Can quantum phase estimation on a fault-tolerant processor compute band structure topological invariants (Z₂ invariants) for Bi₂Se₃ at material-accurate Hamiltonians faster than classical DFT+GW calculations for equivalent accuracy?Quantum Advantage for Nuclear Structure Simulation
Investigating quantum algorithms for computing nuclear energy levels and reaction cross-sections using nuclear shell model and ab initio nuclear structure approaches, estimating resource requirements for fault-tolerant advantage.
Research Question: What are the fault-tolerant quantum resource requirements for computing the ground state energy and excitation spectrum of ²⁸Si using the nuclear shell model, and at which nuclear mass number does quantum simulation become computationally advantageous over classical shell model diagonalisation?Quantum Machine Learning Research Topics
The frontier intersection of quantum computing and artificial intelligence
Quantum machine learning (QML) is one of the most actively debated sub-fields in quantum computing — generating both extraordinary excitement and significant scepticism. The fundamental question is whether quantum computers can provide genuine speedups for machine learning tasks, beyond the quantum speedups achievable for the underlying linear algebra subroutines. The dequantisation results of Ewin Tang and others demonstrated that several proposed quantum machine learning algorithms can be matched by classical sampling-based algorithms, cooling some of the early enthusiasm. Nevertheless, the field contains genuine open problems and potential quantum advantages for specific learning tasks, particularly those involving quantum data.
Quantum Kernel Methods: Advantage on Quantum-Generated Data
Investigating whether quantum kernel methods — which use quantum circuits to define feature maps in exponentially large Hilbert spaces — provide genuine classification advantages over classical kernel methods on datasets with underlying quantum structure.
Research Question: Do quantum kernel classifiers trained on quantum phase classification tasks outperform classical SVM with RBF kernels on datasets generated by quantum many-body systems with known phase structure, and does this advantage require quantum data access?Quantum Convolutional Neural Networks for Phase Classification
Designing and benchmarking quantum convolutional neural networks (QCNNs) for classifying quantum phases of matter from measurement data, evaluating whether the QCNN’s translationally invariant architecture provides an inductive bias advantage for quantum data.
Research Question: Does the quantum convolutional neural network architecture achieve higher accuracy with fewer parameters than fully connected parameterised quantum circuits for classifying the phase diagram of the 1D cluster-Ising model from classical shadow measurements?Quantum Generative Adversarial Networks for Molecular Data Generation
Developing and evaluating quantum GANs for generating molecular structure data — representing molecular geometries and property distributions — comparing generated data quality and diversity against classical deep generative models.
Research Question: Do quantum generative adversarial networks trained on molecular graph datasets produce valid molecular structures with higher property diversity than classical SMILES-based RNN generators when training data is limited to 10,000 molecules?Quantum Federated Learning: Privacy Guarantees and Communication Overhead
Designing quantum federated learning protocols that leverage quantum communication to enhance differential privacy guarantees while reducing classical communication rounds, evaluating the privacy-accuracy-communication trade-off.
Research Question: Can quantum key distribution-secured gradient aggregation in federated learning provide stronger differential privacy guarantees than classical DP-SGD at equivalent model accuracy, and what quantum communication rate is required?Quantum Transfer Learning: Pre-Training on Quantum Hardware
Investigating whether hybrid quantum-classical networks pre-trained on quantum data can be fine-tuned on classical datasets to achieve better generalisation than purely classical networks, and identifying the conditions for positive transfer.
Research Question: Does pre-training a hybrid quantum-classical network on quantum phase classification tasks provide a useful parameter initialisation for classical image classification fine-tuning, compared to random initialisation or classical pre-training?Quantum Advantage in Reinforcement Learning: Sample Complexity
Theoretically and empirically investigating whether quantum reinforcement learning agents can achieve quadratic or polynomial reductions in the number of environment interactions (sample complexity) required to learn optimal policies for specific environments.
