**Elevated Temperature Performance of Multiple-Blended Binder Concretes**

Haider M. Owaid, Roszilah Hamid and Mohd Raihan Taha

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64415

#### **Abstract**

Concretes that contain binary-blended binders (BBB) and ternary-blended binders (TBB) incorporating thermally activated alum sludge ash (AASA), silica fume (SF), ground-granulated blast-furnace slag (GGBS) and palm oil fuel ash (POFA) are exposed to temperatures as high as 800 °C. The water-binder ratio of the multiple-blended binder (MBB) concretes was 0.30, and the total binder and polypropylene (PP) fibre contents were 493 and 1.8 kg/m3 , respectively. The elevated temperature performance of the MBB concretes is evaluated in terms of the mass loss, compressive strength, ultrasonic pulse velocity (UPV) and surface cracks. The concrete strength deteriorated significantly due to elevated temperature up to 800 °C, but the residual strength of the BBB containing 15 % AASA was higher than that of the control and 20 % AASA concretes. Hightemperature exposure decreased measured UPV values. The concrete weight loss was more pronounced for TBB concretes. The elevated temperature performance of all of the TBB concretes was better than that of the BBB concretes with the same AASA replacement levels. It was observed that PP fibres help reduce spalling. BBB concrete containing 15 % AASA combined with either SF or GGBS or POFA exhibits superior performance at elevated temperature than Portland cement concrete at the same mix design proportion.

**Keywords:** elevated temperature, alum sludge, thermal activation, pozzolanic materi‐ als, multiple-blended binders

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

### **1. Introduction**

Concrete may be exposed to elevated temperatures upon the outbreak of fire or when they are located near furnaces or reactors. The exposure of concrete to high temperatures (above 200 °C) from accidental fire or elevated temperatures in industrial plants leads to high internal tensile stresses that can cause cracks and damage to concrete structures. The compressive strength of concrete also decreases significantly when subjected to elevated temperatures. Therefore, the residual compressive strength (RCS) is an appropriate factor to consider when assessing the strength of concrete after accidental fire exposure. It is vital for concrete that has been ex‐ posed to fire to maintain a high RCS to preserve the safety of the whole structure. The mix proportion, curing period, aggregate type and presence of pozzolanic materials are all fea‐ tures that affect the performance of concrete at elevated temperatures. The use of pozzolanic materials such as metakaolin (MK), silica fume (SF), fly ash (FA), ground-granulated blastfurnace slag (GGBS) and palm oil fuel ash (POFA) in concrete mixtures has been described as an efficient technique for improving the strength and other material properties of concrete. However, the temperature resistance of concrete incorporated with pozzolanic materials such as SF and MK was found to be lower than the resistance of control concrete [1, 2]. Over the last decade, therefore, there has been extensive research into the performance of high-strength concrete (HSC) that incorporate FA, GGBS and SF at elevated temperatures, such as those produced by fire. Phan and Carino [3] compiled experimental reports of the mechanical properties of concrete when exposed to rapid heating, as occurs in a fire. They found that the material properties of HSC vary with temperature differently than do those of normalstrength concrete (NSC) in the range between room temperature and approximately 450 °C and noted that the differences narrow at temperatures above 450 °C. Morsy and Shebl [4] found that a composition including 15 % MK and 5 % SF demonstrated improved fire resistance. It appeared that the MK had a more pronounced effect on the residual compressive strength of the con‐ crete. Behnood and Ziari [5] showed that the addition of SF had no significant effect on the relative residual compressive strength when the concrete was subjected to temperatures of 100 and 200 °C, but that the amount of SF had considerable influence on the residual compressive strength when the concrete was subjected to temperatures of 300 and 600 °C.

Phan et al. [6] reported that high-performance concrete (HPC) with higher original strength (lower w/c) and with SF retained more residual strength after exposure to elevated temperature than did those HPC with lower original strength (higher w/c) and without SF. There was a significant reduction in the weight of the specimen and the relative strength of the concrete at elevated temperatures (200–1200 °C) [7]. Ghandehari et al. [8] evaluated the residual mechan‐ ical properties of HPC after exposure to elevated temperatures by using SF. They found that after heating the concrete to 200 °C, the strength of all of the concrete samples showed a slight improvement compared with the strength of concrete at 100 °C. Mohammad et al. [9] reported the residual compressive strength of concrete containing 20 % POFA after exposure to elevated temperatures and subsequent cooling. They found that there was a continuous decrease in the residual compressive strength with increasing temperature. The highest reductions were observed in ordinary Portland cement (OPC) concrete: 22.5, 33 and 78 % at 300, 500 and 800 °C, respectively. They also found that residual performance was higher in POFA concrete than in the OPC concrete.

**1. Introduction**

88 High Performance Concrete Technology and Applications

Concrete may be exposed to elevated temperatures upon the outbreak of fire or when they are located near furnaces or reactors. The exposure of concrete to high temperatures (above 200 °C) from accidental fire or elevated temperatures in industrial plants leads to high internal tensile stresses that can cause cracks and damage to concrete structures. The compressive strength of concrete also decreases significantly when subjected to elevated temperatures. Therefore, the residual compressive strength (RCS) is an appropriate factor to consider when assessing the strength of concrete after accidental fire exposure. It is vital for concrete that has been ex‐ posed to fire to maintain a high RCS to preserve the safety of the whole structure. The mix proportion, curing period, aggregate type and presence of pozzolanic materials are all fea‐ tures that affect the performance of concrete at elevated temperatures. The use of pozzolanic materials such as metakaolin (MK), silica fume (SF), fly ash (FA), ground-granulated blastfurnace slag (GGBS) and palm oil fuel ash (POFA) in concrete mixtures has been described as an efficient technique for improving the strength and other material properties of concrete. However, the temperature resistance of concrete incorporated with pozzolanic materials such as SF and MK was found to be lower than the resistance of control concrete [1, 2]. Over the last decade, therefore, there has been extensive research into the performance of high-strength concrete (HSC) that incorporate FA, GGBS and SF at elevated temperatures, such as those produced by fire. Phan and Carino [3] compiled experimental reports of the mechanical properties of concrete when exposed to rapid heating, as occurs in a fire. They found that the material properties of HSC vary with temperature differently than do those of normalstrength concrete (NSC) in the range between room temperature and approximately 450 °C and noted that the differences narrow at temperatures above 450 °C. Morsy and Shebl [4] found that a composition including 15 % MK and 5 % SF demonstrated improved fire resistance. It appeared that the MK had a more pronounced effect on the residual compressive strength of the con‐ crete. Behnood and Ziari [5] showed that the addition of SF had no significant effect on the relative residual compressive strength when the concrete was subjected to temperatures of 100 and 200 °C, but that the amount of SF had considerable influence on the residual compressive

strength when the concrete was subjected to temperatures of 300 and 600 °C.

Phan et al. [6] reported that high-performance concrete (HPC) with higher original strength (lower w/c) and with SF retained more residual strength after exposure to elevated temperature than did those HPC with lower original strength (higher w/c) and without SF. There was a significant reduction in the weight of the specimen and the relative strength of the concrete at elevated temperatures (200–1200 °C) [7]. Ghandehari et al. [8] evaluated the residual mechan‐ ical properties of HPC after exposure to elevated temperatures by using SF. They found that after heating the concrete to 200 °C, the strength of all of the concrete samples showed a slight improvement compared with the strength of concrete at 100 °C. Mohammad et al. [9] reported the residual compressive strength of concrete containing 20 % POFA after exposure to elevated temperatures and subsequent cooling. They found that there was a continuous decrease in the residual compressive strength with increasing temperature. The highest reductions were observed in ordinary Portland cement (OPC) concrete: 22.5, 33 and 78 % at 300, 500 and 800 °C,

Demirel and Keleştemur [10] demonstrated that adding pozzolanic materials (finely ground pumice (FGP) and SF) to concrete decreased both unit weight and compressive strength. Morsy et al. [11] evaluated the effects of high temperature on cement paste mixes containing 15–30 % MK and 5–15 % SF OPC replacement. They concluded that the best performance was achieved with cement paste containing 10 and 15 % SF, which increased in strength at 400 °C by 39 and 48 % but decreased in strength at 600 °C by 33 and 43 %, respectively. Rahel et al. [12] studied the replacement of cement by high-volume fly ash combined with colloidal nanosilica to produce high-strength mortars after exposure to temperatures of 400 and 700 °C. They found that high-strength mortars that have equivalent residual strength after exposure to 700 °C to that of control (unheated) cement mortar specimens can be produced by replacing cement with high-volume fly ash and by using colloidal nanosilica. Because of the increasing use of HSC in columns, resistance to spalling has become one of the crucial components of effective fire resistance [13]. Surface spalling occurs when a low-permeability paste is subjected to a high rate of heating and the vapour pressure in the pores consequently develops stresses greater than the material's tensile strength [14].

The internal stresses in compression members make them more vulnerable to spalling. In particular, there is a greater risk that HPC, with its lower permeability, will spall at high temperature compared with conventional concrete. To combat the spalling effect in HSC, it is necessary to add polypropylene (PP) fibres to the concrete mixes. Polypropylene fibres melt at approximately 160–170 °C and become capable of producing moisture escape channels to release the vapour pressure. One major consideration in the design of buildings is the safety of the occupants in case of an outbreak of fire. As such, thorough knowledge of the behaviour of all construction materials is required before incorporating them into structural elements. Additionally, the growing prevalence of engineering structures characterized as large span, high rise and ultra-high rise has necessitated a continual increase in performance requirements.

Water treatment plants (WTPs) produce waste residual sludge, alum sludge (AS), when aluminium sulphate is used as a coagulant in the process of making drinking water for human consumption. Thermally activated alum sludge ash (AASA) is a new pozzolanic material that is acquired from the calcination of AS at 800 °C. Owaid et al. [15] studied the feasibility of using AASA as a pozzolanic material for replacing cement in binary-blended binder (BBB) and ternary-blended binder (TBB) concretes that incorporate SF, GGBS and POFA. According to their results, AASA exhibits pozzolanic behaviour and can be classified as a Class natural (N) pozzolan [15]. The BBB containing 15 % AASA increased the compressive and tensile strength of concrete up to 85.3 and 5.38 MPa at 28 days, respectively, but further increases in AASA content gradually reduced these strengths [15]. The mechanical properties of the ternary combinations are better than those of the binary mixes at the same AASA replacement levels [15].

The performance of multiple-blended binder (MBB) concretes incorporating AASA, SF, GGBS and POFA at elevated temperature is presented in this chapter. The fire resistance of MBB concretes with thermally activated alum sludge ash (AASA) and pozzolanic materials as replacements for cement in both binary and ternary blends can be used to determine the suitable application of the concrete.

## **2. Experimental programme**

## **2.1. Materials**

The cement used in the concrete mixtures was ordinary Portland cement (OPC) type I from Orang Kuat Berhad, which conforms to ASTM C150-1992 [16].The chemical composition and physical properties of OPC are as reported previously [15]. Alum sludge (AS) is the raw material used in the present research that was obtained from the drinking water purification process. The AS was collected at the ABASS Consortium water treatment plant and then oven dried at 105 °C for 24 h. The dried sludge was crushed and sieved through a 10 mm sieve to remove coarse and foreign particles.

In a previous study by Owaid et al. [15], an effective way of preparing thermally activated alum sludge ash (AASA) by thermal activation of AS was to incinerate dry alum sludge in a laboratory electric furnace at 800 °C for a period of 2 h with the heating rate of 5 °C/min. The chemical composition and physical properties of the AS and AASA are as given in Ref. [15]. Three types of pozzolanic materials, namely, SF, GGBS and POFA, were employed as a partial replacement of OPC by weight in different combinations of binary and ternary cementitious blends. The type of condensed silica fume (SF) was Force 10,000 D microsilica. The chemical and physical properties of these materials have previously been reported [15].

Fine and coarse aggregates obtained from local sources were in accordance with the ASTM standard. The local natural sand used as a fine aggregate had a maximum aggregate size of 4.75 mm and a fineness modulus of 2.89. The maximum size of the local coarse aggregate (crushed granite) was 10 mm; its specific gravity was 2.64, and its water absorption value was 0.48 %. The fine aggregate had a specific gravity of 2.61 and a water absorption value of 0.72 %. The superplasticizer (SP) used in this study was an aqueous solution of modified polycar‐ boxylate-based superplasticizer (ViscoCrete-2044). The specific gravity of 1.08 was utilized to achieve the desired workability in all HPC mixtures. The polypropylene (PP) fibres used in this study were obtained from Timuran Engineering and bore the brand name of fibrillated polypropylene fibre. They were white and 12.19 mm long, and they had a specific gravity of 0.9. The PP fibres were used to eliminate the spall effect for all specimens that were subjected to high temperatures and were added at an amount of 1.8 kg/m3 for concrete mixes.

#### **2.2. Mix proportions and preparation of specimens**

The mix proportions were based on recommendations by Owaid et al. [15]. **Table 1** shows the mix proportions for both categories of concrete. Nine types of concrete mixtures were prepared to explore the effects of elevated temperatures on the properties of MBBC specimens containing AASA in both binary- and ternary-blended cement with the same binder content of 493 kg/m3 , and the ratio of water-binder (w/b) was kept at 0.30. The aggregates used were in accordance with ASTM standards and comprised crushed granite gravel with a nominal maximum size of 10 mm and local natural sand with a maximum size of 4.75 mm. The superplasticizer used was an aqueous solution of modified polycarboxylates with two OPC mass fractions, 1.5 and 1.8 %.


FA\* stands for fine aggregates; CA\* for coarse aggregates; SP\*\*\* for superplasticizer; PP\*\*\*\* for polypropylene fibres.

**Table 1.** Concrete mix proportions [15].

replacements for cement in both binary and ternary blends can be used to determine the

The cement used in the concrete mixtures was ordinary Portland cement (OPC) type I from Orang Kuat Berhad, which conforms to ASTM C150-1992 [16].The chemical composition and physical properties of OPC are as reported previously [15]. Alum sludge (AS) is the raw material used in the present research that was obtained from the drinking water purification process. The AS was collected at the ABASS Consortium water treatment plant and then oven dried at 105 °C for 24 h. The dried sludge was crushed and sieved through a 10 mm sieve to

In a previous study by Owaid et al. [15], an effective way of preparing thermally activated alum sludge ash (AASA) by thermal activation of AS was to incinerate dry alum sludge in a laboratory electric furnace at 800 °C for a period of 2 h with the heating rate of 5 °C/min. The chemical composition and physical properties of the AS and AASA are as given in Ref. [15]. Three types of pozzolanic materials, namely, SF, GGBS and POFA, were employed as a partial replacement of OPC by weight in different combinations of binary and ternary cementitious blends. The type of condensed silica fume (SF) was Force 10,000 D microsilica. The chemical

Fine and coarse aggregates obtained from local sources were in accordance with the ASTM standard. The local natural sand used as a fine aggregate had a maximum aggregate size of 4.75 mm and a fineness modulus of 2.89. The maximum size of the local coarse aggregate (crushed granite) was 10 mm; its specific gravity was 2.64, and its water absorption value was 0.48 %. The fine aggregate had a specific gravity of 2.61 and a water absorption value of 0.72 %. The superplasticizer (SP) used in this study was an aqueous solution of modified polycar‐ boxylate-based superplasticizer (ViscoCrete-2044). The specific gravity of 1.08 was utilized to achieve the desired workability in all HPC mixtures. The polypropylene (PP) fibres used in this study were obtained from Timuran Engineering and bore the brand name of fibrillated polypropylene fibre. They were white and 12.19 mm long, and they had a specific gravity of 0.9. The PP fibres were used to eliminate the spall effect for all specimens that were subjected

The mix proportions were based on recommendations by Owaid et al. [15]. **Table 1** shows the mix proportions for both categories of concrete. Nine types of concrete mixtures were prepared to explore the effects of elevated temperatures on the properties of MBBC specimens containing AASA in both binary- and ternary-blended cement with the same binder content

, and the ratio of water-binder (w/b) was kept at 0.30. The aggregates used were

for concrete mixes.

and physical properties of these materials have previously been reported [15].

to high temperatures and were added at an amount of 1.8 kg/m3

**2.2. Mix proportions and preparation of specimens**

suitable application of the concrete.

90 High Performance Concrete Technology and Applications

**2. Experimental programme**

remove coarse and foreign particles.

**2.1. Materials**

of 493 kg/m3

The content of polypropylene fibres was 1.8 kg/m3 for all of the concrete mixtures that contained AASA to eliminate the spall effect. All concrete materials were mixed in a rotating pan mixer for approximately 5 min to conform with the mixing process described in ASTM C192-2002 [17]. The mixtures were cast into specimens by using 100 mm standard cube moulds and were compacted with a vibrating table to reduce the air voids content in mixes. The specimens were subsequently covered to prevent evaporative water loss. After casting, the moulded specimens were left in the casting room at a temperature of 26 °C for 24 h. After demoulding, the specimens were cured in a water tank for 28 days. Later, they were removed from the tank and left in the laboratory to air and cure naturally for up to 56 days, under similar conditions of temperature and relative humidity.

#### **2.3. Heating and cooling regimens**

After a curing period of 56 days, the concrete specimens were conveyed to an electrical furnace. The specimens were kept in the furnace for 3 h at maximum temperature; the rate of temper‐ ature increase in the automatic electric furnace was 5 °C/min [18], as observed in **Figure 1**. Subsequently, the power was shut off, and the specimens remained in the furnace until the temperature dropped to room temperature to prevent the specimens from experiencing thermal shock. For this research study, the concrete specimens were heated in an electric furnace to 400, 600 and 800 °C.

**Figure 1.** Automatic electric furnace used to heat concrete specimens.

#### **2.4. Testing procedures**

The residual properties of the unheated control mix were compared with the properties of the specimens for multiple-blended binder (MBB) concretes containing OPC, AASA and pozzolanic materials. Also, the crack patterns on the surface of the multiple-blended binder concretes were inspected after the heated specimens had cooled down. An electronic digital balance with an accuracy of ±0.1 g was used to determine the concrete mass loss (*M***loss**) before and after each heating temperature. The calculation of the concrete mass loss of specimens was based on Eq. (1); subsequently, specimens were taken out from furnace and weighed (Wd). After the concrete specimens had cooled to room temperature, they were removed from the furnace to determine their mass loss, residual compressive strength and ultrasonic pulse velocity of MBB concrete specimens. For each type of concrete, the residual properties were subsequently compared with the properties of the unheated control specimens. Additionally, crack patterns on the surface of the MBB concrete specimens were inspected after the heated specimens had cooled down. An electronic digital balance with an accuracy of ±0.1 g was used to determine the concrete mass loss (*M***loss**) before and after each heating step and the final weight (Wd). Calculation of the concrete mass loss of specimens was based on Eq. (1):

$$M\_{\text{loss}} = \frac{M\_{\text{initial}} - M\_{\text{load}}}{M\_{\text{initial}}} \tag{1}$$

where *M***initial** and *M***heated** are the initial mass (before heating) and heated mass (after heating) weighed in air, respectively. The compressive strength of the concrete was determined by crushing three 100 mm3 cubes for each mix. The test was carried out according to BS EN 12390-3 [19]. The compression load was applied at a rate of 3 kN/s using a compression machine with a capacity of 5,000 kN. The residual compressive strength (RCS) was calculated using the following Eq. (2):

$$RCS = \frac{\sigma\_{\text{elav}}}{\sigma\_{2\text{6}}} \times 100\tag{2}$$

where *σelev* is the compressive strength (MPa) of the cubes subjected to elevated temperature and *σ*26 is the compressive strength (MPa) of the cubes kept at room temperature (26 °C).

The portable equipment for ultrasonic non-destructive indicating test (known as PUNDIT) was used to measure ultrasonic velocity (V), in accordance with BS 1881: Part 203 (1986) [20]. Ultrasonic velocity (V) was determined by measuring the ultrasonic pulse transmission time by means of the direct transmission method. The ultrasonic pulse transmission time is the time required for longitudinal vibrations of ultrasonic frequency to travel a known distance through the material. Wave speed was calculated as:

$$V = \frac{L}{T} \tag{3}$$

where *L* is the transmission distance (m), *T* is the transmission time in the concrete (s), and *V* is the velocity of pulse transmission in the concrete (m/s) [20].

