Cement Soundness Sampling Inspection and Testing Scheme

Cement soundness is a critical quality parameter that determines whether hardened cement paste will maintain a stable volume without delayed expansion, cracking, or distortion. This article presents a complete cement soundness sampling inspection and testing scheme, covering lot definition, representative sampling procedures, Le Chatelier boiling tests, autoclave expansion tests, and acceptance criteria based on GB, ISO, EN, and ASTM standards. It serves as a practical reference for concrete producers, construction sites, supervision engineers, and accredited third-party laboratories.

Technical Overview

Cement soundness (volume stability) refers to the ability of cement paste to undergo uniform volume change during setting and hardening without cracking, warping, or disintegration. Unsound cement can cause delayed expansion in concrete structures, leading to map cracking, spalling, and long-term durability failure. The principal causes of unsoundness are excessive free calcium oxide (fCaO), excessive free magnesium oxide (periclase), and excessive sulfur trioxide (SO3), which react slowly with water after the cement has hardened and generate expansive hydration products.

A soundness sampling inspection and testing scheme is a structured quality control program that defines how inspection lots are established, how statistically representative samples are drawn, which test methods are applied, and how results are judged for lot acceptance or rejection. The scheme is primarily governed by the following standards:

• GB 175 — Common Portland Cement (product requirements, including mandatory soundness conformity)
• GB 12573 — Sampling of Cement (sampling tools, methods, and sample quantities)
• GB/T 1346 — Test methods for water requirement of normal consistency, setting time and soundness of the portland cement (equivalent principles to EN 196-3)
• GB/T 750 — Autoclave method for soundness of Portland cement (comparable to ASTM C151)
• GB/T 176 — Chemical analysis of cement (SO3 determination, comparable to ASTM C114 and EN 196-2)
• ISO 2859-1 / GB/T 2828.1 — Attribute sampling plans for lot-by-lot inspection where statistical acceptance schemes are required

Testing Samples

Lot Definition

An inspection lot should consist of cement from the same manufacturer, same variety, same strength grade, and same batch (factory) number. Typical acceptance lot sizes on construction projects are 200 tons or less for bagged cement and 500 tons or less for bulk cement. Cement delivered in smaller quantities, from different batch numbers, or stored beyond three months must be re-inspected as separate lots.

Bagged Cement Sampling

Using random number tables or a randomization method, select no fewer than 20 bags from different positions and layers of the stack. From each selected bag, draw at least 1 kg of cement with a sampling tube inserted diagonally through the bag. Combine all increments into a single mixed sample of no less than 12 kg to ensure representativeness across the entire lot.

Bulk Cement Sampling

For cement delivered in bulk tankers or stored in silos, take increments from at least five evenly distributed sampling points or at five evenly spaced time intervals during discharge from the transport stream. Each increment should be no less than 1 kg, combined to form a mixed sample of no less than 12 kg. Continuous mechanical samplers may be used where available.

Sample Handling and Retention

The mixed sample is homogenized and then divided by splitting or cone-and-quartering into two equal portions of at least 6 kg each. One portion is submitted for testing, and the second is sealed in a moisture-proof container as a retention sample for no less than 90 days to permit retesting or dispute resolution. Each container must be clearly labeled with the manufacturer, cement variety, strength grade, batch number, sampling date, sampling location, and name of the sampler.

Testing Items

The following items constitute the core testing scope of the scheme:

• Water requirement for normal consistency (prerequisite test for paste preparation)
• Soundness by the boiling method (Le Chatelier apparatus) — the mandatory acceptance item
• Autoclave expansion test — required when magnesium oxide content is elevated or when specified by contract or design
• Sulfur trioxide (SO3) content — supplementary chemical verification of a known cause of delayed expansion

Testing Methods

Determination of Normal Consistency

Perform the test in a laboratory maintained at 20 ± 2 °C with relative humidity not less than 50%. Using the Vicat apparatus, prepare a cement paste with the measured water content that allows the standard plunger to settle to 5–7 mm from the base plate. The resulting water-to-cement ratio defines normal consistency and is used to prepare all subsequent soundness specimens.

