Using Fines from Recycled High-Quality Concrete as a Substitute for Cement

Abstract

Concrete manufacturing and recycling must evolve to meet sustainability and carbon reduction demands. While the focus is often on reusing coarse aggregates, fine fractions are also produced during recycling. This study explores using ground fine fractions (0/2) as a partial cement substitute. The fines were characterized for their mineralogical, chemical, and physical properties, and experiments were conducted on pastes and mortars with 0% to 30% cement substitution, including isothermal calorimetry and strength tests. Two concrete mixes—a reference mix with natural aggregates and CEM I, and a mix with 10% concrete fines replacing CEM I—using recycled sand and coarse aggregates were tested for compressive strength, carbonation, shrinkage, and freeze–thaw resistance. The results indicated that the recycled concrete had a comparable strength to the reference and a slightly reduced durability in freeze–thaw conditions. In terms of shrinkage, recycled concrete with 10% concrete fines had an increased drying shrinkage and a lower autogenous shrinkage due to the water retention capacity of the recycled aggregates.

1. Introduction

One of the most prominent building materials worldwide is concrete due to its advantages in terms of ease of use and affordability. Nevertheless, the production process has a significant environmental impact which is starting to outweigh the many advantages. The manufacturing process of cement, one of the primary ingredients of concrete, is known to produce up to 8% of the world’s anthropogenic CO₂ emissions [1]. Concrete manufacturing must be modified to satisfy the growing demand for sustainability and reduce carbon emissions. In order to produce concrete that is more sustainable, recycling concrete may be essential.

The concrete used for recycling is usually obtained from construction and demolition waste or from production waste of the precast industry. Currently, coarse recycled concrete aggregates (grain size > 4 mm) are the primary focus of the concrete recycling sector. The coarse fraction is thought to be the most suitable for utilization in concrete due to the fact that it resembles natural aggregates the most [2]. This preference is also visible at the normative level. For instance, the European standard, NBN EN 206 [3], already allows partially replacing natural coarse aggregates with recycled aggregates. During the processing of waste concrete to obtain recycled coarse aggregates, up to 50 wt.% of recycled fine aggregates and 5 to 10 wt.% recycled fines are produced [4,5]. Mostly, the recycled fine fraction is used for landfilling or in geotechnical applications. This is due to the fact that more residual cement paste is present in the fine fraction in comparison to the coarse one. The inclusion of this cement paste results in a higher porosity as compared to primary extracted aggregates. The increased porosity presents the problem of higher water absorption, leading to inferior performance in the fresh state. It also requires extra use of cement if a certain strength is desired [4]. However, due to the high amount of fine fraction that is produced during the recycling process, more and more research is focusing on the use of this fine fraction in concrete. In the literature, it was found that more attention should be given to the composition and the grain size distribution of the aggregates. Furthermore, they found that with a replacement rate up to 50%, the impact on strength was limited, and that it increases the desiccation shrinkage of concrete [2].

As previously mentioned, during the recycling process, 5 to 10 wt.% of recycled fines are produced. These are particles with a maximum grain size of 63 µm. Furthermore, recycled fines can be obtained from crushing the recycled fine aggregates in a ball mill. The most common possibilities for using recycled fines are as follows: (1) as a supplementary cementitious material; (2) as inert filler; or (3) as feedstock for the production of clinker [4]. It was observed by Lidmila et al. and Šeps and Broukalová that after grinding the fines in a mill, they had some reactive capacity due to the presence of unhydrated cement grains [6,7]. However, many studies concluded that recycled fines exhibit limited reactivity, since there is only a small amount of non-hydrated cement grains present [4,8]. Several studies have shown that it is possible to dehydrate hydrated cement by applying heat treatment (at 500 °C) in order to increase the reactivity capacity again [9,10]. Another purpose of the recycled fines is to use them as an inert filler material, which can lead to better particle packing and subsequently to a denser structure. In order to have this filler effect, the recycled fines should ideally be smaller than the cement particles and spherically shaped [8]. The last option is to use the recycled fines as a raw material for Portland clinker. Many studies have confirmed that recycled fines could replace SiO₂ and limestone without having a large effect on the composition of the clinker [11,12,13]. In order to achieve this, the recycled fines should have a high CaO content [4].

