This paper presents the development of lightweight volcanic pumice concrete (VPC) using pumice as aggregate illustrating 21 concrete mixtures. Fresh, mechanical, and durability properties of VPC mixtures such as slump, air content, compressive strength, tensile strength,
Keywords: durability; fire resistance; lightweight concrete; mixture design; strength.
INTRODUCTION
Research has been conducted worldwide on a large number of natural and artificial lightweight aggregates to manufacture mortar and concrete.1-9 Natural lightweight aggregates include diatomite, pumice, volcanic scoria, sawdust, oil palm shells, bottom ash, and starch-based aggregate. Artificial aggregates include expanded shale, slate, perlite, sintered fly ash, bonded fly ash, and vermiculite. Beside these aggregates, solidified blast furnace slag is also used for the manufacture of lightweight concrete. Use of natural lightweight aggregates instead of processed artificial aggregates can significantly reduce the cost of such concretes.
Pumice is a natural material of volcanic origin produced by the release of gases during the solidification of lava.10 World pumice (and related materials) production was 14.4 million metric tones (Mt) [28.800 million pound (Mlb)]) in 2004.11 Globally, Italy remains the dominant producer of pumice, with production estimated to be 4.6 Mt (9200 Mlb) per year. Other leading countries are Chile, Ecuador, Ethiopia, France, Germany, Greece, Spain, Turkey, and the U.S. In 2004, pumice and pumicite consumption in the U.S. was 1.1 Mt (2200 Mlb) with an estimated value of 26 million U.S. dollars. The main use for pumice is as an aggregate in lightweight building blocks and assorted building products.11 Volcanic pumice (VP) has also been used as aggregate in the production of lightweight concrete. So far, the use of VP was dependent on the availability and limited to the countries where it is locally available or easily imported.
In Europe, pumice was used in ancient Rome over 2000 years ago and many notable pumice structures are still standing today.12 In Europe, pumice concrete constitutes approximately 3% of the total lightweight concrete consumption with 70% of the total consumption in Germany. Lightweight concrete made with pumice and pozzolanic cement with volcanic ash/lime (developed in Mexico by the builders of an ancient culture, the Totonacas) has survived more than 2000 years and provides an example of a low strength concrete and very long-term performance.13,14 The use of pumice and perlite as additives was found to provide excellent resistance to freezing and thawing of cement pastes, mortar, and concrete.15
Lightweight concrete with a 28-day compressive strength of 55 MPa (7977 psi) and a dry unit weight in the range of 1700 to 2100 kg/m[sup]3[/sup] (106 to 131 lb/ft3) was produced by incorporating Turkish pumice aggregate.16 Prefabricated lightweight pumice concrete infill panels were investigated as a retrofit alternative of building frames subjected to quasistatic loading.17 Investigation of reinforced concrete made with lightweight aggregate (such as pumice and slag) revealed their satisfactory resistance against salt attack/steel corrosion and their suitability of using in load-bearing and enclosure structures.18 Precast panels made with lightweight pumice concrete (35 and 25% lighter-than-normalweight and lightweight expanded shale concrete, respectively) having a 28-day compressive strength of 24 MPa (3500 psi) were satisfactorily used in the Rockwood residence on Hayden Island, Oregon.19 The 73-story World Trade Tower in downtown Los Angeles is the tallest building west of the Mississippi River, and was built using structural lightweight pumice concrete.2,4
This study was undertaken to investigate the potential use of VP in lightweight aggregate concrete production. Durability of concrete is one of its most important desirable properties and it is essential that the concrete made with VP should be capable of preserving its durability under actual environmental conditions as well as at elevated temperatures. This paper presents the results of a comprehensive experimental investigation studying mixture design, strength, durability, and fire resistance characteristics of lightweight volcanic pumice concrete (VPC).
RESEARCH SIGNIFICANCE
Pumice can be found in many places around the world where volcanoes erupt. Although it has been used successfully in many countries, finding new and improved ways to build with pumice is becoming widespread. This paper focuses on the use of VP in lightweight concrete production with particular emphasis on Papua New Guinea (PNG). Volcanic materials such as VP are found abundantly in PNG due to frequent volcanic eruption. The use of VP as a construction material can lead to low-cost construction. Although this research is based on PNG, it provides useful information to manufacture lightweight concrete incorporating VP from other parts of the world. The recommendations of this paper can be of special interest to owners and engineers considering the production and use of VPC in construction.
EXPERIMENTAL INVESTIGATION
The investigation was based on the VP obtained from the Tavurvur and Vulcan craters located in the Rabaul area of the East New Britain province of PNG. The Rabaul area is situated in the worldwide earthquake and volcanic zone known as the Belt of Fire. Table 1 presents the chemical properties of VP. Locally manufactured ASTM Type I portland cement (PC), whose chemical and physical properties are shown in Table 1, was used.