## **3. Results and discussion**

**Figure 1.** Automatic electric furnace used to heat concrete specimens.

92 High Performance Concrete Technology and Applications

The residual properties of the unheated control mix were compared with the properties of the specimens for multiple-blended binder (MBB) concretes containing OPC, AASA and pozzolanic materials. Also, the crack patterns on the surface of the multiple-blended binder concretes were inspected after the heated specimens had cooled down. An electronic digital balance with an accuracy of ±0.1 g was used to determine the concrete mass loss (*M***loss**) before and after each heating temperature. The calculation of the concrete mass loss of specimens was based on Eq. (1); subsequently, specimens were taken out from furnace and weighed (Wd). After the concrete specimens had cooled to room temperature, they were removed from the furnace to determine their mass loss, residual compressive strength and ultrasonic pulse velocity of MBB concrete specimens. For each type of concrete, the residual properties were subsequently compared with the properties of the unheated control specimens. Additionally, crack patterns on the surface of the MBB concrete specimens were inspected after the heated specimens had cooled down. An electronic digital balance with an accuracy of ±0.1 g was used to determine the concrete mass loss (*M***loss**) before and after each heating step and the final

weight (Wd). Calculation of the concrete mass loss of specimens was based on Eq. (1):

*M M <sup>M</sup> M*

*loss*

*initial heated*


cubes for each mix. The test was carried out according to BS EN 12390-3

*initial*

where *M***initial** and *M***heated** are the initial mass (before heating) and heated mass (after heating) weighed in air, respectively. The compressive strength of the concrete was determined by

[19]. The compression load was applied at a rate of 3 kN/s using a compression machine with

**2.4. Testing procedures**

crushing three 100 mm3

The saturated surface-dry specimens were heated to 400, 600 and 800 °C for 3 h and subse‐ quently cooled to room temperature (26 °C). **Figures 2**–**11** show the results obtained from the tests (losses in weight, compressive strength and ultrasonic pulse velocity) of MBB concrete specimens subjected to elevated temperatures. The finding is analyzed and discussed below.

#### **3.1. Concrete mass loss for different elevated temperatures**

**Figure 2** and **Table 2** present the percentage of the mass loss relative to the original weight (weight before heating) of the mixtures of OPC, AASA, SF, GGBS and POFA in the binaryand ternary-blended concrete with increasing temperature. It can be observed that the mass loss in all specimens showed a gradual increase from 3.72 to 7.41 % with the increase in temperature from 400 to 800 °C. The results also show that the mass loss increased with increasing amounts of AASA and pozzolanic materials. It is observed that the mass loss was less than 7.5 % for the specimens that did not exceed 800 °C. After being subjected to temper‐ atures of 400, 600 and 800 °C, the mass losses of OPC concrete were 3.78, 5.24 and 6.46 %, respectively. Additionally, the mass losses of AASA 15 % and AASA 20 % concrete were measured to be 3.98, 5.41 and 6.84 % and 4.1, 5.82 and 7.06 %, respectively. This small difference in the mass loss has not caused significant effects when the AASA replacement ratios are at 15 and 20 %.

**Figure 2.** Mass loss of BBB and TBB concrete mixtures subjected to different elevated temperatures.

Mass loss occurs in the specimens due to the loss of water. Because of the loss of bound water from the cement paste, air voids are formed in the concrete. The structural integrity of the specimens deteriorates corresponding to the increase in mass loss with increasing temperature. Similar observations were reported by Janotka and Nurnbergerova [21]. The rate of concrete mass loss was slower when the heating temperature increased from 400 to 800 °C. These findings of the current study agree with observations by Hanaa et al. [22], who reported that approximately 70 % of the water contained in the concrete had evaporated at 300 °C. **Figure 2** shows that the incorporation of SF, GGBS and POFA at 6, 20 and 15 % by total binder

weight, respectively, resulted in a decrease in mass loss of the pozzolanic material concrete with increasing temperature. The observations show that there were higher mass losses in SF concrete compared with GGBS and POFA concretes. The mass loss associated with the SF, GGBS and POFA concretes was 3.93, 3.72 and 3.81 % at 400 °C; 5.53, 5.39 and 5.42 % at 600 °C; and 6.82, 6.59 and 6.68 % at 800 °C,respectively. Similar observations were reported in various studies [9, 10, 23].As observed, all of the ternary binder mixtures shown in **Figure 2** lost more mass compared with the OPC concrete. In this study, the weight losses in the concrete containing both AASA and SF were higher than the losses in AASA + GGBS and AASA + POFA concrete with an increasing temperature from 400 to 800 °C. This result can be attributed to the specific gravity of the selected pozzolanic materials, with the specific gravity of SF being the lower than that of GGBS and POFA. The average mass loss of the ternary-blend binder concrete was 4.2 % at 400 °C, 5.8 % at 600 °C and 7.3 % at 800 °C.

in the mass loss has not caused significant effects when the AASA replacement ratios are at 15

**Figure 2.** Mass loss of BBB and TBB concrete mixtures subjected to different elevated temperatures.

Mass loss occurs in the specimens due to the loss of water. Because of the loss of bound water from the cement paste, air voids are formed in the concrete. The structural integrity of the specimens deteriorates corresponding to the increase in mass loss with increasing temperature. Similar observations were reported by Janotka and Nurnbergerova [21]. The rate of concrete mass loss was slower when the heating temperature increased from 400 to 800 °C. These findings of the current study agree with observations by Hanaa et al. [22], who reported that approximately 70 % of the water contained in the concrete had evaporated at 300 °C. **Figure 2** shows that the incorporation of SF, GGBS and POFA at 6, 20 and 15 % by total binder

and 20 %.

94 High Performance Concrete Technology and Applications


**Table 2.** Percentage of the mass loss relative to the original weight of the BBB and TBB mixtures.

#### **3.2. Residual compressive strength of concrete subjected to elevated temperatures**

**Figures 3**–**5** show the results of residual compressive strength (RCS) measurements of the specimens for multiple-blended binder concretes containing OPC, AASA and pozzolanic materials that were subjected to elevated temperatures followed by cooling in ambient air until the 56th day after the heating process. **Figures 6**–**8** indicate the relative residual compressive strength (ratio of residual compressive strength after elevated temperature to initial compres‐ sive strength at room temperature) of the concrete specimens. The residual strength of the MBB concrete specimens decreased as the temperature was increased. **Figure 3** shows that regard‐ less of the presence of PP fibres, the strength of AASA concrete deceased when the temperature was increased from 400 to 800 °C. Hence, there was a significant increase in the compressive strength of the binary concrete blends with 15 % AASA at room temperature (26 °C) compared with the compressive strength of the control mix and concrete mixture with 20 % AASA. The increase in compressive strength of the concrete mixture with 15 % AASA was approximately 12.4 % compared with the compressive strength of the control mix. However, the concrete with 20 % AASA cement exhibited reduced compressive strength.

**Figure 3.** Influence of AASA on the compressive strength of concrete subjected to different elevated temperatures.

The change in the strength of concrete specimens appeared to follow a common trend. Initially, as the temperature was increased to 400 °C, the strength decreased relative to that at room temperature; the relative decrease was approximately 14.2–17.7 %, as shown in **Table 3**. This effect could be due to variations in pore structure, including porosity and pore size distribu‐ tion, or to an increase in pore diameter [24]. According to several researchers [25–27], this reduction is primarily due to the release of free and physically bound water from the pores of the hydrates and to the first stage of dehydration in the hydrated products (calcium hydroxide, calcium silicate hydrates and calcium aluminosilicate hydrates) and the breakdown of tobermorite gel. The concrete containing 15 % AASA performed better and showed higher residual strength compared with the control-OPC and 20 % AASA concretes. A severe loss in strength was observed in the concrete as the temperature increased from 400 to 600 °C. The average strength loss was of 42.6 % (**Table 3**). The quick loss in compressive strength for concrete mixtures has been attributed to the dense microstructure of this type of concrete, which is the direct cause of the excessive build-up of vapour pressure. This pressure produces large cracks in the specimens during heating. The decomposition of calcium hydroxide and calcium carbonate in cement paste, which occurs from 430 to 600 °C, is an additional reason for the loss of strength [26, 28].


**Table 3.** Percentage of the strength loss relative to the original weight of the BBB and TBB mixtures.

increase in compressive strength of the concrete mixture with 15 % AASA was approximately 12.4 % compared with the compressive strength of the control mix. However, the concrete with

**Figure 3.** Influence of AASA on the compressive strength of concrete subjected to different elevated temperatures.

for the loss of strength [26, 28].

The change in the strength of concrete specimens appeared to follow a common trend. Initially, as the temperature was increased to 400 °C, the strength decreased relative to that at room temperature; the relative decrease was approximately 14.2–17.7 %, as shown in **Table 3**. This effect could be due to variations in pore structure, including porosity and pore size distribu‐ tion, or to an increase in pore diameter [24]. According to several researchers [25–27], this reduction is primarily due to the release of free and physically bound water from the pores of the hydrates and to the first stage of dehydration in the hydrated products (calcium hydroxide, calcium silicate hydrates and calcium aluminosilicate hydrates) and the breakdown of tobermorite gel. The concrete containing 15 % AASA performed better and showed higher residual strength compared with the control-OPC and 20 % AASA concretes. A severe loss in strength was observed in the concrete as the temperature increased from 400 to 600 °C. The average strength loss was of 42.6 % (**Table 3**). The quick loss in compressive strength for concrete mixtures has been attributed to the dense microstructure of this type of concrete, which is the direct cause of the excessive build-up of vapour pressure. This pressure produces large cracks in the specimens during heating. The decomposition of calcium hydroxide and calcium carbonate in cement paste, which occurs from 430 to 600 °C, is an additional reason

20 % AASA cement exhibited reduced compressive strength.

96 High Performance Concrete Technology and Applications

**Figure 4.** Influence of pozzolanic materials on the compressive strength of concrete subjected to different elevated tem‐ peratures.

All types of the MBB concretes showed severe deterioration at high temperatures between 600 and 800 °C temperature ranges, with further reduction in the strength. This reveals that the decomposition of C─S─H greatly affects the loss in the strength of concrete with severe deterioration for all mixtures of concrete [1]. The average loss of strength for the heated specimens was 66.1 %, compared with the strength of the unheated specimens. Thus, the sharp reduction in strength may be due to the formation of microcracks in the specimen, which weakens the interfacial transition zone and bonding between the aggregate and the cement paste. Therefore, the contractions of the paste lead to appear the cracks following the loss of water and expansion of the aggregate [29, 30].

Concrete mixtures that contained AASA exhibited extensive cracking and spalling, and their residual compressive strength was less than that of the control mixture. These effects are attributed to the presence and amount of filler additives in concrete mixtures that produce very dense transition zones between aggregates and paste due to their ultra-fine particles as filler materials and their pozzolanic reactions. During the expansion of the aggregates and contraction of the paste, higher stress concentrations are produced in the transition zone. These stresses worsen the bonding between aggregate and paste that contains filler additives compared with that of the control mixture [27]. The residual strength at 800 °C ranges from 23 to 29 % relative to unheated controls, as shown in **Figure 3**.

**Figure 4** presents the changes in the compressive strength of concrete that contained 6, 20 and 15 % of SF, GGBS and POFA, respectively, with increasing temperature from 400 to 800 °C. The observations show a decrease in the strength of the pozzolanic concretes at temperatures from 26 to 400 °C. This drop in strength is quantified as a 14.8–16.1 % reduction of the original strength, as given in **Table 3**. Once again, the concrete containing pozzolanic materials performed better and exhibited higher residual strength [1, 27].

Generally, it can be concluded that the loss in strength of pozzolanic concrete is caused by the dense microstructure in this type of concrete, which causes the build-up of high internal pressures due to the water-vapour transition in the water interlayer. The observations also reveal a severe loss in strength for all four types of concrete at temperatures ranging from 400 to 600 °C. This loss recorded is 41.3, 43.3, 42.1 and 42.6 % of the initial values for OPC, SF6, GGBS and POFA concretes, respectively (see **Table 3**). When the concrete specimens are subjected to high temperatures, the cement paste contracts and the aggregates expand. This response causes the transition zone to weaken and results in bonding between aggregates and cement paste. As a result, this process as well as the chemical decomposition of hydrated products causes severe deterioration and loss of strength in concrete after exposure to high temperatures. All the concrete containing pozzolanic materials exhibited severe deterioration up to 800 °C, and the average loss in strength was 64.1 % because of the decomposition of C─S─H gel. A similar observation has been reported by Demirel and Kelestemur [10]. From their results, it is shown that the pozzolanic material exerts a considerable influence on the residual strength. The concrete mixture containing SF performed poorly compared with the concrete mixtures containing GGBS and POFA. Although the addition of SF increased the initial strength of the concrete, there was a considerable compressive strength loss when the concrete was subjected to high temperatures. This decrease in strength likely arose from the very dense structure of SF concrete, which resulted in a build-up of vapour pressure due to the evaporation of physically and chemically bound water.

All types of the MBB concretes showed severe deterioration at high temperatures between 600 and 800 °C temperature ranges, with further reduction in the strength. This reveals that the decomposition of C─S─H greatly affects the loss in the strength of concrete with severe deterioration for all mixtures of concrete [1]. The average loss of strength for the heated specimens was 66.1 %, compared with the strength of the unheated specimens. Thus, the sharp reduction in strength may be due to the formation of microcracks in the specimen, which weakens the interfacial transition zone and bonding between the aggregate and the cement paste. Therefore, the contractions of the paste lead to appear the cracks following the loss of

Concrete mixtures that contained AASA exhibited extensive cracking and spalling, and their residual compressive strength was less than that of the control mixture. These effects are attributed to the presence and amount of filler additives in concrete mixtures that produce very dense transition zones between aggregates and paste due to their ultra-fine particles as filler materials and their pozzolanic reactions. During the expansion of the aggregates and contraction of the paste, higher stress concentrations are produced in the transition zone. These stresses worsen the bonding between aggregate and paste that contains filler additives compared with that of the control mixture [27]. The residual strength at 800 °C ranges from 23

**Figure 4** presents the changes in the compressive strength of concrete that contained 6, 20 and 15 % of SF, GGBS and POFA, respectively, with increasing temperature from 400 to 800 °C. The observations show a decrease in the strength of the pozzolanic concretes at temperatures from 26 to 400 °C. This drop in strength is quantified as a 14.8–16.1 % reduction of the original strength, as given in **Table 3**. Once again, the concrete containing pozzolanic materials

Generally, it can be concluded that the loss in strength of pozzolanic concrete is caused by the dense microstructure in this type of concrete, which causes the build-up of high internal pressures due to the water-vapour transition in the water interlayer. The observations also reveal a severe loss in strength for all four types of concrete at temperatures ranging from 400 to 600 °C. This loss recorded is 41.3, 43.3, 42.1 and 42.6 % of the initial values for OPC, SF6, GGBS and POFA concretes, respectively (see **Table 3**). When the concrete specimens are subjected to high temperatures, the cement paste contracts and the aggregates expand. This response causes the transition zone to weaken and results in bonding between aggregates and cement paste. As a result, this process as well as the chemical decomposition of hydrated products causes severe deterioration and loss of strength in concrete after exposure to high temperatures. All the concrete containing pozzolanic materials exhibited severe deterioration up to 800 °C, and the average loss in strength was 64.1 % because of the decomposition of C─S─H gel. A similar observation has been reported by Demirel and Kelestemur [10]. From their results, it is shown that the pozzolanic material exerts a considerable influence on the residual strength. The concrete mixture containing SF performed poorly compared with the concrete mixtures containing GGBS and POFA. Although the addition of SF increased the initial strength of the concrete, there was a considerable compressive strength loss when the concrete was subjected to high temperatures. This decrease in strength likely arose from the

water and expansion of the aggregate [29, 30].

98 High Performance Concrete Technology and Applications

to 29 % relative to unheated controls, as shown in **Figure 3**.

performed better and exhibited higher residual strength [1, 27].

**Figure 5.** Influence of ternary-blended binders on the compressive strength of concrete subjected to different elevated temperatures.

**Figure 5** illustrates the comparison of the residual compressive strengths of ternary-blend binder concrete specimens and the control specimens that were subjected to elevated temper‐ ature. It is clear that the residual compressive strength decreased as the treatment temperature increased to 800 °C. The observations show that the performance of the ternary blends with AASA and pozzolanic materials in terms of compressive strength is better than the perform‐ ance of binary blends with AASA for the same replacement levels of the unheated specimen (26 °C). **Figure 5** shows that the OPC + AASA + SF mix exhibited the highest strength, followed by OPC + AASA + POFA and OPC + AASA + GGBS. This result is due to the transformation of calcium hydroxide, which leads to the formation of calcium silicate hydrate on the surface of the aggregate particles. This transformation occurs because the average particle size of SF is very small compared with the particle sizes of other pozzolanic materials. This has led to the refinement of grains in ternary mixtures that contain AASA. From 26 to 400 °C, a decrease was observed in the performance of ternary blends in terms of the loss of compressive strength in mixtures containing AASA and pozzolanic materials. The loss in strength is from 16.2 to 17.7 % of the original strength, as shown in **Table 3**.

The OPC + AASA + SF mix exhibited the highest loss in strength, followed by OPC + AASA + POFA and OPC + AASA + GGBS. This loss is due to the ultra-fine particles of SF that are used as fillers, along with the pozzolanic reactions of the particles. All concretes that contain ternary blends of AASA and pozzolanic materials lose their strength at a faster rate when they are subjected to temperatures ranging from 400 to 600 °C. The loss is from 44.2 to 45.8 % of the original strength. At these temperatures, the dehydration of the cement paste results in its gradual disintegration. Because the paste tends to shrink and aggregates tend to expand at high temperatures, the bond between the aggregate and the paste is weakened, thereby reducing the strength of the concrete. At 800 °C, the average strength loss is 67.7 % for the ternary-blend binder concrete (see **Table 3**). The test results revealed that all of the tested concretes deteriorated at a temperature over 600 °C, as indicated in previous studies [10, 31].

**Figure 6.** Relative compressive strength of control and AASA specimens subjected to different elevated temperatures.

**Figures 6**–**8** show the values of the relative compressive strengths of the concrete mixtures containing OPC, AASA and pozzolanic materials in the form of binary and ternary blends after being subjected to high temperatures of 400, 600 and 800 °C. The relative strength was calculated as the percentage of strength retained by the concrete relative to the strength of the unheated specimen (26 °C). **Figure 6** shows that the relative compressive strength of the AASA concrete increases slightly when it is heated to 400 °C and then decreases slightly at 600 and 800 °C compared with the control concrete. At 400 °C, the relative compressive strength of concrete that contains 15 % AASA drops by 85.8 %, whereas that of the control-OPC and 20 % AASA concretes drop by 83.7 and 82.3 %, respectively. Finally, it is observed that a sharp reduction in relative strength occurs when the temperature increases beyond 600 and 800 °C due to the loss of crystal water, which leads to the reduction of the Ca(OH)2 content, morpho‐ logical changes and the formation of microcracks. The average relative compressive strengths of concrete mixtures are approximately 57.3 and 33.8 % when the concrete is subjected to temperatures of 600 and 800 °C, respectively. Thus, the compressive strength of concrete decreases significantly when the temperature rises beyond 400 °C. Similar results were obtained by Arioz [7].

The OPC + AASA + SF mix exhibited the highest loss in strength, followed by OPC + AASA + POFA and OPC + AASA + GGBS. This loss is due to the ultra-fine particles of SF that are used as fillers, along with the pozzolanic reactions of the particles. All concretes that contain ternary blends of AASA and pozzolanic materials lose their strength at a faster rate when they are subjected to temperatures ranging from 400 to 600 °C. The loss is from 44.2 to 45.8 % of the original strength. At these temperatures, the dehydration of the cement paste results in its gradual disintegration. Because the paste tends to shrink and aggregates tend to expand at high temperatures, the bond between the aggregate and the paste is weakened, thereby reducing the strength of the concrete. At 800 °C, the average strength loss is 67.7 % for the ternary-blend binder concrete (see **Table 3**). The test results revealed that all of the tested concretes deteriorated at a temperature over 600 °C, as indicated in previous studies [10, 31].