In this paper, recycled concrete fines from Devagro’s (Waregem, Belgium) concrete recycling plant are evaluated as a potential partial cement substitute. First, the fines are characterized, and their chemical and physical properties are determined. Subsequently, the reactivity and strength development of pastes and mortars with the recycled fines is studied using isothermal calorimetry as well as flexural and compressive strength tests. Lastly, two different concrete mixtures are created: a reference and a mix in which recycled aggregates are partially used in place of the natural aggregates and recycled fines are introduced to replace 10% of the cement. Tests for compressive strength, shrinkage, porosity, and freeze–thaw resistance were carried out to assess the durability of these two mixtures. This paper presents additional results and further discussions of the research included in [14].

2. Materials

This study examines the fine fraction of recycled concrete to partially replace cement. Prefabricated concrete paver waste, including form-imperfect pavers and leftover concrete from the manufacturing process, is the source of the recycled fines. In these concrete pavers, the binder is a combination made up of 30% blast furnace slag (Ecocem, Moerdijk, The Netherlands) and 70% CEM I 52.5 R (Holcim, Nijvel, Belgium). The concrete is up to a year old when recycled. The following steps are involved in the manufacturing of the recycled aggregates: First, the concrete pavers are broken down into smaller fragments and a preliminary screening process is performed to remove dirt and organic materials. Next, the material is crushed, and the aggregates are washed. Finally, they are divided into different fractions. The 0/6 mm fraction is sieved until it becomes a 0/2 mm fraction, after which it is crushed in a ball mill, with a volume of 70 L and stainless-steel balls as grinding medium, for 5 h to achieve the necessary maximum grain size of 63 µm in order to create the recycled fines. In order to examine the impact of the fines on durability, two concrete mixtures were created: a reference and a mix in which recycled fines are substituting for 10% of the cement (RC10).

Table 1. Mix designs of reference and RC10 concretes. All quantities are expressed in kg/(m³ of concrete).

Materials	Reference	RC10
CEM I 52.5 N	310	280
Recycled fines	-	30
Limestone 4/20	1030	-
Recycled coarse aggregate 4/20	-	802
Sea sand 0/2	345	-
Sand 0/4	515	-
River sand 0/7	-	660
Recycled sand 0/6	-	262
Water	179	224
TechniFlow 92	1.5	-
PowerFlow EVO503	-	7

displays the composition of the mixes. The aggregates fractions are here described with the designation d/D according to the standard EN 12620 [15], with d corresponding to the minimum grain size and D corresponding to the maximum grain size. Recycled sand and coarse aggregates, possessing water absorption rates of 7% and 5.4%, respectively, replaced 27.8% of the sand and all of the coarse aggregates in the RC10 mix. To compensate for the higher water absorption of recycled materials, the RC10 concrete mix was mixed using more water. Furthermore, two different PCE-based (polycarboxylates-ether) superplasticizers, TechniFlow 92 and PowerFlow EVO503 (MC-Bauchemie, Westmeerbeek, Belgium), were utilized to improve the workability of the mixes, with the former admixture being more suitable for use with natural aggregates, whereas the latter worked well with recycled aggregates. Even though the recipes of reference and RC10 concretes are substantially different, the effective water-to-cement ratio remained the same, and both mixtures had comparable workability.

The concretes were mixed at Devagro’s concrete plant and cast in an uncontrolled atmosphere in order to maintain realistic casting circumstances. Three cylinders with a diameter of 100 mm and a height of 200 mm, six prisms (100 × 100 × 400 mm³), and fifteen cubes (150 × 150 × 150 mm³) were cast for each mixture. The specimens were compacted by placing them on a vibration table and in order to stop moisture from evaporating and then covered with plastic foil after casting. After 24 h, the specimens were demolded.

