Computational Materials Science Platform

From Hypothesis
to Atomic Structure

Unified access to scientific literature, materials databases, DFT compute, and an AI research assistant — designed for materials scientists.

Scroll to explore

The Problem

Research Shouldn't Require
15 Open Tabs

Modern materials research demands jumping between literature databases, property repositories, simulation tools, and analysis software — each with its own interface, format, and workflow.

Hours of context-switching. Data trapped in incompatible formats. Manual copy-paste from paper tables to spreadsheets. Critical insights lost between tools.

material.codes connects your entire research stack into a single, AI-powered workflow.

One unified interface
Automated data pipelines
AI-assisted synthesis

Core Platform

Four Modules. One Workflow.

Every tool a materials scientist needs — tightly integrated so results from one module flow directly into the next.

Literature Intelligence

Semantic search across 2M+ indexed papers from Semantic Scholar and OpenAlex. Extract structured data, compare experimental results, and get AI-synthesized summaries with citations.

"perovskite solar cell efficiency above 25% — last 2 years"

Materials Datasets

Unified access to Materials Project, AFLOW, OQMD, and Citrination. Query by composition, structure, or property. Download structures in any format for downstream simulation.

"cubic perovskites with bandgap 1.2–1.8 eV, stable at 300K"

Compute Engine

Submit DFT calculations to Quantum ESPRESSO or ABINIT without writing input files. Automated pseudopotential selection, k-point grids, and convergence testing. Results parsed automatically.

"optimize geometry of mp-2815 with PBE+U, U=3.5 on Fe"

AI Assistant

A domain-aware research assistant that orchestrates all four modules in response to natural language queries. Maintains conversation context and cites primary sources.

"compare Li-ion cathode candidates for high-temp applications"

Live Demo

Watch It Work

Ask a research question in plain English. material.codes orchestrates the full pipeline automatically.

material.codes — research terminal
Initializing...

Animation loops automatically · Actual latency depends on compute queue

Open by Design

MCP Protocol Support

material.codes implements the Model Context Protocol (MCP), letting you connect our data tools directly to Claude, Cursor, or any MCP-compatible AI client.

Query Materials Project, run literature searches, or launch DFT jobs from inside your own AI workflows — without leaving your tool.

Connect once, use everywhere — Claude, Cursor, Copilot
API key authentication, rate-limited per tier
Full tool schema documentation available on request
{ "mcpServers": { "material-codes": { "url": "https://mcp.material.codes/sse", "transport": "sse", "headers": { "Authorization": "Bearer YOUR_API_KEY" } } } }

Available tools: search_literature · query_materials · run_dft · get_properties

Connected Sources

Built on Real Data

material.codes indexes and unifies the most trusted sources in computational materials science.

Materials Project AFLOW OQMD Citrination Semantic Scholar OpenAlex ABINIT Quantum ESPRESSO JARVIS-DFT Alexandria
2M+
Indexed Papers
150K+
Materials Entries
12
Property Databases
4
Compute Codes

Research Scenarios

Built for Real Research Questions

From high-throughput screening to targeted property engineering — see how material.codes handles complex research workflows.

Battery Material Screening

Identify promising Li-ion cathode candidates from first-principles data with application-specific voltage and capacity constraints.

"Find layered oxide cathodes with average voltage > 3.8V, capacity > 180 mAh/g, and thermal stability above 200°C"
  • Queries Materials Project + AFLOW for matching compositions
  • Cross-references electrochemical data from literature
  • Ranks candidates by synthesizability scores
  • Exports structures for DFT geometry relaxation

Top Result

LiNi0.8Mn0.1Co0.1O2 (NMC811)

V̄ = 3.86V 199 mAh/g Td = 215°C

Pipeline Steps

Literature search (847 papers)
Materials DB query (23 candidates)
DFT validation (in queue)

Semiconductor Band Engineering

Design alloy compositions to hit target band gaps for photovoltaic and LED applications using high-throughput DFT screening.

"InGaN alloys with direct bandgap between 2.6 and 3.0 eV for blue LED emission, lattice-matched to GaN substrate"
  • Retrieves composition-property relationships from AFLOW
  • Applies Vegard's law interpolation with bowing corrections
  • Generates band structure plots from existing calculations
  • Flags lattice mismatch strain for selected compositions

Optimal Composition

In0.18Ga0.82N

Eₐ = 2.77 eV Direct gap Δa/a = 0.8%

Screening Results

Compositions screened: 156
Within bandgap range: 23
Lattice-matched (<1%): 8

Catalyst Surface Analysis

Identify active sites, compute adsorption energies, and screen surface terminations for heterogeneous catalysis applications.

"CO2 reduction on Cu surfaces — compare (100), (110), (111) facets for formate vs CO selectivity"
  • Constructs surface slabs from bulk structure
  • Computes surface energies and identifies stable terminations
  • Submits NEB calculations for reaction barriers
  • Correlates results with selectivity trends from literature

Recommended Facet

Cu(100) — formate pathway

ΔEₐₐ = -0.42 eV Ea = 0.67 eV

Adsorption Energies (eV)

Cu(100)-0.42 (HCOO*)
Cu(110)-0.31 (CO*)
Cu(111)-0.19 (CO*)
Early Access · Limited Spots

Join the Waitlist

Be among the first researchers to access material.codes. Priority access for academic users.

No spam. We'll only reach out when your access is ready.

You're on the list!

We'll notify you at when your access is ready.

280+ researchers waitlisted
MIT, Stanford, ETH Zürich, NIMS