The coarse aggregates used were 12.5 and 20 mm (1/2 and 3/4 in.) maximum size crushed gravel aggregate (GA) and volcanic pumice aggregate (VPA). The fine aggregates used were fine VPA and local river sand (Table 2). Clean drinking water was used for the concrete mixtures.
VP collected from PNG sources consists of lumps of various sizes ranging from 2 to 64 mm (0.08 to 2.5 in.) but bigger blocks up to 1 m[sup]3[/sup] (35.3 ft3) in size are also available. Crushed VPA of various grading can be produced from these sources. The particle size distributions of VPA and sand were determined according to ASTM C 136. Grading of VPA and sand (used in this study) meet the requirement of lightweight aggregate for structural concrete as per ASTM C 330 (Table 2).
The properties of GA and VPA are compared in Table 3. The bulk density (both dry rodded and loose, determined as per ASTM C 29) and specific gravity (determined as per ASTM C 127) suggest that coarse volcanic pumice aggregate is much lighter than gravel aggregate. Coarse pumice aggregate also has high water absorption (determined as per ASTM C 127), which indicates its high degree of porosity compared with gravel aggregate. As per ASTM C 330, VPA satisfies the requirement of the lightweight coarse aggregate for structural concrete as the ovendry density falls within the range of 560 to 1120 kg/m[sup]3[/sup] (35 to 70 lb/ft3). The water absorption in coarse VPA, however, is higher (27 to 32%) than the range of 5 to 20% as normally occurred in other lightweight aggregates.3 The specific gravity and water absorption of sand and fine VPA (determined as per ASTM C 128) are also presented in Table 3.
Mixture design of volcanic pumice concrete (VPC)
Mixture proportioning was conducted according to ACI 211.2-98.20 An extensive series of tests were conducted to develop suitable VPC mixtures. Developed VPC mixtures are classified into three series:
Series A-VPA used as replacement of coarse crushed gravel aggregate: VPC mixtures were developed by combining different percentages (ranging from 0 to 100%) of 12.5 mm (1/2 in.) maximum size VPA with 12.5 mm (1/2 in.) maximum size crushed gravel aggregate. Mixture details are presented in Table 4.
Series B-VP as coarse as well as both coarse and fine aggregates: VPC mixtures were developed by using pumice aggregate as both fine and coarse (with 20 mm [3/4 in.] maximum size as per ASTM C 330) aggregates having no sand as well as using 20 mm (3/4 in.) maximum size pumice aggregate with sand. Mixture proportions of concrete mixtures in this series are presented in Table 5.
Series C-VP as coarse aggregate: VPC mixtures having varying cement (C) content and water-cement ratio (w/c) were developed by using 20 mm (3/4 in.) maximum size pumice aggregate and sand. Mixture design of VPC mixtures are compared with normal density concrete (NC) mixtures using 20 mm (3/4 in.) maximum size gravel aggregate and sand in Table 6.
Concrete mixing, test specimens, curing conditions, and testing details
All 21 concrete mixtures were prepared in a laboratory counter-current mixer. The high water absorption of VPA in the initial stages of mixing can cause balling-up of cement and a loss of slump. To avoid this, the VPA was made saturated surface dry before mixing of concrete. The mixing sequence consisted of 3 minutes of mixing the aggregates with water followed by an additional 3 minutes of mixing with cement.
A comprehensive series of tests on fresh, mechanical, and durability properties of concrete mixtures such as slump, air content, compressive strength fcu, tensile strength ft, density, elastic modulus (E), drying shrinkage (DS), water permeability, and fire resistance were conducted. The schedules of tests carried out on different series are summarized as follows.
* Series A and B: slump, air content, 28-day compressive/tensile strength, 28-day density, 28-day E, DS, and water permeability.
* Series C: slump, air content, 28-day compressive/tensile strength, 28-day density, 28-day E, DS, water permeability, and fire resistance.
The slump of fresh concrete mixtures was determined as per ASTM C 143. Air contents for NC and VPC mixtures were determined by pressure meter as per ASTM C 231 and air meter as per ASTM C 173/173M, respectively.
Twenty-eight-day compressive and indirect tensile strengths were determined by crushing 100 mm (4 in.) cubes as per BS 1881: Part 120:1983 and by splitting 100 x 200 mm (4 x 8 in.) cylinders as per ASTM C 496; 100 x 200 mm (4 x 8 in.) cylinders were also cast for modulus of elasticity tests. Three specimens were tested for each test at each age and mean values were reported. The specimens were removed from molds after 24 hours of casting and then placed in a water tank at 23 ± 2 °C (73 °F) for 28 days. For measuring density, 100 x 200 mm (4 x 8 in.) cylinders were air dried in the laboratory for 28 days. The 28-day air dry density was then calculated by dividing the weight of the cylinder by its volume.
