Quantum ESPRESSO App: Workflow Platform

• Quantum ESPRESSO App is a web-based platform that delivers integrated, automated DFT workflows through a modular, plugin-based architecture.
• It employs a wizard-style Input-Process-Output model on AiiDAlab to streamline simulation setup, execution, error recovery, and interactive results analysis.
• By unifying functionalities for electronic structure, spectroscopy, and post-processing, it lowers barriers for advanced, reproducible first-principles simulations.

Searching arXiv for papers on the Quantum ESPRESSO app and related QE ecosystem applications. The Quantum ESPRESSO app is a web-based platform built on AiiDAlab that provides user-friendly graphical interfaces with automated DFT workflows for Quantum ESPRESSO, with the stated aim of addressing barriers such as software installation, input preparation, high-performance computing setup, and output analysis (Wang et al., 25 Jul 2025). It is organized around a modular Input-Process-Output model and a plugin-based architecture, and it combines a browser-accessible interface with the AiiDA workflow engine so that simulation setup, execution, provenance capture, error handling, and interactive results analysis are integrated into a single environment (Wang et al., 25 Jul 2025). Within the broader Quantum ESPRESSO landscape—an open-source distribution of interoperable codes for first-principles materials modeling rather than a single monolithic executable—the app serves as a workflow and usability layer over an established electronic-structure ecosystem (Giannozzi et al., 2021). The app’s demonstrated scope includes plugins for electronic band structures, projected density of states, phonon, infrared/Raman, X-ray and muon spectroscopies, Hubbard parameters (DFT+$U$+$V$), Wannier functions, and post-processing tools (Wang et al., 25 Jul 2025).

1. Definition and software context

The app is presented as a turn-key interface to QE workflows built on AiiDAlab, where the front end is browser-based and the backend is provided by AiiDA (Wang et al., 25 Jul 2025). AiiDA supplies workflow orchestration, communication with local or remote machines, scheduler interaction, file transfer, database-backed storage of data and metadata, and a provenance graph connecting every input, calculation, output, and workflow decision (Wang et al., 25 Jul 2025). The app therefore does not simply generate input files for pw.x; it exposes complete, managed workflows over the Quantum ESPRESSO stack (Wang et al., 25 Jul 2025).

This positioning reflects the structure of Quantum ESPRESSO itself. Quantum ESPRESSO is described as an integrated suite of open-source computer codes for simulations based on density-functional theory, density-functional perturbation theory, and related methodologies, with a modular ecosystem of executables and interoperable packages rather than a single narrowly defined program (Giannozzi et al., 2017). A later architectural review similarly characterizes Quantum ESPRESSO as an open-source distribution of interoperable codes for first-principles materials modeling, emphasizing a layered software organization and broad interoperability with external tools and workflow engines (Giannozzi et al., 2021). The app is therefore best understood as a workflow-native interface layer over this ecosystem, rather than as an alternative electronic-structure engine.

A plausible implication is that the term “Quantum ESPRESSO app” is most precise when used for the AiiDAlab-based application described in 2025, while the wider Quantum ESPRESSO ecosystem also contains many specialized applications and modules such as EPW, HP, turboEELS, and QE-GIPAW (Wang et al., 25 Jul 2025, Poncé et al., 2016, Timrov et al., 2022, Timrov et al., 2021).

2. Architecture and execution model

A central design choice is the Input-Process-Output (IPO) model, introduced to provide a common structure across distinct simulation tasks (Wang et al., 25 Jul 2025). In this model, Input covers structure selection, editing, protocols, simulation options, and resource selection; Process corresponds to one or more AiiDA workflows and calculations; and Output comprises summaries, plots, tables, 3D viewers, and exportable data products (Wang et al., 25 Jul 2025). The app implements this through a wizard-style interface with four main steps: selecting a structure, configuring the workflow, choosing computational resources and submitting, and monitoring and viewing results (Wang et al., 25 Jul 2025).

