Skip to content
BY 4.0 license Open Access Published by De Gruyter Open Access October 12, 2022

Investigation of mechanical activation effect on high-volume natural pozzolanic cements

  • Abdul Vahap Korkmaz EMAIL logo and Huseyin Fırat Kayıran
From the journal Open Chemistry

Abstract

Mechanical activation is one of the most preferred methods to increase the reactivity of mineral, mining, and industrial wastes or materials with low pozzolanic content in cement production. The mechanical activation process of such materials takes a long time and requires high grinding energy. Although it seems like an advantage to using the materials in cement production by gaining reactivity thanks to mechanical activation, mechanical activation is a long duration and expensive process, causing an increase in cement production costs. In this study, a hybrid method is proposed for the mechanical activation of materials in a shorter duration and with lower energy consumption. A roller press and a hammer grinder were integrated before the laboratory type ball mill to realize the mechanical activation processes. To perform the mechanical activation processes in this way, two different hybrid grinding methods were tested and compared. Both methods activated volcanic tuff samples with low pozzolanic properties were investigated particle microstructures, energy consumption differences, mechanical activation durations and their effects on the cement performance, and mortar microstructures. The hybrid grinding method integrated with roller press and ball mill was determined to be the best mechanical activation method.

1 Introduction

Both natural pozzolans and residue from industrial, commercial, mining, or agricultural activities have been used as construction and building materials in the global cement and concrete industry for years. The cement industry has focused on natural pozzolan materials, such as volcanic tuff, zeolite, perlite, pumice, and diatomite [1,2,3,4,5,6] besides industrial by-products, such as silica fume, fly ash, coal mining waste, and furnace slag [7,8,9,10]. In recent years, as artificial pozzolanic materials, composite bonding rice ash [11], bamboo leaf ash [12], ground brick waste [12], sugarcane pulp ash [13], palm oil ash [14], soda sludge [15], activated charcoal washing waste [16], bottle glass [17], and different biomass ashes have become the most popular pozzolans [18]. As a non-renewable natural resource, the gradual depletion of pozzolanic materials will cause a reduction in sustainable cement production in the future. It will be inevitable to use industrial mineral wastes or materials with low pozzolanic properties together in cement production, which can be an alternative to pozzolanic materials.

The most successful and easily applicable method used to improve the pozzolanic activity of materials is mechanical activation. Mechanical activation may cause changes in the physical and chemical properties of powders, such as tiny particles of dust, crystal structures, and powder surface properties, to improve the reactivity of materials. Mechanical activation can increase the reactivity of particles and enable them to be used in composite cement production. There are limited numbers of available research effect of mechanical activation on the improvement of the pozzolanic activity of natural sources and different mineral phases [19]. Wei et al. [20] showed that mechanical activation influenced in increasing the reactivity of vanadium residues, and the amount of reactive silicon and aluminum in vanadium residues gradually increased at the end of mechanical activation. Cheng et al. [21], in studies on the pozzolanic activity of iron residues, proved that activated iron residues as a result of mechanical activation may be used as a cement mineral additive in the concrete industry. Wu et al. [22] declared that the lattice distortion on the surfaces of the minerals increased with the increase in the mechanical activation duration and the pozzolanic activities of the minerals improved as a result of the mechanical activation. Yao et al. [23] concluded that mechanical activation in the presence of calcium hydroxide (CH) made quartz powders have sufficient pozzolanic activity to react with CH at room temperature.

The highest energy consumption in the mining and cement sector occurs in the grinding of materials. To proceed in mechanical activation, feldspar powders were ground for 200 min [24], iron ore residues were ground for 90 min [21], muscovite was ground for 200 min [25], gold mine and siliceous gold ore residues were ground for 80 min [19], granular copper slag was ground for 180 min [26], and calcium phosphate chemistry material was ground for 1,440 min [27]. Long mechanical activation durations bring extra grinding costs and a waste of time to the process. Most of the studies carried out in recent years have focused on using energy efficiently and reducing energy costs. In this study, a hybrid grinding system has been designed to both reduce energy costs and make maximum use of mechanical activation.

Schönert [28] has shown that the most energy-efficient method of crushing particles is to compress them between two plates. Microcracks formed on the inner and outer surfaces of the mineral by very high compression forces in the particle bed facilitate the grindability of the material for subsequent grinding operations. Besides, the most generally used mechanical activation types of equipment in the laboratory are a vibration mill, ball mill, and prototype machines. However, there are almost no studies on equipment design to improve pozzolanic activity by mechanical activation methods.

In this study, two different methods were proposed to investigate the activation effect of various grinding equipment on minerals, shorten the mechanical activation time, and reduce grinding costs. For mechanical activation, two different systems tried to be created by placing the rotary crusher and hammer crusher separately in front of the conventional ball mill. The mechanical activation processes of pozzolanic materials were investigated by using two different grinding methods by means of a roller press and hammer crusher integrated into the laboratory ball mill.

In this study, the hybrid grinding method was proposed to investigate the activation effect of various grinding equipment on minerals, shorten the mechanical activation duration, and reduce grinding costs. The hybrid grinding method integrated with roller press and ball mill was determined to be the best mechanical activation method for high-volume natural pozzolanic cements. This hybrid application has shown that the grinding process with small grinding ratio and step-by-step completion is more energy-saving and short grinding duration than that with large one-time completion. The cement roller press causes internal tension and cracks in the particles in the process of mechanical activation to achieve the purpose of grinding. The biggest reason for this is that the high pressure load in the roller press directly affects the material layer through the double rollers, and as a result of the rubbing of the grains against each other during the grinding process, the grains break with their own energy. Thus, the roller press performs pre-mechanical activation, reducing the grain size of the materials, a high energy utilization rate, and a significant energy saving effect. Therefore, the roller press may play the role of fine crushing and coarse grinding in the ball mill. A large number of cracks may occur in all materials in the roller press, which may greatly reduce the energy consumption of the materials in the next grinding process.