Research Question: Do quantum reinforcement learning agents using Grover-based policy search achieve polynomial reductions in sample complexity compared to classical deep Q-network agents on tabular MDPs with known quantum structure, and does this advantage persist under function approximation?Expressibility and Entanglement Capacity of Quantum Circuit Ansätze
Quantifying the expressibility (coverage of Hilbert space by random circuits from the ansatz) and entanglement generation capacity of parameterised quantum circuit families, and relating these properties to learning performance on machine learning benchmarks.
Research Question: Do parameterised quantum circuit families with higher expressibility (as measured by the Haar-random fidelity distribution KL divergence) achieve better generalisation on quantum machine learning tasks, or does expressibility correlate with the onset of barren plateaus?Quantum Boltzmann Machines: Training and Application to Anomaly Detection
Implementing and benchmarking quantum Boltzmann machines — which use quantum annealing or stochastic quantum sampling to accelerate training of Boltzmann machine generative models — for anomaly detection in cybersecurity and fraud datasets.
Research Question: Does quantum Boltzmann machine training using D-Wave quantum annealing produce generative models with higher log-likelihood on held-out network intrusion data than classically trained restricted Boltzmann machines with equivalent model capacity?Quantum Neural Networks: Capacity and Trainability at Scale
Investigating the generalisation capacity of quantum neural networks — parameterised quantum circuits used as function approximators — and whether they exhibit double descent and other phenomena characteristic of overparameterised classical neural networks.
Research Question: Do parameterised quantum circuits used as function approximators exhibit a double-descent generalisation curve as circuit width and depth increase, and does overparameterisation improve trainability by reducing barren plateau frequency?Quantum-Enhanced Bayesian Optimisation for Drug Discovery
Integrating quantum computing into Bayesian optimisation pipelines for molecular property optimisation — using quantum circuits as surrogate models or for quantum speedup of the acquisition function optimisation — and benchmarking against classical Gaussian process Bayesian optimisation.
Research Question: Does quantum circuit-based surrogate modelling in Bayesian optimisation achieve fewer function evaluations to find molecules exceeding a target bioactivity threshold compared to Gaussian process Bayesian optimisation on the QM9 molecular dataset?Quantum Data Encoding: Amplitude vs. Basis vs. Angle Encoding Trade-offs
Systematically comparing quantum data encoding strategies — amplitude encoding, basis encoding, angle encoding, and more complex feature maps — in terms of their expressibility, circuit depth requirements, and sensitivity to measurement noise for classification tasks.
Research Question: Which quantum data encoding strategy for classical tabular datasets produces the best accuracy-circuit depth trade-off for a binary classification benchmark suite, and does the optimal encoding depend more on dataset dimensionality or class separability geometry?Quantum Natural Language Processing: Compositional Syntax and Meaning
Investigating DisCoCat and related grammatically-informed compositional models of natural language semantics that map linguistic structure onto quantum circuits, evaluating their performance on NLP tasks compared to classical transformer baselines.
Research Question: Do quantum NLP models based on compositional distributional semantics encoded as quantum circuits achieve competitive accuracy on the RelClause sentence classification task compared to classical LSTM baselines at equivalent parameter counts on current quantum hardware?Quantum Networking & Communication Research Topics
Quantum internet, entanglement distribution, and quantum repeater networks
Quantum Repeater Network Architecture: Entanglement Purification Rate
Designing and simulating quantum repeater network architectures that use entanglement swapping and purification to distribute high-fidelity Bell pairs over metropolitan-scale distances, optimising for secret key rate in QKD applications.
Research Question: What quantum memory coherence time and entanglement generation rate are required for a two-node quantum repeater to achieve a secret key rate exceeding 1 kbit/s for BB84 QKD over a 200 km metropolitan fibre link?Quantum Network Routing: Entanglement Distribution Under Link Uncertainty
Designing routing algorithms for quantum networks where entanglement links are probabilistic (Bell state generation succeeds with probability p), optimising for end-to-end entanglement distribution rate and fidelity across multi-hop paths.