In this study, the chemical and physical properties of the recycled fines are determined. Then, the reactivity and strength development of pastes and mortars containing the fines are studied. These small-scale experiments were conducted in order to determine which replacement rate of cement by recycled fines will be used in the concrete mix. Lastly, tests for compressive strength, shrinkage, porosity, and freeze–thaw resistance were carried out to assess the durability of the two concrete mixtures.

3. Test Methods

3.1. Characterization of the Recycled Fines and Mortar Testing

3.1.1. Particle Size Distribution (PSD)

The particle size distribution (PSD) of the recycled fines and CEM I 52.5 N was determined using a laser diffractometer (Mastersizer 2000), employing both a dry and a wet dispersion unit. Both materials were tested as delivered and no drying was performed before the tests. First of all, for each material three tests were conducted with the dry dispersion unit with a feed rate of 20% and a pressure of 2 bars. For verification, one sample of each material was tested in the wet dispersion unit. Here, approximately 1 g of the material was added to 50 mL of isopropanol, creating a suspension. Then, the mixture was vibrated for 10 min to avoid agglomeration of particles. Finally, 1 mL of this mixture was pipetted and tested.

3.1.2. Particle Shape

The particle shape of the recycled fines is analyzed using the Occhio Flowcell FC200M-HR, coupled with the Callisto 3D software to receive and decipher the obtained data. This software uses the following equations to calculate the roundness (R) and circularity (C) in 2D:

$$R={\displaystyle \frac{4\times A}{F_{max}^2\times \pi }}$$

(1)

$$C={\displaystyle \frac{\sqrt{4\times \pi \times A}}{P}}$$

(2)

3.1.3. Particle Density and Water Absorption

The pycnometer technique, as specified in the standard NBN EN 1097-6 [16], was used to evaluate the particle density and water absorption on the 0/2 fraction prior to milling.

3.1.4. X-Ray Diffraction Analysis

With the use of X-ray diffraction (XRD) in combination with Rietveld refinement, a qualitative and quantitative phase analysis was performed. In order to do this, the sample of fines was mixed thoroughly with 10 wt.% Zinc oxide (ZnO). This oxide is used as an internal standard to make the quantification of amorphous phases possible. For the identification of the phases and the Rietveld refinement, the Profex XRD software (version 5.4.0) was used.

3.1.5. X-Ray Fluorescence

Using X-ray fluorescence (XRF), the elemental composition of the recycled fines was determined. To achieve this, three pellets were created by combining approximately 3 to 4 g of the binding agent C₄₂H₈₃ON from SpectroBlend (Chemplex, FL, USA) with 18 g of recycled fines.

3.1.6. Isothermal Calorimetry

An isothermal calorimetry test was performed on the fines at a temperature of 20 °C to examine their reactivity. Tests were conducted for seven different cement replacement levels: 0%, 5%, 10%, 15%, 20%, 25%, and 30%. Paste with a mass of 14 g was prepared for each sample, with a w/b of 0.5. After the paste was combined outside of the calorimeter, it was tested in a glass vial. For every sample, the heat progression was registered for three days in a row.

3.1.7. Flexural and Compressive Strength of Mortars

To have a first indication of the strength development in mortars with recycled fines, three mortar mixtures with varying percentages of cement replacement—0%, 10%, and 25%—were created and tested in order to examine the impact of recycled fines on the strength. The proportion of sand to cement in each mix was 3:1, and the water-to-binder ratio (w/b) was 0.5. CEN standard sand with a particle size of 0/2, and CEM I 52.5 N were utilized for these mixtures. At 7, 28, and 90 days of age, one prism (40 × 40 × 160 mm³) was tested using the procedure described in the standard NBN EN 196-1 [17].