The frontend is implemented in the AiiDAlab ecosystem, using Jupyter notebooks together with the appmode plugin to expose notebook-based interfaces as web applications (Wang et al., 25 Jul 2025). The software can be deployed locally, using aiidalab-launch to retrieve a Docker image and run a containerized instance, or in the cloud, where the paper describes a public deployment on Microsoft Azure orchestrated with Kubernetes and using JupyterHub with OAuth2 for multi-user authentication and authorization (Wang et al., 25 Jul 2025). The implementation also follows the Model-View-Controller (MVC) pattern, uses traitlets to realize Observer and Mediator patterns, and supports lazy loading (Wang et al., 25 Jul 2025).

The backend depends on aiida-quantumespresso, which provides AiiDA calculation classes for Quantum ESPRESSO executables and workflows that compose them into higher-level tasks (Wang et al., 25 Jul 2025). This is consistent with broader Quantum ESPRESSO software engineering trends, where reusable libraries, standardized APIs, and structured XML/HDF5 data models have been emphasized to support interoperability, workflow systems, and external interfaces (Giannozzi et al., 2017, Giannozzi et al., 2021).

3. Workflow design, automation, and resource management

The user workflow begins with structure preparation. Structures can be uploaded in formats such as XYZ or CIF, imported from OPTIMADE-compliant databases including the Materials Project, Materials Cloud, and the Alexandria Materials Database, selected from the local AiiDA database, or taken from provided examples (Wang et al., 25 Jul 2025). Once loaded, structures can be edited through tools that include primitive/conventional/supercell transformations, periodicity definition, surface and molecular setup, atom removal and tagging, point defects in 2D materials, and bond distortions and random displacements to explore metastable defect structures (Wang et al., 25 Jul 2025).

Workflow configuration then exposes basic, advanced, and property-specific settings (Wang et al., 25 Jul 2025). A key feature is the use of predefined computational protocols that balance speed and accuracy, so that parameterization does not have to begin from hand-tuned low-level settings (Wang et al., 25 Jul 2025). The paper states that the examples use the aiida-quantumespresso balanced protocols and the SSSP v1.3.0 efficiency pseudopotential family (Wang et al., 25 Jul 2025). Users may also choose pseudopotentials from SSSP or PseudoDojo, or upload custom pseudopotentials (Wang et al., 25 Jul 2025).

The app delegates execution to AiiDA Code instances representing executables on local or remote machines, and users can specify resources such as node and CPU counts before submission (Wang et al., 25 Jul 2025). AiiDA then handles file transfer, scheduler interaction, remote execution, monitoring, and retrieval (Wang et al., 25 Jul 2025). This reflects a broader Quantum ESPRESSO deployment model in which the simulation engine is routinely coupled to workflow systems and HPC schedulers rather than used only through direct command-line interaction (Giannozzi et al., 2017, Giannozzi et al., 2021).

The automation layer includes automatic $\mathbf{k}$-point mesh generation, automatic restart logic, and error recovery (Wang et al., 25 Jul 2025). Two concrete recovery examples are stated explicitly: if a calculation fails to achieve electronic convergence, the workflow can adjust parameters such as reducing the mixing ratio; and if a calculation exceeds its walltime, the workflow can restart from the last output structure and charge density (Wang et al., 25 Jul 2025). The app also inherits AiiDA’s caching mechanism, allowing reuse of previous results when identical inputs are submitted (Wang et al., 25 Jul 2025).

This suggests that the app’s primary technical contribution is not a new electronic-structure formalism, but a formalization of best-practice workflow behavior around Quantum ESPRESSO calculations.

4. Provenance, FAIR principles, and reproducibility

The app explicitly extends FAIR principles to simulations, workflows, and analyses (Wang et al., 25 Jul 2025). Every calculation is stored in the AiiDA provenance graph, which records inputs, outputs, metadata, workflow nodes, dependencies, and recovery or restart decisions (Wang et al., 25 Jul 2025). If workflow settings change during error recovery, those changes are also preserved in the provenance record (Wang et al., 25 Jul 2025).