Research Question: Do adaptive routing algorithms for quantum networks that select paths based on real-time link success probability estimates achieve end-to-end entanglement distribution rates closer to theoretical bounds than static shortest-path routing?Quantum Memory Platforms for Network Applications: AFC vs. DLCZ Comparison
Comparing atomic frequency comb (AFC) quantum memories in rare-earth-doped crystals with DLCZ (Duan-Lukin-Cirac-Zoller) atomic ensemble memories for quantum network applications, evaluating storage time, multimode capacity, and retrieval efficiency.
Research Question: Do AFC quantum memories in Pr³⁺:Y₂SiO₅ achieve multimode storage capacity and retrieval efficiency sufficient for entanglement distribution in the quantum internet at network clock rates compatible with practical QKD applications?Continuous-Variable QKD: Security in the Finite-Size Regime
Proving composable security of continuous-variable quantum key distribution (CV-QKD) protocols — which use coherent states and homodyne detection rather than single photons — in the finite key size regime, and evaluating practical performance at metropolitan distances.
Research Question: Does Gaussian-modulated coherent state CV-QKD with finite-size security corrections achieve positive secret key rates at 100 km fibre distances under collective attacks with experimentally achievable excess noise levels?Quantum Internet Protocol Stack: Link Layer and Network Layer Design
Designing protocol specifications for the quantum internet link layer and network layer — analogous to classical TCP/IP but adapted for quantum entanglement distribution — and simulating their performance using quantum network simulation tools.
Research Question: What entanglement generation and purification protocol specifications at the quantum network link layer achieve optimal end-to-end Bell pair fidelity for distributed quantum computing applications across a simulated 10-node quantum network?Satellite-Based Quantum Key Distribution: Micius Mission Extensions
Extending the analysis of the Micius satellite QKD demonstrations to evaluate the feasibility of ground-to-geostationary-orbit QKD and multi-node satellite quantum networks for intercontinental key distribution.
Research Question: What are the photon loss and pointing accuracy requirements for achieving positive secret key rate QKD between ground stations and geostationary (36,000 km) orbit satellites using the adaptive optics beam compensation demonstrated in low-Earth-orbit by Micius?Device-Independent Quantum Cryptography: Loophole-Free Bell Tests at Distance
Designing and analysing device-independent QKD protocols that derive security from Bell inequality violations without trusting the quantum devices — requiring loophole-free Bell tests at km-scale distances to be practical.
Research Question: What detection efficiency and visibility thresholds are required for loophole-free Bell tests between NV centre nodes connected by 1 km fibre to generate device-independent secret keys at a positive rate, and are these achievable with current technology?Quantum Network Simulation: NetSquid and SeQUeNCe Platform Benchmarking
Systematically benchmarking and validating quantum network simulation platforms (NetSquid, SeQUeNCe, SimulaQron) against experimental data from installed quantum network testbeds, and extending their capabilities for simulating quantum internet protocols.
Research Question: Do existing quantum network simulation platforms (NetSquid, SeQUeNCe) accurately reproduce experimentally measured entanglement distribution rates and fidelities from the Delft quantum network testbed, and what physical noise model details are most critical for accuracy?Blind Quantum Computing: Delegation of Quantum Computation with Privacy
Designing and evaluating practical blind quantum computing protocols that allow a client with limited quantum capability to delegate computation to a quantum server while keeping both the algorithm and data private, analysing verification and security properties.
Research Question: Can measurement-based blind quantum computing protocols using single-qubit resource states enable a classical client to verifiably delegate universal quantum computation to an untrusted server with composable security and overhead polynomial in the circuit size?Quantum-Secured Industrial IoT: Quantum Random Number Generators at Edge
Evaluating quantum random number generators (QRNGs) based on quantum vacuum fluctuations or photon arrival times for deployment in industrial IoT security applications, assessing their certification against NIST randomness test suites.