Three 75 x 75 x 285 mm (3 x 3 x 11.2 in.) drying shrinkage specimens were cast for each concrete mixture. The shrinkage specimens were cured under water for 7 days and then transferred to a 23 ± 2 °C (73 °F), 50 ± 5% relative humidity room where the shrinkage was monitored using a vertical length comparator according to ASTM C 157 every week for a total of 12 weeks.
Two 100 x 200 mm (4 x 8 in.) cylinder specimens were cast for each concrete mixture for permeability tests. The 56-day water permeability was determined after 1 day of moist curing and 55 days of air curing by applying 1.4 MPa (200 psi) of water pressure in a hydraulic permeability test apparatus.
For fire resistance, 100 mm (4 in.) concrete cube specimens were used. The specimens were removed from molds after 24 hours of casting and then placed in a water tank at 23 ± 2 °C (73 °F) for 28 days. At an age of 28 days, the specimens were heated in an electric furnace up to 200, 400, 600, and 800 °C (392, 752, 1112, and 1472 °F) for different durations of 0, 0.5, 1, and 2 hours. The heating rate was set at 80 °C/minute (176 °F/minute) to reach the desired temperature. The rate of heating, however, is lower than the standard fire rate of temperature rise specified by ASTM E 119a, which is approximately 600 °C (1112 °F) in the first 6.7 minutes. The specimens were then allowed to cool naturally to room temperature outside the furnace. The heating/cooling rate depends on the lightweight properties (especially thermal conductivity) of the material. Because of lower thermal conductivity of pumice aggregate, additional time (lower rate of heating) is required for heating VPC specimens. A compressive strength test was performed according to BS 1881: Part 120:1983. Three specimens were tested at each stage and average values are reported. There are three test methods available for finding the residual compressive strength of concrete at elevated temperatures: stressed, unstressed, and unstressed residual strength test. The first two types of test are suitable for accessing the strength of concrete during high temperatures. The unstressed residual test adopted in the current study allowed the assessment of residual properties of concrete after cooling down to room temperature. One should note that the strength measured in this way is the smallest because the lack of transient creep and partial rehydration of the cement during and after cooling induce further damage in the concrete mass.21 Crack patterns in heated concrete cubes were examined.
RESULTS AND DISCUSSION
Fresh properties of VPCs
The fresh properties such as slump, air content, and segregation characteristics of concrete mixtures were investigated. The results are presented in Tables 4 through 6.
Due to lower aggregate density, structural low-density concrete does not slump as much as NC with the same workability. It is reported that a low-density mixture with a slump of 50 to 75 mm (2 to 3 in.) can be placed under a condition that would require a slump of 75 to 125 mm (3 to 5 in.) for normal density concrete.1-4 Lower slump values for VPC compared with NC were confirmed from the current study (Series A, Table 4 and Series C, Table 6). To make a VPC with 50 to 100% VPA (as replacement of crushed gravel aggregate) having satisfactory workability, the range of slump values should be between 50 and 75 mm (2 and 3 in.) (as illustrated from the current study of Series A, Table 4). With higher slumps in excess of 125 mm (5 in.), the large VPA particles tend to float to the surface, making finishing difficult. For VPC mixtures in Series B and C, slump values range between 53 and 70 mm (2 and 2.75 in.) (Tables 5 and 6). All concrete mixtures were found to be cohesive and workable as well as showed no segregation.
The air content of concrete mixtures increases with the increase of VPA content (Series A, Table 4). In the air content of Series B, VPC mixtures ranges between 4.1 and 4.7% with VPC Mixture 2 having pumice as both coarse and fine aggregates showing higher content (Table 5). In Series C (Table 6), the air content of VPC mixtures (ranging between 2.7 and 3.6%) was higher compared with NC (ranging between 1.6 and 2.1%). This can be attributed to the presence of porous VPA in VPC.