The paper maps specific FAIR aspects onto the implementation. Findability is supported through automatic provenance capture, export as AiiDA archive, deposition in Materials Cloud Archive, DOI assignment there, and plugin discoverability through AiiDA entry points and the app’s plugin management page (Wang et al., 25 Jul 2025). Accessibility is achieved through browser-based cloud deployment and local Docker-based execution without manual assembly of the full software stack (Wang et al., 25 Jul 2025). Interoperability is treated not only as data-format compatibility but also as architectural generality: although the present implementation supports only Quantum ESPRESSO, the authors state that the framework is intended as a template for other first-principles engines (Wang et al., 25 Jul 2025). Reusability follows from open-source distribution, modular plugins, exportable archives, and provenance-linked analysis states (Wang et al., 25 Jul 2025).

A particularly specific feature is that interactive plot and widget state—including settings such as axis ranges, colormaps, and structure-viewer choices—is stored together with the relevant AiiDA node (Wang et al., 25 Jul 2025). This goes beyond the reproducibility of raw simulation data and extends to reproducibility of the analysis presentation itself.

Within the wider Quantum ESPRESSO software trajectory, this emphasis on provenance-rich workflows is consistent with earlier recognition of AiiDA as a primary orchestration environment for Quantum ESPRESSO-based automated calculations (Giannozzi et al., 2017).

5. Scientific capabilities and plugin ecosystem

The app is organized around a plugin-based architecture, allowing distinct scientific capabilities to be added while preserving a common IPO interface (Wang et al., 25 Jul 2025). The paper demonstrates plugins spanning electronic structure, vibrational spectroscopy, X-ray spectroscopy, muon spectroscopy, Hubbard-parameter calculations, Wannierization, post-processing, and charge analysis (Wang et al., 25 Jul 2025).

Electronic structure

The electronic structure plugin automates band structure, projected band structure (“fat bands”), and total/projected density of states (DOS/PDOS) workflows (Wang et al., 25 Jul 2025). For band structures, the app automatically generates reciprocal-space paths, using tools such as seekpath for 3D crystals and symmetry-aware logic for low-dimensional systems (Wang et al., 25 Jul 2025). For PDOS, it automates the sequence of SCF, NSCF, total DOS, and PDOS calculations (Wang et al., 25 Jul 2025). Outputs include interactive band and DOS plots aligned to the Fermi level, exportable figures, and numerical data (Wang et al., 25 Jul 2025).

Vibrational spectroscopy

The aiidalab-qe-vibroscopy plugin runs VibroWorkChain, orchestrating workflows from aiida-vibroscopy and aiida-phonopy (Wang et al., 25 Jul 2025). It uses a finite-difference phonon approach and can also compute dielectric properties for non-analytic corrections when needed (Wang et al., 25 Jul 2025). Supported modes include full phonon characterization, phonons without NACs, $\Gamma$-point IR/Raman mode calculations, and dielectric-property workflows (Wang et al., 25 Jul 2025). Outputs include phonon dispersion, phonon DOS, thermodynamic quantities, IR and Raman spectra, dielectric tensor, Born effective charges, and 3D animations of vibrational modes (Wang et al., 25 Jul 2025). When phonon data are present, the plugin can also compute inelastic neutron scattering (INS) dynamical structure factors using Euphonic (Wang et al., 25 Jul 2025).

X-ray spectroscopy

The XAS plugin automates XANES calculations using XSpectra, including core-hole treatments, symmetry analysis for inequivalent sites, and lifetime broadening (Wang et al., 25 Jul 2025). The XPS plugin computes core-level binding energies using the Delta Kohn-Sham ($\Delta$KS) method (Wang et al., 25 Jul 2025). These workflows package what would otherwise be multiple manually coordinated QE and post-processing stages.