100 High Performance Concrete Technology and Applications

**Figure 6.** Relative compressive strength of control and AASA specimens subjected to different elevated temperatures.

**Figures 6**–**8** show the values of the relative compressive strengths of the concrete mixtures containing OPC, AASA and pozzolanic materials in the form of binary and ternary blends after being subjected to high temperatures of 400, 600 and 800 °C. The relative strength was calculated as the percentage of strength retained by the concrete relative to the strength of the unheated specimen (26 °C). **Figure 6** shows that the relative compressive strength of the AASA concrete increases slightly when it is heated to 400 °C and then decreases slightly at 600 and 800 °C compared with the control concrete. At 400 °C, the relative compressive strength of

**Figure 7** shows that the average relative compressive strengths of concrete specimens subjected to temperatures at 400, 600 and 800 °C with 6, 20 and 15 % of SF, GGBS and POFA were 84.2, 57.3 and 35.5 % of the strengths at room temperature, respectively. From the results, it can be concluded that below 400 °C, the relative residual compressive strength does not change significantly but that it does drop significantly at temperatures above 400 °C [24].

**Figure 7.** Relative compressive strength of control and pozzolanic material specimens subjected to different elevated temperatures.

**Figure 8** shows that the relative compressive strength of TBB concrete exhibits a significant decrease after being subjected to elevated temperatures. The relative compressive strength of mixtures OPC + AASA + SF, OPC + AASA + GGBS and OPC + AASA + POFA is approximately 82.3, 83.8 and 82.2 %, respectively, at 400 °C compared with 26 °C. A significant reduction of the relative compressive strength for mixtures OPC + AASA + SF, OPC + AASA + GGBS and OPC + AASA + POFA occurred at 600 °C. The loss recorded for each of the mixtures was 54.1, 55.7 and 54.7 %, respectively. The reduction of compressive strength of concrete is primarily attributed to the evaporative loss of free and physically bound water [11, 28]. After the temperature was increased up to 800 °C, the relative compressive strengths of OPC + AASA + SF, OPC + AASA + GGBS and OPC + AASA + POFA concretes were approximately 30.2, 33.8 and 32.7 % of the control concrete. It is clear that the compressive strength of concrete decreased significantly when the temperature was raised above 400 °C, as reported in several studies [32, 33]. The loss in strength was due to the excessive build-up of vapour pressure, which produced large cracks in the specimens [12]. Moreover, the binder products in cement paste dehydrate at this temperature, which causes a reduction in strength. However, specimens that contained AASA and SF had lower relative compressive strength compared with those of the ternaryblend binder specimens that contained GGBS and POFA materials.

**Figure 8.** Relative compressive strength of control and ternary-blended binder specimens subjected to different elevat‐ ed temperatures.

#### **3.3. Ultrasonic pulse velocity of concrete subjected to elevated temperatures**

mixtures OPC + AASA + SF, OPC + AASA + GGBS and OPC + AASA + POFA is approximately 82.3, 83.8 and 82.2 %, respectively, at 400 °C compared with 26 °C. A significant reduction of the relative compressive strength for mixtures OPC + AASA + SF, OPC + AASA + GGBS and OPC + AASA + POFA occurred at 600 °C. The loss recorded for each of the mixtures was 54.1, 55.7 and 54.7 %, respectively. The reduction of compressive strength of concrete is primarily attributed to the evaporative loss of free and physically bound water [11, 28]. After the temperature was increased up to 800 °C, the relative compressive strengths of OPC + AASA + SF, OPC + AASA + GGBS and OPC + AASA + POFA concretes were approximately 30.2, 33.8 and 32.7 % of the control concrete. It is clear that the compressive strength of concrete decreased significantly when the temperature was raised above 400 °C, as reported in several studies [32, 33]. The loss in strength was due to the excessive build-up of vapour pressure, which produced large cracks in the specimens [12]. Moreover, the binder products in cement paste dehydrate at this temperature, which causes a reduction in strength. However, specimens that contained AASA and SF had lower relative compressive strength compared with those of the ternary-

**Figure 8.** Relative compressive strength of control and ternary-blended binder specimens subjected to different elevat‐

ed temperatures.

blend binder specimens that contained GGBS and POFA materials.

102 High Performance Concrete Technology and Applications

The ultrasonic pulse velocities (UPVs) of the specimens of binary and ternary AASA and pozzolanic material MBB concretes that were subjected to different elevated temperatures are given in **Figures 9**–**11**. Each data point represents the average of three measurements. With regard to strength, the UPV values of both the binary and ternary AASA and the pozzolanic material MBB concrete specimens decreased with increasing temperatures. However, the rate of reduction in UPV was slightly different from that of strength. **Figure 9** reveals that the UPV values of the binary concrete blend with 15 % AASA were greater than those of the specimens with 20 % AASA and the control concrete at room temperature (26 °C). However, the values of the ultrasonic pulse velocities for AASA concrete were lower than those for the control concrete with increasing temperature, and there was a notable reduction in UPV after the specimens were subjected to elevated temperatures (i.e. higher than 400 °C).

**Figure 9.** UPV of control and AASA specimens subjected to different elevated temperatures.

Clearly, the transmission of pulse waves through a concrete mass is greatly influenced by the microcracking of the concrete. Thus, the decrease in pulse velocity with increasing temperature is a sensitive measure of the progress of cracking in the material [23, 27, 34]. **Figure 9** also shows that the UPV displayed a continuous drop when the temperature was raised. It was noted that the reduction of pulse velocity for all concrete mixtures was due to the content of PP fibres in the concrete mixtures. When the temperature was above 162 °C, the melting point of the PP fibres, these fibres created more randomly distributed pathways or voids in the concrete specimens. Moreover, thermal expansion and dehydration of the concrete due to high temperatures tend to cause the formation of fissures in the concrete. With more fissures, the cracks or micro-pathways delayed the pulse velocity in the concrete [35]. Therefore, micro‐ cracks reduce the UPV, resulting in low UPV values. The UPV values of OPC, AASA15 and AASA20 concrete mixtures were approximately 3173, 3189 and 3113 m/s at 400 °C; 2397, 2364 and 2224 m/s at 600 °C; and 1986, 1886 and 1832 m/s at 800 °C, respectively.

As illustrated in **Figure 10**, replacing part of the OPC with pozzolanic materials with 6, 20 and 15 % of SF, GGBS and POFA at 400, 600 and 800 °C caused a reduction in the UPV values. The specimens deteriorated due to increasing temperature, particularly at 800 °C, as reported by 23, 27, and the UPV values decreased substantially with increasing temperature. The measured UPVs of SF6, GGBS20 and POFA15 concretes after exposure to 400, 600 and 800 °C were approximately 3092, 3188 and 3151 m/s at 400 °C; 2287, 2332 and 2352 m/s at 600 °C; and 1829, 1963 and 1942 m/s at 800 °C, respectively.

**Figure 10.** UPV of control and pozzolanic material specimens subjected to different elevated temperatures.

**Figure 11** presents the measured UPV of TBB concretes in comparison to control concrete. **Figure 11** shows that there was a noticeable decrease in the UPV of the ternary blends of concrete with AASA and pozzolanic material after being subjected to temperatures higher than 400 °C. This decrease in the UPV values of the concrete specimens that have been subjected to high temperatures is due to the degeneration of the C─S─H gel at temperatures above 600 °C, which increases the amount of air voids and decreases the transmission speed of sound waves through the specimens. The decrease in UPV values was observed to be higher for SF-entrained concrete specimens, especially at 600 and 800 °C. A similar conclusion was reported by Demirel and Kelestemur [10]. This decrease results from the formation of a more porous structure due to the decomposition of the C─S─H gel, which is more abundant in samples containing SF.

the concrete mixtures. When the temperature was above 162 °C, the melting point of the PP fibres, these fibres created more randomly distributed pathways or voids in the concrete specimens. Moreover, thermal expansion and dehydration of the concrete due to high temperatures tend to cause the formation of fissures in the concrete. With more fissures, the cracks or micro-pathways delayed the pulse velocity in the concrete [35]. Therefore, micro‐ cracks reduce the UPV, resulting in low UPV values. The UPV values of OPC, AASA15 and AASA20 concrete mixtures were approximately 3173, 3189 and 3113 m/s at 400 °C; 2397, 2364

As illustrated in **Figure 10**, replacing part of the OPC with pozzolanic materials with 6, 20 and 15 % of SF, GGBS and POFA at 400, 600 and 800 °C caused a reduction in the UPV values. The specimens deteriorated due to increasing temperature, particularly at 800 °C, as reported by 23, 27, and the UPV values decreased substantially with increasing temperature. The measured UPVs of SF6, GGBS20 and POFA15 concretes after exposure to 400, 600 and 800 °C were approximately 3092, 3188 and 3151 m/s at 400 °C; 2287, 2332 and 2352 m/s at 600 °C; and 1829,

and 2224 m/s at 600 °C; and 1986, 1886 and 1832 m/s at 800 °C, respectively.

**Figure 10.** UPV of control and pozzolanic material specimens subjected to different elevated temperatures.

**Figure 11** presents the measured UPV of TBB concretes in comparison to control concrete. **Figure 11** shows that there was a noticeable decrease in the UPV of the ternary blends of concrete with AASA and pozzolanic material after being subjected to temperatures higher than

1963 and 1942 m/s at 800 °C, respectively.

104 High Performance Concrete Technology and Applications

**Figure 11.** UPV of control and ternary-blended binder specimens subjected to different elevated temperatures.

#### **3.4. Effect of elevated temperatures in multiple-blended-binder concretes on UPV and quality of concrete**

The heating regimens of UPV measured after exposure to elevated temperatures could be divided into three stages, ranging from 26 to 400 °C, 400 to 600 °C and 600 to 800 °C. The effects of elevated temperatures on the quality of MBB concretes containing AASA, SF, GGBS and POFA are illustrated in **Table 4**. Whitehurst [36] provided varying ranges of UPV ratings to describe the quality of concrete. For concrete of excellent quality, the UPV must be greater than 4500 m/s; for good-quality concrete, the UPV must be in the range of 3500–4500 m/s; for medium-quality concrete, the UPV should be in the range of 3000–3500 m/s, while for poorquality concrete, the UPV is in the range of 2000–3000 m/s; and finally, for very poor-quality concrete, the UPV is less than 2000 m/s.

**Table 4** illustrates the effects of elevated temperatures in MBB concretes that contain AASA, SF, GGBS and POFA on the quality of concrete within different heating regimens (°C). This table evaluates the quality of concrete processed at different temperatures for all of the types of mixtures that were subjected to the heating regimens of 26–400, 400–600 and 600–800 °C. The findings indicate that the specimens of the multiple-blended binder concretes degraded from excellent- to good- or medium-quality concrete, from medium- to poor-quality concrete and from poor- to almost very poor-quality concrete, respectively (see **Table 4**). The reason for this degradation is that the quality of concrete depends on the compressive strength. It is observed that the UPV values show a falling trend. The UPV values decrease with decreasing compressive strength for all of the mixtures after being subjected to temperatures of 400, 600 and 800 °C. **Table 5** tabulates the reduction of the UPV values for all concrete types. The UPV values for all mixtures at room temperature (26 °C) range from 4582 to 4743 m/s. Thus, all types of concrete produced at this temperature are classified as excellent-quality concrete, as shown in **Figures 9**–**11**.


**Table 4.** The effect of elevated temperatures on types of MBB concrete specimens.


**Table 5.** Reduction of the UPV relative to the original weight of the BBB and TBB mixtures.

#### **3.5. Surface observations of concrete specimens**

table evaluates the quality of concrete processed at different temperatures for all of the types of mixtures that were subjected to the heating regimens of 26–400, 400–600 and 600–800 °C. The findings indicate that the specimens of the multiple-blended binder concretes degraded from excellent- to good- or medium-quality concrete, from medium- to poor-quality concrete and from poor- to almost very poor-quality concrete, respectively (see **Table 4**). The reason for this degradation is that the quality of concrete depends on the compressive strength. It is observed that the UPV values show a falling trend. The UPV values decrease with decreasing compressive strength for all of the mixtures after being subjected to temperatures of 400, 600 and 800 °C. **Table 5** tabulates the reduction of the UPV values for all concrete types. The UPV values for all mixtures at room temperature (26 °C) range from 4582 to 4743 m/s. Thus, all types of concrete produced at this temperature are classified as excellent-quality concrete, as shown

**Concrete quality at different heating regimens (°C)** 

Control-OPC Medium Poor Very poor AASA15 Medium Poor Very poor AASA20 Medium Poor Very poor SF6 Medium Poor Very poor GGBS20 Medium Poor Very poor POFA15 Medium Poor Very poor AASA20 SF6 Medium Poor Very poor AASA20 GGBS20 Medium Poor Very poor AASA20 POFA15 Medium Poor Very poor

**Table 4.** The effect of elevated temperatures on types of MBB concrete specimens.

**Mix description (%) Reduction of the UPV values at different heating regimens (°C)** 

Control-OPC 3173 2397 1986 AASA15 3189 2364 1886 AASA20 3113 2224 1832 SF6 3092 2287 1829 GGBS20 3188 2332 1963 POFA15 3151 2352 1942 AASA20 SF6 3024 2218 1783 AASA20 GGBS20 3082 2270 1854 AASA20 POFA15 3119 2243 1887

**Table 5.** Reduction of the UPV relative to the original weight of the BBB and TBB mixtures.

**26–400 400–600 600–800** 

**26–400 400–600 600–800** 

in **Figures 9**–**11**.

106 High Performance Concrete Technology and Applications

**Mix description (%) Concrete quality** 

A thorough visual inspection was performed to evaluate the visible signs of cracking and spalling on the surface of the specimens after being subjected to elevated temperatures. The surface cracks began to appear after the specimens were subjected to elevated temperatures higher than 400 °C and continued to grow until the final rise in temperature up to 800 °C. There was no visible cracking or spalling for concrete specimens in the 26–400 °C temperature range, as shown in **Figure 12** (**a** and **b**). When the temperature was increased to approximately 600 °C, **Figure 12 (c)**, a network of visible fine surface cracks began to appear extensively and become even more pronounced at 800 °C (see **Figure 12 (d)**). In addition, the presence of PP fibres, which were used in all of the concrete mixtures, reduced or eliminated the risk of explosive spalling in all of the MBB concrete that contained AASA, SF, GGBS and POFA. During the process of rapid temperature increase at approximately 162 °C, polypropylene fibres melt and produce escape channels for vapour. The vapour produced in the specimens due to high temperature can be released without any build-up pressure. Hence, this might be the reason for the absence of explosive spalling in the multiple-blended binder concretes with PP fibres [37]. Generally, crack formations, propagations and pattern were similar in all of the multipleblended binder concrete control-OPC specimens and specimens containing AASA, SF, GGBS and POFA. However, there were noticeable differences in terms of sizes of cracks, including in the length, width and depth of the specimens.

**Figure 12.** Typical crack patterns observed in multiple-blended binder concretes at different temperatures: (a) 26 °C, (b) 400 °C, (c) 600 °C and (d) 800 °C.

## **4. Conclusion**

This study examines the behaviour of multiple-blended-binder concretes containing AASA and pozzolanic materials to form binary and ternary blends of cement at high temperatures, including the loss in weight, compressive strength and the reduction in ultrasonic pulse velocity as well as observations of surface characteristics of samples. The following conclusions can be drawn from the experimental results presented in this paper.


**8.** A visual inspection of the surface of the specimens that were subjected to high temperatures in the range of 26–400 °C revealed no visible cracking or spalling on these specimens. However, at 600 °C, visible networks of fine surface cracks were observed on some of the specimens. When the temperature was raised to 800 °C, all of the specimens showed visible spalling and cracking. However, the use of PP fibres reduced or eliminated the risk of explosive spalling and cracking for all specimens of multiple-blended-binder concrete.

## **Acknowledgements**

**4. Conclusion**

20 % replacement levels.

108 High Performance Concrete Technology and Applications

and POFA.

800 °C.

This study examines the behaviour of multiple-blended-binder concretes containing AASA and pozzolanic materials to form binary and ternary blends of cement at high temperatures, including the loss in weight, compressive strength and the reduction in ultrasonic pulse velocity as well as observations of surface characteristics of samples. The following conclusions

**1.** The mass loss decreases when the concrete is subjected to elevated temperature. That is, the mass losses of OPC concrete were recorded at 3.78, 5.24 and 6.46 %, and the mass losses of concrete mixtures of AASA 15 % and AASA 20 % were recorded at 3.98, 5.41 and 6.84  % and 4.1, 5.82 and 7.06% at 400, 600 and 800 °C, respectively. The mass loss of OPC concrete was slightly less than that of the concrete mixtures of AASA 15 % and AASA

**2.** As for the replacement levels of SF, GGBS and POFA concrete, the observations revealed a decrease in mass loss with increasing temperature. The loss in mass for the mixtures was in the range of 3.72–6.82 %. All of the ternary binder mixtures tended to show more mass loss compared with OPC concrete. The average mass loss of the concrete of ternary binder

**3.** The residual compressive strengths of specimens were lower than those of the control concrete for all mixtures. The performance of the residual compressive strength of binaryblended concrete mixture with 15 % AASA was observed to be better compared with the performance of the residual compressive strength of control concrete and the binary-

**4.** The pozzolanic concretes containing SF6, GGBS20 and POFA15 exhibited better perform‐ ance at elevated temperatures than did the control-OPC concrete. The concrete mixture containing SF performed poorly compared with the concrete mixtures containing GGBS

**5.** Based on the experimental results obtained from the ternary mixtures with AASA and pozzolanic materials, the relative compressive strength of concrete exhibited a significant decrease after the specimens were subjected to elevated temperature between 400 and

**6.** The heated binary and ternary-blended-binder concrete specimens showed decreased UPV values with increasing temperature. UPV values were reduced considerably for concrete specimens that were subjected to temperatures between 400 and 800 °C, which indicates that the physical state of the concrete specimens deteriorated rapidly when the

**7.** At high temperatures, the quality of the MBB concretes that contained AASA, SF, GGBS and POFA degraded from excellent to good or medium quality, from medium- to poorquality concrete and from poor quality to inferior (very poor) quality, respectively,

because the quality of concrete depends on the compressive strength.

can be drawn from the experimental results presented in this paper.

mixtures was 4.2 % at 400 °C, 5.8 % at 600 °C and 7.3 % at 800 °C.

blended concrete mixture with 20 % AASA.

temperature was raised above 400 °C.

The authors would like to acknowledge the financial support from Universiti Kebangsaan Malaysia through grants AP-2015-011 and DIP-2014-019.

## **Author details**

Haider M. Owaid1,2, Roszilah Hamid2\* and Mohd Raihan Taha2

\*Address all correspondence to: roszilah@ukm.edu.my

1 Department of Civil Engineering, Faculty of Engineering, University of Babylon, Babil, Iraq

2 Department of Civil and Structural Engineering, Faculty of Engineering and Built Environ‐ ment, Universiti Kebangsaan Malaysia, Bangi, Selangor Darul Ehsan, Malaysia

## **References**


[20] British Standard Institution. BS 1881–203. Testing Hardened Concrete. Recommenda‐ tions for Measurement of Velocity of Ultrasonic Pulses in Concrete. London: British Standards Institution (BSI); 1986.

[6] Phan LT, Lawson JR, Davis FL. Effects of elevated temperature exposure on heating characteristics, spalling and residual properties of high performance concrete. Materi‐

[7] Arioz O. Effects of elevated temperatures on properties of concrete. Fire Safety Journal.

[8] Ghandehari M, Ali B, Mostafa K. Residual mechanical properties of high strength concretes after exposure to elevated temperatures. Journal of Materials in Civil

[9] Mohammad I, Mohamed IE, Muhammad B. Influence of elevated temperatures on physical and compressive strength properties of concrete containing palm oil fuel ash.

[10] Demirel B, Kelestemur O. Effect of elevated temperature on the mechanical properties of concrete produced with finely ground pumice and silica fume. Fire Safety Journal.

[11] Morsy MM, Shebl SS, Rashad AM. Effect of fire on microstructure and mechanical properties of blended cement pastes containing metakaolin and silica fume. Asian

[12] Rahel KI, Hamid R, Taha MR. Fire resistance of high-volume fly ash mortars with nanosilica addition. Construction and Building Materials. 2012;36:779–786.

[14] Caldarone M. High-Strength Concrete: A Practical Guide. London: Taylor & Francis

[15] Owaid HM, Hamid R, Taha MR. Influence of thermally activated alum sludge ash on the engineering properties of multiple-blended binders concretes. Construction and

[16] ASTM C150. Standard Specification for Portland Cement. Annual Book of ASTM Standard. Vol. 04.02. Philadelphia: America Society for Testing and Materials; 1992.

