Utilization of waste Autoclaved Aerated Concrete as lighting material in the structure of a green roof

Construction and Building Materials 69 (2014) 351–361

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Utilization of waste Autoclaved Aerated Concrete as lighting material in the structure of a green roof Franco Bisceglie a,⇑, Elisa Gigante b, Marco Bergonzoni b a b

Department of Chemistry, University of Parma, Parco Area delle Scienze, 17/A, 43124 Parma, Italy Department of Civil, Environmental, Land Management Engineering and Architecture, DICATeA, University of Parma, Parco Area delle Scienze, 181/A, 43124 Parma, Italy

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Reuse of waste AAC with natural soil

to create a neutral substrate for green roofs.  AAC–peat mixture respects the UNI 11235 standard.  AAC–peat mixture is similar to the mixture with natural stones and natural soil.  Utilization of granular waste AAC reduces the consumption of natural materials.  AAC cost, obtained by waste materials, is lower than the natural materials one.

a r t i c l e

i n f o

Article history: Received 2 April 2014 Received in revised form 27 June 2014 Accepted 23 July 2014 Available online 15 August 2014 Keywords: Autoclaved Aerated Concrete Waste management Green roof Chemical analysis Physical analysis

Green Roof

Waste Autoclaved Aerated Concrete a b s t r a c t Usually, green roofs are made with natural materials, as lapillus or pumice rock, which have the same porous characteristic of the granular AAC. To verify if this substitution was a good hypothesis, we have carried out chemical and physical analysis on a mixture of 70% of soil and natural peat and 30% of granular AAC. We compared all the results with natural green roof characteristics, finding a good connection between these two groups of values. In fact the pH value of the water extract is of 7.23; the organic matter is less than 4.08; the apparent density is 459.2 kg/m2; the demand for high water retention capacity is completely satisfied by the value of 222.62% of the mass of water absorbed relative to the mass of the dry sample. For this reason, we think that the introduction of granular waste AAC within the structure of a green roof could help to reduce industrial wastes and respects the European ideas of a sustainable future. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction During the end of XX century, the European Union published new directives on environment to promote a sustainable future with prevention and reduction of the waste production. These new directives pressed all manufacturing sectors to research new way to reuse or recycle their production waste. With this aim we ⇑ Corresponding author. Tel.: +39 0521905418; fax: +39 0521905557. E-mail address: [email protected] (F. Bisceglie). http://dx.doi.org/10.1016/j.conbuildmat.2014.07.083 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.

analyzed waste of Autoclaved Aerated Concrete (AAC) produced by an Italian company, to reuse it as lighting material within the structure of green roofs. Similar studies are also in progress in other countries. 1.1. Autoclaved Aerated Concrete Autoclaved Aerated Concrete – AAC – (detail shown in Fig. 1) is a special kind of concrete, where cement, lime, water, sand and a blowing agent are mixed. It was born in the 20s in Sweden as an

352

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

and non-crystalline CSH, growth of semi-crystals of tobermorite and recrystallization of tobermorite in a solid phase. During the cooking period, the presence of aluminum influences different aspects of the final product of tobermorite: it has a significant effect on the average length of the chains of silicate and on the stability of the CSH structures; it reduces quartz and lime reactivity, replacing within the structures of the oxides.

1.2. Green roofs

Fig. 1. Detail of a brick made in Autoclaved Aerated Concrete. High porosity characterizes its light structure.

alternative material used for buildings. Subsequently, it spread worldwide as an eco-friendly material and with a negligible environmental impact because it is completely mineral and produced with greatly abundant components. Porosity is the most interesting feature, whereby this material presents: – – – – – –

Lightness. Ease of transport and installation. Ductility. Thermal insulation properties. Acoustic insulation properties. Transpiring properties.

In the AAC production process, during the hardening period, aluminum is added and a gas develops within the structure of the concrete [1]. The pH reaction environment is basic and aluminum, due its amphoteric behavior, melts and oxidizes [2], allowing gaseous hydrogen development:

2Al ðsÞ þ 3CaðOHÞ2 þ 6H2 O ! 3CaO  Al2 O3  6H2 O þ H2 ðgÞ Later, porous blocks, still moist, are cooked within autoclaves with quite low temperature, 190 °C, and high-pressure, 12 bar, for about 14 h. In this time, there is a hydrothermal treatment [3] and the formation of tobermorite, whose chemical formula is

