A materials database is a structured, searchable collection of materials information, for example composition, mechanical and physical properties, curves and standard data. Good materials databases also provide the context for each value: material condition, test conditions, unit and source. Without this context, a property value can hardly be assessed for design or simulation.
This article covers materials databases as a data source. How companies manage their own and external materials data across processes and systems is described in the article on materials data management.
Types of materials data
Materials databases differ mainly in the type of data they contain:
| Type | Typical content | Example |
|---|---|---|
| Standard and registration data | designations, chemical composition limits, information specified in standards or registered with associations | Register of European Steels in StahlDat SX; Teal Sheets with the designations and chemical composition limits registered with the Aluminum Association |
| Test data | measured values and curves from test programs | flow curves, hardenability curves, thermophysical test data in StahlDat SX |
| Manufacturer data | information from material manufacturers’ data sheets | Polymaster with data from manufacturer data sheets |
| Time-dependent data | creep rupture strength and creep behavior at high temperatures | creep data for iron, nickel and cobalt base alloys |
| Simulation data | calculated properties for compositions and conditions | temperature-dependent properties from JMatPro |
| Allowables | statistically based design values | MMPDS for metallic aerospace materials |
These types cannot be mixed at will. A minimum value specified in a standard, a typical value from a data sheet and a statistically based design value answer different questions, even if they describe the same property.
Allowables: statistically based values
In aviation, design values must be based on enough tests and established on a statistical basis. The US certification regulation 14 CFR § 25.613, for example, requires values that are met with 99 % probability at 95 % confidence for single-load-path structure, and 90 % probability at 95 % confidence for redundant structure.
MMPDS (Metallic Materials Properties Development and Standardization) publishes corresponding values for metallic materials, including A-basis and B-basis values. The values are recognized by the FAA, DoD and NASA within their limitations. Matplus provides access to MMPDS via mmpds.eu.
The basis values differ in their statistical basis. The A-basis value is the lower of the statistically calculated T99 value, which at least 99 % of the population is expected to equal or exceed with 95 % confidence, and the specification minimum (S-basis). The B-basis value is based correspondingly on the T90 value. The S-basis value is the minimum value specified in the material specification. Typical values from data sheets, by contrast, are not statistically based design values.
Curves, temperature and time
Many properties are not single values but curves or multidimensional data fields. StahlDat SX, for example, contains hardenability curves, flow curve models, thermophysical test data and more than 400 CCT and TTT diagrams. The JAHM database provides temperature-dependent properties such as Young’s modulus, thermal expansion and thermal conductivity, according to its provider for more than 18,900 materials in 184,000 data curves with 50 properties.
At high temperatures, time also plays a role. When steels and high-temperature alloys are loaded for longer periods above about 0.4 of their absolute melting temperature, strength and deformation become time-dependent. Room-temperature strength values do not describe this behavior; design then requires creep data.
Criteria for selecting a materials database
- Coverage: does the database contain the required material groups, conditions and product forms?
- Properties: are the required properties available, including temperature-dependent values or curves, not only single values at room temperature?
- Type of values: is it clear whether values are standard values, test values, typical values, calculated values or allowables?
- Provenance: are source, edition of the standard, test conditions and, where relevant, sample size stated?
- Cross-referencing designations: can materials be found and compared via national and international designations? For steel, the material number serves this purpose; for copper materials, for example, the Copper Key.
- Maintenance: is the database updated when standards change or new materials are added?
- Use: can data be exported, for example to CAE systems, and transferred into your own systems? Which usage rights apply?
Data quality and provenance
What matters for assessing a value is less the database than the information on the individual record. Important questions are: Which material in which condition was tested? Under which conditions? Is the value a single value, a mean, a minimum or statistically based? Which source and which edition does it come from?
For steel, the unique material number helps to assign data from different sources to the same steel grade. Calculated values from materials simulation should remain labeled as such.
Matplus materials databases
Matplus offers a range of materials databases, including:
- Matglobe®: reference data for engineering, design and modeling
- MMPDS: statistically based design allowables for metallic materials
- StahlDat SX: Register of European Steels (Stahl-Eisen-Liste) as well as sheet steel data, hardenability curves, phase transformation data, flow curves and thermophysical test data
- Polymaster: polymers and polymer composites with data from manufacturer data sheets
- AluSelect: mechanical, physical and chemical properties of widely used aluminum alloys
- Teal Sheets: alloy designations registered with the Aluminum Association, with chemical composition limits for more than 500 wrought aluminum alloys
- Copper Key: comparison of national and international standards, material data sheets and properties of copper materials, offered by the Kupferverband e.V. (German Copper Association) and technically implemented by Matplus
- Creep data: creep data for iron, nickel and cobalt base alloys; the underlying 2019 book edition covers 199 iron-based, 77 nickel-based and 23 cobalt-based alloys
- JAHM database: temperature-dependent material properties for FEA and CAD