Research Question: Do chip-scale quantum random number generators based on photon arrival time vacuum fluctuations pass NIST SP 800-90B entropy estimation tests at generation rates compatible with TLS session key generation for industrial IoT gateway applications?Quantum Information Theory Research Topics
Entropy, channel capacity, resource theory, and the foundations of quantum advantage
Quantum Channel Capacity Under Correlated Noise
Computing and bounding the quantum capacity of quantum channels subject to correlated (non-Markovian) noise — which violates the product channel assumptions used in standard quantum Shannon theory — and identifying channel families where correlations increase or decrease capacity.
Research Question: Does temporal correlation in quantum channel noise increase or decrease the quantum channel capacity compared to the memoryless channel with equivalent average error rate, and does this depend on the correlation structure?Resource Theory of Magic: Characterising Non-Clifford Resources
Developing the resource theory of magic states — the non-Clifford quantum states required for universal quantum computation — characterising monotones, conversion rates between magic states, and free operations under Clifford-circuit restrictions.
Research Question: What is the regularised mana (a magic state monotone) of the T state, and does regularisation change the conversion rate between T states and CCZ states under stabiliser operations, with implications for magic state distillation overhead?Classical Shadows: Efficient Many-Body State Tomography
Developing and analysing classical shadow protocols for efficiently estimating many-body observables of quantum states from a polynomial number of measurements, and applying these to quantum simulation and quantum chemistry benchmarking.
Research Question: Do derandomised classical shadow protocols using locally scrambled ensembles achieve provably lower sample complexity than Pauli-basis randomised measurement for estimating k-body reduced density matrices of n-qubit states?Quantum Advantage in Communication Complexity
Proving exponential quantum advantages in communication complexity for distributed computing problems — identifying functions where quantum protocols with entanglement require exponentially fewer bits than classical randomised protocols.
Research Question: Can quantum communication complexity advantages for the disjointness and inner product functions under entanglement assistance be extended to naturally occurring distributed computing tasks in database query and privacy-preserving computation?Entanglement Measures: Squashed Entanglement Computability
Investigating computational methods for approximating the squashed entanglement — one of the most operationally meaningful but computationally hardest entanglement measures — for specific families of quantum states, and connecting it to quantum channel capacity.
Research Question: Is there a polynomial-time classical algorithm to approximate the squashed entanglement of quantum states in the symmetric subspace to within additive ε, and does this class of states have a quantum advantage in entanglement distillation?Quantum Computational Pseudorandomness: T-Designs and PRU Constructions
Constructing efficient approximate unitary t-designs and pseudorandom unitary (PRU) ensembles — computationally indistinguishable from Haar-random unitaries — with applications to quantum cryptography, benchmarking, and property testing.
Research Question: Can random quantum circuits of depth O(n log n) form approximate unitary 4-designs on n qubits, and do these circuits provide quantum pseudorandomness against polynomial-time quantum adversaries assuming post-quantum one-way functions?Quantum Thermodynamics: Work Extraction from Correlated Quantum Systems
Investigating the second law of thermodynamics for quantum systems — deriving work extraction limits for quantum states with correlations and entanglement, and understanding how quantum correlations modify classical thermodynamic bounds.
Research Question: What is the maximum extractable work from an entangled two-qubit state through local operations and classical communication (LOCC), and how does this compare to the classical work extraction bound set by the free energy difference?Quantum Error Bounds via Decoupling Theorems
Applying decoupling theorems — which characterise when quantum information becomes independent of a reference system — to derive tight one-shot capacity bounds for quantum channels and establish quantum information-theoretic security proofs.
Research Question: Do single-shot quantum capacity bounds derived from decoupling theorems provide tighter estimates of quantum communication rate for short block lengths than asymptotic capacity formulas for practically relevant quantum channels?Quantum Complexity Theory: The Power of One Clean Qubit (DQC1)
Investigating the DQC1 model of quantum computation — which uses only one pure qubit with all others maximally mixed — and characterising its computational power relative to BQP and classical complexity classes, with connections to quantum sensing and NMR computing.