2 Material and methods

In experimental studies, volcanic tuff samples with low pozzolanic content, industrial clinker and gypsum stone samples were used. The reason why materials with low pozzolanic properties are preferred is to see the effect of the hybrid grinding system on increasing the pozzolanic properties of materials.

2.1 Chemical and mineralogical characterization of the materials

Activation is required to improve the reactivity and pozzolanic properties of some minerals and mineral wastes. Mechanical activation aims to change the reactivity or physical and chemical properties of a solid material. Generating new surfaces can improve the specific surface area of materials and accelerate the hydration rate through phase transformation and particle size reduction. Also, mechanical activation will lead to structural changes and unstable states, thereby changing the chemical composition of materials. For this reason, it is important to know the chemical and mineralogical properties of materials before activation in experimental studies.

Chemical analyses of taken from two different regions volcanic tuffs and Portland cement clinker were performed according to the TS EN 197-1 [29] and TS EN 196-2 [30]. The sample results are given in Table 1. The X-ray diffraction method was used under a Riguku Ultima IV X-ray Diffractometer to determine the mineralogical compositions of the materials. In this study, Blaine test was performed following TS EN 196-6 [31] standard while the pozzolanic activity test was conducted according to TS 25 [32] standard. In addition, reactive SiO2 tests were performed according to TS 369 standard [33] (Table 1).

Table 1

Chemical compositions of the materials

Chemical compounds Volcanic tuff green (%) Volcanic tuff Beige (%) Clinker (%) TS 25
Pozzolana standards (%)
SiO2 58.62 69.96 20.58 >70
Al2O3 18.77 12.71 5.34
Fe2O3 6.05 0.97 3.29
CaO 6.30 2.99 65.88
MgO 2.53 1.21 1.93 <5
Na2O 0.38 0.07 0.29
K2O 2.76 1.93 0.95
Cl 0.03 0.07 0
TiO2 1.08 0.30 0
P2O5 0.87 0.00 0
Cr2O3 0.1 0.01 0
MnO 0.04 0.04 0
SO3 0.02 0.15 1.26 <3
Loss on ignition 2.35 9.15 0.15 <5
Total 99.90 99.40 99.67
Moisture 2.20 8.20 0.20
Reactive silica 25.22 27.76
Pozzolanic activity (MPa) 6.30 6.40

The green volcanic tuff (GVT) samples consist of plagioclase (labrador andesine), pyroxene, more minor olivine, and opaque minerals. As secondary minerals, small amounts of chlorite, rutile, apatite, and sericite are present in the GVT sample. As a result of examinations, it was observed that GVT consists of plagioclase dominated by more than 60% andesine, 30% pyroxene, 7% opaque mineral, 2% olivine, and 1% secondary minerals.

2.2 Specimen preparation and grinding procedures

Many materials, especially siliceous materials and wastes, are mechanically activated using high-energy grinding. In other words, high energy costs and high grinding time are needed for grinding such materials. When the literature studies are examined, it has been seen that the time required for mechanical activation of materials that can be used as cement additives varies between 60 and 180 min on average. Standard laboratory ball mills were used for mechanical activation processes in academic studies. In this study, it was thought and aimed that the mechanical activation time would be reduced because a rotary grinder or hammer crusher was added before the standard laboratory type ball mill. For this reason, grinding times were chosen shorter (15, 30 and 45 min) compared to other scientific studies. A hybrid system is created by integrating a roller press before the grinding mill to reduce the mechanical activation duration and grinding energy costs of the materials. As a result of mechanical activation, the particle size will decrease and the specific surface area of the materials will increase, and the pozzolanic activities of the activated materials will increase in a shorter time at a lower cost.

The rotary press is equipped with a movable roller and a fixed roller. A manual drive is mounted so that each of the rollers rotates in reverse at a slow speed. There is a gap of 1 mm between the moving roller and the stationary roller. When the raw material enters the gap between the two rollers vertically from the upper chamber, the material particles in contact with the roller surface are directly pressurized from the roller surface, and two rotating rollers compress the particles inside the cavity. The laboratory type manual driven roller press device is shown in Figure 1a and c.

Figure 1 
                  Image of roller press (a) and the hammer crusher (b) used in experiments and operating principle of roller press (c) and the hammer crusher (d).
Figure 1

Image of roller press (a) and the hammer crusher (b) used in experiments and operating principle of roller press (c) and the hammer crusher (d).

A simple design has been realized by establishing a direct connection between the motor shaft and the rotor with the hammers. The hammer mill is fed regularly by a manual feeder. Particle size is reduced using a beater rotor that captures the materials. The mixing process continues until the particles are expelled from the perforated sieve surrounding the rotor. Then, the product is collected in a hopper. Laboratory type manual-driven roller press device is shown in Figure 1b and d.

The GVT, Beige volcanic tuff (BVT), Portland cement clinker, and natural gypsum samples were first ground in a 22 l cylindrical laboratory type ball mill [305 mm × 305 mm (Q × L); 12.70, 19.05, 25, 30, and 36 mm in diameter with balls rotating at 70 rpm]. The material samples were ground in 15, 30, and 45 min.

The same laboratory type ball mill was also used to produce composite cement samples. The GVT and BVT samples were ground in the laboratory mill for an equal time (30 min) by balls. Then, the particle size distribution of ground GVT and BVT samples for an equal time (30 min) was determined using a Sympatec HELOS model particle size tester. Energy consumption readings were made by attaching an electricity meter to the laboratory mill.