Le Chatelier Boiling Method

Fill two clean, calibrated Le Chatelier molds with paste at normal consistency in two layers on oiled glass plates, cover with a second glass plate, and mark the indicator pointers. Cure the assemblies for 24 ± 2 hours in a moist cabinet at 20 ± 1 °C and relative humidity of at least 90%. After demolding, measure the initial pointer distance of each specimen to the nearest 0.5 mm. Place the specimens in a boiling tank, bring the water to boiling within 30 ± 5 minutes, and maintain boiling for 3 h ± 5 minutes. After cooling to ambient temperature, measure the final pointer distance. The expansion of each specimen is the increase in pointer distance. If the difference between the two specimens exceeds 4 mm, the test shall be repeated immediately on the same sample. The cement is judged sound when the average expansion is no greater than 5.0 mm.

Alternative Pat Test

Where permitted, the pat method may be used as an alternative: a pat approximately 70–80 mm in diameter with a 10 mm thick center is cured for 24 ± 2 hours and boiled under the same regime. The cement passes if no visible cracking occurs and the pat does not warp, as checked with a straight edge. In case of dispute, the Le Chatelier method is the referee procedure.

Autoclave Expansion Test

To evaluate expansion caused by magnesium oxide, mold 25 mm × 25 mm prisms from paste at normal consistency, cure for 24 hours, measure the initial length with a comparator, then subject the prisms to saturated steam at approximately 2.0 MPa (about 215–216 °C) for 3 hours in an autoclave. After depressurization, cooling, and measurement of final length, calculate linear expansion. A typical acceptance limit is 0.80% or as specified in the applicable product standard.

SO3 Determination

Determine sulfur trioxide content by gravimetric barium sulfate precipitation or instrumental methods in accordance with GB/T 176. Typical chemical limits are 3.5% for ordinary Portland cement and up to 4.0% for slag cement, as specified in GB 175. Elevated SO3 combined with unsoundness indicators warrants autoclave and chemical verification.

Acceptance and Retest Rules

A lot passes when the boiling soundness result is 5.0 mm or less and all supplementary indicators conform. If the result fails, a second sample may be drawn from the sealed retention portion and retested in parallel. If both results conform, the lot is accepted; if either fails, the lot is rejected or downgraded. For large projects, attribute sampling plans with a specified acceptance quality limit (AQL) per ISO 2859-1 may be applied to define sample sizes and acceptance numbers. Referee testing should be entrusted to accredited third-party laboratories.

Applications

This sampling inspection scheme is applicable to a wide range of engineering and industrial scenarios, including:

• Incoming quality control of cement at ready-mixed concrete plants and mortar producers
• Factory acceptance and periodic type verification at precast concrete element manufacturers
• Site material acceptance testing for building, road, bridge, and hydraulic construction projects
• Supervision and consultant verification of contractor material declarations
• Quality dispute arbitration and failure investigations involving cracked or distorted concrete works
• Conformity assessment support by accredited third-party laboratories for domestic and international projects

FAQ

What is the minimum sample quantity required for a soundness test?

A combined mixed sample of no less than 12 kg is required, drawn from at least 20 bags for bagged cement or at least five points or intervals for bulk cement, which is then split into a test portion and a sealed retention portion.

How long must retention samples be kept?

Retention samples are sealed in moisture-proof containers and stored for no less than 90 days, allowing retesting in the event of a failed result or a later quality dispute.

What causes cement to fail the soundness test?

Failure is most commonly caused by excessive free calcium oxide, high free magnesium oxide, or excessive SO3 content, all of which produce delayed expansive reactions after the cement has hardened.

What is the difference between the boiling test and the autoclave test?

The boiling (Le Chatelier) test accelerates the hydration of free calcium oxide and detects expansion within 3 hours of boiling. The autoclave test applies high-pressure steam at about 2.0 MPa for 3 hours and is used to evaluate expansion caused by magnesium oxide, which reacts more slowly and requires more severe acceleration.

What should be done if a lot fails the soundness test?

Retest a second portion of the original retained sample. If the retest also fails, the lot must be rejected, returned to the supplier, or restricted from structural use, and the failure should be documented and reported through the project quality management system.

Can this scheme be adapted for statistical lot acceptance?

Yes. Attribute sampling plans with defined AQL values in accordance with ISO 2859-1 or GB/T 2828.1 can be layered onto the lot definition to establish sample sizes, acceptance numbers, and rejection numbers for large-volume supply contracts.