Mechanical properties of VPCs
The 28-day compressive strength and the 28-day density decrease with the increase of percentage of VPA due to the replacement of strong gravel aggregate by relatively weak VPA and the decrease of mortar content (Series A, Table 7). Mixture 2 (in Series A) also shows lower strength than Mixture 1 due to a high percentage of comparatively weaker VPA and comparatively lower mortar content. All VPCs (Series A) with a VPA replacement within the range of 50 to 100% of coarse gravel aggregate by volume developed strength in excess of 15 MPa (2175 psi) ranging between 23 and 35 MPa (3336 and 5076 psi) and an air dry density ranging between 1850 and 2150 kg/m[sup]3[/sup] (115 and 134 lb/ft3) to satisfy the criteria for semi-lightweight structural concrete (Table 7).22 Much lighter VPC (conforming to lightweight structural concrete with a density less than 1850 kg/m[sup]3[/sup] [115 lb/ft3]), however, was obtained by using VPA as both fine and coarse aggregates as per the grading requirement suggested by ASTM C 330 (Series B, Mixtures 2 and 3, Table 8). VPC mixtures in Series C also satisfy the criteria for semi-lightweight structural concrete except 12 MPa (1740 psi) VPC (Table 9). Lower strength VPC, however, can be used in the manufacture of lightweight building blocks and other building products. In Series C, the 28-day compressive strength of VPC mixtures is lower (ranges between 12 and 30 MPa [1740 and 4350 psi]) compared with NC (ranges between 21 and 51 MPa [3045 and 7395 psi]) as presented in Table 9.
The 28-day tensile strength decreases from 3.7 to 2.6 MPa (537 to 377 psi) for Mixture 1 and from 3.4 to 2.2 MPa (493 to 319 psi) for Mixture 2 with the increase of VPA from 0 to 100% (Series A, Table 7). The 28-day tensile strength of Series B mixtures ranges between 2.1 and 2.6 MPa (305 and 377 psi) with the lowest strength for VPC having VPA as both coarse and fine aggregates (Table 8). The 28-day tensile strength of VPC mixtures in Series C is lower (ranges between 1.0 and 2.8 MPa [145 and 406 psi]) compared with NC (ranges between 1.9 and 4.3 MPa [276 and 624 psi]) (Table 9).
The 28-day elastic modulus decreases from 18.2 to 10.5 GPa (2.64 × 106 to 1.52 × 106 psi) for Mixture 1 and from 17.7 to 10.1 GPa (2.57 × 106 to 1.46 × 106 psi) for Mixture 2 with the increase of VPA from 0 to 100% (Series A, Table 7). The 28-day elastic modulus of Series B mixtures ranges between 9.5 and 12.2 GPa (1.38 × 106 to 1.77 × 106 psi) with the lowest elastic modulus for VPC having VPA as both coarse and fine aggregates (Table 8). The 28-day elastic modulus of Series C VPC mixtures (ranges between 7.8 and 12.4 GPa [1.13 × 106 and 1.8 × 106 psi]) is lower compared with NC (ranges between 18.0 and 20.4 GPa [2.61 × 106 and 2.96 × 106 psi]) (Table 9). The elastic modulus of lightweight aggregate concrete usually ranges between 40 and 80% of normal concrete of the same strength.23 Lower E-values for VPCs indicate that there is a need for studies on structural behaviors (to assess the viability of application of VPCs in actual construction) so that service stage parameters, such as, deflections and crack widths, are within allowable limits.
Shrinkage and permeability of VPCs
Drying shrinkage-Aggregates with high absorption properties are associated with high shrinkage in concrete and this is confirmed from the increase in drying shrinkage with the increase of the amount VPA in Series A mixtures (Fig. 1). The shrinkage in VPC with 100% VPA is found to be approximately 22% (mean value) higher than those in the representative normal concrete (0% VPA). The 12-week drying shrinkage in VPC with 100% VPA was approximately 600 microstrain compared with approximately 450 microstrain in NC (0% VPA) (Fig. 1 and Table 7). The 12-week drying shrinkage for VPC in Series B ranges between 570 and 680 microstrain with the highest value shrinkage for VPC having VPA as both coarse and fine aggregates (Table 8). The 12-week drying shrinkage of VPC in Series C is lower (ranges between 530 and 660 microstrain) compared with NC (ranges between 410 and 450 microstrain) (Table 9). It is reported that the shrinkage of lightweight concrete can be 50% greater than NC.24 Generally, shrinkage increases with the increase of w/c and with the decrease of aggregatecement ratio (A/C) of concrete.10 For the current series of tests on VPC in Series A, as the percentage of VPA is increased from 0 to 100% by volume, the A/C is decreased from 4.1 to 1.7 by mass (Tables 4 and 7). As a consequence, the increase in shrinkage with the increase of VPA is justified. High initial drying shrinkage and comparatively low tensile strength may lead to the danger of shrinkage cracking. The danger of shrinkage cracking, however, can be compensated by the lower modulus of elasticity of VPC.
Water permeability-Permeability refers to the amount of water migration through concrete when the water is under pressure or the ability of concrete to resist water penetration. In Series A, water permeability of concrete is increased from approximately 3.6 × 10-10 cm/second (1.4 × 10-10 in./second) to approximately 13 × 10-10 cm/second (5.1 × 10-10 in./second) when VPA content is increased from 0 to 100% by volume (Table 7). The increase in permeability with the increase of VPA is due to the replacement of coarse gravel aggregate by comparatively porous VPA. As expected, Mixture 2 shows higher permeability than Mixture 1 due to higher VPA content and comparatively lower density.