2.3 Study on the grindability and energy cost estimation

Using an industrial-sized horizontal ball mill (in the range of 50–25 mm ball diameter) produced composite types of cement with GVT and BVT additives at 10%, 30% and 50%. The energy consumption value of each composite cement sample was calculated and unit energy consumption costs were compared.

2.4 Microscopic examination under scanning electron microscope (SEM)

SEM image analyses of composite types of cement including 10 and 30% GVT and 10 and 30% BVT were performed under a JEOL JSM 5600 brand SEM device in the Adaçal Industrial Minerals laboratory.

2.5 Preparation of cement mortar samples

The blended types of cement were designated according to the pozzolan content present. Cement samples with additives of 10, 30, and 50% by weight were produced. Mixing ratios consist of series in which only GVT and BVT are used. Given in TS EN 196-1 (2016), water, cement, and sand values to be used in 1 m3 were calculated using 225 g water, 450 g cement, and 1,350 g standard Rilem sand. Compressive strength measurement was performed with a Tony technique at the loading rate of 20–40 N/mm2/s considering TS EN 19-1 [34]. The setting times of mixtures were tested considering TS EN 196-3 [35].

The specific surface area of clinker and prepared composite cement samples was measured using an ATOM technical brand Blaine instrument considering TS EN 196-6 [31]. The specific gravity of the pieces was measured by a Le Chatelier pycnometer using kerosene.

3 Results and discussion

3.1 Evaluation of the pozzolanic properties of volcanic tuffs

Volcanic tuffs are widely used as a substitution material for pozzolanic cement production in Turkey since these rocks are widely available all over the country in high reserves. Pozzolanic activities of volcanic tuffs found in Turkey range from 5 to 11 MPa [37,38]. The main reason for choosing materials with low pozzolanic properties in the study is the desire to produce high volume pozzolanic cements by increasing their pozzolanic properties with mechanical activation. Unfortunately, not every cement producer has high pozzolanic materials, and they pay high costs for transportation or supply. By increasing the pozzolanic properties of materials with low pozzolanic properties found in nature, an additional added value will be provided to cement production and competition with other cement companies will decrease.

Chemical composition of a natural pozzolan is a guiding step in terms of its pozzolanic property. However, most of the volcanic rocks do not always exhibit pozzolanic activity although they chemically have standard values. It is well known that the pozzolanic activity mainly depends on the amount of amorphous/reactive silica and alumina in the composition of the pozzolanic material. Pozzolanic activity is also affected by other characteristics, such as specific surface area and particle size distribution of the pozzolans [36]. Volcanic tuffs are widely used a substitution material for pozzolanic cement production in Turkey since these rocks are widely available all over the country in high reserves. Pozzolanic activities of volcanic tuffs found in Turkey ranges from 4 to 11 MPa [37,38].

According to TS EN 196-5 [39], pozzolanic materials should contain more than 25% reactive SiO2 by weight. In addition, sum of SiO2 + Al2O3 + Fe2O3 should be at least 70% for pozzolanic materials [36].

GVT and BVT provide sufficient amount of reactive SiO2, which are 25.28 and 27.47. Pozzolanic activities of the GVT and BVT were found as 6.3 and 6.4 MPa. Sum of SiO2 + Al2O3 + Fe2O3 is 76.87 and 82.25 for volcanic tuff samples, respectively. It can be seen from the pozzolanic activity tests that the GVT and the BVT provide the required standard values despite their low activity (Table 1).

3.2 Grinding of the Materials

The energy consumption differences as a result of grinding the materials in the ball mill are shown in Figure 2. The energy consumed during the grinding process was determined by ENTES MPR-53S brand Watt-meter. Time-dependent energy consumption differences of GVT, BVT, and clinker are shown. The highest energy consumption increase occurred in clinker, then in BVT, and then in GVT.

Figure 2 
                  Energy consumption differences of raw materials (a) and blended cement samples (b).
Figure 2

Energy consumption differences of raw materials (a) and blended cement samples (b).

Depending on the grinding time, the percentage of grinding energy consumption of clinker is higher than that of GVT and BVT. It was observed that the percentage of grinding energy consumption of BVT was higher than that of GVT (Figure 2a). The energy consumption differences that occur during the production of cement with additives increased with the increase in the amount of additives. The use of high volumes of GVT and BVT with 50% additives maximized the difference in grinding energy consumption (Figure 2b). As a result of the increase in the amount of additives, the increase in the grinding energy consumption differences of the cements produced with both materials is in harmony with Figure 2a.

3.3 Particle size distribution of raw materials

Roller presses are designed according to the bed grinding principle. The high-pressure load acts directly on the material layer by means of double rollers, which causes internal tension in the particles and causes cracks in the material, making the grinding work easier. Roller presses increase the output of the ball mill system by 30–50 and 20–35% of the crushed materials are smaller than 0.08 mm, and 65–85% are smaller than 2 mm [40]. The hammer crusher can crush materials of 600–1,800 mm size into materials under 25 or 25 mm. To better see the mechanical activation effect in the laboratory type hammer crusher used in the study, it is aimed to reduce the materials below 4,000 µm. GVT, BVT, and clinker samples were grinded separately in roller press and hammer crusher. Sieve analysis results for both crusher types are shown in Figure 3. The images of the products were obtained as a result of the size reduction process by passing the materials through a roller press and hammer crusher are shown in Figure 4.

Figure 3 
                  Particle size of the roller press (a) and the hammer crusher (b).
Figure 3

Particle size of the roller press (a) and the hammer crusher (b).

Figure 4 
                  Roller press output products clinker (a), GVT (b), BVT (c), hammer output products clinker (d), GVT (e), and BVT (f).
Figure 4

Roller press output products clinker (a), GVT (b), BVT (c), hammer output products clinker (d), GVT (e), and BVT (f).