3.2. Tests for Strength and Durability on Concrete

3.2.1. Compressive Strength of Concrete

The compressive strength of the two mixtures was evaluated following the NBN EN 12390-3 standard [18]. Three concrete cubes with dimensions 150 × 150 × 150 mm³ were tested after 2, 7, 28, 56, and 90 days of curing for each mix.

3.2.2. Water Absorption Under Vacuum

Following the procedure specified in NBN B 24-213 [19], the porosity of both concrete mixtures was measured by means of water absorption under vacuum. Cylinders (height 200 mm, diameter 100 mm) were cast, demolded after one day, and brought to the lab to cure in a climate-controlled chamber (20 °C, 95% RH). Three discs (50 mm in height) were cut from each cylinder at 115 days of age. Following that, the discs were put in a vacuum tank and left there for two hours. The specimens were submerged after two hours when the vacuum tank was gradually filled with water, and the vacuum was maintained for one more hour. Then, the pressure in the tank was returned to atmospheric pressure. The specimens were kept submerged in water until the mass increase was less than 0.1 m% throughout the course of 24 h. The saturated samples were weighed in air (m_s) and under water (m_w) once a steady mass was attained. The samples were then put in an oven set to 40 °C, and they were weighed every 24 h until they attained a constant mass (m_dry,40) (<0.1 m% over 24 h). After that, the samples were heated to 105 °C in an oven until they attained a consistent mass (m_dry,105). The following formulas can be used to calculate the capillary and open porosity:

$$CP={\displaystyle \frac{m_s-m_{dry,40}}{m_s-m_w}}\times 100$$

(3)

$$OP={\displaystyle \frac{m_s-m_{dry,105}}{m_s-m_w}}\times 100$$

(4)

3.2.3. Carbonation

The resistance to carbonation for both mixtures is determined following the standard NBN EN 12390-12 [20]. For each mixture, three prisms (100 × 100 × 400 mm³) were made and demolded after one day before being transported to the laboratory. Then, the samples were submerged in water until they reached 28 days of age. Subsequently, the prisms were dried in a climate-controlled room at 20 °C and 60% RH for 14 days. After preconditioning, the specimens were placed in a carbonation chamber with a carbon dioxide (CO₂) concentration of 2%. Carbonation depth measurements are performed on the specimens after 0, 7, 28, and 70 days in the carbonation chamber. In order to do this, a slice is split from the prism and sprayed with phenolphthalein. The measured depth at 0 days, immediately after preconditioning, shows whether natural carbonation had occurred before placing the samples in the chamber.

3.2.4. Shrinkage

Concrete prisms (100 × 100 × 400 mm³) were used to assess autogenous and drying shrinkage in accordance with NBN EN 12390-16 [21]. Six prisms were cast for each concrete mix, demolded 24 h later, and then coated in plastic foil to keep them from absorbing moisture from the environment while being brought to the lab. In a climate-controlled room in the laboratory (20 °C, 60% RH), the prisms used to evaluate autogenous shrinkage were covered with aluminium tape to keep them from drying. Then, two DEMEC-points with a spacing of 200 mm were attached on each of the three prism sides that had been exposed to the formwork during casting (Figure 1). Lastly, a DEMEC (demountable mechanical strain gauge) was used to precisely measure the starting distance between the two points. After 1, 7, 14, 28, 42, 56, and 70 days, the distance between the two points was measured again.

3.2.5. Freeze–Thaw Resistance

For both mixes, the freeze–thaw resistance was determined as described in the standard CEN/TS 12390-9 [22]. To achieve this, cylinders (200 mm in height, 100 mm in diameter) were cast, demolded after 24 h, and then brought to the laboratory. The samples were then submerged in water until they were 7 days old. After that, they were kept in a climate-controlled chamber at 20 °C and 60% relative humidity until they were 25 days old. Three 50 mm high discs were cut from each cylinder at the age of 21 days. After 25 days of age, the discs were glued with epoxy resin into a 70 mm-high PVC tube with a 104 mm internal diameter (see Figure 2). A layer of demineralized water was applied to the samples after 28 days, saturating them for three days. Following this, the demineralized water was replaced with a 3 mm layer of 3% NaCl solution and the samples were put in a freeze–thaw chamber for 56 cycles, each lasting 24 h, with a temperature range of −20 °C to 20 °C. The samples were removed from the chamber after 7, 14, 28, 42, and 56 cycles. The material that had scaled off the specimen surface was collected, dried at 110 °C, and weighed once a consistent mass was attained.