[17] ASTM C192/C 192M. Standard Practice for Making and Curing Concrete Test Speci‐ mens in the Laboratory. West Conshohocken, PA: ASTM International; 2002.

[18] Culfik MS, Ozturan T. Mechanical properties of normal and high strength concretes subjected to high temperatures and using image analysis to detect bond deteriorations.

[19] British Standard Institution European Standard. BS EN 12390-3. Testing Hardened

Concrete. Compressive Strength of Test Specimens. London: BSI; 2002.

Construction and Building Materials. 2010;24:1486–1493.

Journal of Civil Engineering (Building and Housing). 2008;9(2):93–105.

[13] Buchanan A. Structural Design for Fire Safety. UK: Wiley Chichester; 2002.

Construction and Building Materials. 2011;25:2358–2364.

als and Structural. 2001;34(236):83–91

2007;42(8):516–522.

110 High Performance Concrete Technology and Applications

2010;45:385–391.

Group; 2008.

Building Materials. 2014;61(30):216–229.

Engineering. 2010;20:59–64.


**High Performance Concrete Technology and Applications**

[34] Yang H, Lin Y, Hsiao C, Liu JY. Evaluating residual compressive strength of concrete at elevated temperatures using ultrasonic pulse velocity. Fire Safety Journal.

[35] Yuksel I, Siddique R, Ozkan O. Influence of high temperature on the properties of concretes made with industrial by-products as fine aggregate replacement. Construc‐

[36] Whitehurst EA. Soniscope tests concrete structures. Journal of the American Concrete

[37] Chen B, Liu J. Residual strength of hybrid-fiber reinforced high strength concrete after exposure to high temperature. Cement and Concrete Research. 2004;34:1065–1069.

2009;44:121–130.

Institute. 1951;47:443–440.

112 High Performance Concrete Technology and Applications

tion and Building Materials. 2011;25:967–972.

## **Energy-Efficient Technologies in Cement Grinding**

## Ömürden Genç

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64427

#### **Abstract**

In this chapter an introduction of widely applied energy-efficient grinding technologies in cement grinding and description of the operating principles of the related equip‐ ments and comparisons over each other in terms of grinding efficiency, specific energy consumption, production capacity and cement quality are given. A case study per‐ formed on a typical energy-efficient Horomill® grinding technology, is explained. In this context, grinding circuit is introduced and explanations related to grinding and classification performance evaluation methodology are given. Finally, performance data related to Horomill® and high-efficiency TSV™ air classifier are presented.

**Keywords:** Barmac Vertical Shaft Impact Crusher (VSI), High-pressure grinding rolls, Vertical roller mills, CKP pre-grinder, Cemex® mill, Horomill®, TSV™ separator, Grinding, Classification, Energy, Cement

## **1. Introduction**

Cement is an energy-intensive industry in which the grinding circuits use more than 60 % of the total electrical energy consumed and account for most of the manufacturing cost [1]. The requirements for the cement industry in the future are to reduce the use of energy in grinding and the emission of CO2 from the kilns. In recent years, the production of composite cements has been increasing for reasons concerned with process economics, energy reduction, ecology (mostly reduction of CO2 emission), conservation of resources and product quality/diversity. The most important properties of cement, such as strength and workability, are affected by its specific surface and by the fineness and width of the particle-size distribution. These can be modified to some extent by the equipment used in the grinding circuit, including its configu‐ ration and control.

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Performance of grinding circuits has been improved in recent years by the development of machinery such as high-pressure grinding rolls (HPGR) (roller presses), Horomills, highefficiency classifiers and vertical roller mills (VRM) for clinker grinding which are more energy efficient than machinery which has been in common use for many years such as tube mills. Energy-efficient equipments such as high-pressure grinding rolls, vertical roller mills, CKP pre-grinders, Cemex® mills and Horomills® are used at both finish grinding of cement and raw material-grinding stages due to higher energy consumption of conventional multi-compart‐ ment ball milling circuits. Multi-compartment ball mills can be classified as:


Multi-compartment ball mills and air separators have been the main process equipments in clinker grinding circuits in the last 100 years. They are used in grinding of cement raw materials (raw meal) (i.e. limestone, clay, iron ore), cement clinker and cement additive materials (i.e. limestone, slag, pozzolan) and coal. Multi-compartment ball mills are relatively inefficient at size reduction and have high specific energy consumption (kWh/t). Typical specific energy consumption is 30 kWh/t in grinding of cement. Barmac-type crushers found application as a pre-grinder in cement grinding circuits operating with ball mills to reduce the specific energy consumption of ball mill-grinding stage [2]. An overview of technical innovations to reduce the power consumption in cement plants was given by Fujimoto [1].

In this chapter, operating principles of high-pressure grinding rolls, Horomill®, vertical roller mills, CKP pre-grinders and Cemex® mills which are widely applied in finish grinding of cement are briefly explained in addition to the advantages and disadvantages over each other.

## **2. Energy-efficient grinding systems**

#### **2.1. Barmac VSI crusher**

The Barmac rock-on-rock crusher has a rotor that acts as a high-velocity, dry stone pump, hurling a continuous rock stream into a stone-lined crushing chamber. Broken rock about 30– 50 mm in diameter enters the top of the machine from a feeder set and is accelerated in the rotor to be discharged into the crushing chamber at velocities of up to 85 m/s. Collision of highspeed rocks, with rocks falling in a separate stream or with a rock-lined wall, causes shattering. The product is typically gravel and sand-sized particles. Barmac crushers are available from 75 to 600 kW. The product-size distribution can be controlled by the rotor speed [3]. A schematic of a Barmac-type VSI crusher is given in **Figure 1** [4].

#### **2.2. High-pressure grinding rolls (HPGR)**

High-pressure grinding rolls (roller presses) are used in both raw material and cement grinding. The principle of the HPGR is shown in **Figure 2**.

**Figure 1.** Barmac VSI crusher.

Performance of grinding circuits has been improved in recent years by the development of machinery such as high-pressure grinding rolls (HPGR) (roller presses), Horomills, highefficiency classifiers and vertical roller mills (VRM) for clinker grinding which are more energy efficient than machinery which has been in common use for many years such as tube mills. Energy-efficient equipments such as high-pressure grinding rolls, vertical roller mills, CKP pre-grinders, Cemex® mills and Horomills® are used at both finish grinding of cement and raw material-grinding stages due to higher energy consumption of conventional multi-compart‐

Multi-compartment ball mills and air separators have been the main process equipments in clinker grinding circuits in the last 100 years. They are used in grinding of cement raw materials (raw meal) (i.e. limestone, clay, iron ore), cement clinker and cement additive materials (i.e. limestone, slag, pozzolan) and coal. Multi-compartment ball mills are relatively inefficient at size reduction and have high specific energy consumption (kWh/t). Typical specific energy consumption is 30 kWh/t in grinding of cement. Barmac-type crushers found application as a pre-grinder in cement grinding circuits operating with ball mills to reduce the specific energy consumption of ball mill-grinding stage [2]. An overview of technical innovations to reduce

In this chapter, operating principles of high-pressure grinding rolls, Horomill®, vertical roller mills, CKP pre-grinders and Cemex® mills which are widely applied in finish grinding of cement are briefly explained in addition to the advantages and disadvantages over each other.

The Barmac rock-on-rock crusher has a rotor that acts as a high-velocity, dry stone pump, hurling a continuous rock stream into a stone-lined crushing chamber. Broken rock about 30– 50 mm in diameter enters the top of the machine from a feeder set and is accelerated in the rotor to be discharged into the crushing chamber at velocities of up to 85 m/s. Collision of highspeed rocks, with rocks falling in a separate stream or with a rock-lined wall, causes shattering. The product is typically gravel and sand-sized particles. Barmac crushers are available from 75 to 600 kW. The product-size distribution can be controlled by the rotor speed [3]. A

High-pressure grinding rolls (roller presses) are used in both raw material and cement

ment ball milling circuits. Multi-compartment ball mills can be classified as:

the power consumption in cement plants was given by Fujimoto [1].

schematic of a Barmac-type VSI crusher is given in **Figure 1** [4].

grinding. The principle of the HPGR is shown in **Figure 2**.

**•** Single-compartment ball mills

**•** Two- or three-compartment ball mills

116 High Performance Concrete Technology and Applications

**2. Energy-efficient grinding systems**

**2.2. High-pressure grinding rolls (HPGR)**

**2.1. Barmac VSI crusher**

**Figure 2.** Principle of compressive size reduction.

The material between the rolls is submitted to a very high pressure ranging from 100 to 200 MPa. Special hard materials are used as protection against wear, for example, Ni-hard linings to protect the rollers. During the process, cracks are formed in the particle, and fine particles are generated. Material is fed into the gap between the rolls, and the crushed material leaves as a compacted cake. The energy consumption is 2.5–3.5 kWh/t and about 10 kWh/t when recycling of the material is used. The comminution efficiency of a HPGR is better than ball mills such that it consumes 30–50 % of the specific energy as compared to a ball mill. Four circuit configurations of HPGR can be used in grinding of raw materials, clinker and slag such as [5]:


**Figure 4.** Semifinish-grinding options.

Application of HPGR in cement grinding circuits and the effects of operational and design characteristics of HPGR on grinding performance were discussed by Aydoğan [6]. HPGR arrangements and semifinish-grinding options are given in **Figures 3** and **4**.

## **2.3. Vertical roller mills (VRM)**

**Figure 3.** HPGR arrangements.

118 High Performance Concrete Technology and Applications

**Figure 4.** Semifinish-grinding options.

Vertical roller mills have a lower specific energy consumption than tumbling mills and require less space per unit and capacity at lower investment costs. Vertical roller mills are developed to work as air-swept grinding mills. Roller mills are operated with throughput capacities of more than 300 t/h of cement raw mix (Loesche mill, Polysius® double roller mill, Pfeiffer® MPS mill). Loesche roller mill and Polysius® roller mills are widely applied in cement raw material grinding. Schematical view of a Pfeiffer MPS mill is given in **Figure 5** [7], and a view from inside of a vertical roller mill is given in **Figure 6**.

**Figure 5.** Schematical view of a Pfeiffer MPS mill [7].

**Figure 6.** A view from the interior of a vertical roller mill.

#### *2.3.1. Loesche vertical roller mill*

A cross section of a Loesche mill with a conical rotor-type classifier is shown in **Figure 7**. The pressure arrangement of the grinding rolls is hydraulic. The mill feed is introduced into the mill from above, falling centrally upon the grinding plate; then it is thrown by centrifugal force underneath the grinding rollers. A retention ring on the periphery of the grinding table forms the mill feed into a layer called the grinding bed. The ground material spills over the rim of the retention ring. Here an uprising airstream lifts the material to the rotor-type classifier located at the top of the mill casing where the coarse particles are separated from the fines. The coarse particles drop back into the centre of the grinding compartment for further size reduction, whereas the fines together with the mill air leave the mill and the separator. The separator controls the product sizes from 400 to 40 μm. The moisture of the mill feed (cement raw material) can amount to 15–18 %. The fineness of the mill product can be adjusted in the range between 94 and 70 % passing 170 mesh. Capacities up to 400 t/h of cement raw mix are recorded [8].

#### *2.3.1.1. Cement quality*

Better product quality can be achieved as compared to the ball mill product due to the better options for separate grinding. For example, in additive cement production, the blast furnace slag has to be ground to Blaine values of 5,000 cm2 /g. Water demand and setting times are similar to that of a ball mill cement under comparable conditions [9].

**Figure 6.** A view from the interior of a vertical roller mill.

120 High Performance Concrete Technology and Applications

slag has to be ground to Blaine values of 5,000 cm2

similar to that of a ball mill cement under comparable conditions [9].

A cross section of a Loesche mill with a conical rotor-type classifier is shown in **Figure 7**. The pressure arrangement of the grinding rolls is hydraulic. The mill feed is introduced into the mill from above, falling centrally upon the grinding plate; then it is thrown by centrifugal force underneath the grinding rollers. A retention ring on the periphery of the grinding table forms the mill feed into a layer called the grinding bed. The ground material spills over the rim of the retention ring. Here an uprising airstream lifts the material to the rotor-type classifier located at the top of the mill casing where the coarse particles are separated from the fines. The coarse particles drop back into the centre of the grinding compartment for further size reduction, whereas the fines together with the mill air leave the mill and the separator. The separator controls the product sizes from 400 to 40 μm. The moisture of the mill feed (cement raw material) can amount to 15–18 %. The fineness of the mill product can be adjusted in the range between 94 and 70 % passing 170 mesh. Capacities up to 400 t/h of cement raw mix are

Better product quality can be achieved as compared to the ball mill product due to the better options for separate grinding. For example, in additive cement production, the blast furnace

/g. Water demand and setting times are

*2.3.1. Loesche vertical roller mill*

recorded [8].

*2.3.1.1. Cement quality*

#### *2.3.2. Polysius® vertical roller mill (drying grinding roller mill)*

A mill feed arrangement conveys the raw material to the grinding bowl. Two double rollers (representing four grinding rollers) are put in motion by the revolving grinding bowl. The double rollers are independently mounted on a common shaft; they move and adjust them‐

selves to the velocity of the grinding bowl as well as to the thickness of the grinding bed. Thus, rollers are in permanent contact with the grinding bed. A hydropneumatic arrange‐ ment transfers the grinding pressure to the rollers. The disintegrated mill feed is shifted to the grinding bowl rim from where a gas stream emerging from the nozzle ring surrounding the grinding bowl carries the material upwards to the separator. The coarses precipitated in the separator gravitate centrally back to the grinding bowl, whereas the fines are collected in the electric precipitator. A raw material moisture of up to 8 % can be dried when utilizing the preheater exit gases only. If hot air from an air heater is also supplied, then a raw mate‐ rial moisture of up to 18 % can be handled [8]. The power requirement is 10–20 % lower than a ball mill, depending upon the grindability and moisture content of the raw material [10]. Other types of roller mills such as ball race mill (Fuller-Peters mill) and Raymond bowl-type ring mill are used in coal grinding.

### **2.4. CKP vertical pre-grinder**

The CKP pre-grinder has been under development by Chichibu Cement and Kawasaki Heavy Industries since 1987. It has been commissioned by Technip under licence since 1993. The system is applied widely for clinker grinding and has also been used on raw material grinding. In operation, material is fed through the inlet chute onto the grinding table centre, spread out to the grinding path by the centrifugal force arising from the table rotation, before being compressed and ground by the rollers. The preground material drops down out of the periphery of the table to the bottom of the casing and is discharged by the scrapers through the discharge chute. Grinding principle of the CKP system is shown in **Figure 8**. Typical CKP application is given in **Figure 9** [11].

**Figure 8.** The grinding principle of the CKP [11].

**Figure 9.** Typical CKP pre-grinding circuit.

selves to the velocity of the grinding bowl as well as to the thickness of the grinding bed. Thus, rollers are in permanent contact with the grinding bed. A hydropneumatic arrange‐ ment transfers the grinding pressure to the rollers. The disintegrated mill feed is shifted to the grinding bowl rim from where a gas stream emerging from the nozzle ring surrounding the grinding bowl carries the material upwards to the separator. The coarses precipitated in the separator gravitate centrally back to the grinding bowl, whereas the fines are collected in the electric precipitator. A raw material moisture of up to 8 % can be dried when utilizing the preheater exit gases only. If hot air from an air heater is also supplied, then a raw mate‐ rial moisture of up to 18 % can be handled [8]. The power requirement is 10–20 % lower than a ball mill, depending upon the grindability and moisture content of the raw material [10]. Other types of roller mills such as ball race mill (Fuller-Peters mill) and Raymond

The CKP pre-grinder has been under development by Chichibu Cement and Kawasaki Heavy Industries since 1987. It has been commissioned by Technip under licence since 1993. The system is applied widely for clinker grinding and has also been used on raw material grinding. In operation, material is fed through the inlet chute onto the grinding table centre, spread out to the grinding path by the centrifugal force arising from the table rotation, before being compressed and ground by the rollers. The preground material drops down out of the periphery of the table to the bottom of the casing and is discharged by the scrapers through the discharge chute. Grinding principle of the CKP system is shown in **Figure 8**. Typical CKP

bowl-type ring mill are used in coal grinding.

122 High Performance Concrete Technology and Applications

**2.4. CKP vertical pre-grinder**

application is given in **Figure 9** [11].

**Figure 8.** The grinding principle of the CKP [11].

Main advantages of the CKP pre-grinders are stated by Dupuis and Rhin [11] as follows:


#### **2.5. Cemex® ring roller mill**

F.L.Smidth has developed this cement grinding system which is a fully air-swept ring roller mill with internal conveying and grit separation. This mill is a major improvement of the cement grinding systems known today which are ball mill, roller press (HPGR)/ball mill, vertical roller mill and closed-circuit roller press for finish grinding. Views of mill interior are given in **Figures 10** and **11**. Cemex® grinds the material by compressing it between a ring and a roller. The roller rotates between dam rings fitted on the sides of the grinding ring, ensuring uniform compaction and grinding. The mill rotates at a subcritical speed, and scooping devices at both ends of the ring ensure effective internal conveying of the material being ground. The material leaves the scooping devices at various points, which ensures good distribution of the material in the airstream between the air inlets and outlets. The process air enters through two inlets at either end of the mill and leaves through an outlet at either end of the mill. The air passes the falling material and carries the finer particles to Sepax® separator, in which the final classification of the product takes place. The oversize particles are returned from Sepax® to Cemex® for further grinding. Due to this unique combination of internal grit separation and air-swept material conveying to Sepax®, no external mechanical conveyor is needed, which makes the installation very compact and simple. The airflow rate through the mill is relatively low, the only lower limitation being the need for sufficient internal grit separation and conveying of the preseparated material to the final classification in Sepax® separator [12].

**Figure 10.** F.L.Smidth Cemex® mill grinding [12].

**Figure 11.** F.L.Smidth Cemex® mill [12].

Main purposes in designing of the ring roller mill (Cemex®) can be summarized as follows:


Grinding tests by the F.L.Smidth company have shown that Cemex® produces cement which meets the requirements of the standard specifications while enabling substantial savings in grinding energy consumption compared to the traditional ball mill systems. Due to the more energy-efficient grinding process, Cemex® ground cement will usually have a steeper particlesize distribution curve than corresponding ball mill cements. Consequently, when ground to the same specific surface (Blaine), Cemex® cement will have lower residues on a 32 or 45 μm sieve and tend to have a faster strength development. Grinding of cement to a lower Blaine value will reduce the specific power consumption [12]. A comparison of typical specific energy consumption of Cemex® mill with conventional multi-compartment ball mill grinding and HPGR pre-grinding closed-circuit operations is given in **Table 1**.


**Table 1.** Comparison of typical specific energy consumption of Cemex® mill with conventional multi-compartment ball mill grinding and HPGR pre-grinding closed-circuit operations.

Some of the advantages of Cemex® mill can be summarized as follows:


makes the installation very compact and simple. The airflow rate through the mill is relatively low, the only lower limitation being the need for sufficient internal grit separation and conveying of the preseparated material to the final classification in Sepax® separator [12].

Main purposes in designing of the ring roller mill (Cemex®) can be summarized as follows:

**•** To reduce the energy consumption of the mill fan by reducing the air consumption in the

Grinding tests by the F.L.Smidth company have shown that Cemex® produces cement which meets the requirements of the standard specifications while enabling substantial savings in

**•** To reduce the wear on the mill elements by applying pressures on the grinding bed

**•** Simple and compact design to reduce the external mill load recirculation

**•** Simple and easy control of product quality and mill operation

**Figure 10.** F.L.Smidth Cemex® mill grinding [12].

124 High Performance Concrete Technology and Applications

**Figure 11.** F.L.Smidth Cemex® mill [12].

grinding process

**•** Simple mechanical design

**•** Simple and easy change of product type

**•** To reduce the specific energy consumption of grinding


## *2.5.1. Cement quality*

As it was stated in the literature, grinding tests have shown that Cemex® produces cement which meets the requirements of standard specifications while enabling substantial savings in grinding energy consumption compared to the traditional ball mill systems. Due to the more energy-efficient grinding process, Cemex® ground cement will usually have a steeper particlesize distribution curve than corresponding ball mill cements. Consequently, when ground to the same specific surface (Blaine), Cemex® cement will have lower residues on a 32 or 45 μm sieve and tend to have a faster strength development. When grinding to a 28-day-strength target, Cemex® cement can be ground to a lower Blaine value, which further reduces specific power consumption [12].