93.89% of clinker, 98.15% of GVT, and 98.25% of BVT are smaller than 2 mm. In addition, 74.4% of clinker, 85.14% of GVT, and 68.45 of BVT are smaller than 0.085 mm. After the materials are pulverized by the roller press, all or some of them with good extruding effect will enter the next ball mill for further grinding. This system has advantages, such as simple operations, less equipment, wide particle size distribution, and stable powder performance.

95.64% of the clinker, 97.80% of the GVT, and 88.53 of the BVT are smaller than 2 mm from the materials obtained by passing through the hammer crusher. In addition, 74.92% of clinker, 78.69% of GVT and 59.80% of BVT are smaller than 0.085 mm. According to both methods, GVT was broken more easily than the other two materials. The reason why clinker breaks more easily than BVT is due to the fracture properties of clinker and the fact that it produces finer powder material by forming cracks more easily in its structure after each compression

3.4 Surface morphology of crushed materials

Particle size significantly affects particle shape. Because of the different particle shapes, the direction of the laser scattering changes more [21]. Therefore, supplementing the laser particle size with an SEM analysis will be more effective in determining material particle shapes and particle sizes.

According to both methods, the microstructures of the broken materials were examined according to the sem image. First, the raw image of the GVT and BVT is shown in Figure 5a and b. It is seen that the raw GVT has a more porous structure than the BVT. This, of course, will bring some advantages in favor of GVT in the crushing and grinding processes of materials. The changes and effects in the surface morphology resulting from crushing the raw materials with a roller press and hammer crusher are shown in Figure 5c–f. As a result of the grinding of raw materials in both roller press and hammer crusher, both particle sizes and surface morphologies were strongly affected in both grinding systems, the sharp edge of the large particles started to disappear and the amount of the fine-ground particles gradually increased. It is seen that the BVT has a tighter structure. This gives us clues that BVT can be broken down and ground more difficult than GVT. BVT particles exhibited coarser and sharp angular particles in both grinding stages. In addition, ground BVT powders showed a longitudinal fracture in both grinding sets. Round pieces are less than GVT. GVT particles showed rounder and finer particles in both grinding sets. The particle properties of GVT tend to be smoother in both grinding sets. In both grinding systems, the sharp angle and long flake-like particles of GVT were all transformed into small and spherical particles, showing a good indication of homogenization (in terms of particle size and shape). When evaluated in terms of crushing systems, it was seen that grinding with a roller press was slightly more effective than a hammer grinder. It has been thought that the biggest reason for this is that by applying pressure between the rollers in the roller press, the minerals come into contact with each other, making the grinding process a little easier.

Figure 5 
                  SEM images of grinded materials according to crusher types; raw GVT (a); raw BVT (b) (under 2 cm); GVT roller (c); BVT roller (d); GVT hammer (e); and BVT hammer (f).
Figure 5

SEM images of grinded materials according to crusher types; raw GVT (a); raw BVT (b) (under 2 cm); GVT roller (c); BVT roller (d); GVT hammer (e); and BVT hammer (f).

High-strength mechanical forces can change the particle shape of mineral admixtures and affect the crystal structure of minerals. Such effects transform some of the mechanical energy into the surface activation energy of the mineral particles at the end of the grinding period and contribute to developing the pozzolanic activity of natural or artificial minerals [41]. Residual materials that have undergone mechanical activation can become as active as other pozzolanic materials, such as fly ash, kaolin, and slag [42,43,44].

3.5 Particle size distribution of blended cement samples

The materials coming out of the roller press were then ground in a ball mill with clinker at different rates and other times. To produce high volumes of cement with GVT and BVT, cements with 30 and 50% additives have been made. The produced types of cement were ground for 15, 30, and 45 min. The particle size distribution of ground cements is shown in Figures 6 and 7.

Figure 6 
                  30% BVT (a) and 30% GVT (b) blended types of cement ground at different durations.
Figure 6

30% BVT (a) and 30% GVT (b) blended types of cement ground at different durations.

Figure 7 
                  50% BVT (a) and 50% GVT (b) blended types of cement ground at different durations.
Figure 7

50% BVT (a) and 50% GVT (b) blended types of cement ground at different durations.

The particle size of materials is significantly reduced by mechanical activation, creating a new specific surface area. This case may be due to the natural defects (capillary crack, void, alteration, etc.) in the raw materials subjected to the action of mechanical impact [45,46]. Although the grinding times of GVT and BVT were shortened by the hybrid grinding method, their crystal structures changed. When clinker, GVT, and BVT are grinding, they show a macroscopic decrease in the particle size and a macroscopic increase in the specific surface area as the defects in the microstructure of the materials increase. As a result of mechanical activation, the activation energy on the particle surfaces increases by changing the particles’ physical, mineralogical, and pozzolanic properties, which enormously improve pozzolanic activity [47].

The particle size and frequency distribution curves for GVT and BVT activated by the hybrid grinding methods are different. Powder diameters of 30% GVT-blended cements are lower than 30% BVT-blended cements’ powder diameters at 15, 30, and 45 min. Likewise, powder diameters of 50% GVT-blended cements are lower than 50% BVT-blended cements’ powder diameters at 15, 30. and 45 min (Table 2).

