Cement autoclave soundness testing is the authoritative method for evaluating the volume stability of hydraulic cement by exposing standardized paste prisms to saturated steam at approximately 2.0 MPa. This accelerated test reveals delayed expansion caused by slowly hydrating free magnesium oxide and free calcium oxide, defects that cannot be reliably detected by ordinary boiling tests. This guide explains the testing principle, sample requirements, procedures in accordance with ASTM C151 and GB/T 750, evaluation criteria, and practical applications in cement quality control.
Technical Overview
Soundness refers to the ability of hardened cement paste to maintain a stable volume over time without cracking, distortion, or disintegration. Unsoundness in cement is primarily caused by free calcium oxide (free lime) and free magnesium oxide (periclase), which hydrate very slowly after the surrounding paste has already hardened. The hydration of these oxides produces expansive reaction products that generate internal stress, leading to delayed cracking and loss of structural integrity.
The standard boiling test (Le Chatelier apparatus per EN 196-3, or the pat test per GB/T 1346) is effective for detecting expansion caused by free lime, but it is insufficient for magnesium oxide, because periclase crystals hydrate too slowly even at 100°C. The autoclave method overcomes this limitation by subjecting hardened cement paste specimens to saturated steam at 2.0 MPa gauge pressure, corresponding to a temperature of approximately 215°C. Under these severe conditions, the hydration of free magnesium oxide is dramatically accelerated, and any potential for delayed expansion is revealed within hours rather than years.
- Governing test methods: ASTM C151 (Standard Test Method for Autoclave Expansion of Hydraulic Cement), GB/T 750 (Autoclave Method for Soundness of Cement), and ASTM C490 for length-change apparatus
- Specification limits: acceptance criteria defined in ASTM C150 and GB 175 for various cement types
- Core apparatus: high-pressure steam autoclave, prism molds measuring 25 mm × 25 mm × 280 mm (or 285 mm per ASTM), and a length comparator with an effective gauge length of 250 mm
- Test severity: saturated steam at 2.0 ± 0.05 MPa maintained for 3 hours
Testing Samples
Cement Paste Prisms
The standard specimen for autoclave soundness testing is a cement paste prism with a cross-section of 25 mm × 25 mm and a total length of 280 mm (285 mm under ASTM C151), fitted with stainless steel gauge studs at both ends to provide an effective gauge length of 250 mm. The paste is prepared using the water content corresponding to the standard consistency of the cement, determined in accordance with ASTM C187 or GB/T 1346. A minimum of two or three companion prisms are typically molded from each sample batch to allow for retesting and result averaging.
Sampling Requirements
Cement samples shall be representative of the delivery lot or production batch, obtained using recognized sampling procedures such as ASTM C183 or GB 12573. The sample must be protected from moisture and contamination, and testing should be carried out promptly. For factory production control, samples are usually drawn from the cement silo discharge point or packaged units at defined intervals.
Specimen Conditioning
Immediately after molding, specimens are placed in a moist cabinet or curing room at 20 ± 1°C (GB) or 23 ± 2°C (ASTM) with a relative humidity of at least 90%. The prisms are demolded after approximately 24 hours of moist curing, at which point they possess sufficient strength for handling and initial length measurement.
Testing Items
The autoclave soundness test determines the following measurable quantities:
- Initial length of each hardened paste prism after 24 hours of moist curing
- Final length of each prism after high-pressure steam treatment and cooling
- Autoclave expansion, expressed as a percentage of the 250 mm effective gauge length
- Visual observations of cracking, warping, curving, or edge disintegration after autoclaving
- Qualitative judgment of cement soundness: qualified (expansion within the specified limit) or unqualified (expansion exceeding the limit)
Testing Methods
Specimen Preparation and Initial Measurement
Cement paste is mixed at standard consistency water content, molded into the prisms, and cured in the moist cabinet for approximately 24 hours. After demolding, each specimen is placed on the comparator with gauge studs seated in the measuring anvils, and the initial length is recorded to the nearest 0.01 mm. The comparator must be verified against a reference invar bar before each measurement session, and specimens should be rotated to confirm seating stability.