The water permeability of VPC mixtures in Series B ranges between 11.6 × 10-10 and 13.2 × 10-10 cm/second (4.6 × 10-10 and 5.2 × 10-10 in./second) with the highest value for VPC having VPA as both coarse and fine aggregates (Table 8). The water permeability of VPC mixtures in Series C (ranges between 9.8 × 10-10 and 14.2 × 10-10 cm/second [3.9 × 10-10 and 5.6 × 10-10 in./second]) is lower compared with NC (ranges between 2.9 × 10-10 and 4.2 × 10-10 cm/second [1.1 × 10-10 and 1.7 × 10-10 in./second]) (Table 9).
Permeability of concrete is a function of permeability of paste, permeability and gradation of aggregate, paste-aggregate transition zone, and paste to aggregate proportion. Permeability also depends on w/c (increases with the increase of w/c) and initial curing conditions.25 To be watertight, structural concrete should have a w/c of not more than 0.48 for exposure to fresh water and not more than 0.44 for exposure to seawater.26 A mature good-quality concrete has a permeability of 1 × 10-10 cm/second (0.4 × 10-10 in./second).3 The higher permeability of VPC will allow higher moisture movement and has the harmful effect of corrosion (in the presence of chlorides) needing special care for protection of reinforcement.
Comparative study of mechanical and durability properties of NC and VPC
Table 10 compares the mechanical and durability properties of 20 and 30 MPa (2900 and 4350 psi), NC, and VPC. For a particular 28-day compressive strength, the 28-day tensile strength of VPC is similar, the 28-day density of VPC is approximately 22% lower, the 28-day E of VPC is 32 to 46% lower, the 12-week drying shrinkage of VPC is 18 to 28% higher, and the 56-day permeability of VPC is approximately 136% higher compared with NC.
The range of properties of Series A, B, and C concretes (presented in Table 10) show the potential of designing VPC mixtures incorporating volcanic pumice from Papua New Guinea with mechanical and durability characteristics conformable to other lightweight concrete using aggregates such as expanded slag, expanded clay, expanded slate, sintered fly ash, and scoria.2,10,23,24 The E values of concrete made with expanded clay aggregate vary from 10 to 16 GPa when compressive strength varies from 15 to 30 MPa (2175 to 4350 psi).23 Concrete made with lightweight aggregates having open textured and irregular surface can produce a shrinkage of 1000 microstrain.24
Fire resistance of VPC
Fire resistance of concrete mixtures was investigated in terms of residual strength. Residual strength decreases with the increase of temperature for a particular fire duration (Fig. 2 through 5). Residual strength also decreases with the increase of fire duration at a constant fire temperature (Fig. 2 through 5). In Fig. 4 through 5, St and S0 represent residual strength at t and 0 hours of fire duration, respectively. Residual strength at 400 °C (752 °F) for VPC is approximately 90% (81 to 86% for NC), 81 to 83% (76 to 79% NC), and 67 to 70% (62 to 66% for NC) of original strength when subjected to fire for a duration of 0.5, 1, and 2 hours, respectively. Residual strength at 600 °C (1112 °F) for VPC is approximately 80 to 81% (71 to 72% for NC), 71 to 77% (67 to 69% for NC), and 57% (48 to 52% for NC) of original strength when subjected to fire for a duration of 0.5, 1, and 2 hours, respectively. Residual strength at 800 °C (1472 °F) for VPC is approximately 70 to 76% (52 to 59% for NC), 42 to 48% (24 to 29 for NC), and 27 to 29% (17 to 19% for NC) of original strength when subjected to fire for a duration of 0.5, 1, and 2 hours, respectively. Residual strength of VPC is higher compared with NC when subjected to fire for a particular duration at different temperatures or at a particular temperature for different fire duration.
The test results indicated that each temperature range and duration of fire had a distinct pattern of strength loss. Rate of loss of strength is significantly higher at 800 °C (1472 °F) compared with 400 and 600 °C (752 and 1112 °F), especially beyond 0.5 hours of fire duration for both NC and VPC. Between 600 and 800 °C (1112 and 1472 °F) at 2 hours of fire duration, residual strength is significantly reduced from 57 to 28% for VPC and 50 to 18% for NC. At high temperature, the dehydration of cement paste results in its gradual disintegration. Because the paste tends to shrink and the aggregate expands at high temperatures of above 600 °C (1112 °F), the bond between the aggregate and the paste is weakened resulting in the greater reduction of strength as confirmed from test results.27,28 The deterioration of strength at elevated temperatures for such concretes can be attributed to the coarsening of the pore structure and the increase in pore diameter.27,28
Generally, the strength loss in VPC is lower compared with NC when the temperature is varied from 0 to 800 °C (1472 °F) and fire duration ranged between 0 and 2 hours. For instance, at 800 °C (1472 °F), the residual strength of VPC is 18, 17, and 10% higher compared with NC when subjected to fire for a duration of 0.5, 1, and 2 hours, respectively. This is an indication of better performance of VPCs in retaining the strength at elevated temperature as compared with NC. This can be attributed to the less dense pore structure of VPC (compared with NC) due to the presence of comparatively porous and lightweight VPA. Because the evaporation of physically absorbed water starts at 80 °C (176 °F), comparatively denser pore structure of NC allows the build-up of higher pore pressure by water vapors (compared with VPC) forming thermal cracks and subsequent higher reduction of strength at elevated temperatures.