Research Question: Does DQC1 (deterministic quantum computation with one clean qubit) contain problems that are hard for classical randomised polynomial time but do not require full BQP computation, and can these problems be realised on near-term hardware without full state preparation?Quantum Sensing and Metrology: Heisenberg Limit in Noisy Environments
Designing quantum sensing protocols that approach the Heisenberg limit (1/N scaling of phase estimation uncertainty with N probe states) in the presence of decoherence, using error-corrected probes and adaptive measurement strategies.
Research Question: Can quantum error-corrected sensing protocols maintain Heisenberg-limited sensitivity for gravitational wave detection when quantum memory decoherence is included in the noise model, and at what decoherence rate does the advantage over the standard quantum limit disappear?How to Choose Your Quantum Computing Research Topic
Selecting a quantum computing research topic is one of the most consequential decisions of an early research career — because quantum computing is technical enough that topic switching carries a high cost in ramp-up time. The following framework helps you identify the right starting point by aligning your genuine strengths with the field’s genuine needs.
Identify Your Mathematical and Physical Foundations
Quantum computing research demands specialised prerequisites that differ sharply by sub-field. Before choosing a topic, assess your actual — not aspirational — competence in the foundational areas. Be honest: a PhD topic in quantum error correction requires genuine fluency in the stabiliser formalism, group theory, and coding theory. A topic in quantum algorithm design requires solid quantum circuit formalism, amplitude analysis, and complexity theory. A hardware topic requires physical intuition about the relevant quantum system.
- Strong in linear algebra and quantum formalism → Algorithms, QML, QEC theory, quantum information theory
- Strong in physics and experimental methods → Hardware, simulation, quantum optics
- Strong in cryptography and mathematics → Post-quantum crypto, quantum communication protocols
- Strong in machine learning and software → QML, NISQ algorithm development, benchmarking, circuit compilation
- Strong in computer science and complexity → Quantum complexity, algorithm analysis, verification
Access the Current Open Problem Literature
The primary literature repositories for quantum computing research are the arXiv quant-ph archive — where virtually all quantum computing preprints appear before or simultaneously with journal publication — and the proceedings of top venues including Physical Review Letters, Nature Physics, Quantum (the community-owned open-access journal), npj Quantum Information, and IEEE Transactions on Quantum Engineering. Survey papers in these venues explicitly enumerate open problems. For student researchers, the “Perspectives” and “Roadmap” articles in npj Quantum Information and the review articles in Reviews of Modern Physics provide authoritative overviews of the field’s open challenges.
Track the most cited papers from the last 12 months in your target area, read their “limitations” and “future work” sections carefully, and identify three or four specific open questions that (a) the authors themselves identify as important, (b) appear in multiple recent papers’ future work sections, and (c) you have the background to approach. The intersection of these is your research frontier.
Assess Hardware and Software Access Realistically
Quantum computing research requires specific resources, and a mismatch between your research question and your available resources will derail even the most intellectually compelling project. Use this resource map to assess your realistic access:
- Quantum simulators (free, sufficient for most student research): Qiskit Aer, Cirq, PennyLane, QuTiP, Strawberry Fields — run on standard laptops or HPC clusters, enabling circuits of 20–50 qubits for most research purposes
- Real NISQ hardware (free tier available): IBM Quantum (up to 127 qubits free, open access via Qiskit), through the IBM Quantum Experience platform
- Commercial hardware (requires funding or institutional access): IonQ, Quantinuum, Rigetti via Amazon Braket; Google Quantum AI via academic partnership; D-Wave via LEAP cloud platform
- HPC resources for tensor network simulation: Most universities provide HPC cluster access for students; classical simulation of quantum circuits up to ~50 qubits is feasible with proper parallelisation
Formulate a Precise, Falsifiable Research Question
Use the Research Question Builder below to transform your topic into a precise, answerable question. In quantum computing, a good research question specifies: the exact quantum method or protocol; the specific physical system or computational problem; the measurable performance criterion; the comparison baseline; and the specific experimental or theoretical methodology. A question without all five elements is not yet a research question — it is a research direction.