Muon spectroscopy

The muon spectroscopy plugin interfaces with aiida-muon to predict muon stopping sites and associated local environments (Wang et al., 25 Jul 2025). It can also compute muon spin polarization functions using UNDI (Wang et al., 25 Jul 2025).

Hubbard parameters

The Hubbard parameters plugin interfaces with aiida-hubbard and QE’s hp.x to compute on-site $U$ and inter-site $V$ parameters for DFT+$U$+$V$ using density-functional perturbation theory (Wang et al., 25 Jul 2025). This connects directly to the dedicated Quantum ESPRESSO HP code, which was introduced as a QE component for computing Hubbard parameters through DFPT rather than supercell-based linear response (Timrov et al., 2022). The app exposes these workflows in one-shot and self-consistent modes and returns summary tables and structure-linked visualization of the resulting interactions (Wang et al., 25 Jul 2025).

Wannier functions

The Wannierization plugin automates workflows for maximally localized Wannier functions (MLWFs) and related analyses (Wang et al., 25 Jul 2025). The paper states that it supports selected columns of the density matrix (SCDM) and projectability disentanglement Wannier functions (PDWF) schemes, and provides DFT/Wannier band comparison, Fermi surface interpolation, and de Haas–van Alphen frequency calculations (Wang et al., 25 Jul 2025).

Post-processing and charge analysis

The post-processing plugin uses pp.x and Critic2 to visualize quantities such as charge density, spin density, wavefunctions, electrostatic potential, integrated local density of states, LDOS at chosen energies, and STM images (Wang et al., 25 Jul 2025). The Bader plugin computes atom-resolved charges from the charge density using the Bader code (Wang et al., 25 Jul 2025).

Taken together, these plugins show that the app is not limited to a narrow subset of Quantum ESPRESSO functionality. Instead, it operates as a federating interface across multiple QE-derived and QE-adjacent workflows, much as Quantum ESPRESSO itself is a federation of interoperable components (Giannozzi et al., 2017, Wang et al., 25 Jul 2025).

6. Relation to the wider Quantum ESPRESSO application ecosystem

A common misconception is that the “Quantum ESPRESSO app” is synonymous with Quantum ESPRESSO as a whole. The software literature instead describes Quantum ESPRESSO as an open-source distribution of interoperable codes with many specialized modules and companion applications (Giannozzi et al., 2021). The AiiDAlab app is one layer in that landscape: it provides a browser-based, workflow-managed entry point to the ecosystem (Wang et al., 25 Jul 2025).

This broader ecosystem includes specialized applications that solve domain-specific problems downstream of standard DFT/DFPT calculations. EPW is integrated in and distributed with Quantum ESPRESSO as a module for electron–phonon coupling, transport, and superconducting properties using maximally localized Wannier functions (Poncé et al., 2016). EPIq is a Quantum ESPRESSO ecosystem application for post-processing DFPT electron–phonon data with Wannier interpolation, supporting quantities such as adiabatic and non-adiabatic phonons, superconducting properties, Raman spectra, and excited-carrier lifetimes (Marini et al., 2023). HP is a QE component for first-principles Hubbard $U$ and $V$0 calculations using DFPT (Timrov et al., 2022). turboEELS is a QE-integrated TDDFPT application for finite-$V$1 charge-excitation spectroscopy (Timrov et al., 2021). qeirreps is a dedicated post-processing program for Quantum ESPRESSO that extracts irreducible representations of Bloch states for topological diagnosis (Matsugatani et al., 2020). QERaman is a Quantum ESPRESSO-based application for first-order resonance Raman spectra (Hung et al., 2023). QE-CONVERSE is a standalone package compatible with recent QE releases for orbital magnetization, NMR chemical shifts, and EPR parameters (Fioccola et al., 6 Mar 2025).