Table 2

Grinding time particle size relationship

15 min 30 min 45 min
Materialas D10 D50 D97 D10 D50 D97 D10 D50 D97
GVT 30 1.07 8.76 46.06 0.96 7.27 47.77 0.83 5.07 44.02
BVT 30 1.08 15.89 72.55 0.82 5.35 50.43 0.85 6.62 55.74
GVT 50 1.29 14.60 60.40 1.14 14.64 57.69 0.92 8.89 50.27
BVT 50 0.94 10.93 70.51 0.85 6.46 57.51 0.89 8.28 59.97

In the ball mill grinding method, the materials are exposed to the combined effect of extrusion and shear force; in contrast, in the hybrid grinding system (ball mill + roller press), the materials are additionally exposed to the extrusion force. The literature studies show that the stress produced by applying pure extrusion is 5 times the stress produced by the shear force in the grinding process of the granular material. After grinding with a roller press, fine particles smaller than 90 μm make up about 20–30%. At the same time, many microcracks are observed in the coarse particles. The microcracks accelerate the mechanical activation of the materials and contribute to the reduction of grinding energy. After applying the hybrid grinding method, the specific surface area of the powders increases significantly and the powder particle size decreases significantly, as activated GVT and BVT powders are structurally damaged under mechanical impact. Comparing the ball grinding methods resulting in the grinding duration, the specific surface area of the GVT-blended cement and BVT-blended cements produced by mechanical activation by the hybrid grinding processes is the largest. This advantage is due to the different working principles of other equipment for materials that require mechanical activation. As a result, with the increase in the mechanical activation process time, the phase composition changes, the powder particle size decreases, and the powder specific surface area increases. The time required for mechanical activation was halved with the hybrid grinding method.

3.6 Surface morphology of hybrid grinding material products

The micromorphology of BVT-blended cements and GVT-blended cements has changed dramatically after being ground by hybrid grinding equipment. Under mechanical force, spherical particles and dense irregular larger particles form sheetlike or blocklike configurations to varying degrees; the particles are rapidly ground finer. The original spherical particles give way to compact, irregularly shaped particles with sharp edges and corners showing good dispersion [48]. By comparing the output from the ball mill and hybrid grinding method at different grinding times, we found that over time, the large irregular particles collided, fractioned, and impacted each other under the action of the roller press, whereby further refined them into smaller unstable particles showing no evidence of the structure of the original spherical particles. Many spherical particles in the GVT and BVT become almost invisible and a large number of small particles become thoroughly intermixed number of irregular polygonal particles form. We also found that the BVT-blended cement particles produced by hybrid grinding systems have sharp edges and corners, In contrast, GVT-blended cement particles produced by the hybrid grinding systems have relatively smooth edges and corners (Figure 8).

Figure 8 
                  Fifteen-minute grinding time of 50% GVT-blended cement (a) and 50% BVT-blended cement (b); 30 min grinding time of 50% GVT-blended cement (c) and 50% BVT-blended cement (d); and 45 min grinding time of 50% GVT-blended cement and (e) 50% BVT-blended cement (f).
Figure 8

Fifteen-minute grinding time of 50% GVT-blended cement (a) and 50% BVT-blended cement (b); 30 min grinding time of 50% GVT-blended cement (c) and 50% BVT-blended cement (d); and 45 min grinding time of 50% GVT-blended cement and (e) 50% BVT-blended cement (f).

3.7 Hydraulic properties

The initial setting time of Portland reference cement mortar is 157 min and the final setting time is 200 min. Portland reference cement has a water content of 28% and a volume expansion of 1 mm. The hydraulic properties of the cement samples were compared with the reference cement.

Initial and final setting times of GVT-blended cements at 10 and 30% admixture rates were also shorter than the reference cement. The setting initial and final times of the cements with 10, 30, and 50% BVT were longer than the reference cement (Figure 9). It has been observed that the final setting times of GVT-blended cements and BVT-blended cement pastes increase with the increase in the amount of additives. It has been observed that the final setting times of GVT-blended cements are lower than Portland reference cement and BVT-added cements at 10–30% additive ratios. Such similar effects are also valid for cements with zeolite additives. As the zeolite additive ratio increases, the setting times are shortened [49].

Figure 9 
                  The hydration properties of the blended cements.
Figure 9

The hydration properties of the blended cements.

An inverse decreasing trend in setting times was observed, possibly due to the higher surface area of the blended cements due to the increased replacement ratio. The prolongation of the setting time with natural pozzolanic materials is attributed to the increase in the water/Portland cement ratio [50]. In addition, it was observed that the volume expansion of the pastes did not increase with the increase in the substitution level and found the result of the decrease in the free CaO content of the blended cements to be consistent. The water content of cements with volcanic additives was found higher than Portland reference cement and GVT cements in all additive ratios.

3.8 Mechanical and microstructural properties

The mechanical performance of the blended cements mortars was checked with compared to the reference Portland cement (RPC). As the amount of additive material in the blended cements increased, a consistent decrease was observed in the mortars’ compressive strength and strength development rate (Figure 10). The compressive strength of the blended mortars produced by increasing the natural pozzolanic content from 10 to 50% is lower than the corresponding RPC at all test ages. The compressive strength of GVT-blended mortars are higher than the compressive strength of BVT-blended mortars at all ages. It is expected that the strength differences between Portland cement and the corresponding high volume BVT and GVT cements will decrease at longer ages.

Figure 10 
                  Compressive strength results of blended cements.
Figure 10

Compressive strength results of blended cements.

Although the BVT is very reactive compared to GVT, it is seen that the early and final strength value is low. This is thought to be due to the grindability properties of the materials. Since the BVT was not sufficiently ground, it remained coarse (BVT; +45 µm, >25%). Due to the finer grinding of the GVT, its pozzolanic activity increased and developed a better strength.

The compressive strength development properties of blended cement were influenced not only by the pozzolanic activity and the chemical and mineralogical structure of the clinker but also by the particle size distribution resulting from the mutual effect between the blended cement components. The particle size distribution of pozzolanic-blended cements is more affected by pozzolanic materials’ relative hardness than clinker. A natural pozzolan that is harder than clinker contributed to finer grinding of the clinker in the hybrid system, while a softer pozzolan resulted in a relatively coarser grinding of the clinker. GVT is harder than BVT but easier to grind.

