Proceedings of the International rilem conference Materials, Systems and Structures in Civil Engineering 2016

105

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

EVALUATION OF CONCRETE’S RESISTANCE TO PHYSICAL 

SULFATE SALT ATTACK 

 

Semion Zhutovsky 

(1)

, R. Douglas Hooton 

(1)

 

 

(1) University of Toronto, Canada 

 

 

 

 

 

 

 

 

Abstract 

Physical sulfate salt attack (PSA) is one of the most severe and rapid deterioration 

mechanisms that can take place in concrete. Yet there is no standard method for evaluation of 

concrete resistance to PSA. Testing of concrete’s resistance to PSA has two main aspects – 

exposure conditions and evaluation of deterioration. Evaluation of concrete’s resistance to 

PSA often requires a long time, which is inappropriate for evaluation of concrete mixtures for 

a construction project. Many studies report sulfate resistance based on subjective visual 

ratings, which is inadequate as a durability design criterion. The objective of this research is 

to identify exposure conditions and deterioration evaluation methods suitable for standard 

testing that can be used for rapid comparison of mixture compositions, durability design, and 

analysis of life cycle cost. Various methods for assessment of deterioration were applied using 

selected exposure. It was found that 100 thermal cycles between 5 and 30 °C immersed in 

30% sodium sulfate solution were sufficient to assess the resistance of a range of mortar 

mixtures.  The most suitable techniques for evaluation of deterioration rate was mass loss, 

while fundamental resonance frequency and ultrasonic pulse velocity were found to be 

unsuitable. 

 

 

1. Introduction 

 

Physical sulfate salt attack (PSA) is one of the most severe and rapid deterioration 

mechanisms that can take place in concrete. Salt scaling, salt weathering, or salt hydration are 

the terms that are often used for PSA [1]. PSA is sometimes confused with chemical sulfate 

attack [2], [3]. Unlike in sulfate attack of chemical origin, no chemical interaction between 

sulfate salts and cement minerals is involved in PSA [3]. Sulfate attack of concrete has been 

the subject of extensive research, though the mechanisms of PSA were often overlooked [4]. 

A recent Portland Cement Association report on the results of testing concrete resistance to 

---

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

sulfate attack declares that the PSA damage of concretes can be significantly more extensive 

than the damage caused by chemical sulfate attack [5].   

 

It has been demonstrated by previous research that water to binder (w/b) ratio is the most 

important parameter for the resistance to sulfate attack [5]. The current North American 

standards limit w/b ratio to 0.40 and 0.45 in Canada [6] and USA [7], respectively, for the 

severe conditions of sulfate exposure. However, the reduction of w/b ratio does not prevent 

deterioration, though it enhances the resistance to sulfate attack [8]. Supplementary 

cementitious materials (SCM) are often used to make concrete tolerant for sulfate rich 

environment. While SCMs improve the resistance to chemical sulfate attack by diluting 

aluminates in the binder, some literature sources report that they may even increase the 

susceptibility to PSA [5], [9], [10]. So there is a controversy about the resistance of different 

binders and mixture designs to PSA.  

 

  

Currently, there are no standard methods for testing concrete resistance to PSA [2]. Existing 

standard test methods ASTM C452 [11] and ASTM C1012 [12] were prepared for testing the 

resistance of binders to chemical sulfate attack. In these standards, the expansion of mortar 

bars is used as a measure of deterioration, while PSA is typically not associated with 

expansion. The form PSA damage is usually the gradual surface scaling, much like the 

damage caused by freezing and thawing [3]. This is because PSA causes damage by means of 

the cycles of crystallization, dissolution and phase transitions of sulfate salts [13]. Testing of 

concrete’s resistance to PSA has two main aspects – exposure conditions and evaluation of 

deterioration. Historically, field testing was used for testing sulfate resistance of concretes [5], 

[9], [8], [14]. Such tests could take from 8 to 40 years. Such long time is inappropriate for 

evaluation of concrete mixtures for a construction project. Many such studies report sulfate 

resistance based on visual ratings, which are subjective and inadequate as a design criterion 

for durability. Thus it is of great interest to identify exposure conditions and deterioration 

evaluation methods suitable for accelerated standard testing that can be used for rapid 

comparison of mixture compositions, durability design, and analysis of life cycle cost. 

  

Different accelerated exposure conditions for the testing of PSA resistance of concrete are 

reported in the literature [13], [15], [16]. Typically, the most significant damage in PSA is 

caused by phase transitions between sulfate salts: thenardite ( 

Na2SO4) and mirabilite 

( 

Na2SO4·10H2O) [13]. To activate this mechanism, either relative humidity (RH) or 

temperature variations are needed. Thus, PSA testing exposures can be categorized into three 

groups: wetting and drying cycles, partial submerged samples, and fully submerged samples. 

In first group, changes in RH are triggered by wetting and drying cycles, inducing the phase 

changes between mirabilite and thenardite. However, change of RH inside concrete or mortar 

samples is slow, and significant time is required to reach RH equilibrium inside a sample. For 

this reason, there will always be a moisture gradient through the sample cross-section, which 

results in partial conversion of thenardite to mirabilite, and the testing time required to 

achieve the damage level needed to quantify PSA resistance in a wide range of mixtures is 

long. In partially submerged samples, the rate of deterioration may be fast, but the damage is 

localized in the evaporation zone, which makes it difficult to quantify the resistance to PSA. 

On the other hand, thermal cycling of fully submerged samples seems to be a very promising 

PSA exposure, because thermal equilibrium can be achieved quickly. When temperature is 

---