A critical additional consideration in quantum computing: distinguish between theory questions (proving properties, deriving bounds, establishing algorithms) and experiment questions (implementing and benchmarking on real hardware or simulators). Theory questions require mathematical proof; experiment questions require careful empirical design and honest comparison with state-of-the-art classical methods. Many of the most impactful quantum computing papers of recent years have been experiment questions that critically evaluated earlier theoretical claims.
Quantum Research Question Builder
Transform any quantum computing topic into a precise, publishable research question
Research Methodology Guide: How Quantum Computing Research Is Done
Quantum computing research employs four distinct methodological paradigms, and the most impactful papers often combine at least two of them. Understanding which paradigm your research question calls for — and what standards of evidence each requires — is essential for producing credible, publishable work.
Proving theorems about algorithms, protocols, or information-theoretic limits. Requires rigorous proof (not simulation evidence). Standard: results reproducible by any reader with the mathematics. Venue: Physical Review Letters, Quantum, journal papers.
Simulating quantum circuits classically using Qiskit Aer, Cirq, QuTiP, or tensor network methods. Reproducible, accessible, can simulate noise models. Limitation: restricted to ~50 qubits for statevector simulation; larger for shallow circuits with tensor networks.
Running circuits on real quantum processors (IBM, IonQ, etc.). Captures real device noise, limited by hardware availability, decoherence, and connectivity. Requires noise characterisation, error mitigation, and statistically valid sampling over many circuit executions.
Computing the classical or quantum resources (qubits, gates, circuit depth, runtime) required for fault-tolerant quantum algorithms to achieve a specified task at a specified accuracy. Uses formal quantum resource estimation frameworks (Q#, Qualtran) to project requirements.
Rigour Standards by Research Type
Pre-Submission Research Checklist for Quantum Computing Papers
- Research question is specific, falsifiable, and names the exact quantum method, system, metric, and baseline
- All theoretical claims are mathematically proven or precisely stated as conjectures with evidence
- All empirical results include shot counts, repetitions, uncertainty measures, and hardware specifications
- Classical baselines are state-of-the-art and fairly implemented — not strawman comparisons
- Noise model is explicitly specified for simulation results; hardware calibration data cited for experimental results
- Resource requirements (qubit count, gate depth, T gate count) are explicitly stated for all algorithms
- Quantum advantage claims are honest — “advantage over best known classical method” not “advantage over all possible classical methods”
- Code and circuit definitions are open-source or reproducible from paper specifications
- Limitations section honestly identifies the noise regimes, system sizes, or problem structures where the method fails
- Paper is targeted at an appropriate venue given the contribution type (theory: Quantum journal; experiment: PRX Quantum; algorithms: QIP proceedings)
Tools, Simulators, and Hardware Access for Quantum Research
| Tool / Platform | Type | Best For | Access |
|---|---|---|---|
| Qiskit (IBM) | SDK + Simulator + Hardware | Superconducting qubit algorithms, circuit optimisation, real hardware runs, full-stack development from circuit to result. Industry standard for NISQ research. | Free (open source); IBM Quantum hardware free via IBM Quantum Network |
| Cirq (Google) | SDK + Simulator | NISQ algorithm research, QAOA, noise simulation, near-term hardware-aware circuit construction. Used by Google internally for quantum experiments. | Free (open source) |
| PennyLane (Xanadu) | QML Framework | Quantum machine learning, variational algorithms, automatic differentiation of quantum circuits. Integrates with PyTorch and TensorFlow for hybrid QML workflows. | Free (open source) |
| QuTiP | Open Systems Simulator | Open quantum systems, Lindblad dynamics, quantum optics, hardware physics simulation. Best for modelling decoherence and dissipation in physical qubit systems. | Free (open source) |
| Strawberry Fields (Xanadu) | Photonic QC Simulator | Continuous-variable and photonic quantum computing, Gaussian boson sampling, photonic circuit simulation and optimisation. | Free (open source) |
| Q# (Microsoft) | QC Language + Resource Estimator | Fault-tolerant algorithm development, quantum resource estimation (qubit count, T gate depth) for algorithms targeting fault-tolerant hardware. Azure Quantum integration. | Free (VS Code extension) |