Autoclave Exposure Procedure
The prisms are positioned vertically on a perforated rack inside the autoclave, spaced apart so that all surfaces are exposed to steam. Sufficient water is added, and the autoclave is sealed and heated. When steam begins to discharge from the vent valve, the valve is held open for approximately 10 minutes to expel entrained air, ensuring a pure saturated steam atmosphere. The pressure is then raised at a controlled rate so that the gauge pressure reaches 2.0 ± 0.05 MPa within the standard time window of 45 to 75 minutes, and this pressure is maintained for 3 hours. The safety relief valve and pressure gauge must be within calibration, and only trained personnel may operate the pressure vessel.
Cooling and Final Measurement
At the end of the 3-hour holding period, the heat source is switched off and the autoclave is allowed to cool so that the gauge pressure returns to atmospheric over a controlled period of roughly one hour, avoiding thermal shock to the specimens. Once the pressure is fully released, the autoclave is opened and the prisms are immersed in water at the standard laboratory temperature until thermal equilibrium is reached. The final length of each specimen is measured on the same comparator within 15 minutes of removal, recorded to the nearest 0.01 mm, and any visible defects are documented.
Calculation and Acceptance Criteria
Autoclave expansion is calculated as the difference between the final and initial lengths, divided by the effective gauge length of 250 mm and multiplied by 100, with the result reported to the nearest 0.01%. Under ASTM C150, the maximum permissible autoclave expansion for most portland cements is 0.80%. Under GB 175, when the magnesium oxide content of the cement exceeds 5.0%, the autoclave test becomes mandatory and the expansion must not exceed 0.80%. If a specimen shows curving or edge cracking without measurable over-expansion, the result is treated with caution and a retest is normally performed. Accredited third-party laboratories typically require consistent results from duplicate prisms before issuing a final conformity judgment.
Applications
Autoclave soundness testing is applied across the entire cement and concrete supply chain:
- Factory quality control at cement plants, particularly those using raw materials with elevated magnesium carbonate content
- Conformity assessment of ordinary portland cement, blended cement, and specialty cement against national and international specifications
- Verification of soundness when magnesium oxide content falls in the critical range between 5.0% and 6.0%, where the boiling test alone is not conclusive
- Durability evaluation for mass concrete structures such as dams, foundations, and hydraulic works, where delayed expansion would be catastrophic
- Research and development of magnesium oxide-based expansive additives used to compensate for shrinkage in mass concrete
- Independent inspection, export certification, and dispute resolution performed by accredited third-party laboratories
- Qualification of new cement formulations, alternative fuels and raw materials, and clinker production adjustments
FAQ
What is the difference between the boiling test and the autoclave test?
The boiling test, performed with the Le Chatelier apparatus or the pat method, accelerates only the hydration of free calcium oxide because it operates at atmospheric pressure and 100°C. Free magnesium oxide hydrates far too slowly at 100°C to be detected, so the autoclave test raises the temperature to approximately 215°C under 2.0 MPa saturated steam, forcing periclase to hydrate within hours and exposing potential delayed expansion.
Why does free magnesium oxide cause cement to be unsound?
Periclase crystals hydrate to form brucite, a reaction accompanied by a significant increase in solid volume. Because this reaction continues slowly for months or years after the cement paste has hardened and gained rigidity, the localized volume growth generates internal expansive stress that produces map cracking, edge crumbling, and eventual loss of strength.
What is the acceptance criterion for autoclave expansion?
The most widely applied limit is 0.80% maximum expansion, as specified in ASTM C150 for portland cements and in GB 175 for Chinese cements when the test is required. Certain specialty or blended cement specifications may impose stricter limits, so the relevant product standard should always be consulted.
When is the autoclave test mandatory?
Under GB 175, the autoclave test is mandatory when the magnesium oxide content of the cement exceeds 5.0%. It is also required in contractual specifications for mass concrete projects, whenever soundness disputes arise, and whenever a cement producer changes raw materials or clinker chemistry in a way that could affect magnesia hydration behavior.
How long does a complete test take?
A full test cycle, from paste mixing through 24 hours of moist curing, the approximately 6-hour pressurized autoclave cycle including heating and cooling, and final measurement, normally takes about 30 hours. Laboratories therefore typically report results within two working days of sample receipt.