It is required that the structural concrete should preserve good fire rating characteristics over a desired length of time. The resistance of concrete to fire and to elevated temperature depends on many factors such as duration of fire, temperature, moisture conditions, concrete mixture (leaner or richer), and aggregate properties.29 The strength of concrete decreases when exposed to a temperature in excess of 35 °C (95 °F) and under conditions allowing loss of moisture content. The moisture content has a significant effect on the strength of concrete in the temperature range of approximately 28 to 600 °C (82 to 1112 °F). The loss of strength is attributed to the loss of moisture and the degree of moisture loss depends on the duration of fire.
The 30 MPa (4350 psi) concrete, particularly VPC, shows lower strength retaining capacity compared with 20 MPa (2900 psi) concrete predominantly after 0.5 hour of fire exposure at 800 °C (1472 °F). The relatively high A/C and associated compact/dense internal microstructure of high strength concrete make it difficult for water vapor to transport and release (in concrete), causing more internal stress build-up, which leads to the reduction of fire resistance and subsequent spalling at high temperatures compared with low strength concrete.30,31 The lower capillary voids (associated with higher density) in high strength concrete also causes most of the water to be absorbed, which leads to the higher loss of strength.
At 800 °C (1472 °F), during the initial 0.5 hour of exposure, VPC specimens experienced considerable cracks as well as surface spalling. The color of specimens also changed. Surface features of VPC specimens exposed to 600 °C (1112 °F) also showed color changes as well as some edge cracks but not as severe as those exposed to 800 °C (1472 °F). Typical deterioration of concrete cubes subjected to elevated temperatures is shown in Fig. 6.
Lightweight VPC should exhibit more fire resistance characteristics than ordinary aggregate concrete due to lesser tendency to spall and loss of lesser proportion of its original strength with the rise in temperature.32 Lower thermal conductivity of VPC can induce high temperature gradients (rapid rise in surface temperature), causing hot surface layers tending to separate and spall from the cooler interior concrete layer. On the other hand, lower thermal conductivity also helps interior layers of VPC to remain cool for longer duration and to preserve their strength despite visible surface damage.32 This can be attributed to the higher strength retaining capacity of VPC compared with NC.
CONCLUSIONS
The mixture design characteristics including fresh and hardened properties as well as durability of lightweight VPC are described. The fire resistance of VPC subjected to high temperatures up to 800 °C (1472 °F) for a duration of up to 2 hours is also presented. The performance of VPC mixtures is compared with representative NC mixtures. The following conclusions are drawn from the study:
1. The 28-day compressive strength and the 28-day density decreases with the increase of the percentage of VPA as replacement of normal gravel aggregate. All VPC mixtures developed in this study satisfy the requirement of semi-lightweight structural concrete with strength in excess of 15 MPa (2175 psi) and an air-dry density ranging between 1850 and 2150 kg/m[sup]3[/sup] (115 and 134 lb/ft3). Much lighter VPC conforming to lightweight structural concrete with density less than 1850 kg/m[sup]3[/sup] (115 lb/ft3) is also obtained by using VPA as both fine and coarse aggregates. Design charts developed can be used as guidelines for the mixture design of VPC for similar condition and type of mixture;
2. The 56-day water permeability of concrete increases with the increase of the percentage of VPA. The water permeability of VPC mixtures in this study (ranges between 9.8 × 10-10 and 14.2 × 10-10 cm/second [3.9 × 10-10 and 5.6 × 10-10 in./second]) is higher compared with NC (ranges between 2.9 × 10-10 and 4.2 × 10-10 cm/second [1.1 × 10-10 and 1.7 × 10-10 in./second]). The higher permeability suggests that special care be taken to prevent corrosion of reinforcement due to higher moisture movement, if VPC is used in outdoor environment;
3. The 12-week drying shrinkage of VPC (less than 600 microstrain for most of the mixtures) is higher than those in the representative NC. VPC may be prone to shrinkage cracking due to its higher drying shrinkage and comparatively lower tensile strength compared with NC. The danger of shrinkage cracking, however, can be compensated by the lower modulus of elasticity (30 to 50% lower than NC);
4. The deterioration of both NCs and VPCs increases with the increase of temperature up to 800 °C (1472 °F) and duration of fire up to 2 hours due to substantial reduction in residual strength. VPC showed better performance showing higher residual strength and strength retaining capacity at elevated temperatures compared with NC; and
5. The research confirms the viability of producing VPC mixtures incorporating VP from PNG with strength, durability (in terms of water permeability and drying shrinkage), and fire resistance characteristics conformable to other lightweight concretes as well as some commercial VPC. Further investigations on many important parameters including corrosion of steel and carbonation, however, are needed to assess the suitability of application of VPCs in actual structures.
