138 Damage to concrete structures
5.3.1.1.2 Alkali carbonate reaction
The alkali carbonate reaction is less common, although several cases have
been reported. Some dolomitic aggregates containing CaMg(CO 3 ) 2 can
react with hydroxyl and alkali ions according to a complex reaction mechanism, which is not yet fully understood. Two main subsequent reactions
can be considered in the case of ACR:
CaMg(CO 3 ) 2 + 2NaOH → Mg(OH) 2 + CaCO 3 + Na 2 CO 3
(5.3)
Ca(OH) 2 + Na 2 CO 3 → CaCO 3 + 2NaOH
(5.4)
The first reaction is the so-called dedolomitization, transforming dolomite (CaMg(CO 3 ) 2 ) into brucite (Mg(OH) 2 ) and calcite (CaCO 3 ). This
process is not expansive, so no expansion cracks will be formed. On the
contrary, a reduction of 5.1% of the total solid volume is obtained.
The second reaction, a dissolution–precipitation process, describes the
formation of secondary calcite, which is deposited in the voids of the interfacial transition zone (ITZ) surrounding the dolomite aggregate. Although
the absolute volume increases with 10.2%, no expansion stresses are
obtained as the products are formed in the voids of the ITZ. The given
simplified reaction shows that Na + ions are regenerated through the watersoluble NaOH, maintaining a high alkalinity, favouring a potential alkali
silica reaction.
The reaction process further proceeds near the dolomite surface forming
a narrow rim of hydrotalcite (6MgO∙Al 2 O 3 ∙CO 2 ∙12H 2 O) and additional
calcite, while increasing the porosity of the dolomite.
As the cause of this damage process is typically ascribed to the dedolomitization (reaction of dolomite with hydroxyl ions and subsequent crystallisation of brucite), it is commonly considered to be a different type
of alkali aggregate reaction in comparison with alkali silica reaction.
However, the damage process due to the alkali carbonate reaction (ACR)
is controversial.
In the 1990s, Katayama stated that ACR, in fact, was a combination
of expansive ASR of amorphous silica and harmless dedolomitization of
dolomitic aggregate (Katayama 1992). By means of petrographic analysis in combination with SEM-EDS analysis, Katayama (2010) recently was
able to give more evidence for this statement. Dolomitic aggregates did not
develop expansion cracks unless ASR was involved. It is the alkali silica
gel which is responsible for the crack formation in concrete. This is further confirmed by Grattan-Bellew et al. (2010). For ACR-susceptible aggregates, Grattan-Bellew et al. found a good correlation between the amount
of quartz in the aggregates and the expansion of the concrete prisms, referring to the more important role of ASR. They finally conclude that the
5.3.1.1.2 Alkali carbonate reaction
The alkali carbonate reaction is less common, although several cases have
been reported. Some dolomitic aggregates containing CaMg(CO 3 ) 2 can
react with hydroxyl and alkali ions according to a complex reaction mechanism, which is not yet fully understood. Two main subsequent reactions
can be considered in the case of ACR:
CaMg(CO 3 ) 2 + 2NaOH → Mg(OH) 2 + CaCO 3 + Na 2 CO 3
(5.3)
Ca(OH) 2 + Na 2 CO 3 → CaCO 3 + 2NaOH
(5.4)
The first reaction is the so-called dedolomitization, transforming dolomite (CaMg(CO 3 ) 2 ) into brucite (Mg(OH) 2 ) and calcite (CaCO 3 ). This
process is not expansive, so no expansion cracks will be formed. On the
contrary, a reduction of 5.1% of the total solid volume is obtained.
The second reaction, a dissolution–precipitation process, describes the
formation of secondary calcite, which is deposited in the voids of the interfacial transition zone (ITZ) surrounding the dolomite aggregate. Although
the absolute volume increases with 10.2%, no expansion stresses are
obtained as the products are formed in the voids of the ITZ. The given
simplified reaction shows that Na + ions are regenerated through the watersoluble NaOH, maintaining a high alkalinity, favouring a potential alkali
silica reaction.
The reaction process further proceeds near the dolomite surface forming
a narrow rim of hydrotalcite (6MgO∙Al 2 O 3 ∙CO 2 ∙12H 2 O) and additional
calcite, while increasing the porosity of the dolomite.
As the cause of this damage process is typically ascribed to the dedolomitization (reaction of dolomite with hydroxyl ions and subsequent crystallisation of brucite), it is commonly considered to be a different type
of alkali aggregate reaction in comparison with alkali silica reaction.
However, the damage process due to the alkali carbonate reaction (ACR)
is controversial.
In the 1990s, Katayama stated that ACR, in fact, was a combination
of expansive ASR of amorphous silica and harmless dedolomitization of
dolomitic aggregate (Katayama 1992). By means of petrographic analysis in combination with SEM-EDS analysis, Katayama (2010) recently was
able to give more evidence for this statement. Dolomitic aggregates did not
develop expansion cracks unless ASR was involved. It is the alkali silica
gel which is responsible for the crack formation in concrete. This is further confirmed by Grattan-Bellew et al. (2010). For ACR-susceptible aggregates, Grattan-Bellew et al. found a good correlation between the amount
of quartz in the aggregates and the expansion of the concrete prisms, referring to the more important role of ASR. They finally conclude that the