The cement used for this study was the Type I/II Ordinary Portland Cement (OPC) which conforms to the ASTM C150 specifications [30]. In this study, two fly ashes were used in the majority of the mixtures including a low-calcium (LC) and high-calcium (HC) fly ash. The two fly ashes conform to the ASTM C618 [31] and were procured from plants in Texas, USA. The compositions of all cementitious materials used are given in Table 1, together with the calculated Bogue composition for the OPC.

The coarse and fine aggregates used for this research were of crushed and manufactured limestone for the coarse and fine aggregate, respectively. The materials were procured from a local source in central Texas. The physical properties are highlighted in Table 2.

A total of 23 concrete mixtures were produced with the concretes divided into

three series of mixtures designated LC, HC, and CONT each varying in water-to-binder (W/binder) contents (0.50 and 0.45) and total cementitious contents (310, 340, 370, and 400 kg/m³). The first two series consisted of mixtures with either a low-calcium (LC) or high-calcium (HC) fly ash used as a cement clinker replacement ranging from 15% to 50% at 15% intervals by mass of total cementitious content. Lastly, a final series included control mixture (CONT) with no SCMs replacement utilizing the Type I/II OPC with varying cementitious contents. The mixture proportions together with fresh concrete properties are given in Table 3.

For mixing, the mixer was initially cleaned by rinsing and draining the water. Aggregating batching was done by mixing the aggregates well in the mixer for approximately five minutes to ensure uniformity. Coarse aggregates were added first and were allowed to mix for three minutes, following by fine aggregates for two minutes. Approximately 70% of the total water was added and mixed for another three minutes. Subsequently, the cementitious materials were added and allowed to mix for five minutes following by the remaining water for another three minutes. The mixture was allowed to rest for another three minutes after removing from the mixer. The mixing procedure was followed in accordance with ASTM C192 [32].

The carbonation chamber was large enough to accommodate 50 prisms when specimens were placed vertically. One mechanical fan was placed in the chamber to ensure air circulation. A humidity probe placed inside the chamber provided information of the humidity conditions inside throughout the test using a data acquisition system. The prisms were positioned in the chamber in such a manner that all specimens were spaced at a minimum of 5 mm between prisms and chamber walls as shown in Figure 1. For this experiment, the RH of the chamber was controlled and monitored at 57% ± 5% through the use of an industrial dehumidifier attached to the chamber; the dehumidifier reduces the humidity in the enclosed environment by removing water vapor from the air through ventilation ducts and were attached in such a way that the inflow and outflow of CO₂ through the system was maintained throughout the test. Lastly, the CO₂ level inside the chamber was adjusted and maintained at 4.0% ± 0.5% by volume through the use of a gas regulator and oxygen tank attached to a solenoid valve that would trigger the system to adjust when levels were beyond the threshold. A digital display was placed above the carbon tank which indicated the CO₂ level and the temperature inside the carbonation chamber. Figure 2 demonstrates the entire accelerated carbonation setup.

Each mixture included a set of 100 × 200 mm concrete cylinders for determining compressive strength at various ages, and prismatic specimens of 100 × 100 × 350 mm (4 × 4 × 14 inches) to determine the carbonation depth under accelerated conditions. For carbonation test, a pair of specimens were prepared per mix and demolded after being stored at 23˚C ± 3˚C (73˚F ± 5˚F) under wet burlap and plastic for 24 hours. Thereafter, one prism per concrete mix was immediately prepared for transfer to the carbonation chamber, while the second prism was placed in a moist curing chamber for additional 6 days (7-day total moist cure) to explore the influence of curing age on the rate of carbonation. In the case of compression testing, twelve 100 × 200 mm (4 × 8 inches) cylinders were cast per mix and standard procedure according to ASTM C31 was followed for making the specimens. The cylinders were allowed to cure in the laboratory at 23˚C ± 3˚C (73˚F ± 5˚F) under wet burlap and plastic for 24 hours prior to being demolded. Thereafter, samples were transferred to the moist curing chamber and tested after desired age of curing was obtained (2, 7, 28, and 90 days).

Following an initial 1 day and 7 days of moist curing, the carbonation prisms were allowed to air dry in a controlled environment (23˚C, 50% RH) for a minimum of 3 hours to allow removal of moisture from the concrete as well as easier control of the relative humidity in the chamber during testing; the short drying period minimized the length and curing age for which each specimen was intended to be subjected to (1 and 7-day moist cure) prior to testing. While it is likely the specimens were still in relatively saturated conditions upon placement into the chamber, the attached dehumidifier allowed for faster control of the relative humidity conditions to reach equilibrium between the specimens and surrounding environment. The prisms were then vertically placed into the carbonation chamber with CO₂ level of 4.0% ± 0.5% by volume, temperature of 23˚C ± 2˚C, and relative humidity (RH) of 57% ± 5%.