ACKNOWLEDGMENTS
The contributions of L. Mol, a MPhil student supervised by the first author and all other members of the first author's research team at Papua New Guinea University of Technology are highly appreciated.
ACI Materials Journal, V. 104, No. 5, September-October 2007.
MS No. M-2006-061.R2 received July 2, 2006, and reviewed under Institute publication policies. Copyright © 2007, American Concrete Institute. All rights reserved, including the making of copies unless permission is obtained from the copyright proprietors. Pertinent discussion including authors' closure, if any, will be published in the July-August 2008 ACI Materials Journal if the discussion is received by April 1, 2008.
REFERENCES
1. Hossain, K. M. A., "Properties of Volcanic Ash and Pumice Concrete," IABSE Report No. 81, Zurich, Switzerland, 1999, pp. 145-150.
2. Hossain, K. M. A., "Properties of Volcanic Scoria Based Lightweight Concrete," Magazine of Concrete Research, V. 56, No. 2, 2002, pp. 111-120.
3. Kostmatka, S. H.; Kerkhoff, B.; Panarese, W. C.; Macleod, N. F.; and McGrath, R. J., "Design and Control of Concrete Mixtures," Engineering Bulletin 101, 7th Canadian Edition, Cement Association of Canada, 2002, 356 pp.
4. Hossain, K. M. A., "Experimental Investigation on the Fire Durability of Volcanic Pumice Concrete," Research Report No. CE-SE-07-2000, Department of Civil Engineering, University of Technology, Papua New Guinea, 2000, 95 pp.
5. Singh, M., and Garg, M., "Perlite-Based Building Materials-A Review of Current Applications," Construction and Building Materials, V. 5, No. 2, June 1991, pp. 75-81.
6. Kohno, K.; Okamoto, T.; Isikawa, Y.; Sibata, T.; and Mori, H., "Effects of Artificial Lightweight Aggregate on Autogenous Shrinkage of Concrete," Cement and Concrete Research, V. 29, No. 4, 1999, pp. 611-614.
7. Al-Khaiat, H., and Haque, M. N., "Effect of Initial Curing on Early Strength and Physical Properties of a Lightweight Concrete," Cement and Concrete Research, V. 28, No. 6, 1998, pp. 859-866.
8. Bouguerra, A.; Ledhem, A.; de Barquin, F.; Dheilly, R. M.; and Queneudec, M., "Effect of Microstructure on the Mechanical and Thermal Properties of Lightweight Concrete Prepared from Clay, Cement, and Wood Aggregate," Cement and Concrete Research, V. 28, No. 8, 1998, pp. 1179-1190.
9. Lu-shu, K.; Man-qing, S.; Xing-Sheng, S.; and Yun-xiu, L., "Research on Several Physico-Mechanical Properties of Lightweight Aggregate Concrete," International Journal of Lightweight Concrete, V. 2, No. 4, 1980, pp. 185-191.
10. Neville, A. M., Properties of Concrete, 4th Edition, Wiley, Harlow, UK, 1996, 844 pp.
11. Founie, A., "Pumice and Pumicite," U.S. Geological Survey Minerals Year Book, 2005, pp. 128-129.
12. Grasser, K., and Minke, G., "Building with Pumice," Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ) GmbH, Eschburn, Germany, 1990, 64 pp.
13. Cabrera, J. G.; Rivera-Villarreal, R.; and Ravindrarajah, R. S., "Properties and Durability of a Pre-Columbian Lightweight Concrete," Fourth CANMET/ACI International Conference on Durability of Concrete, SP-170, V. M. Malhotra, ed., American Concrete Institute, Farmington Hills, MI, 1997, pp. 1215-1230.
14. Rivera-Villarreal, R., and Cabrera, J. G., "Microstructure of Two-Thousand-Year-Old Lightweight Concrete," High-Performance Concrete: Performance and Quality of Concrete Structures, SP-186, V. M. Malhotra, P. Helene, L. R. Prudencio Jr., and D. C. C. Dal Molin, eds., American Concrete Institute, Farmington Hills, MI, 1999, pp. 186-200.