### **2.6. Horomill®**

Horomill® is a ring roller mill which is a joint development by the French plant manufacturer FCB Ciment and the Italian cement producer Buzzi Unicem Group [13]. Horomill® can be used in grinding of:


#### *2.6.1. Horomill® design and operational principle*

The Horomill® (horizontal roller mill) consists of a horizontal shell equipped with a grinding track in which a roller exerts grinding force. The shell rotates faster than the critical speed which leads to centrifuging of the material. The main feature is the roller inside the shell which is rotated by the material freely on its shaft without a drive. Operating principle is schemati‐ cally shown in **Figure 12**. Material is fed to the mill by gravity. There are scrapers located in the upper part of the shell. Scrapers cover the entire length of the mill and scrape off the material which falls onto the adjustable panel of the material advance system. Position of the material advance system which is sloping towards the discharge end could be changed in such a way that material could advance slower or faster, and thus it determines the number of passage of material under the roller which means the adjustment of circulating load. Grinding pressures change within a range of 500–800 bars. Concave and convex geometries of the grinding surfaces lead to angles of nip two or three times higher than in roller presses resulted in a thicker layer of ground material [14].

**Figure 12.** Operating principle of FCB Horomill®.

Horomill® mainly consists of three zones:

**•** Feeding

**•** High grinding capacity.

*2.5.1. Cement quality*

power consumption [12].

**2.6. Horomill®**

in grinding of:

**•** Minerals and coal

of ground material [14].

**•** Cement quality meets prevailing standards.

126 High Performance Concrete Technology and Applications

**•** Same or better strengths than cement from ball mill.

**•** Cement raw materials (i.e. limestone, clay, iron ore, etc.)

*2.6.1. Horomill® design and operational principle*

As it was stated in the literature, grinding tests have shown that Cemex® produces cement which meets the requirements of standard specifications while enabling substantial savings in grinding energy consumption compared to the traditional ball mill systems. Due to the more energy-efficient grinding process, Cemex® ground cement will usually have a steeper particlesize distribution curve than corresponding ball mill cements. Consequently, when ground to the same specific surface (Blaine), Cemex® cement will have lower residues on a 32 or 45 μm sieve and tend to have a faster strength development. When grinding to a 28-day-strength target, Cemex® cement can be ground to a lower Blaine value, which further reduces specific

Horomill® is a ring roller mill which is a joint development by the French plant manufacturer FCB Ciment and the Italian cement producer Buzzi Unicem Group [13]. Horomill® can be used

The Horomill® (horizontal roller mill) consists of a horizontal shell equipped with a grinding track in which a roller exerts grinding force. The shell rotates faster than the critical speed which leads to centrifuging of the material. The main feature is the roller inside the shell which is rotated by the material freely on its shaft without a drive. Operating principle is schemati‐ cally shown in **Figure 12**. Material is fed to the mill by gravity. There are scrapers located in the upper part of the shell. Scrapers cover the entire length of the mill and scrape off the material which falls onto the adjustable panel of the material advance system. Position of the material advance system which is sloping towards the discharge end could be changed in such a way that material could advance slower or faster, and thus it determines the number of passage of material under the roller which means the adjustment of circulating load. Grinding pressures change within a range of 500–800 bars. Concave and convex geometries of the grinding surfaces lead to angles of nip two or three times higher than in roller presses resulted in a thicker layer

**•** Cement clinker and cement additive materials (i.e. limestone, slag, pozzolan, etc.)


In the grinding zone, the cylindrical roller transfers the grinding power onto the material. Material bed in the mill is generated by the centrifugal effect.

As compared to hybrid systems, Horomilling resulted in lower energy consumptions with energy savings of 30–50 % for the same product quality. Noise generated is lower than conventional ball mill. They are smaller and compact units. Frictional forces in the Horomill grinding are kept at its minimum, and hence wear is due to the lack of differential speed between the material and the grinding ring. Horomill® is designed for closed-circuit finish grinding when compared with an HPGR. Bed thickness is two or three times the roll press (HPGR) [15].

It also has the flexibility of a vertical roller mill in grinding of different materials. A larger angle of nip draws the material bed into the grinding gap and reduces wear as compared to vertical roller mills. The recirculation of material within a vertical roller mill is very high. The recycle ratios are 15 or more, but it is practically impossible to measure the recycle ratios in a mill operating on the airflow principle. Material bed passes many times through the stressing gap, and it is possible to adjust the number of stressing during operation in a Horomill®. Also an internal bypass can be implemented if some of the ground material is returned from the mill outlet to the inlet. The external recycle ratio of a Horomill® connected in a closed circuit lies between four and eight and is therefore lower than with a roller press (HPGR) and vertical roller mill [14]. A comparison of the angles of nip of material is given in **Figure 13** [15]. A photograph of an industrial scale Horomill® [13] is shown in **Figure 14**.

**Figure 13.** Comparison of the angles of nip [15].

**Figure 14.** An industrial scale Horomill® [13].

## *2.6.2. Typical Horomill® grinding application*

Typical industrial scale Horomill® grinding and classification closed circuit are given in **Figure 15**. The circuit includes an elevator, a conveyor to the TSV™ classifier, a finishedproduct recovery filter at the TSV™ outlet and an exhauster. The rejects from the TSV™ classifier are returned by gravity to the mill inlet. The main features of the plant are as follows [15]:


**Figure 15.** Flowsheet of Trino's Horomill® plant [15].

**Figure 13.** Comparison of the angles of nip [15].

128 High Performance Concrete Technology and Applications

**Figure 14.** An industrial scale Horomill® [13].

**•** Horomill® diameter: 2,200 mm

**•** Nominal-circulating load: 140 t/h

**•** TSV™ classifier for classification

[15]:

*2.6.2. Typical Horomill® grinding application*

**•** Horomill®-installed power: 600 kW at variable speed

**•** Circuit nominal rate in CP42.5R cement production: 25 t/h at 3,200 Blaine

Typical industrial scale Horomill® grinding and classification closed circuit are given in **Figure 15**. The circuit includes an elevator, a conveyor to the TSV™ classifier, a finishedproduct recovery filter at the TSV™ outlet and an exhauster. The rejects from the TSV™ classifier are returned by gravity to the mill inlet. The main features of the plant are as follows

#### *2.6.3. Typical Horomill® grinding and classification circuit (case study)*

An industrial sampling survey was carried out during CPP-30R (pozzolanic portland cement) production around the Horomill® grinding and classification circuit given in **Figure 16**. Sampling points of the circuit are shown in a simplified flowsheet (**Figure 16**). Horomill® was closed circuited with a TSV™-type dynamic separator in the circuit.

**Figure 16.** Simplified flowsheet of a Horomill® grinding and classification circuit.

#### *2.6.4. Performance evaluation methodology*

Prior to sampling surveys, steady-state conditions were verified by examining the variations in the values of variables in the control room. When steady-state condition was achieved in the circuit, sampling was started, and sufficient amount of samples were collected from each point as shown in **Figure 16**. Due to the physical limitations, dried pozzolan stream was not sampled. Samples collected after stopping the belt conveyors by stripping the material from a length between 3 and 5 m is shown in **Table 2**. The operation during sampling was closed to steady-state conditions. Important variables of the operation were recorded in every 5 min in the control room. Average values of the control room data were used in the mass balance calculations. Mass balance calculations were carried out using JKSimMet computer program. Design parameters of the Horomill are presented in **Table 3**.

#### *2.6.4.1. Laboratory studies*

A combination of sieving and laser-sizing techniques was used for the determination of the whole particle-size distributions for each sample. SYMPATEC® dry laser sizer was used to determine the particle-size distribution of subsieve sample of 149 μm for each sample. Size distribution of +149 μm material was determined by dry sieving using a Ro-Tap. The entire size distribution for each sample was calculated using the sieving results obtained from the top size (50.8 mm) down to 149 μm and laser results obtained for the subsieve sample of −149 μm.


**Table 2.** Typical sample amounts taken after stopping the belt conveyors during survey.


**Table 3.** Design parameters of the Horomill®.

#### *2.6.4.2. Mass balance calculations*

*2.6.4. Performance evaluation methodology*

130 High Performance Concrete Technology and Applications

*2.6.4.1. Laboratory studies*

**Table 3.** Design parameters of the Horomill®.

−149 μm.

Design parameters of the Horomill are presented in **Table 3**.

Prior to sampling surveys, steady-state conditions were verified by examining the variations in the values of variables in the control room. When steady-state condition was achieved in the circuit, sampling was started, and sufficient amount of samples were collected from each point as shown in **Figure 16**. Due to the physical limitations, dried pozzolan stream was not sampled. Samples collected after stopping the belt conveyors by stripping the material from a length between 3 and 5 m is shown in **Table 2**. The operation during sampling was closed to steady-state conditions. Important variables of the operation were recorded in every 5 min in the control room. Average values of the control room data were used in the mass balance calculations. Mass balance calculations were carried out using JKSimMet computer program.

A combination of sieving and laser-sizing techniques was used for the determination of the whole particle-size distributions for each sample. SYMPATEC® dry laser sizer was used to determine the particle-size distribution of subsieve sample of 149 μm for each sample. Size distribution of +149 μm material was determined by dry sieving using a Ro-Tap. The entire size distribution for each sample was calculated using the sieving results obtained from the top size (50.8 mm) down to 149 μm and laser results obtained for the subsieve sample of

**Sampling points Swept length (m)** Pozzolan feed 5.0 m Clinker + gypsum feed 3.0 m

**Horomill#3® Value** Inside diameter (m) 3.64 Roller diameter (m) 1.82 Roller/track width (m) 1.365 Nominal pressure (at cylinder) (bar) 220 Type of motor Slip ring Installed motor power (kW) 2500 Mill shell speed (rpm) 35.9

**Table 2.** Typical sample amounts taken after stopping the belt conveyors during survey.

Some errors are inevitable in any sampling operation. These errors result from dynamic nature of the system, physical conditions at particular point, random errors, measurement errors and human errors. Mass balancing involves statistical adjustment of the raw data to obtain the best fit estimates of flow rates. In this context, by using the particle-size distributions and the control room data, an extensive mass-balancing study was performed around Horomill®#3 circuit. Tonnage flow rates (t/h) and particle sizes of the streams are calculated by JKSimMet mass balance software. The success of the mass balance was checked by plotting the experimental and calculated (mass-balanced) particle-size distributions as shown in **Figure 17**. These results plotted in a 45° line indicate the quality of both sampling operation and laboratory studies.

**Figure 17.** Comparison of mass-balanced and experimental particle-size data of each sample across the grinding cir‐ cuit.

According to the result of mass balance calculations, if there had been a statistically significant difference between experimental and calculated values (scattering data), the data would have been rejected and not be used for performance evaluation studies. In this research, data obtained as a result of sampling and experimental studies were found to be in a satisfactorily good fit. Mass balance model of the circuit with the calculated tonnage flow rates (t/h) in every stream and fineness as 45 μm% residue is shown in **Figure 18**.

**Figure 18.** Calculated flow rates (t/h) and fineness after mass balancing around the circuit.

F80 and P80 particle-size values from the mass-balanced size distributions can be used to calculate the ratio of size reduction which can be given by Eq. (1):

$$S.R = \frac{F\_{80}}{P\_{80}}\tag{1}$$

where F80 is the 80 % passing size of the Horomill® feed determined as 1.06 mm and P80 is the 80 % passing size of the Horomill® discharge determined as 0.56 mm. It means that the ratio of size reduction is 1.88.

Using the F80 (13.21 mm) and P80 (0.024 mm) size values from the mass-balanced size distri‐ butions of the fresh feed and the TSV® fine, the ratio of the overall size reduction was calculated as 550.42 by Eq. (1):

$$S.R = \frac{F\_{80}}{P\_{80}} = \frac{13.2 \text{ l mm}}{0.024 \text{ mm}} = 550.42$$

Circulating factor (CF) can be defined by Eq. (2)

$$\text{C.F} = \frac{\text{Mill feed} \left(\text{t/h}\right)}{\text{Total fresh feed} \left(\text{t/h}\right)} \tag{2}$$

$$\text{C.F} = \frac{887.23}{100.66} = 8.811$$

and recycling factor (RF) can be defined by Eq. (3)

$$\text{R.F} = \frac{\text{TSV reject } \text{(t\% h)}}{\text{TSV time (t\%)}} \tag{3}$$

$$\text{R.F} = \frac{799.08}{100.66} = 7.94$$

Circulating and recycling load percentages are determined as 881 and 794 %, respectively.

#### *2.6.4.3. Specific energy consumption (Ecs) calculation*

F80 and P80 particle-size values from the mass-balanced size distributions can be used to

80 80

where F80 is the 80 % passing size of the Horomill® feed determined as 1.06 mm and P80 is the 80 % passing size of the Horomill® discharge determined as 0.56 mm. It means that the ratio

Using the F80 (13.21 mm) and P80 (0.024 mm) size values from the mass-balanced size distri‐ butions of the fresh feed and the TSV® fine, the ratio of the overall size reduction was calculated

> 13.21 mm . 550.42 0.024 mm

> > ( )

( ) ( )

( )

= (2)

TSV fine t/h <sup>=</sup> (3)

*<sup>P</sup>* == =

Total fresh feed t/h

887.23 C.F 8.81 100.66 = =

799.08 R.F 7.94 100.66 = =

Circulating and recycling load percentages are determined as 881 and 794 %, respectively.

TSV reject t/h R.F

80 80

Mill feed t/h C.F

*<sup>F</sup> S R*

Circulating factor (CF) can be defined by Eq. (2)

and recycling factor (RF) can be defined by Eq. (3)

*<sup>P</sup>* <sup>=</sup> (1)

. *<sup>F</sup> S R*

calculate the ratio of size reduction which can be given by Eq. (1):

132 High Performance Concrete Technology and Applications

of size reduction is 1.88.

as 550.42 by Eq. (1):

Horomill® motor power (2,126 kW) is the average operating mill motor power reading from the control room during the sampling survey and used in the calculation. Total fresh feed tonnage is the dry tonnage amount used in the mass balance calculations represented by the TSV fine stream tonnage flow rate which is 100.66 t/h. Thus, the specific energy consumption (Ecs) can be calculated by Eq. (4):

$$\mathbf{E\_{cs}} = \frac{\text{Mill power (kW)}}{\text{Total fresh feed} \left(\text{t/h}\right)} \tag{4}$$

$$\mathbf{E\_{cs}} = \frac{2126}{100.66} = 21.12 \quad kWh/m$$

When the final cement tonnage is considered which is 105.53 t/h, specific energy consumption (Ecs) is calculated by Eq. (5):

$$\mathbf{E}\_{\rm cs} = \frac{\text{Mill power } \left( \mathbf{kW} \right)}{\text{Final element } \left( \mathbf{t} \,\mathrm{h} \right)} \tag{5}$$

$$E\_{\rm cs} = \frac{2126}{105.53} = 20.15 \text{ } kWh/m$$

#### *2.6.4.4. Tromp curve of the TSV® separator*

The performance of any classifier, in terms of size separation, is represented by an efficiency (TROMP) curve. An example for a classifier is shown in **Figure 19**. It describes the proportion of a given size of solids which reports to the coarse product. Mass-balanced particle-size distributions and tonnage flow rates around the separator were used to evaluate the perform‐ ance of the separator. Percentage of any fraction in the feed pass to the coarse product (%) is defined as partition coefficient and expressed by Eq. (6):

$$P = \frac{U u\_i}{F f\_i} \tag{6}$$

where U is the separator coarse tonnage (t/h), F the separator feed tonnage (t/h), ui is the % of size fraction (i) in separator coarse and fi is the % of size fraction (i) in separator feed.

Actual TROMP curve established for TSV® is presented in **Figure 19**.

**Figure 19.** Actual TSV® TROMP curve.

The d50 size corresponds to 50 % of the feed passing to the coarse stream. It is therefore the size which has equal probability of passing to either coarse or fine streams. When this size is decreased, the fineness of the product increases. The operational parameters that affect the cut size are rotor speed and separator air velocity. Cut size for the TSV® was determined as 23.33 μm. The percentage of the lowest point on the tromp curve is referred as the bypass. It is the part of the feed which directly passes to the coarse stream (separator reject) without being classified. Bypass value is a function of the separator ventilation and separator feed tonnage. The bypass value of TSV® was 23.29 % which indicated a consistent performance for this separator. Fish-hook effect (β) is the portion of fines returning back into separator reject stream. When there is incomplete feed dispersion at the separator entry, or even within the classifica‐ tion zone, aggregates of fine particles may be classified as coarse particles and thus report to the coarse stream. Fish-hook amount of TSV® was 1.58 % which also indicated how effectively it is operating:

Fish hook 24.87 23.29 1.58% -= - =

The sharpness of separation was defined as d25/d75

where d75 is the particle size whose 75 % is reported to the separator reject and d25 is the particle size whose 25 % is reported to the separator reject.

For the TSV®, parameter values determined from the TROMP curve are d75 as 32.36 μm and d25 as 10.50 μm.

The range of this parameter *k* (acuity) depends on the type of separator. This parameter can be calculated by Equation 7 as 0.32:

$$
\kappa = \left[\frac{d\_{25}}{d\_{\gamma s}}\right] \tag{7}
$$

$$
\kappa = 0.32
$$

Usually, for TSV®-type separator, it is between 0.55 and 0.7. When the normal range for sharpness (*k*) parameter is considered, it is found to be not in the normal range [16]. When the normal range for sharpness (*k*) parameter is considered, it was found to be not in the normal range. The imperfection of separation is defined by Equation 8, and I was calculated as 0.47:

k

$$I = \left[\frac{d\_{\gamma s} - d\_{2s}}{2d\_{s0}}\right] \tag{8}$$

*I* = 0.47

The value of I indicated that separation performance is sufficiently good.

**Figure 19.** Actual TSV® TROMP curve.

134 High Performance Concrete Technology and Applications

it is operating:

d25 as 10.50 μm.

The sharpness of separation was defined as d25/d75

size whose 25 % is reported to the separator reject.

The d50 size corresponds to 50 % of the feed passing to the coarse stream. It is therefore the size which has equal probability of passing to either coarse or fine streams. When this size is decreased, the fineness of the product increases. The operational parameters that affect the cut size are rotor speed and separator air velocity. Cut size for the TSV® was determined as 23.33 μm. The percentage of the lowest point on the tromp curve is referred as the bypass. It is the part of the feed which directly passes to the coarse stream (separator reject) without being classified. Bypass value is a function of the separator ventilation and separator feed tonnage. The bypass value of TSV® was 23.29 % which indicated a consistent performance for this separator. Fish-hook effect (β) is the portion of fines returning back into separator reject stream. When there is incomplete feed dispersion at the separator entry, or even within the classifica‐ tion zone, aggregates of fine particles may be classified as coarse particles and thus report to the coarse stream. Fish-hook amount of TSV® was 1.58 % which also indicated how effectively

Fish hook 24.87 23.29 1.58% -= - =

where d75 is the particle size whose 75 % is reported to the separator reject and d25 is the particle

For the TSV®, parameter values determined from the TROMP curve are d75 as 32.36 μm and


**Table 4.** Operational characteristics of Horomill® and Polysius® HPGR/Polysius® two-compartment ball mill and classification closed-circuit operations at the same cement production type.

*2.6.5. Operational results from an industrial scale Horomill® grinding and HPGR/two-compartment ball mill and classification closed circuit*

Typical operating conditions for the Horomill® and two-compartment ball mill grinding with HPGR pre-crushing and classification circuits are compared in **Table 4** for the same production type. As can be seen from **Table 4**, Horomill® production configuration has resulted in energy savings of 50 % as compared to HPGR/two-compartment ball milling configuration [16].

## *2.6.6. Comparison of different grinding technologies*

Typical specific energy consumption comparison between Horomill® product and HPGR hybrid system for pozzolanic cement with a 4,200 Blaine is as follows [13]:

**Figure 20.** Compressive strength on mortar [15].


It was also reported that concrete workability from a portland cement with a 3,200 Blaine which is a Horomill® product is equal or better than an equivalent ball mill product. Mortar and concrete strengths are always higher as shown in **Figures 20** and **21**. The closed-circuit recirculation factor is noted as about six in Horomill® grinding [17]. A comparison between the grinding systems and conventional ball mills applied in cement grinding circuits is given in **Table 5**. Grinding efficiencies of different systems in grinding of cement to a fineness according to a Blaine of 3,000 cm2 /g were compared in **Table 6**.