Accordingly, up to 30%, w/w, of raw material can be added to meet the mechanical requirements for cements with BVT. In addition, up to 50% can add w/w raw material for GVT-added cements. It should be noted that the low strength of the blended cement can be attributed to the slow evolution of the pozzolanic reaction, and we can assume that the measurements at 90 days will likely show higher strength.

The GVT-blended mortars were well hydrated and showed a homogeneous distribution and a dense C–S–H structure due to the high GVT particle surface. GVT-blended mortars showed a more compact and void-free structure than BVT-blended mortars at all ages. It is seen that the GVT-blended mortars harden entirely and fill the pores. The porosity of the 50% BVT-blended mortars increased (Figures 11 and 12). As the fineness of the cement increases, the amount of water required for the mortar also increases. Volcanic rocks with high GVT type SiO2 and Al2O3 content fill the capillary cracks and gaps in cement mortars if they are finely ground due to their high-density thanks to their acidic properties. Low-fidelity cements prepared with GVT type volcanic rocks protect and strengthen the connective tissue of mortars and concretes in acidic and sulfated environments [51]. In cement mortar containing 30% GVT, high CSH gel is seen and ettringite needles are very rare. SEM images of cement paste at 50% GVT admixture show very few gaps. As a result of the increase in the amount of pozzolan in blended cements, the decrease in the amount of clinker causes a reduction in alite mineral (C3S), which causes an early decrease in strength. In addition, this situation causes the formation of CH (CH product, which has no effect on strength development in mortars. Thus, the early strength of the pozzolanic-blended cements decreases as the additive ratios increase compared to the pure Portland Cem I cement. In addition, materials with strong pozzolanic properties on the 7th and 28th curing days bind the remaining CH over time and fill the gaps by transforming them into C–S–H gels. Thus, the compressive strength of pozzolanic cement mortars gradually increases. Pozzolanic-containing materials react slowly with free lime in the cement in the first place, reducing the temperature and hydration rate of fresh concrete and providing more uniform crystallization. At the end of these processes, the strength of concrete with pozzolanic properties is positively affected for a long time (≥6 months).

Figure 11 
                  SEM images of 30% GVT (a) and 50% (b) GVT-blended cement paste samples at 28 days.
Figure 11

SEM images of 30% GVT (a) and 50% (b) GVT-blended cement paste samples at 28 days.

Figure 12 
                  SEM images of 30% BVT (a) and 50% (b) BVT-blended cement paste samples at 28 days.
Figure 12

SEM images of 30% BVT (a) and 50% (b) BVT-blended cement paste samples at 28 days.

4 Conclusions

This study investigates the effects of the hybrid effect of different grinding mechanisms on mechanical activation on cement performance. The results of the study are given below.

  • It has been observed that the grain size of the material coming out of the roller press is finer than the grain size of the material coming out of the hammer mill in case of feeding in equal amounts.

  • As a result of ball milling, the highest energy consumption occurred in clinker, followed by BVT and GVT. In addition, the amount of energy required for the production of BVT is higher than the amount of energy necessary for the production of GVT cements.

  • The hybrid grinding system reduced the grinding time required for GVT and BVT activities by 50%. This means that the grinding energy is reduced.

  • The chemical composition and pozzolanic properties of GVT and BVT provided sufficient features as a substitution material for composite cement production. While GVT provides compressive strength up to 50% additive, BVT provides maximum compressive strength with 30% additive.

  • Considering the amount of energy and cost to produce building materials, high-efficiency hybrid milling will be a promising use method for reducing the mechanical activation time and energy amounts of raw materials.

  • The hybrid grinding method can achieve the structural degradation and amorphization degree required to improve the pozzolanic activities of raw materials in a shorter time with lower energy costs.

  • This method is readily applicable to almost all natural materials.

Acknowledgments

The authors report no acknowledgment for this work.

  1. Funding information: The authors report no funding information for this work.

  2. Author contributions: All works done by both authors.

  3. Conflict of interest: The authors declare that there is no conflict of interest.

  4. Ethical approval: The authors declare that there is no ethical issue for this work.

  5. Data availability statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

[1] Worrell E, Martin N, Price L. Potentials for energy efficiency improvement in the US cement industry. Energy. 2000;25(12):1189–214.10.1016/S0360-5442(00)00042-6Search in Google Scholar

[2] Al-Zboon KK, Al-Zou’by J. Effect of volcanic tuff on the characteristics of cement mortar. Eur J Environ Civ Eng. 2016;20(5):520–31.10.1080/19648189.2015.1053151Search in Google Scholar

[3] Perraki T, Kontori E, Tsivilis S, Kakali G. The effect of zeolite on the properties and hydration of blended cements. Cement Concr Compos. 2010;32(2):128–33.10.1016/j.cemconcomp.2009.10.004Search in Google Scholar

[4] Sun D, Wang L. Utilization of paraffin/expanded perlite materials to improve mechanical and thermal properties of cement mortar. Constr Build Mater. 2015;101:791–6.10.1016/j.conbuildmat.2015.10.123Search in Google Scholar

[5] Hossain KM. Volcanic ash and pumice as cement additives: pozzolanic, alkali-silica reaction and autoclave expansion characteristics. Cement Concr Res. 2005;35(6):1141–4.10.1016/j.cemconres.2004.09.025Search in Google Scholar

[6] Degirmenci N, Yilmaz A. Use of diatomite as partial replacement for Portland cement in cement mortars. Constr Build Mater. 2009;23(1):284–8.10.1016/j.conbuildmat.2007.12.008Search in Google Scholar

[7] Song HW, Pack SW, Nam SH, Jang JC, Saraswathy V. Estimation of the permeability of silica fume cement concrete. Constr Build Mater. 2010;24(3):315–21.10.1016/j.conbuildmat.2009.08.033Search in Google Scholar