15. Litvan, G. G., "Further Study of Particulate Admixtures for Enhanced Freeze-Thaw Resistance of Concrete," ACI JOURNAL, Proceedings V. 82, No. 5, Sept.-Oct. 1985, pp. 724-730.
16. Yeginobali, A.; Sobolev, K. G.; Soboleva, S. V.; and Tokyay, M., "High Strength Natural Lightweight Aggregate Concrete with Silica Fume," Sixth CANMET/ACI/JCI Conference: Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, SP-178, V. M. Malhotra, ed., American Concrete Institute, Farmington Hills, MI, 1998, pp. 739-758.
17. Humay, F. K., and Durrani, A. J., "Experimental Study of Perforated Infill Panels for Retrofitting Flat Plates," ACI Structural Journal, V. 98, No. 5, Sept.-Oct. 2001, pp. 727-734
18. Stepanova, V. F., "Protection of Steel Reinforcing by the Passivating Effect of Lightweight Concrete," Durability of Concrete: Second International Conference, SP-126, V. M. Malhotra, ed., American Concrete Institute, Farmington Hills, MI, 1991, pp. 1135-1146.
19. Carmichael, J., "Pumice Concrete Panels," Concrete International, V. 8, No. 11, Nov. 1986, pp. 31-32.
20. ACI Committee 211, "Standard Practice for Selecting Proportions for Structural Lightweight Concrete (ACI 211.2-98)," American Concrete Institute, Farmington Hills, MI, 20 pp.
21. Phan, L. T., "Fire Performance of High Strength Concrete: A Report of the State-of-the-Art," Building and Fire Research Laboratory, National Institute of Standards and Technology, Gaithersburg, MD, 1996, 105 pp.
22. CSA 23.3-94, "Design of Concrete Structures," Canadian Standard Association, Rexdale, ON, Canada, 220 pp.
23. Shideler, J. J., "Lightweight Aggregate Concrete for Structural Use," ACI JOURNAL, Proceedings V. 54, No. 10, Oct. 1957, pp. 299-328.
24. FIP, FIP Manual of Lightweight Aggregate Concrete, 2nd Edition, Surrey University Press, UK, 1983, 4 pp.
25. Whiting, D., "Permeability of Selected Concretes," Permeability of Concrete, SP-108, D. Whiting and A. Walitt, eds., American Concrete Institute, Farmington Hills, MI, 1989, pp. 195-222.
26. ACI Committee 301, "Specifications for Structural Concrete (ACI 301-99)," American Concrete Institute, Farmington Hills, MI, 1999, 41 pp.
27. Chan, Y. N.; Luo, X.; and Sun, W., "Compressive Strength and Pore Structure of High-Performance Concrete after Exposure to High Temperature up to 800 °C," Cement and Concrete Research, V. 30, No. 2, 2000, pp. 247-251.
28. Hossain, K. M. A., and Lachemi, M., "Residual Strength and Durability Characteristics of Volcanic Ash Concretes at High Temperature," ACI Materials Journal, V. 101, No. 6, Nov.-Dec. 2004, pp. 493-500.
29. Sarshar, R., and Khoury, G. A., "Material and Environmental Factors Influencing the Compressive Strength of Unsealed Cement Paste and Concrete at High Temperature," Magazine of Concrete Research, V. 45, No. 162, 1993, pp. 51-61.
30. Castillo, C., and Durrani, A. J., "Effect of Transient High Temperature on High-Strength Concrete," ACI Materials Journal, V. 87, No. 1, Jan.-Feb. 1990, pp. 47-53.
31. Sanjayan, G., and Stocks, L. J., "Spalling of High-Strength Silica Fume Concrete in Fire," ACI Materials Journal, V. 90, No. 2, Mar.-Apr. 1993, pp. 170-173.
32. Hossain, K. M. A., "Fire Durability of Lightweight Volcanic Pumice Concrete with Special Reference to Thin Walled Filled Sections," Durability of Building Material and Components 8, Canadian Institute for Scientific and Technical Information (CISTI), NRC No. 42738, NRC Research Press, 1999, pp. 149-158.
ACI member Khandaker M. A. Hossain is an Associate Professor in the Department of Civil Engineering at Ryerson University, Toronto, ON, Canada. His research interests include blended cement and concrete, structural application of self-consolidating concrete, composite construction, and finite element analysis of reinforced and composite structures.
ACI member Mohamed Lachemi is a Canada Research Chair in Sustainable Construction and a Professor in the Department of Civil Engineering at Ryerson University. He is a member of ACI Committee 231, Properties of Concrete at Early Ages. His research interests include the use of high-performance materials in the built infrastructure including the development and use of self-consolidating concrete in construction.