**Figure 21.** Compressive strength on concrete [15].

*2.6.5. Operational results from an industrial scale Horomill® grinding and HPGR/two-compartment*

Typical operating conditions for the Horomill® and two-compartment ball mill grinding with HPGR pre-crushing and classification circuits are compared in **Table 4** for the same production type. As can be seen from **Table 4**, Horomill® production configuration has resulted in energy savings of 50 % as compared to HPGR/two-compartment ball milling configuration [16].

Typical specific energy consumption comparison between Horomill® product and HPGR

It was also reported that concrete workability from a portland cement with a 3,200 Blaine which is a Horomill® product is equal or better than an equivalent ball mill product. Mortar and concrete strengths are always higher as shown in **Figures 20** and **21**. The closed-circuit recirculation factor is noted as about six in Horomill® grinding [17]. A comparison between the grinding systems and conventional ball mills applied in cement grinding circuits is given

hybrid system for pozzolanic cement with a 4,200 Blaine is as follows [13]:

*ball mill and classification closed circuit*

136 High Performance Concrete Technology and Applications

**Figure 20.** Compressive strength on mortar [15].

**•** Pozzolanic cement 4,200 Blaine: Horomill®, 23.1 kWh/t Hybrid system, 32 kWh/t

**•** Portland cement 3,100 Blaine: Horomill®, 28.3 kWh/t Hybrid system, 39 kWh/t

*2.6.6. Comparison of different grinding technologies*


**Table 5.** Comparison between the grinding systems and conventional ball mills applied in cement grinding circuits.


**Table 6.** Comparison of typical mill-grinding efficiencies.

The efficiency of a two-compartment ball mill is defined to be 1.0. This efficiency reflects the power consumption of the mill only and does not include any auxiliary equipment like conveyors and dust collectors nor the separator.

## **3. Conclusions**

Comparisons between different energy-efficient grinding technologies and applications were presented for production of cement with energy savings. Industrial-scale data related to Horomill® and Polysius® HPGR/two-compartment ball mill circuit provided insights in‐ to the operational and size-reduction characteristics of Horomill® and HPGR/two-compart‐ ment ball mill-grinding process with indications that Horomill® application could produce the same type of pozzolanic portland cement at lower grinding energy requirement. The specific energy consumption figures indicated approximately 50 % grinding energy savings in Horomill® process.

## **Author details**

Ömürden Genç

Address all correspondence to: ogenc@mu.edu.tr

Muğla Sıtkı Koçman University, Department of Mining Engineering, Muğla, Turkey

## **References**


The efficiency of a two-compartment ball mill is defined to be 1.0. This efficiency reflects the power consumption of the mill only and does not include any auxiliary equipment like

Comparisons between different energy-efficient grinding technologies and applications were presented for production of cement with energy savings. Industrial-scale data related to Horomill® and Polysius® HPGR/two-compartment ball mill circuit provided insights in‐ to the operational and size-reduction characteristics of Horomill® and HPGR/two-compart‐ ment ball mill-grinding process with indications that Horomill® application could produce the same type of pozzolanic portland cement at lower grinding energy requirement. The specific energy consumption figures indicated approximately 50 % grinding energy savings

Muğla Sıtkı Koçman University, Department of Mining Engineering, Muğla, Turkey

[1] Fujimoto S. Reducing specific power usage in cement plants, *World Cement*, 1993; 7:

[2] Jankovic A, Valery W, Davis E. Cement grinding optimization. *Minerals Engineering*.

[3] Lynch AJ, *Comminution Handbook*, The Australasian Institute of Mining and Metallurgy,

[4] Barmac Thousand Series Duopactor™ Rock-on-Rock VSI Crushers, Svedala Crushing

[5] Ghosh SN, *Cement and Concrete Science and Technology*, Volume I, Part I, ABI Books

and Screening Brochure, Svedala New Zealand Ltd, New Zealand: 1995.

conveyors and dust collectors nor the separator.

138 High Performance Concrete Technology and Applications

Address all correspondence to: ogenc@mu.edu.tr

**3. Conclusions**

in Horomill® process.

**Author details**

Ömürden Genç

**References**

25-35.

2004; 17: 1075-1081.

Spectrum 21; Australia: 2015.

Private Ltd, First Edition, New Delhi: 1991.


## **Concretes with Photocatalytic Activity**

Magdalena Janus and Kamila Zając

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64779

#### **Abstract**

This chapter is a short review about the modified concretes with photocatalytic activity. In the beginning, the photocatalysis process is explained; the authors are focused on the mechanism of organic contamination and nitrogen oxide decomposition. Next the three main methods for concretes modification are presented: the first group is when the concrete is covered by thin layer of TiO2 materials, e.g., paints or TiO2 suspensions. The second group is the concretes with thick layer of photoactive concrete on the top. The third group constitutes concretes modified in mass with TiO2. The two main methods for photocatalytic activity of the modified concrete determination were shown: an air purification by a nitrogen oxide decomposition and the self-cleaning properties by dyes decomposition. Also in this chapter the mechanical properties of the modified concrete are presented. In the end, the examples of the buildings made of photocatalytic concretes are shown.

**Keywords:** modified concretes, photocatalysis, air purification, mechanical properties

## **1. Introduction**

In this chapter the information about photoactive concretes are shown. For many years the researchers were interested in preparation of modified building materials with the special properties. It was found that using titanium dioxide as additive to such materials gives them self-cleaning, antibacterial and antifungal properties, moreover it was found that in some cases the mechanical properties are also improved. This chapter contain five parts. First, the photocatalysis process is described. The mechanism of hydroxyl radical production during titanium dioxide irradiation is presented. In the literature, mainly the used method could be divided into three groups: the first group is when the concrete is covered by thin layer of TiO2 materials, e.g., paints or TiO2 suspensions. The second group is the concretes with thick layer

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

of photoactive concrete on the top. The third group is the concretes with different weight percent of TiO2 in the mass (substituted cement), then the results of photoactive tests of such materials are presented, mainly focused on the air cleaning from nitrogen oxides. After‐ wards, the mechanical properties of photoactive concretes are presented. In this chapter, TiO2 effect on hydration process of cement, TiO2 effect on the compressive strength, TiO2 effect on carbonation of concretes, the fire resistance of photocatalytic concrete and the influence of TiO2 on abrasion resistance and shrinkage are presented. In the end of the chapter, the examples of building and objects built with using photoactive concretes are presented.

## **2. Description of photocatalysis process**

The beginning of photocatalysis is considered a year of Fujishima and Honda article publication in 1972 [1]. The article presents the results of electrochemical photolysis of water at a semiconductor electrode. Although various definitions and interpretations of the term "photocatalysis" have been proposed, "photocatalysis" or "photocatalytic reaction" is defined by Ohtani [2] as a chemical reaction induced by photoabsorption of a solid material, or "photocatalyst", which remains chemically unchanged during and after the reaction. The principle of photocatalysis is often explained with an illustration like **Figure 1**. An electron (*e*− ) from the valence band (VB) is excited by photoirradiation to a vacant conduction band (CB). After excitation, a positive hole (*h*<sup>+</sup> ) appears in the VB, these electrons and positive holes take part in the reduction and oxidation reactions of compounds adsorbed on the surface of a photocatalyst. The common photocatalyst is TiO2. Then the possible ways for concretes modification are presented.

**Figure 1.** A schematic representation of the electronic structure of semiconducting materials.

The mechanism of the photocatalysis is summarized in Eqs. (1)–(12). First the production of electrons (*e*<sup>−</sup> ) and holes (*h*<sup>+</sup> ) in conduction band and valence band occurs (Eq. (1)). The photogenerated holes that escape direct recombination (Eqs. (3) and (4)) reach the surface of TiO2 and react with surface adsorbed hydroxyl groups or water to form trapped holes (Eq. (2)). The trapped hole (≡TiO•) is usually described as a surface-bound or adsorbed OH• radical OH*ads* • . According to (Eq. (7)), OH• generates at the surface of semiconductor and leaves the surface to bulk solution to form free OH• (OH *free* • ). If electron donors (*Redorg*) are present at the TiO2 surface, electron transfer may occur according to Eqs. ((5), (6) and (8)). In aerated systems, oxidative species, such as O2 •− and H2O2 generate from the reduction site [3]:

charge-carrier generation:

$$\text{TiO}\_2 + h\text{v} \to h^\* + e^- \tag{1}$$

hole trapping:

of photoactive concrete on the top. The third group is the concretes with different weight percent of TiO2 in the mass (substituted cement), then the results of photoactive tests of such materials are presented, mainly focused on the air cleaning from nitrogen oxides. After‐ wards, the mechanical properties of photoactive concretes are presented. In this chapter, TiO2 effect on hydration process of cement, TiO2 effect on the compressive strength, TiO2 effect on carbonation of concretes, the fire resistance of photocatalytic concrete and the influence of TiO2 on abrasion resistance and shrinkage are presented. In the end of the chapter, the examples of

The beginning of photocatalysis is considered a year of Fujishima and Honda article publication in 1972 [1]. The article presents the results of electrochemical photolysis of water at a semiconductor electrode. Although various definitions and interpretations of the term "photocatalysis" have been proposed, "photocatalysis" or "photocatalytic reaction" is defined by Ohtani [2] as a chemical reaction induced by photoabsorption of a solid material, or "photocatalyst", which remains chemically unchanged during and after the reaction. The principle of photocatalysis is often explained with an illustration like **Figure 1**. An electron

) from the valence band (VB) is excited by photoirradiation to a vacant conduction band

take part in the reduction and oxidation reactions of compounds adsorbed on the surface of a photocatalyst. The common photocatalyst is TiO2. Then the possible ways for concretes

) appears in the VB, these electrons and positive holes

building and objects built with using photoactive concretes are presented.

**Figure 1.** A schematic representation of the electronic structure of semiconducting materials.

**2. Description of photocatalysis process**

142 High Performance Concrete Technology and Applications

(CB). After excitation, a positive hole (*h*<sup>+</sup>

modification are presented.

(*e*−

$$\text{H}^+ + \equiv \text{Ti}^{\text{IV}}\text{OH} \rightarrow \left\{ \rightleftharpoons \text{Ti}^{\text{IV}}\text{OH}^\* \right\}^+ \rightarrow \equiv \text{Ti}^{\text{IV}}\text{O}^\* + \text{H}^+ \tag{2}$$

charge-carrier recombination:

$$h^\* + e^- \to heat \tag{3}$$

$$\text{e}^- + \equiv \text{Ti}^{\text{IV}} \text{O}^\* + \text{H}^\* \rightarrow \equiv \text{Ti}^{\text{IV}} \text{OH} \tag{4}$$

charge transfer at the oxidation site:

$$h^{+} + Red\_{org} \to \text{Ox}\_{org} \tag{5}$$

$$\mathbf{H} \equiv \mathbf{T} \mathbf{i}^{\rm{IV}} \mathbf{O}^\* + \mathbf{R} \mathbf{e} d\_{\rm{org}} \to \mathbf{Ox}^{\cdot}\_{\rm{org}} \tag{6}$$

$$\text{H}^\* + \text{H}\_2\text{O} \rightarrow \text{H}\_2\text{O}^{\cdot \*} \rightarrow \text{H}^\* + \text{OH}^\cdot \tag{7}$$

$$\text{OH}^{\cdot} + \text{Red}\_{\text{org}} \rightarrow \text{Ox}^{\cdot}\_{\text{org}} \tag{8}$$

charge transfer at the reduction site:

$$\text{e}^- + \text{O}\_{2(ads)} \to \text{O}\_2^{\text{-}} \tag{9}$$

$$\rm{O}\_{2}^{\cdot -} + e^{-} \left( + 2H^{\ast} \right) \rightarrow \rm{H}\_{2}\rm{O}\_{2} \tag{10}$$

$$\rm{O}\_{2}^{\ast-} + \rm{H}\_{2}\rm{O}\_{2} \rightarrow \rm{OH}^{\ast} + \rm{OH}^{\cdot} + \rm{O}\_{2} \tag{11}$$

$$\text{CH}\_2\text{O}\_2 + h\nu \rightarrow 2\text{OH}^\* \tag{12}$$

It is necessary to say that the photocatalysis process depends on: type and concentration of photocatalysts, type of eliminated contaminations, energy and intensity of used light. To determine the best possible conditions to perform photocatalytic process, all of these aspects are essential to consider. In the case of modified concrete, to have photocatalytic activity, most of tests are focused on air purification especially nitrogen oxide reduction and self-cleaning properties are tested during dyes decomposition.

For laboratory test of modified concrete activity, the nitrogen oxides are used. Below equations showed the mechanism of photocatalytic NOx oxidation on active concrete under UV illumi‐ nation [4]. First the charge-carrier generation occurred (Eq. (1)).

The adsorption of the reactants onto the photocatalyst surface takes place:

$$\text{TiO}\_2 + \text{H}\_2\text{O} \leftrightarrow \text{TiO}\_2 - \text{H}\_2\text{O} \tag{13}$$

$$\rm{TiO}\_2 + \rm{O}\_2 \leftrightarrow \rm{TiO}\_2 - \rm{O}\_2 \tag{14}$$

$$\text{TiO}\_2 + \text{NO} \leftrightarrow \text{TiO}\_2-\text{NO} \tag{15}$$

$$\rm{TiO}\_2 + \rm{NO}\_2 \leftrightarrow \rm{TiO}\_2 - \rm{NO}\_2 \tag{16}$$

OH• radicals produced according Eqs. (7) and (12) take part in the nitrogen oxide oxidation, as follows:

$$\text{NO} + \text{OH}^\cdot \rightarrow \text{HNO}\_2^\cdot \tag{17}$$

$$\rm HNO\_2 + OH^\cdot \rightarrow NO\_2 + H\_2O \tag{18}$$

$$\rm{NO}\_2 + \rm{OH}^\* \rightarrow \rm{NO}\_3^- + \rm{H}^- \tag{19}$$

## **3. The methods of modified concrete preparation**

( )

• • O H O OH OH O 2 22 <sup>2</sup>

properties are tested during dyes decomposition.

144 High Performance Concrete Technology and Applications

as follows:

nation [4]. First the charge-carrier generation occurred (Eq. (1)).

The adsorption of the reactants onto the photocatalyst surface takes place:

It is necessary to say that the photocatalysis process depends on: type and concentration of photocatalysts, type of eliminated contaminations, energy and intensity of used light. To determine the best possible conditions to perform photocatalytic process, all of these aspects are essential to consider. In the case of modified concrete, to have photocatalytic activity, most of tests are focused on air purification especially nitrogen oxide reduction and self-cleaning

For laboratory test of modified concrete activity, the nitrogen oxides are used. Below equations showed the mechanism of photocatalytic NOx oxidation on active concrete under UV illumi‐

OH• radicals produced according Eqs. (7) and (12) take part in the nitrogen oxide oxidation,

• O O <sup>2</sup> *ads* <sup>2</sup> *e*- - + ® (9)

( ) • O 2H H O <sup>2</sup> 2 2 *e* -- + ++ ® (10)


• H O 2OH 2 2 + ® *hv* (12)

TiO H O TiO H O 22 22 +« - (13)

TiO O TiO O 22 22 +« - (14)

TiO NO TiO NO 2 2 +« - (15)

TiO NO TiO NO 22 22 +« - (16)

• NO OH HNO + ® <sup>2</sup>(17)

• HNO OH NO H O <sup>2</sup> +® +2 2 (18)

The methods of concrete preparation can be divided into three main groups (**Figure 2**). The first group is when the concrete is covered by thin layer of TiO2 materials, e.g., paints or TiO2 suspensions (**Figure 2a**). The second group is the concretes with thick layer of photoactive concrete on the top (**Figure 2b**). The third group is the concretes with different weight percent of TiO2 in the mass (substituted cement) (**Figure 2c**).

**Figure 2.** Scheme of possible ways for concrete modification by photocatalysts.

## **3.1. The first group: the concrete is covered by thin layer of TiO2**

Chen and Chu [4] covered the surface of the concrete by a different types of the slurries. The authors tested eight application methods, five of them showed over 95% static NO reduction and over 89% static automobile exhaust (toluene, trimethylbenzene and nitrogen oxide) reduction. The slurries were brushed onto the surface of a previous concrete.

CWB—commercial water-based TiO2.

CWLS—thin slurry with low cement concentration and TiO2 uniformly mixed together.

DIPM—a transparent liquid driveway protector (siliconate, water-based concrete sealer) and TiO2.

TIW—water and TiO2 uniformly mixed.

PUR—PURETI commercial water based TiO2 applied to the surface with a special electrostatic sprayer.

Not only water is used for the slurry preparation, organic solvents are also used. Smits et al. [5] used ethanol to prepare dispersions. Photocatalyst was dispersed in ethanol (50 mg/ml) by sonication. The one layer of coating was performed by applying ethanol slurry on top of the mortar sample with the pipette. The coatings contain an equal amount of TiO2 (24 ± 2 mg) on samples' surface, equivalent to 267 μg/cm2 . Building materials coated with TiO2 show selfcleaning properties as all coated samples are able to remove soot.

#### **3.2. The second group: the concretes with thick layer of photoactive concrete on the top**

These concretes consist two parts, lower layer is unmodified concrete, the top part of concrete consist cement with TiO2. The amount of TiO2 used in the top layer is different for example Folli et al. [6] used about 40 kg/m3 of concrete. Under ideal weather and irradiation conditions, i.e. summer months, the monthly average NO concentration in proximity of the photocatalytic area was around 22% lower than the references are [6]. Fiore et al. [7] tested the concrete covered with photocatalytic cement mortar, with thickness of 3 and 5 mm. The results of the experimental tests have shown that the concrete carbonation depth can be significantly reduced by adopting photocatalytic surface layers. The results have also indicated that the application of titanium dioxide, modifies cementinous materials on the external surface of reinforced concrete elements, improves the corrosion performance of reinforcing bars in presence of carbonation of concrete.

#### **3.3. The third group: the concretes with different weight percent of TiO2 in the mass (TiO2 substituted cement)**

In these examples, the titanium dioxide modified concrete in mass substitute cement. Usually the commercial titanium dioxide is used, for example: P25 (Evonic) or PC-105 (Millennium). The amounts of used photocatalysts: 0.5, 1, 2.5, 5 and 10 wt.% [8–10].

### **4. The results of photoactive tests**

The activity of photoactive concretes is usually tested during an air purification especially NOX removal. The self-cleaning properties are tested during dye removal from modified concretes surface. A lot of air purification tests are conducted according to method ISO 22197-1. The ISO standard employs an inert flat-bed photoreactor system designed to hold 5 × 10 cm2 sample under illumination with UV-A light (irradiance = 1mW/cm2 ). Humidified (RH = 50% at 25°C) air and dry NO at concentration 1.0 ppm with flow 3.0 dm3 /min via mass flow rate controllers (**Figure 3**). The outlet gas stream from the reactor is sampled through a valve attached to a suitable NOx detection system usually based on chemiluminescence [11].

Among the research concerning the transformation of the photocatalytic concrete materials to a larger scale the majority of the experiments refer to NOx reduction, because the compounds are currently one of the main causes of a poor quality in large cities [7–9]. Blocks of the photocatalytic concrete are analysed towards the different features: the thickness of the photocatalytic layer, types of TiO2, content of incorporated photocatalyst. It was also observed that in a real condition the blocks with more porous surfaces showed better results for the rate of NOx degradation [12]. However, it is worth pointing out that apart from NOx the volatile organic compounds (VOCs) are the target pollutants to remove using new concrete elements. Shen et al. [13] have made the attempts towards VOCs degradation during application of photocatalytic pavement.. Even though VOC displayed a significant variability in the removal efficiency, the reduction achieved level of nearly 90%. Other efforts were taken in a case of a durability of titanium dioxide photocatalyst coating for the concrete pavement. Hassan et al. [14] determined abrasion and wear resistance properties of TiO2 coating and its effect on the coating environmental performance. The application of a special tester, which employs a scaled dynamic wheel passing back and forth over the sample, indicated on the acceptable durability and a wear resistance of the prepared photocatalytic concrete.

**Figure 3.** Photoreactor according to ISO 22197-1.

**3.2. The second group: the concretes with thick layer of photoactive concrete on the top**

Folli et al. [6] used about 40 kg/m3

146 High Performance Concrete Technology and Applications

presence of carbonation of concrete.