For each specimen, the age at which the carbonation depth was measured was 28, 56, 63, 70, and 105 days of exposure to CO₂. An approximately 50 mm thick slice was cut off the prism at each stage of testing with a mechanical cutting device which left a freshly broken surface. The split slice is then brushed and the depth of carbonation was determined by spraying the freshly fractured surface with a 0.5% phenolphthalein solution. Following 1 h ± 5 min, a picture of the colored surface was taken and the carbonation depth was determined with a ruler at 5 equidistant points on each of the four sides of the slice giving a total of twenty measurement points (see Figure 3). Measurements were not made within 15 mm of the corners. If the measurement falls to an aggregate, it is extrapolated on both sides of this aggregate. The average carbonation depth for each side was calculated and the average carbonation depth of the concrete mix is reported.

Figure 4 and Figure 5 demonstrate the carbonation patterns for specimens with

and without low- and high-calcium fly ash at 0.50 w/cm ratio moist cured for 7 days, respectively. Based on the information obtained, it is noticeable that carbonation depths increase with time of exposure to CO₂ for all mixtures. However, the carbonation depth over time was more dramatic with increasing amounts of fly ash, especially for mixtures incorporating 50% fly ash. For mixtures incorporating low-calcium fly ash, the rate of CO₂ ingress was more pronounced when compared to high-calcium fly ash concretes. In fact, mixtures incorporating low-calcium fly ash at 50% replacement had such a detrimental effect that specimens were fully carbonated by no later than 56 days of exposure (see Figure 5). It can be clearly inferred that high-calcium fly ash is more resistant against carbonation as compared to low-calcium fly ash. This is likely attributed to the higher calcium oxide (CaO) content present in the chemical composition of high-calcium fly ash. The CaO present in the hydrated cement paste will be transferred into CaCO₃ upon carbonation from CO₂ exposure. Consequently, carbonation rate of concrete is mainly governed by the CO₂ buffer or binding capacity from the availability of calcium bearing hydrated per unit volume of the cement paste as suggested by Leemann et al. [27]. Hence, the extent of chemical interaction (and binding of CO₂) will be reduced compared with OPC concrete [21]. Additionally, the lower portlandite content provides an initial lower pH in the pores solution, in which typically lead to more rapid carbonation of other calcium-bearing hydrates such as calcium silicate hydrates [10] [33]. While the carbonation rate for fly ash mixtures is likely to decrease with increasing curing time, the results demonstrate that the carbonation rate at an early age is significantly faster in insufficiently cured high-volume SCM concrete. The longer the moist curing period for concrete, the higher the degree of hydration of the cement and the lower the porosity and permeability of the concrete is [29].

In general, for mixtures incorporating high-calcium fly ash, the results show that the average carbonation depth across all ages increased by 18%, 54%, and 108% for 15%, 30%, and 50% fly ash replacement in comparison to the control. On the other hand, mixture including low-calcium fly ash showed an increase of 33%, 100%, and 214% for 15%, 20%, and 50% fly ash replacements in comparison to the control (note that 50% low-calcium fly ash mixtures were fully carbonated by 56 days of exposure).

It was also observed that specimens with 1-day moist curing are more prone to carbonation ingress as compared with 7-day moist curing specimens. Figure 6 and Figure 7 describe the effect of 1-day and 7-day moist curing on both low- and high-calcium F fly ashes at different cementitious material contents, and it is evident that longer moist curing period helps the concrete to improve resistance against carbonation. Because the carbonation is related to permeability of concrete, higher carbonation should be expected in inadequately cured concrete, with or without fly ash [7]. Moreover, increasing levels of fly ash, i.e. greater than 30%, significantly increase the rate of carbonation. Several such cases were tested, and it can be clearly inferred that specimens with 1-day moist curing highlight more carbonation depth as compared to specimens with 7 days of moist curing regardless of w/cm ratio or cementitious materials content.

Previous studies suggested that with a decrease in water-cementitious material ratio the carbonation depth will also decrease since the concrete will be denser and less permeable. In the present study, a comparison between low- and high-calcium fly ashes consisting of 0.50 and 0.45 w/cm ratio is demonstrated in Figure 8 and Figure 9. In both cases, 0.5 w/cm ratio specimens indicated higher carbonation depths than 0.45 w/cm ratio specimens with 7-day moist curing. However, this pattern was not consistent throughout all days of testing. Low-calcium fly ash specimens followed an identical trend however the difference between 0.50 and 0.45 w/cm ratio specimens was not significant.

A comparison between different cementitious contents was made at a given w/cm ratio and fly ash replacement. Figure 10 and Figure 11 indicate the comparison between specimens consisting of 340, 370 and 400 kg/m³ at 0.50 w/cm

ratio after 7 days of moist curing. The results indicated a slight decrease in the carbonation depth earlier on however, little significance was observed between the mixtures with extended exposure time in. This pattern was observed in the majority of the cases in which mixtures with increasing total cementitious content, a slight decrease in the carbonation depth was observed. This may be attributed to the availability of the calcium hydroxide (Ca(OH)₂) content per unit volume of the cement paste in each mixture, especially with increasing clinker content. As a result, the binding capacity and thus, chemical reactivity with CO₂ is increased [27].