[8] Canpolat F, Yılmaz K, Köse MM, Sümer M, Yurdusev MA. Use of zeolite, coal bottom ash and fly ash as replacement materials in cement production. Cement Concr Res. 2004;34(5):731–5.10.1016/S0008-8846(03)00063-2Search in Google Scholar

[9] Vegas I, Cano M, Arribas I, Frias M, Rodríguez O. Physical–mechanical behavior of binary cements blended with thermally activated coal mining waste. Constr Build Mater. 2015;99:169–74.10.1016/j.conbuildmat.2015.07.189Search in Google Scholar

[10] Aziz A, Stocker O, El Hassani IE, Laborier AP, Jacotot E, El Khadiri A, et al. Effect of blast-furnace slag on physicochemical properties of pozzolan-based geopolymers. Mater Chem Phys. 2021;258:123880.10.1016/j.matchemphys.2020.123880Search in Google Scholar

[11] Tulashie SK, Ebo P, Ansah JK, Mensah D. Production of Portland Pozzolana cement from rice husk ash. 2021. SSRN 3739628.10.2139/ssrn.3739628Search in Google Scholar

[12] Kolawole JT, Olusola KO, Babafemi AJ, Olalusi OB, Fanijo E. Blended cement binders containing bamboo leaf ash and ground clay brick waste towards the use for sustainable concrete. Cement Concr Compos. 2021;15:103975.10.1016/j.cemconcomp.2021.103975Search in Google Scholar

[13] Kolawole JT, Babafemi AJ, Fanijo E, Paul SC, Combrinck R. State-of-the-art review on the use of sugarcane bagasse ash in cementitious materials. Cement Concr Compos. 2021;118:103975.10.1016/j.cemconcomp.2021.103975Search in Google Scholar

[14] Alabi ST, Mahachi J, Mahachi J. Compressive strength of concrete containing palm oil fuel ash under different curing techniques. Mater Today Proc. 2021;43:1969–72.10.1016/j.matpr.2020.11.426Search in Google Scholar

[15] Wang Q, Li J, Zhang J, Wu P, Lyu X, Hu S, et al. Reuse of the soda sludge dealt with water washing as a supplementary material for the synthesis of clinker binders. J Clean Prod. 2021;295:126433.10.1016/j.jclepro.2021.126433Search in Google Scholar

[16] Frías M, De Rojas MS, García R, Valdés AJ, Medina C. Effect of activated coal mining wastes on the properties of blended cement. Cement Concr Compos. 2012;34(5):678–83.10.1016/j.cemconcomp.2012.02.006Search in Google Scholar

[17] Ali-Boucetta T, Behim M, Cassagnabere F, Mouret M, Ayat A, Laifa W. Durability of self-compacting concrete containing waste bottle glass and granulated slag. Constr Build Mater. 2021;270:121133.10.1016/j.conbuildmat.2020.121133Search in Google Scholar

[18] Bonfim WB, de Paula HM. Characterization of different biomass ashes as supplementary cementitious material to produce coating mortar. J Clean Prod. 2021;291:125869.10.1016/j.jclepro.2021.125869Search in Google Scholar

[19] Yao G, Liu Q, Wang J, Wu P, Lyu X. Effect of mechanical grinding on pozzolanic activity and hydration properties of siliceous gold ore tailings. J Clean Prod. 2019;217:12–21.10.1016/j.jclepro.2019.01.175Search in Google Scholar

[20] Wei B, Zhang Y, Bao S. Preparation of geopolymers from vanadium tailings by mechanical activation. Constr Build Mater. 2017;145:236–42.10.1016/j.conbuildmat.2017.03.234Search in Google Scholar

[21] Cheng Y, Huang F, Li W, Liu R, Li G, Wei J. Test research on the effects of mechanochemically activated iron tailings on the compressive strength of Concrete Constr. Constr Build Mater. 2016;118:164–70.10.1016/j.conbuildmat.2016.05.020Search in Google Scholar

[22] Wu P, Lyu X, Wang J, Hu S. Effect of mechanical grinding on the hydration activity of quartz in the presence of calcium hydroxide. Adv Cement Res. 2017;29(7):269–77.10.1680/jadcr.16.00159Search in Google Scholar

[23] Yao G, Wang Z, Yao J, Cong X, Anning C, Lyu X. Pozzolanic activity and hydration properties of feldspar after mechanical activation. Powder Technol. 2021;383:167–74.10.1016/j.powtec.2021.01.042Search in Google Scholar

[24] Yao G, Cui T, Jia Z, Sun S, Anning C, Qiu J, et al. Effect of anhydrite on hydration properties of mechanically activated muscovite in the presence of calcium oxide. Appl Clay Sci. 2020;196:105742.10.1016/j.clay.2020.105742Search in Google Scholar

[25] Feng Y, Yang Q, Chen Q, Kero J, Andersson A, Ahmed H, et al. Characterization and evaluation of the pozzolanic activity of granulated copper slag modified with CaO. J Clean Prod. 2019;232:1112–20.10.1016/j.jclepro.2019.06.062Search in Google Scholar

[26] Zhang L, Chen H, Deng R, Zuo W, Guo B, Ku J. Growth behavior of iron grains during deep reduction of copper slag. Powder Technol. 2020;367:157–62.10.1016/j.powtec.2019.11.107Search in Google Scholar

[27] TS EN 197-1. Cement-Part 1: general cements-composition. properties and compliance criteria; 2012.Search in Google Scholar

[28] Schönert K. Aspects of the physics of breakage relevant to comminution. Fourth Tewksbury Symposium, University of Melbourne. 1979;3(1):3.30.Search in Google Scholar