**4. The results of photoactive tests**

**substituted cement)**

These concretes consist two parts, lower layer is unmodified concrete, the top part of concrete consist cement with TiO2. The amount of TiO2 used in the top layer is different for example

i.e. summer months, the monthly average NO concentration in proximity of the photocatalytic area was around 22% lower than the references are [6]. Fiore et al. [7] tested the concrete covered with photocatalytic cement mortar, with thickness of 3 and 5 mm. The results of the experimental tests have shown that the concrete carbonation depth can be significantly reduced by adopting photocatalytic surface layers. The results have also indicated that the application of titanium dioxide, modifies cementinous materials on the external surface of reinforced concrete elements, improves the corrosion performance of reinforcing bars in

**3.3. The third group: the concretes with different weight percent of TiO2 in the mass (TiO2**

In these examples, the titanium dioxide modified concrete in mass substitute cement. Usually the commercial titanium dioxide is used, for example: P25 (Evonic) or PC-105 (Millennium).

The activity of photoactive concretes is usually tested during an air purification especially NOX removal. The self-cleaning properties are tested during dye removal from modified concretes surface. A lot of air purification tests are conducted according to method ISO 22197-1. The ISO standard employs an inert flat-bed photoreactor system designed to hold 5 × 10 cm2

controllers (**Figure 3**). The outlet gas stream from the reactor is sampled through a valve attached to a suitable NOx detection system usually based on chemiluminescence [11].

Among the research concerning the transformation of the photocatalytic concrete materials to a larger scale the majority of the experiments refer to NOx reduction, because the compounds are currently one of the main causes of a poor quality in large cities [7–9]. Blocks of the photocatalytic concrete are analysed towards the different features: the thickness of the photocatalytic layer, types of TiO2, content of incorporated photocatalyst. It was also observed that in a real condition the blocks with more porous surfaces showed better results for the rate of NOx degradation [12]. However, it is worth pointing out that apart from NOx the volatile organic compounds (VOCs) are the target pollutants to remove using new concrete elements. Shen et al. [13] have made the attempts towards VOCs degradation during application of photocatalytic pavement.. Even though VOC displayed a significant variability in the removal efficiency, the reduction achieved level of nearly 90%. Other efforts were taken in a case of a durability of titanium dioxide photocatalyst coating for the concrete pavement. Hassan et al.

The amounts of used photocatalysts: 0.5, 1, 2.5, 5 and 10 wt.% [8–10].

sample under illumination with UV-A light (irradiance = 1mW/cm2

at 25°C) air and dry NO at concentration 1.0 ppm with flow 3.0 dm3

of concrete. Under ideal weather and irradiation conditions,

). Humidified (RH = 50%

/min via mass flow rate

**Figure 4.** Separate parking lanes at the Leien of Antwerp with photocatalytic pavement blocks [15].

The scientific group Boonen et al. [15] performed the research in the laboratory and the pilot scale. As a result, the samples were active and the efficiency towards the reduction of NOx increased with a longer contact time, a lower relative humidity and a higher intensity of light. Then they convert the results obtained in the laboratory scale to a real application. 10,000 m2 of a photocatalytic pavement blocks were constructed on the parking lanes of a main road ace in Antwerp (**Figure 4**). The pavement demonstrated a good efficiency and durability towards NOx abatement. Repeated measurements of the concrete pavement blocks confirm the efficiency after more than 5 years of using. It was observed the reduction in the efficiency due to the deposition of the nitrate on the surface. However, the original efficiency could be regained by washing the surface.

Most of the scientific works are focused on the commercial TiO2 as additives to the concretes. Some of the researchers tried to use a modified titanium dioxide as the additive. They found that carbon and nitrogen co-modified titanium added to the cement in the amount of 5%wt increased the photocatalytic activity of the concrete more than addition of the commercial P25. The surface of a modified building material after covering by dyes and after 25 and 100 h of a visible light irradiation in **Figure 5** is presented. The higher activity of TiO2-N,C is explained by the presence of the carbon and the nitrogen. It is a generally claimed that the carbon doping improves the adsorption of the organic pollutants molecules on the catalyst surface. Moreover, the carbon doping can enhance the TiO2 conductivity, as it can facilitate the charge transfer from the bulk to the surface region of TiO2 structure, where the desired oxidation reactions take place [17].

**Figure 5.** The photographs of cement plates stained with RR 198 (azo dye) and treated of Vis irradiation. The compari‐ son of pure cement with cement containing TiO2/N,C—600 (5 wt.%) and commercial P25, towards photocatalytic re‐ sponse.

## **5. The results of mechanical properties of modified concretes**

Photocatalytic concretes as a new functional materials are also studied in detail towards their mechanical properties. For a real wide application, the evaluation of a specific mechanical performance is required. It is worth stressed that the addition of TiO2 into concrete material can influence on some properties as a heat of hydration, a workability, a setting time, a chemical shrinkage, a mechanical strength, an abrasion resistance, a fire resistance, a freeze resistance, a water absorption, etc. [18].

## **5.1. TiO2 effect on hydration process of cement**

The surface of a modified building material after covering by dyes and after 25 and 100 h of a visible light irradiation in **Figure 5** is presented. The higher activity of TiO2-N,C is explained by the presence of the carbon and the nitrogen. It is a generally claimed that the carbon doping improves the adsorption of the organic pollutants molecules on the catalyst surface. Moreover, the carbon doping can enhance the TiO2 conductivity, as it can facilitate the charge transfer from the bulk to the surface region of TiO2 structure, where the desired oxidation reactions

**Figure 5.** The photographs of cement plates stained with RR 198 (azo dye) and treated of Vis irradiation. The compari‐ son of pure cement with cement containing TiO2/N,C—600 (5 wt.%) and commercial P25, towards photocatalytic re‐

Photocatalytic concretes as a new functional materials are also studied in detail towards their mechanical properties. For a real wide application, the evaluation of a specific mechanical performance is required. It is worth stressed that the addition of TiO2 into concrete material can influence on some properties as a heat of hydration, a workability, a setting time, a chemical shrinkage, a mechanical strength, an abrasion resistance, a fire resistance, a freeze resistance,

**5. The results of mechanical properties of modified concretes**

take place [17].

148 High Performance Concrete Technology and Applications

sponse.

a water absorption, etc. [18].

Our concerns about a nature of TiO2 effect on the concrete should be started from the hydration process of the cement. It was proved that nano-TiO2 acts not only as a photocatalyst but it is also a catalyst in the cement hydration reaction. Chen et al. [19] performed the detailed analyses of the hydration process in a case of the cement pastes and the mortars blended with TiO2. The TiO2 particles acted as a potential nucleation sites for the accumulation of the hydration products. The addition of nano-TiO2 powders significantly accelerated the hydration rate and promoted the hydration degree of the cementitious materials at the early ages. Simultaneously, TiO2 was inert and stable during the cement hydration process. Meanwhile, the observed [20] acceleration of the hydration rate and changes of microstructrure (after loading of TiO2 into cement the total porosity of the cement pastes decreased and the pore size distribution was altered) affected the physical and the mechanical properties of the cement-based materials.

## **5.2. TiO2 effect on the compressive strength**

After loading of TiO2 into cement, it was observed that the compressive strength of the mortar was enhanced (at early ages). The initial and final setting time was shortened and more water was required to maintain a standard consistence due to the addition of the smaller nano-TiO2 particles. The relationship between hydration process and photocatalytic cementitious material properties was also by other scientists investigated [20–22]. However, other authors [23] the changes in mechanical properties in modified cementitious materials assigned to other phenomena. Using XRD technique they analysed the orientation index of CH crystal in different cement mortars with nano-TiO2. Test results indicated that when cement was substituted by nano-TiO2 the strength of cement mortar at early ages increased a lot and the fluidity and strength at evening ages decreased. They claimed that the main reason for the improvement of strength is the decrease and modification of orientation index for the nucle‐ ation function, not the increasing amount of hydration products. Experimental date showed the entirely different tendency between the intensity of (0 0 1) crystal plane and that of (1 0 1) crystal plane for various samples without or with various photocatalysts. Namely, it was showed that the orientation index has an obviously effect on the strength of cement mortar.

The effect of loaded TiO2 into cementitious materials might be considered from the point of view of mictrostructural changes as well. Lucas et al. [9] added a photocatalyst to mortars prepared with aerial lime, cement and gypsum binders to determine the way the microstuc‐ tural changes affect the properties of the modified materials. In case of cement based mortar the porosity distribution was different between the mortar without and with TiO2. In the initial material the porosity distribution was divided in two intervals: a set of pores of larger size which ranged between 10 and 60 μm and another group between 0.02 and 1 μm. Up to 1 wt. % titania added to the cement matrix the compressive strength increased. Simultaneously, for these samples the larger pores completely disappeared remaining solely pores between 0.02 and 1 μm and the total porosity was reduced. The cement based mortar showed a mechanical strength reduction with increasing in additive content but the reduction was relatively low. It was explained by emerging a set of nanopores combined with the disappearance of the macropores. It clarified why the mechanical strength did not decrease so significantly. For the maximum TiO2 content (5 wt.%), the nanoporosity increased notably and even with the presence of residual micropores (1.5–2.7 μm), the mechanical strength remained stable. In general, it was concluded that the presence of low size pores, particularly in the range between 1 and 0.1 μm helps to minimize the detrimental effect of the loading of nanoadditives. Beside changes occurring during setting and hardening the microstructure of concrete and mortars evolves in time during the service life of structures [24]. Diamanti et al. [10] focused on mechanical and durability aspects of TiO2-containing photocatalytic concrete by examining mutual influences between TiO2 and concrete components, and their evolution with the material aging. Materials were produced by adding the commercial form of TiO2 (P25) to concrete. In the beginning it was observed that despite the presence of nanoparticles which could play a positive filler effect, a slight decrease in a mechanical strength was observed in TiO2-containing specimens. SEM analysis showed a slight increase in the concrete porosity and to a non-even distribution of TiO2 particles that in some cases were present as clusters.

According to Zhao et al. [25] the compressive strength was reduced with increase of titania content in cementitious composites. A 12% compressive strength reduction was observed when 10 wt.% TiO2 was added to the cement matrix. It was attributed to the flocculation of nano-TiO2 particles which introduced a weak zone as flaws. In most papers is related that both, the strength and water permeability of the photocatalytic concrete are improved by adding TiO2 nanoparticles in the cement paste up to 2.0 wt.% [26], 3.0 wt.% [27] or 4.0 wt.% [28]. Ma et al. [29] studied the effects of nano-TiO2 (NT) on the toughness of hardened mortars. The flexural and tensile strength of cement-based materials with different TiO2 dosage were tested. Results showed that the tensile and flexural strength increased with increasing NT content up to 3 wt.%. The appropriate amount of nanoparticles in mortars significantly improved the crystal orientation of CH between hardened cement pastes and aggregates and grain size of CH is also decreased, both of which can control the crystallization process of hydration products in an appropriate state. In addition, more compact C-S-H gels are formed under the nanometer hydration induction effect, which can significantly improve the mechanical properties of cement mortars. Using too high nanoparticles dosage, drying shrinkage distor‐ tions of mortars are enlarged, leading to more microcracks in the interface of hardened pastes and aggregates. Simultaneously, excess nano-TiO2 is difficult to spread evenly and some internal defects would likely form in mortars.

During examination of new photocatalytic cementitious materials, the mechanical properties have been often estimated by compressive strength values. In order to present the tendency in effects, briefly, the table with results of several works was attached (**Table 1**). To obtain comparative results in each case the effect of analysis of reference material and material with the exemplary TiO2 dosage was shown. Mostly, TiO2 photocatalyst added in relatively low amount increased the compressive strength of cementitious materials. The mechanical properties measured in a form of the compressive strength were enhanced even 82 and 58% after 7 and 28 days of aging, respectively [27]. However, generally, the increase did not exceed the 10–12% determined at 7 days of curing or 18–23% in a case of 28 days.


**Table 1.** Compressive strength of different photoactive cementitious materials.

#### **5.3. TiO2 effect on carbonation of concretes**

maximum TiO2 content (5 wt.%), the nanoporosity increased notably and even with the presence of residual micropores (1.5–2.7 μm), the mechanical strength remained stable. In general, it was concluded that the presence of low size pores, particularly in the range between 1 and 0.1 μm helps to minimize the detrimental effect of the loading of nanoadditives. Beside changes occurring during setting and hardening the microstructure of concrete and mortars evolves in time during the service life of structures [24]. Diamanti et al. [10] focused on mechanical and durability aspects of TiO2-containing photocatalytic concrete by examining mutual influences between TiO2 and concrete components, and their evolution with the material aging. Materials were produced by adding the commercial form of TiO2 (P25) to concrete. In the beginning it was observed that despite the presence of nanoparticles which could play a positive filler effect, a slight decrease in a mechanical strength was observed in TiO2-containing specimens. SEM analysis showed a slight increase in the concrete porosity and to a non-even distribution of TiO2 particles that in some cases were present as clusters.

According to Zhao et al. [25] the compressive strength was reduced with increase of titania content in cementitious composites. A 12% compressive strength reduction was observed when 10 wt.% TiO2 was added to the cement matrix. It was attributed to the flocculation of nano-TiO2 particles which introduced a weak zone as flaws. In most papers is related that both, the strength and water permeability of the photocatalytic concrete are improved by adding TiO2 nanoparticles in the cement paste up to 2.0 wt.% [26], 3.0 wt.% [27] or 4.0 wt.% [28]. Ma et al. [29] studied the effects of nano-TiO2 (NT) on the toughness of hardened mortars. The flexural and tensile strength of cement-based materials with different TiO2 dosage were tested. Results showed that the tensile and flexural strength increased with increasing NT content up to 3 wt.%. The appropriate amount of nanoparticles in mortars significantly improved the crystal orientation of CH between hardened cement pastes and aggregates and grain size of CH is also decreased, both of which can control the crystallization process of hydration products in an appropriate state. In addition, more compact C-S-H gels are formed under the nanometer hydration induction effect, which can significantly improve the mechanical properties of cement mortars. Using too high nanoparticles dosage, drying shrinkage distor‐ tions of mortars are enlarged, leading to more microcracks in the interface of hardened pastes and aggregates. Simultaneously, excess nano-TiO2 is difficult to spread evenly and some

During examination of new photocatalytic cementitious materials, the mechanical properties have been often estimated by compressive strength values. In order to present the tendency in effects, briefly, the table with results of several works was attached (**Table 1**). To obtain comparative results in each case the effect of analysis of reference material and material with the exemplary TiO2 dosage was shown. Mostly, TiO2 photocatalyst added in relatively low amount increased the compressive strength of cementitious materials. The mechanical properties measured in a form of the compressive strength were enhanced even 82 and 58% after 7 and 28 days of aging, respectively [27]. However, generally, the increase did not exceed

the 10–12% determined at 7 days of curing or 18–23% in a case of 28 days.

internal defects would likely form in mortars.

150 High Performance Concrete Technology and Applications

The addition of TiO2 led to a decrease in the resistance to carbonation in time. Increasing in carbonation coefficient was connected with the parallel, and analogous, decreasing in the compressive strength of tested materials. Carbonation has the primary negative consequence of inducing corrosion of an embedded steel reinforcement. It also leads to the changes in the microstructure of cement paste and in the composition in a pore solution [31]. It was showed [10] that on one side the presence of TiO2 influenced the carbonation rate, while, on the other side, the carbonation induced modifications that have influenced the photoactivity of the materials. Self-cleaning efficiency decreased with the material aging, however, certain photoactivity was maintained, and the decreasing can be limited by increasing the initial amount of TiO2. The similar self-cleaning reduction was observed in a several works [20, 30] and was probably due to the shielding of TiO2 nanoparticles by the carbonate precipitates and the obstruction of the surface pores. These work indicated that testing the properties of a freshly produced photocatalytic concrete materials is not indicative of their real behaviour. The experiments carried out after a few weeks, or the months, when the carbonation alters the surface, would give the proper information about the structures built with such materials. Some contradictory date, compared to the above described results, regarding the photocata‐ lytic cement mortars, was obtained by Rao et al. [36]. The researchers used the self-compacting mortars with the addition of nano titanium in an amorphous state and 30% replacement of the cement with the fly ashes. First of all, the family mortars with nano-TiO2 did not show any carbonation effect (or a relatively low carbonation depths) up to 91 days of the exposure in the accelerated carbonation chamber. Generally, in their results, the compressive strength increased with an age and decreased with the nanomaterials addition ratio.

#### **5.4. The fire resistance of photocatalytic concrete**

The next clue issue is a fire resistance of the photocatalytic concrete. The normal concrete is a flameproof and reveals good resistance to the fire, though it is not considered as a refractory material. Biloxi et al. [32] examined the property of a white high performance concrete containing titanium dioxide. The concrete specimens were thermally treated at temperatures of 250, 500 and 750°C in an electrical radiant oven. The test results revealed number of significant variations in the mechanical strengths for the specimens exposed up to 250°C. However, the significant damage was observed for higher treatment temperatures, 500 and 750°C. The most important is that the similar observations were found for photocatalytic as well as for the ordinary concrete made with a similar aggregate. The effects of a damage were in the form of the microcracking. In the specimens treated up to 500°C, SEM images showed cracking of the concrete distributed s randomly along the edges. In the specimens treated up to 750°C, the microcracks were more widespread and of larger dimensions. The only effect of high temperature on photocatalytic concrete, in a contrast to the original concrete, involved loss of photocatalytic capability at 750°C due to the transformation of titanium dioxide from anatase to rutile. Salemi et al. [26] focused on the frost resistance of the concrete containing various nanoparticles. In order to determine the property, the change in compressive strength, the change in the length, the loss of a mass and increasing in a water absorption in the specimens were measured during the cycle of a freezing and the thawing. The strength loss of the concrete containing nanoparticles appeared to be much lower than that of the plain concrete. The concrete containing 2 wt.% of nano-TiO2 showed only 11.5% the strength loss after 300 cycles of freezing and thawing, while the strength loss of the plain concrete after 300 cycles was 100%. It is worth pointing out that the contribution of nano-TiO2 in the improvement of the mechanical properties and durability of the concrete was higher than the other particles (e.g. Al2O3, ZnO2, Fe2O3).

#### **5.5. Influence of TiO2 on abrasion resistance and shrinkage**

The influence of TiO2 presence in a concrete on an abrasion resistance was studied by Li et al. [33]. The measurements were carried out after 28 days of curing in a standard moist room at the temperature 20°C. The test results indicated that the abrasion resistance of concretes containing nanoparticles was a significantly improved. The abrasion resistance of the concrete containing nano-TiO2 in the amount of 1% by weight of a binder increased by 180% for the surface index. However, the effectiveness of TiO2 in the enhancing abrasion resistance increased with the decrease of photocatalyst content (5 wt.% < 3 wt.% < 1 wt.%). The important observation is that the abrasion resistance of the concrete containing nanoparticles increased with the increasing the compressive strength and the relationship appeared to be a linear.

Regarding to the further mechanical properties of the concrete modified with titanium dioxide, it should be mentioned the influence of TiO2 presence on a shrinkage, a workability and the setting time of the cementitious materials. Below was presented a general tendency of the impact. The inclusion of TiO2 in the cementitious matrix increases the chemical shrinkage. The higher TiO2 content resulted in a greater chemical shrinkage. It resulted from the directly relation and a proportionality to the degree of a hydration. The workability decreases with increasing TiO2 content. It is probably related to the higher surface area of TiO2 particles that needs more water to wetting the cement particles. Similarly, as the content of photocatalyst increases as the initial and the final setting time decreases. This is explained by a rapid consumption of a free water speeded up to the bridging process of gaps and as a result, the viscosity increases and the solidification occurs earlier. There is also the possibility that due to the large surface area of photocatalyst a greater availability of nucleation sites is provided and leads to faster hydration rate and shorter setting time [18].

**5.4. The fire resistance of photocatalytic concrete**

152 High Performance Concrete Technology and Applications

(e.g. Al2O3, ZnO2, Fe2O3).