The AiiDAlab Quantum ESPRESSO app differs from these in function. Those codes implement new scientific methods or specialized post-processing algorithms; the app implements a workflow, usability, provenance, and visualization framework over the Quantum ESPRESSO environment (Wang et al., 25 Jul 2025). A plausible implication is that the app can be viewed as an orchestration and accessibility layer that can absorb or expose such specialized workflows through plugins when appropriate.

7. Examples and demonstrated use cases

The paper illustrates the app with several representative examples rather than with a single benchmark system (Wang et al., 25 Jul 2025). For electronic structure, it shows a Se vacancy in a $V$2/graphene heterojunction, where projected bands and PDOS reveal hybridization and in-gap defect states with SOC (Wang et al., 25 Jul 2025). For vibrational spectroscopy, it uses $V$3, including a $V$4 supercell finite-displacement phonon calculation and associated Raman, INS, dielectric, and thermodynamic outputs (Wang et al., 25 Jul 2025). For X-ray spectroscopy, it shows O K-edge XANES of crystalline $V$5 and C 1s XPS shifts in phenylacetylene (Wang et al., 25 Jul 2025). For muon spectroscopy, it shows a candidate muon site in LiF (Wang et al., 25 Jul 2025). For Hubbard-parameter calculations, it shows $V$6 (Wang et al., 25 Jul 2025). For Wannier functions, it uses GaAs (Wang et al., 25 Jul 2025). For post-processing, it shows spin density around an S vacancy in monolayer $V$7 (Wang et al., 25 Jul 2025).

These examples demonstrate breadth rather than numerical benchmarking. The paper’s focus is on showing that the app can package complex, heterogeneous workflows into a consistent interface while maintaining provenance and reproducibility (Wang et al., 25 Jul 2025).

8. Scope, limitations, and future directions

The paper states clearly that the app currently supports only Quantum ESPRESSO (Wang et al., 25 Jul 2025). However, it also argues that the architecture is general and intended as a template for interfaces to other first-principles calculation codes (Wang et al., 25 Jul 2025). The authors mention active exploration of extension toward CP2K and note a broader code-agnostic GUI effort in the PREMISE project (Wang et al., 25 Jul 2025).

Another important boundary is that the app’s scientific scope is inherited from the workflows and plugins it exposes. It is therefore not a complete wrapper over every possible Quantum ESPRESSO executable or external companion code. Instead, it packages selected workflows judged mature enough for reproducible automation and interactive use (Wang et al., 25 Jul 2025). This suggests that the app’s boundaries are defined less by UI design than by the availability of robust AiiDA workflows for the targeted tasks.

The future directions listed in the paper include further modularization of interactive widgets, reuse outside the main app in ordinary Jupyter notebooks or other platforms, and potential decoupling of the UI from the AiiDA backend through more modern UI frameworks and REST APIs (Wang et al., 25 Jul 2025). This indicates that the current implementation is both a production tool and an architectural prototype.

9. Significance

The significance of the Quantum ESPRESSO app lies in how it reframes access to first-principles simulation. Quantum ESPRESSO already provides a broad, high-performance open-source ecosystem for DFT, DFPT, MBPT, spectroscopy, electron–phonon physics, and beyond (Giannozzi et al., 2017, Giannozzi et al., 2021). The app does not replace that computational core. Instead, it provides a web-based, provenance-aware, plugin-extensible workflow environment that lowers the barrier to using advanced Quantum ESPRESSO methods while preserving reproducibility and technical rigor (Wang et al., 25 Jul 2025).

This suggests that the app is most important not because it changes the underlying equations of electronic-structure theory, but because it changes the operational form in which those methods are delivered: from manually assembled command-line workflows to a structured, FAIR, browser-accessible research environment (Wang et al., 25 Jul 2025). In that sense, it occupies the same conceptual space for Quantum ESPRESSO that workflow engines occupy for modern experimental facilities: it standardizes procedure, records provenance, automates recovery, and exposes complex methods through reusable interfaces.