4. Results and Discussion

4.1. Characterization of the Recycled Fines and Mortar Testing

4.1.1. Particle Size Distribution (PSD)

Figure 3 displays the particle size distribution (PSD) of both the cement and recycled fines as measured by the dry dispersion unit of the laser granulometer. Three repetitions were performed for each material, and each repetition is presented as a separate curve in the graph.

Table 2. Value for d10, d50, and d90 for recycled fines and CEM I.

	Recycled Fines	CEM I
d10 [µm]	1.6	2.5
d50 [µm]	24	16
d90 [µm]	159	48

shows the corresponding d10-, d50-, and d90-value for the recycled fines and CEM I. As seen in Figure 3, the curve of CEM I has a higher slope in comparison to the curve of the fines. Consequently, it can be concluded that the size of the recycled fines is less consistent than that of the cement particles and that there is a greater variation in particle size. The elevated d90 value can be ascribed to a milling process that is not optimal. It was previously stated that the fines’ target size is less than 63 µm. These results show that, for CEM I, 97% of the particles are smaller than 63 µm; however, for the recycled fines, only 73% may really be classified as fines. Additional screening, sieving, and milling could resolve this issue because it is likely that the larger particles are cushioned by the smaller ones [4].

4.1.2. Particle Shape

Figure 4 gives the scatter diagram of the recycled fines particles showing the circularity in 2D as a function of the roundness in 2D calculated by the software. Particles of irregular, rectangular, and oval shapes can be found in the recycled fines. The mean values of roundness and circularity are 0.707 and 0.564, respectively. These values correlate with particles that are elongated without sharp corners or edges [23].

4.1.3. Particle Density and Water Absorption

Using the pycnometer method, the particle density of the recycled concrete fine fraction with size 0/2 was determined. The value for the apparent particle density is 2710 kg/m³, 2240 kg/m³ for the oven-dried particle density, and 2410 kg/m³ for the saturated and surface-dried particle density. These obtained values are comparable to those found in the literature for the fine fraction of recycled concrete [24,25]. The density is lower than for CEM I 52.5 N, which has a density of 3150 kg/m³.

The water absorption of the recycled fines after being submerged for 24 h was 7.7%. As found in the literature [26], the least strict standard (DIN 4226-100) requires a maximum water absorption of 13%, whereas the most restrictive standard (KS F2573) sets a limit of 3%. As a result, the fine fraction meets the least strict requirement but not the strictest one.

4.1.4. X-Ray Diffraction Analysis

The XRD analysis and the quantification of the different phases present in the recycled fines are given in Figure 5 and Figure 6, respectively. According to the XRD data, the main phases found in the recycled fines are calcite (CaCO₃) and quartz (SiO₂). The siliceous natural aggregates, such as siliceous river or sea sand, that were probably included in the concrete pavers are the source of the significant quantity of quartz [27]. The substantial amount of amorphous phases is caused by unreacted amorphous binder particles, amorphous cement hydration products (such as C-S-H), and amorphous CaCO₃ from the carbonation of the recycled fines [27]. The early recycling age of the concrete pavers and the consequent presence of unhydrated particles can be used to explain the occurrence of belite, gypsum, and anhydrite [28].

4.1.5. X-Ray Fluorescence

Table 3. Results from XRF analysis on recycled fines.