[29] TS EN 196-2. Cement test methods-Part 2: Chemical analysis of cement. Ankara, Turkey: Turkish Standards Institute; 2013.Search in Google Scholar

[30] TS 196-6. Turkish standards, methods of testing cement - Part 6: Determination of fineness. Ankara: Turkish Standards Institute; 2020.Search in Google Scholar

[31] TS 25. Turkish Standards, Natural pozzolan (Trass) for use in cement and concrete – Definitions, requirements and conformity criteria. Ankara: Turkish Standards Institute; 2015.Search in Google Scholar

[32] TS 25. Trass. Ankara, Turkey: Turkish Standards Institute; 2015.Search in Google Scholar

[33] TS EN 196-1. Methods of testing cement - Part 1: Determination of strength. Ankara, Turkey: Turkish Standards Institute; 2016.Search in Google Scholar

[34] TS EN 196-3. Methods of testing cement-Part 3: determination of setting time and soundness. Ankara, Turkey: Turkish Standards Institute; 2010.Search in Google Scholar

[35] Alp I, Deveci H, Süngün YH, Yilmaz AO, Kesimal A, Yilmaz E. Pozzolanic characteristics of a natural raw material for use in blended cements. Indian J Sci Technol. 2009;33:291.Search in Google Scholar

[36] Biricik H, Sanrı Karapınar I. Pozzolanic activity of central Anatolian volcanic tuff and its usability as admixture in mortar. Adv Cement Res. 2020;32(3):91–100.10.1680/jadcr.18.00033Search in Google Scholar

[37] Korkmaz AV. Mechanical activation of diabase and its effect on the properties and microstructure of Portland cement. Case Stud Const Mater. 2022;16:e00868.10.1016/j.cscm.2021.e00868Search in Google Scholar

[38] TS EN 196-5. Methods of testing cement-Part 5: Pozzolanicity test for pozzolanic cement. Ankara, Turkey: Turkish Standards Institute; 2010.Search in Google Scholar

[39] Cement Plant Equipment; 2022. https://www.cementplantequipment.com/products/cement-mill/cement-roller-press/.Search in Google Scholar

[40] Wang Z, Chu H, Wang J, Feng E, Zhang Y, Lyu X. Mechanical activation of siliceous tailings and its application as cement admixtures. Miner Eng. 2022;177:107366.10.1016/j.mineng.2021.107366Search in Google Scholar

[41] Ilić B, Radonjanin V, Malešev M, Zdujić M, Mitrović A. Effects of mechanical and thermal activation on pozzolanic activity of Kaolin containing mica. Appl Clay Sci. 2016;123:173–81.10.1016/j.clay.2016.01.029Search in Google Scholar

[42] Souri A, Kazemi-Kamyab H, Snellings R, Naghizadeh R, Golestani-Fard F, Scrivener K. Pozzolanic activity of mechanochemically and thermally activated kaolins in cement. Cement Concr Res. 2015;77:47–59.10.1016/j.cemconres.2015.04.017Search in Google Scholar

[43] Ustabaş İ, Kaya A. Comparing the pozzolanic activity properties of obsidian to those of fly ash and blast furnace slag. Constr Build Mater. 2018;164:297–307.10.1016/j.conbuildmat.2017.12.185Search in Google Scholar

[44] Wang Y, Ge Y, Wang X, Chen X, Li Q. The effect of powder activated carbon on mechanical properties and pore structures of cement-based mortars. Constr Build Mater. 2022;316:125798.10.1016/j.conbuildmat.2021.125798Search in Google Scholar

[45] Wu F, Li H, Yang K. Effects of mechanical activation on physical and chemical characteristics of coal-gasification slag. Coatings. 2021;11(8):902.10.3390/coatings11080902Search in Google Scholar

[46] Xiaowei LI, Qi CAO. Research progress on mechanical properties of FRP reinforced seawater coral aggregate concrete materials and structural components. 2022;39(3):926–941.Search in Google Scholar

[47] Uzal B, Turanlı L, Yücel H, Göncüoğlu MC, Çulfaz A. Pozzolanic activity of clinoptilolite: A comparative study with silica fume, fly ash and a non-zeolitic natural pozzolan. Cement Concr Res. 2010;40(3):398–404.10.1016/j.cemconres.2009.10.016Search in Google Scholar

[48] Uzal B, Turanlı L. Blended cements containing high volume of natural zeolites: Properties, hydration and paste microstructure. Cement Concr Compos. 2012;34(1):101–9.10.1016/j.cemconcomp.2011.08.009Search in Google Scholar

[49] Bumanis G, Vitola L, Stipniece L, Locs J, Korjakins A, Bajare D. Evaluation of Industrial by-products as pozzolans: A road map for use in concrete production. Case Stud Const Mater. 2020;13:e00424.10.1016/j.cscm.2020.e00424Search in Google Scholar

[50] Schankoski RA, Pilar R, de Matos PR, Prudencio LRJr, Ferron RD. Fresh and hardened properties of self-compacting concretes produced with diabase and gneiss quarry by-product powders as alternative fillers. Constr Build Mater. 2019;224:659–70.10.1016/j.conbuildmat.2019.07.095Search in Google Scholar

[51] Abali Y, Bayca SU, Targan S. Evaluation of blends tincal waste, volcanic tuff, bentonite and fly ash for use as a cement admixture. J Hazard Mater. 2006 Apr;131(1–3):126–30.10.1016/j.jhazmat.2005.09.031Search in Google Scholar PubMed

Received: 2022-09-04
Revised: 2022-09-14
Accepted: 2022-09-15
Published Online: 2022-10-12

© 2022 Abdul Vahap Korkmaz and Huseyin Fırat Kayıran, published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Downloaded on 8.2.2023 from https://www.degruyter.com/document/doi/10.1515/chem-2022-0217/html
Scroll Up Arrow