| Qualtran (Google) | Resource Estimation Library | Detailed fault-tolerant resource estimation for algorithms including Shor’s, quantum simulation, quantum chemistry at hardware-accurate abstraction levels. | Free (open source) |
| IBM Quantum Network | Real Hardware | Running circuits on 7–127 qubit IBM Heron/Falcon/Eagle processors. Free tier available with limited monthly hours; extended access through university partnerships. | Free tier + IBM Quantum Network academic access |
| Amazon Braket | Hardware Marketplace | Access to IonQ, Rigetti, OQC, QuEra, and D-Wave hardware via unified API. Pay-per-shot/per-task pricing. Useful for multi-hardware comparison studies. | Pay-per-use (free trial) |
| D-Wave Leap | Quantum Annealer | Combinatorial optimisation, QUBO problems, quantum annealing research. 5000+ qubit Advantage processor for large sparse Ising model instances. | Free access (developer programme) |
Getting Started: Recommended Learning Path for Student Researchers
- Mathematical foundations: Nielsen and Chuang, Quantum Computation and Quantum Information — the canonical textbook. Work through Chapters 1–5 before beginning research.
- Algorithm intuition: Scott Aaronson’s lecture notes (available free at scottaaronson.com) provide excellent intuitive grounding for quantum complexity and algorithms.
- Practical circuit skills: Qiskit Textbook (learning.quantum.ibm.com) — a free, interactive, browser-based introduction covering quantum circuits, algorithms, and hardware.
- Current research: Follow arXiv quant-ph daily digest; use Semantic Scholar for citation tracking; attend virtual talks at QIP (Quantum Information Processing), TQC, and the Quantum Computing Colloquium series.
- Community: Join the Qiskit Slack, the Unitary Fund Discord, and the Quantum Computing Stack Exchange for community support and research discussions.
Need Expert Help With Your Quantum Computing Research Paper?
Our specialist team includes quantum information scientists, theoretical physicists, and CS researchers who can support your quantum computing research papers, dissertations, and assignments at every academic level.
Get Professional Help Now →FAQs: Quantum Computing Research Answered
Conclusion: Why Quantum Computing Research Matters and How to Enter the Field
Quantum computing occupies a unique position in the history of science: it is simultaneously a fundamental theoretical investigation into the nature of computation and physical reality, and an engineering race with trillion-dollar implications for national security, pharmaceutical discovery, materials science, and the future of cryptography. The 91 research topics in this guide represent a snapshot of the field’s open frontier as of mid-2025 — but quantum computing moves fast enough that this frontier will look significantly different in two years. What will not change is the intellectual structure of good research: a precise question, a rigorous methodology, honest evaluation against strong baselines, and writing that communicates results with enough clarity for others to reproduce and build on them.
For students approaching quantum computing research for the first time, two pieces of practical advice are worth emphasising. First, invest deeply in the mathematics before you try to contribute to the research literature. Quantum computing is unusual among applied fields in the degree to which superficial engagement with the formalism — writing “quantum” into proposal titles without genuine mastery of the underlying physics and mathematics — is immediately recognisable to experts and produces work that cannot survive peer review. The foundational investment in linear algebra, quantum mechanics, and complexity theory pays compounding returns throughout a quantum computing research career. Second, be honest about the limits of near-term quantum hardware. The most impactful quantum computing papers of recent years have often been those that carefully and honestly characterised what current devices cannot do, rather than those that over-claimed advantage. Scientific credibility is built on honest negative results as much as positive ones — and in quantum computing, where the field is young enough that many claimed advantages have later been dequantised or refuted, intellectual honesty is both an ethical obligation and a strategic career asset.
For expert support with quantum computing research papers, literature reviews, dissertation chapters, and academic assignments at any level, visit Smart Academic Writing and explore our computer science and quantum computing help, research paper writing, dissertation writing, and editing and proofreading services.