Compound	Quantity [%]
SiO2	36.30 ± 1.37
CaO	46.22 ± 1.10
Al2O3	5.91 ± 0.11
MgO	2.17 ± 0.05
Na2O	1.35 ± 0.14
P2O5	0.02 ± 0.01
S	2.01 ± 0.09
K2O	0.50 ± 0.06
TiO2	2.21 ± 0.13
Fe2O3	3.25 ± 0.07
Mn2O3	0.06 ± 0.00

presents the results of the XRF analysis. These results align with the findings obtained with the XRD analysis. Again, the siliceous sand in the concrete pavers can explain the high concentration of SiO₂. Cementitious materials are inherently rich in calcium-based compounds, which are represented as CaO in the XRF results, comprising almost half of the recycled fines (46.22%). However, as suggested by the XRD data, calcium-rich phases can be both crystalline and amorphous. Other elements present in the recycled fines are aluminium, iron, and lower quantities of titanium, magnesium, and sodium. Elemental sulphur, detected using XRF, confirms the possible presence of anhydrite quantified from the XRD pattern. Negligible amounts of potassium, phosphorus, and manganese complete the chemical composition of recycled fines.

4.1.6. Isothermal Calorimetry

The heat flow and cumulative heat flow normalized to the total binder mass obtained via isothermal calorimetry are given in Figure 7 and Figure 8, respectively, for pastes with a cement replacement ranging from 0 to 30%. Figure 7 illustrates how increasing the cement replacement rate results in both a delay and decrease in the hydration peak. Nonetheless, the cumulative heat flow indicates that after 72 h, the total heat generation from the samples with 5% and 10% cement replacement was comparable to that of CEM I 52.5 N, indicating that these replacement rates do not appear to have a detrimental effect on the hydration process at this age.

4.1.7. Flexural and Compressive Strength of Mortars

The flexural and compressive strengths of mortars with 0%, 10%, and 25% cement replacement by recycled fines are displayed in Figure 9 and Figure 10, respectively. It is evident that lower flexural and compressive strengths are obtained with an increasing cement replacement. In measurements performed after 7 days of curing, mortars containing 10% recycled fines showed a somewhat increased compressive strength compared to the reference, suggesting a good early-age strength development. Furthermore, for these mortars, the reduction in flexural and compressive strength after 90 days is restricted to 5.2% and 2.8%, respectively. The reduced strength of the specimens containing recycled fines could possibly be due to their lower reactivity in comparison to cement, since the cement in the recycled concrete was already hydrated, and a larger particle size.

4.2. Tests for Strength and Durability on Concrete

4.2.1. Compressive Strength

Table 4. Compressive strength for REF and RC10 after 2, 7, 28, 56, and 90 days of curing.

	2 Days	7 Days	28 Days	56 Days	90 Days
REF	31.9 ± 1.4	45.8 ± 2.0	57.0 ± 0.3	61.5 ± 0.6	62.3 ± 0.7
RC10	31.2 ± 0.8	44.1 ± 0.6	55.4 ± 1.1	58.9 ± 2.3	59.8 ± 2.1

gives the compressive strength values for REF and RC10, tested after 2, 7, 28, 56, and 90 days. RC10 has a slightly lower compressive strength than the reference at all evaluated ages. However, the largest decrease in strength was found at an age of 56 days, with a strength reduction of only 4.2%. Additionally, a t-test (level of significance = 5%, p > 6.7%) shows that there is no statistically significant difference between the reference and RC10 at any of the tested ages. The desired strength class for both concrete mixes was C30/37. Consequently, both mixes comply with the requirements.

4.2.2. Water Absorption Under Vacuum

Figure 11 shows the capillary and open porosity for REF and RC10. Recycled coarse aggregates, sand, and fines all contributed to an increase in the concrete’s capillary and open porosity of 45.5% and 47.5%, respectively. The use of recycled aggregates in the RC10 mixture instead of natural aggregates in the reference and the decreased reactivity of the binder are probably responsible for this increased porosity.