**5.5. Influence of TiO2 on abrasion resistance and shrinkage**

The next clue issue is a fire resistance of the photocatalytic concrete. The normal concrete is a flameproof and reveals good resistance to the fire, though it is not considered as a refractory material. Biloxi et al. [32] examined the property of a white high performance concrete containing titanium dioxide. The concrete specimens were thermally treated at temperatures of 250, 500 and 750°C in an electrical radiant oven. The test results revealed number of significant variations in the mechanical strengths for the specimens exposed up to 250°C. However, the significant damage was observed for higher treatment temperatures, 500 and 750°C. The most important is that the similar observations were found for photocatalytic as well as for the ordinary concrete made with a similar aggregate. The effects of a damage were in the form of the microcracking. In the specimens treated up to 500°C, SEM images showed cracking of the concrete distributed s randomly along the edges. In the specimens treated up to 750°C, the microcracks were more widespread and of larger dimensions. The only effect of high temperature on photocatalytic concrete, in a contrast to the original concrete, involved loss of photocatalytic capability at 750°C due to the transformation of titanium dioxide from anatase to rutile. Salemi et al. [26] focused on the frost resistance of the concrete containing various nanoparticles. In order to determine the property, the change in compressive strength, the change in the length, the loss of a mass and increasing in a water absorption in the specimens were measured during the cycle of a freezing and the thawing. The strength loss of the concrete containing nanoparticles appeared to be much lower than that of the plain concrete. The concrete containing 2 wt.% of nano-TiO2 showed only 11.5% the strength loss after 300 cycles of freezing and thawing, while the strength loss of the plain concrete after 300 cycles was 100%. It is worth pointing out that the contribution of nano-TiO2 in the improvement of the mechanical properties and durability of the concrete was higher than the other particles

The influence of TiO2 presence in a concrete on an abrasion resistance was studied by Li et al. [33]. The measurements were carried out after 28 days of curing in a standard moist room at the temperature 20°C. The test results indicated that the abrasion resistance of concretes containing nanoparticles was a significantly improved. The abrasion resistance of the concrete containing nano-TiO2 in the amount of 1% by weight of a binder increased by 180% for the surface index. However, the effectiveness of TiO2 in the enhancing abrasion resistance increased with the decrease of photocatalyst content (5 wt.% < 3 wt.% < 1 wt.%). The important observation is that the abrasion resistance of the concrete containing nanoparticles increased with the increasing the compressive strength and the relationship appeared to be a linear.

Regarding to the further mechanical properties of the concrete modified with titanium dioxide, it should be mentioned the influence of TiO2 presence on a shrinkage, a workability and the setting time of the cementitious materials. Below was presented a general tendency of the impact. The inclusion of TiO2 in the cementitious matrix increases the chemical shrinkage. The higher TiO2 content resulted in a greater chemical shrinkage. It resulted from the directly relation and a proportionality to the degree of a hydration. The workability decreases with

## **6. The examples of building and objects built with using photoactive concretes**

In many studies, it was demonstrated that incorporation of titanium dioxide into the concrete materials is very effective solution towards degradation of a various hazardous and the toxic compounds. Therefore, its subsequent application in a real elements appeared to be a prom‐ ising technology for the reduction in the environmental pollution. Among photocatalytic concrete products, it is worth to mention the pavement blocks, the titles, and the walls of the buildings. The essential benefits of photocatalytic concrete elements are that it decomposes chemicals that contribute to soiling and air pollution, keep the concrete cleaner, and a reflect much of the sun's heat, because their white colour [34]. The construction materials might be the easily available medium to distribute photocatalysts over the widest surface area possible. It involves the high efficiency of the materials and simultaneously the increase in the materials costs is limited. Considering the self-cleaning attitude of the concrete materials it should be mentioned the simultaneous occurrence of two effects: photocatalytic degradation of deposits accumulating on their surface and photo-induced superhydrophilicity. As a consequence, the washing away of a reaction product is relatively easy [16]. The scheme of photocatalytic concrete action (example of NOx degradation) was illustrated in **Figure 6** [15].

**Figure 6.** Scheme of photocatalytic air purifying pavement [15].

The first country where TiO2-based cement has been marketed was Japan since the late 1990s [35, 37]. Nowadays, the white cement containing TiO2 is also used for the construction of buildings in Europe [16]. The patent of the application of TiO2 in pavement blocks is held by Murata et al. (Mitsubishi Materials Corporation), as well as Cassar and Pepe (Italcementi S.p.A.) [38, 39]. Before we present the specific examples of buildings built from photocatalytic concrete, which exist in the world, we try to perform the studies focusing on the transformation from laboratory to real scale in a reference to the described materials.

The implementation of products, obtained in the laboratory, into a real scale demand taking into account a lot of parameters. On the final reduction rate of the pollutants influence the geometrical situation, the speed of the traffic, the speed and the direction of the wind, the temperature, humidity, etc. Namely, it is important that the exhaust gaseous pollutants stay in the contact with the photoactive surface during a certain period. Moreover, the real conditions require additional aspects of the new products connected with their multiple usages. In the case of a concrete pavement blocks, e.g., TiO2 is placed in the whole thickness of wearing layer of the pavers. It means that even some abrasion takes place by the traffic, new TiO2 will be present at the surface to maintain the photocatalytic activity. Another possibility is using a double layered concrete with addition of TiO2: in the mass and/or as dispersion on the surface [15]. The implementation efforts of the scientific group realizations [32] were taken in Belgium. Photocatalytic cementitious materials have been applied on the side walls and the roof of the Leopold II tunnel in Brussels. The states of the object before and after photocatalytic renovation were presented in **Figure 6**. About 100 m in length of the photocatalytic materials was applied. Inside the tunnel was observed the effect on the air pollution (NOx, VOC's, CO2, O3, etc.). A dedicated UV lightning system was installed inside the tunnel. which could be modulated (on/off) to directly see the action of the photocatalytic walls (see **Figure 7**).

**Figure 7.** Inside view of test site within Leopold II tunnel in Brussels **(a)** before renovation, **(b)** after renovation with using photocatalytic walls [15].

Folli et al. [6] reported the results of a field test study concerning the use of photocatalytic paving elements in Denmark to decrease NOx pollution. The large scale studies were preceded by the experiments in the lab. The test area was in a Copenhagen central street located close to the Central Railway Station. The test area involved 200 m long × 2 (both side of the road) sidewalk pavers. Hundred meter were built from ordinary concrete blocks and 100 m from concrete blocks containing titanium dioxide as a photocatalyst. Over the entire year, the daily average NO concentration was maintained to very low values (below 40 ppb) in the area paved with the concrete elements containing TiO2. The important aspect is that seasonal variation was observed. NO conversion decreased with increasing relative humidity as a result of competition between water and NO for catalytic sites. Meanwhile, NO conversion increased with increasing temperature due to a higher diffusivity of the gaseous pollutants towards the photocatalytic surface.

Murata et al. (Mitsubishi Materials Corporation), as well as Cassar and Pepe (Italcementi S.p.A.) [38, 39]. Before we present the specific examples of buildings built from photocatalytic concrete, which exist in the world, we try to perform the studies focusing on the transformation

The implementation of products, obtained in the laboratory, into a real scale demand taking into account a lot of parameters. On the final reduction rate of the pollutants influence the geometrical situation, the speed of the traffic, the speed and the direction of the wind, the temperature, humidity, etc. Namely, it is important that the exhaust gaseous pollutants stay in the contact with the photoactive surface during a certain period. Moreover, the real conditions require additional aspects of the new products connected with their multiple usages. In the case of a concrete pavement blocks, e.g., TiO2 is placed in the whole thickness of wearing layer of the pavers. It means that even some abrasion takes place by the traffic, new TiO2 will be present at the surface to maintain the photocatalytic activity. Another possibility is using a double layered concrete with addition of TiO2: in the mass and/or as dispersion on the surface [15]. The implementation efforts of the scientific group realizations [32] were taken in Belgium. Photocatalytic cementitious materials have been applied on the side walls and the roof of the Leopold II tunnel in Brussels. The states of the object before and after photocatalytic renovation were presented in **Figure 6**. About 100 m in length of the photocatalytic materials was applied. Inside the tunnel was observed the effect on the air pollution (NOx, VOC's, CO2, O3, etc.). A dedicated UV lightning system was installed inside the tunnel. which could be modulated (on/off) to directly see the action of the photocatalytic walls (see **Figure 7**).

**Figure 7.** Inside view of test site within Leopold II tunnel in Brussels **(a)** before renovation, **(b)** after renovation with

Folli et al. [6] reported the results of a field test study concerning the use of photocatalytic paving elements in Denmark to decrease NOx pollution. The large scale studies were preceded by the experiments in the lab. The test area was in a Copenhagen central street located close to the Central Railway Station. The test area involved 200 m long × 2 (both side of the road) sidewalk pavers. Hundred meter were built from ordinary concrete blocks and 100 m from concrete blocks containing titanium dioxide as a photocatalyst. Over the entire year, the daily

using photocatalytic walls [15].

from laboratory to real scale in a reference to the described materials.

154 High Performance Concrete Technology and Applications

The experiments concerning a full-scale demonstration of air purifying pavement were also by Ballari and Brouwers [40] presented. In Hengelo, The Netherlands, the full width of the street was provided with concrete pavement containing TiO2 over a length of 150 m. The NOx concentration in the modified street and in control street together with weather parameters was measured. The results were directly connected with the weather conditions. Generally, the NOx concentration was 19% lower in comparison to the reference system. However, considering only afternoons or under high radiation and low relative humidity the value was 28% or 45% lower than in case of reference, respectively. The proposed solutions are promising techniques to reduce a number of air contaminants, especially at sites with a high level of pollution, such as: highly trafficked canyon streets or road tunnels.

**Figure 8.** Dives in Misericordia Church in Rome **(a)**, zoom insight **(b)** [41].

Self-cleaning elements are mostly used in white concrete buildings. The first real project of building with the self-cleaning activity through TiO2 in cementitious materials was started in 1996 during realization of church Dives in Misericordia in Rome, Italy. The project was completed in 2003 by Italcementi S.p.A.—an Italian cement company (architect Richard Meier). The photos of the project were shown in **Figure 8**. It was found that over a six-year monitoring period, only a slight difference was observed between the white exterior and interior walls. The next clue example of objects built from photocatalytic cementitious material is Cité de la Musique et des Beaux- Arts in Chambéry, presented in **Figure 9**, which was completed in 2000. Monitoring for approximately five years indicated that in the Chambéry City Hall the primary colour remained almost constant in different facade position (on West, North, East and South) [42, 43]. It is impressive that according to Fujishima and Zhang [44] by 2003, self-cleaning TiO2-based tiles had been used in over 5000 buildings in Japan. Among them the most famous is the Maru Building, located in Tokyo's main business district.

**Figure 9.** Cité de la Musique et des Beaux- Arts in Chambéry [45].

## **7. Conclusions**

The presented studies show that prepared concretes have photocatalytic activity and may purified an air from for example nitrogen oxides or the volatile organic compounds. Applica‐ tion of titanium dioxide into cement increased the mechanical properties of obtained concretes. Despite these advantage, some disadvantages unfortunately still exist.

1. Sometimes the by-products produce during photocatalytic decomposition of contamination are more toxic than substrates, it is possible to eliminate this by strong adsorption of byproducts on the surface of photocatalysts until its overall mineralization.

2. The commercial photoactive cements are mainly activated under UV light irradiation; the researchers tried to find the photocatalyst active under visible light irradiation.

3. Increasing the weight addition of photocatalyst into cement, increasing its photocatalytic activity but mainly when the addition is higher than 5 wt.% the mechanical properties of modified cement decreased.

4. A price of the commercial photoactive cements (such asTioCEM®, Górażdże, Poland, TX Active®, Italcement Group, Italy) are still from eight to ten times more expensive than pure cement.

Beside these disadvantages, the advantages of new concretes as: air purification, better mechanical properties and self-cleaning properties, makes photoactive concretes the future building materials.

## **Author details**

[42, 43]. It is impressive that according to Fujishima and Zhang [44] by 2003, self-cleaning TiO2-based tiles had been used in over 5000 buildings in Japan. Among them the most famous

The presented studies show that prepared concretes have photocatalytic activity and may purified an air from for example nitrogen oxides or the volatile organic compounds. Applica‐ tion of titanium dioxide into cement increased the mechanical properties of obtained concretes.

1. Sometimes the by-products produce during photocatalytic decomposition of contamination are more toxic than substrates, it is possible to eliminate this by strong adsorption of by-

2. The commercial photoactive cements are mainly activated under UV light irradiation; the

3. Increasing the weight addition of photocatalyst into cement, increasing its photocatalytic activity but mainly when the addition is higher than 5 wt.% the mechanical properties of

4. A price of the commercial photoactive cements (such asTioCEM®, Górażdże, Poland, TX Active®, Italcement Group, Italy) are still from eight to ten times more expensive than pure

Beside these disadvantages, the advantages of new concretes as: air purification, better mechanical properties and self-cleaning properties, makes photoactive concretes the future

Despite these advantage, some disadvantages unfortunately still exist.

products on the surface of photocatalysts until its overall mineralization.

researchers tried to find the photocatalyst active under visible light irradiation.

is the Maru Building, located in Tokyo's main business district.

156 High Performance Concrete Technology and Applications

**Figure 9.** Cité de la Musique et des Beaux- Arts in Chambéry [45].

**7. Conclusions**

modified cement decreased.

cement.

building materials.

Magdalena Janus\* and Kamila Zając

\*Address all correspondence to: mjanus@zut.edu.pl

Faculty of Civil Engineering and Architecture, West Pomeranian University of Technology, Szczecin, Poland

## **References**


Applied Catalysis B: Environmental. 2003;10:205–216. doi:10.1016/ S0926-3373(02)00303-X

[21] Senff L, Tobaldi DM, Lemes-Rachadel P, Labrincha JA, Hotza D. The influence of TiO2 and ZnO powder mixtures on photocatalytic activity and rheological behavior of cement pastes. Construction and Building Materials. 2014;64:191–200. doi:10.1016/ j.conbuildmat.2014. 04.121

[9] Lucas SS, Ferreira VM, Barroso de Aguiar JL. Incorporation of titanium dioxide nanoparticles in mortars—influence of microsturcture in the hardened state properties and photocatalytic activity. Cement Concrete Research. 2013;43:112–120. doi:10.1016/

[10] Diamanti MV, Lollini F, Pedeferri MP, Bertolini L. Mutual interactions between carbonation and titanium dioxide photoactivity in concrete. Building and Environ‐

[11] Mills A, Elouali S. The nitric oxide ISO photocatalytic reactor system: measurement of NOx removal activity and capacity. Journal of Photochemistry and Photobiology A:

[12] Staub de Melo JV, Trichês G, Gleize PJP, Villena J. Development and evaluation of the efficiency of photocatalytic pavement blocks in the laboratory and after one year in the field. Construction and Building Materials. 2012;37:310–319. doi:10.1016/j.conbuildmat.

[13] Shen S, Burton M, Jobson B, Haselbach L. Pervious concrete with titanium dioxide as a photocatalyst compound for greener urban road environment. Construction and

[14] Hassan MM, Dylla H, Mohammad LN, Rupnow T. Evaluation of the durability of titanium dioxide photocatalyst coating for concrete pavement. Construction and

[15] Boonen E, Beeldens A. Photocatalytic Road: from lab tests to real scale applications. European Transport Research Review. 2013;5:79–89. doi:10.1007/s12544-012-0085-6

[16] Spasiano D, Marotta R, Malato S, Fernandez-Ibañez P, Di Somma I. Solar photocatal‐ ysis: materials, reactors, some commercial, and pre-industrialized applications. A comprehensive approach. Applied Catalysis B: Environmental. 2015;170–171:90–123.

[17] Janus M, Zatorska J, Czyżewski A, Bubacz K, Kusiak-Nejman E, Morawski A. Selfcleaning properties of cement plates loaded with N,C-modified TiO2 photocatalysts.

[18] Rashad AM. A synopsis about the effect of nano—titanium dioxide on some properties of cementitious materials—a short guide for civil engineer. Reviews on Advanced

[19] Chen J, Kou SC, Poon CS. Hydration and properties of nano-TiO2 blended cement composites. Cement and Concrete Composites. 2012;34:642–649. doi:10.1016/j.cemcon‐

[20] Lackhoff M, Prieto X, Nestle N, Dehn F, Niessner R. Photocatalytic activity of semi‐ conductor-modified cement—influence of semiconductor type and cement ageing.

Applied Surface Science. 2015;330:200–206. doi:10.1016/j.apsusc.2014.12.113

Building Materials. 2010;24:1456–1461. doi:10.1016/j.conbuildmat.2010.01.009

Building Materials. 2012;35:874–883. doi:10.1016/j.conbuildmat.2012.04.097

ment. 2013;62:174–181. doi:10.1016/j.buildenv.2013.01.023

Chemistry. 2015;305:29–36. doi:10.016/j.photochem.2015.03.002

j.cemconres.2012.09.007

158 High Performance Concrete Technology and Applications

2012.07.073

doi:10.1016/j.apcatb.2014.12.050

Materials Science. 2015;40:72–88.

comp.2012. 02.009


[47] Behfarnia K, Keivan A, Keivan A. The effects of TiO2 and ZnO nanoparticles on physical and mechanical properties of normal concrete. Asian J Civil Eng. 2013;14:517–531.

[33] Li H, Zhang MH, Ou JP. Abrasion resistance of concrete containing nano-particles for

[34] Puzenat E. Photocatalytic self-cleaning materials: principles and impact on atmosphere. The European Physical Journal Conferences. 2009;1:69–74. doi:10.1140/epiconf/

[35] Fujishima A, Zhang X, Tryk DA. TiO2 photocatalysis and related surface phenomena. Surface Science Reports. 2008;63:515–582. doi:10.1016/j.surfrep.2008.10.001

[36] Rao S, Silva P, de Brito J. Experimental study of the mechanical properties and dura‐ bility of self-compacting mortars with nano materials (SiO2 and TiO2). Construction and Building Materials. 2015;96:508–517. doi:10.1016/j.conbuildmat.2015.08.049

[37] Guo S, Wu ZB, Zhao WR. TiO2—based building materials: above and beyond tradi‐ tional applications. Chinese Science Bulletin. 2009;54:1137–1142. doi:10.1007/

[38] Murata Y, Tawara H, Obata H, Murata K. NO-cleaning paving block. EP-patent

[39] Cassar L, Pepe C. Paving tile comprising an hydraulic binder and photocatalyst

[40] Ballari MM, Brouwers HJH. Full scale demonstration of air-purifying pavement. Journal of Hazardous Materials. 2013;254–255:406–414. doi:10.1016/j.hazmat.

[41] Lea-Catherine Szacka. Panorama Living Italian Style [Internet]. 2011. Available from: http://www.panoramitalia.com/en/lifestyle/design-architecture/modern-italian-

[42] Cassar L, Pepe C, Tognon G, Guerrini GL, Amadelli R. White cement for architectural concrete, possessing photocatalytic properties. In: Proceeding 11th International Congress on the Chemistry of Cement, The Cement and Concrete Institute of South

[43] Guerrini GL, Plassais A, Pepe C, Cassar L. Use of photocatalytic cementitious materials for self-cleaning applications. In: RILEM International Symposium on Photocatalysis

[44] Fujishima A, Zhang X. Titanium dioxide photocatalysis: present situation and future approaches. Comptes Rendus Chimie. 2006;9:750–760. doi:10.1016/j.crci.2005.02.055

[45] IDBAT Prefabrication. Available from: http://www.idbat.com/#Realisations/Educa‐

[46] Noorvand H, Ali AAA, Demiboga R, Farzadnia N, Noorvand H, Incorporation of nano TiO2 in black rice husk ash mortars. Constr Build Mater. 2013;47:1350–1361. DOI:

Environment and Construction Materials, Florence, Italy, 2007;55:219–226

0786283A1, Mitsubishi Materials Corporation, Japan, 1997.

particles. EP-patent 1600430A1, Italcementi S.p.A., Italy, 1997

architecture/561/ [Accessed: 20.04.2016]

tion-et-Formation [Accessed: 250.04.2016]

10.1016/j.conbuildmat.2013.06.066

Africa, Durban, 2003;4:2012–2021

pavement. Wear. 2006;260:1262–1266. doi:10.1016/j.wear.2005.08.006

e2009-00911-2

160 High Performance Concrete Technology and Applications

s11434-009-0063-0

2013.02.012

[48] Shekari AH, Razzaghi MS. Influence of nano particles on durability and mechanical properties of high performance concrete. Procedia Engineering. 2011;14:3036–3041. DOI:10.1016/j.proeng.2011.07.382