4.2.3. Carbonation

The carbonation depth as a function of the exposure time is shown in Figure 12. The slope of the linear regression line represents the carbonation rate K_AC in mm/√days. The carbonation rates for REF and RC10 are equal to 0.7481 and 0.7892, respectively. Therefore, the RC10 concrete only has a slightly higher carbonation rate in comparison to the reference. In the literature, it is found that the addition of recycled powder as a cement replacement increases the carbonation depth [29]. This reduced carbonation resistance can be explained by the combination of a higher SiO₂ and a lower CaO content present in the recycled fines with respect to cement. This can cause a decrease in Ca(OH)₂, which decreases the alkalinity of the material and therefore increases its susceptibility to carbonation. However, it is also mentioned in the literature that recycled aggregates can have a beneficial effect on the carbonation resistance due to the reduced effective w/c ratio caused by the water absorption of the recycled aggregates in case this is not compensated for by adding more water. This can lead to a decreased porosity of the cement matrix, increasing the resistance to carbonation [30].

4.2.4. Shrinkage

Figure 13 shows the total, drying, and autogenous shrinkage of both REF and RC10. The results show that after 70 days, the RC10 mix showed a larger total shrinkage than the reference. On the other hand, the concrete with 10% recycled fines and recycled aggregates expands rather than shrinks in terms of autogenous shrinkage. This is most likely caused by the recycled aggregates’ greater porosity and capacity to retain water, which can promote internal curing and prevent autogenous shrinkage [31]. On the contrary, RC10 shows a greater drying shrinkage, most likely due to the water present in the recycled aggregates evaporating.

4.2.5. Freeze–Thaw Resistance

Figure 14 displays the mass of scaled material after 7, 14, 28, 42, and 56 freeze–thaw cycles with de-icing salts. The limit value for scaling after 56 cycles, as stated in EN 12390-9, is 1 kg/m², indicating that none of the mixtures has a good freeze–thaw resistance [22]. Given that the mix designs correspond to exposure class XF1, this was anticipated. In comparison to the reference mix, adding recycled fines, sand, and coarse aggregates to the concrete results in an increase in the scaling by an average of 77.8%. Prior research in the literature has also reported such significant drop in freeze–thaw resistance [32]. The decrease may be explained by the hydration products’ lower Ca(OH)₂ concentration, which can lead to fast carbonation at the surface and increase the risk of freeze–thaw scaling [33].

5. Conclusions

This paper examined the utilization of recycled concrete fines as a partial substitute for cement. This involved a characterization of the powder and its reactivity, as well as a study of the durability of concrete with a 10% cement substitution by fines and a partial substitution of natural aggregates by recycled aggregates. From this research, the following conclusions can be made:

• It is possible to conclude that the recycled fines are less homogeneous in size than cement, likely caused by a milling procedure that is not optimized.
• For replacement rates up to 10%, the total heat production during hydration after 72 h was comparable to a CEM I, and the strength reduction in mortars containing up to 10% recycled fines is restricted.
• When compared to a reference concrete that has natural aggregates and no fines, the recycled concrete with 10% fines and partly recycled aggregates had a positive impact on autogenous shrinkage. This can be explained by the recycled aggregates’ capacity to retain water, which allows them to act as a water reservoir during hydration.
• After 56 days, the compressive strength had decreased by just 4.2% for a concrete mix with (partially) recycled fine and coarse aggregates and 10% cement replacement by recycled fines.
• The inclusion of 10% recycled fines and recycled aggregates only had a limited negative impact on the carbonation resistance of concrete.
• After 56 freeze–thaw cycles, both the reference mix and the recycled concrete lacked sufficient freeze–thaw resistance. For the recycled concrete, the scaling increased by 77.8%.

Based on these findings, it can be concluded that recycled concrete fines are suitable for partial cement replacement. Further research should be conducted to optimize the milling process to obtain the fines. Furthermore, optimizing the mix design can significantly increase the strength and durability properties of the concrete.