5CaO  6SiO2  5H2 O In other words, crystals of calcium silicates, which strongly influence mechanical properties of the final product, are formed. During the initial cooking stage, the silica-based compounds, highly reactive, such as amorphous silica, improve the polymerization of CSH compounds, preceding the direct formation of tobermorite. On the other hand, the presence of aluminum delays the initial formation of the CSH-based compounds reducing the solubility of quartz, but accelerating a direct formation of tobermorite. Tobermorite formation was studied in 2011 [4], where scientists studied a sample of AAC, using X-ray diffraction, during the hardening period. The results suggested different reaction paths for the formation of tobermorite in an AAC system. In one of them, tobermorite appears when the temperature reaches 190 °C, in other words, when amorphous gel-based CSH begin to decrease. The amount of tobermorite is maximum at the end of the autoclave process. For this reason, it is possible to believe that the most of crystalline tobermorite phase results from the CSH amorphous phase. Recently, Houston et al. reported that the formation of tobermorite proceeds through three stages [5]: formation of amorphous

A green roof is a building roof partially or totally covered by a garden. Its main aim is recovering the most important areas within the city that otherwise, would be sterile, transforming the districts appearance just not only in an aesthetical way but also in terms of quality of life [6]. In general, there are two types of green roofs systems: the intensive and extensive green roof [7]. Intensive green roofs are designed with deep substrates, at least 30 cm, and they can withstand a wide range of plants, that require, however, a frequent maintenance. Due the high thickness of the layer, it has a great heat insulating property, but it must be designed with a vapor barrier layer. In the winter season, in fact, water vapor passes from the inside of the heated building to outside, finding the obstacle of an impermeable barrier. In this case, it could condense, moistening the waterproof layer that would lose its main feature. To avoid condensation, this steam must be locked and retained before reaching the insulating layer. Extensive green roof is composed of draining material and it is thinner (5–12 cm), than the intensive one and it can support the growth of a vegetation resisting to drought and extreme conditions, so requiring less maintenance. Due its lightness (60–250 kg/m2), this system requires a small, or even non-existent, additional support structure and it has great potential to extend in width. For each part of the garden structure, detailed studies are carried out, to ensure that dangerous puddles do not form, or to create a good quality cultivation layer and rich in nutrients for plants. Typically, a mixture of different material [8,9] forms it: an organic part of soil and a draining part of natural materials, such as lapillus and pumice stones. From the reduction of negative environmental effects point of view, green roofs offer many opportunities. The main benefits of a vegetated roof are:  It cools down the temperature of the air, absorbed by the atmosphere, through plants evapotranspiration, allowing the reduction of the urban ‘‘heat island’’ effect [10].  It provides passive cooling to the built environment [11] that needs 25% less of conditioned air during in summer and, consequently, provides the reduction of CO2 emissions. It increases the thermal resistance of the roof, important benefit in winter [12]. The presence of various layers works as filter, reducing the CO2 crossing.  It reduces the annual rainwater runoff from roofs, typically 60–70% less [13]. The progressive overbuilding has reduced the drainage capacity of urban areas, which render the sewers obsolete. With new roofs transformed into green roofs, the water absorption increases during raining events, thus preventing the overflow of sewerage systems [14].  It contributes to local biodiversity, providing a perfect habitat for invertebrates that could live in soil layers and for groundnesting birds, such as skylarks.  It improves the life quality of communities developed in high-density areas, contributing to a greener urban environment and reducing the transmission of noise inside the building.

353

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

The goal of this study is to replace the natural lightening materials (lapillus and pumice stone) of a green roof structure with granular waste AAC. To do this, we have verified that the soil– concrete mixture has characteristics similar to the soil–natural material mixture. 2. Materials Examined materials are AAC and sphagnum peat brown. AAC is Autoclaved Aerated Concrete rejected by the manufacturing process of an Italian company, AirBeton SpA, committed to dispose of it. This company produces two kinds of different bricks, first, for their density, which gives to the brick different basic characteristics. In Tables 1 and 2 there is a summary of the main characteristics of this two kind of bricks. During the production process of these bricks, three main critical events bring to the fracture of the final product: 1. ‘‘delay cracks’’: bricks arrive in cutting step with a consistence greater than the designed one, so fractures propagate in the tangential direction of maximum stress; 2. ‘‘beforehand cracks’’: bricks arrive in the cutting step when they are still too wet; immediately after the passage of the wire, the material compacts due to the weight of the upper layers, and then finally solidify in autoclave; 3. ‘‘hydrogen cracks’’: hydrogen bubbles align within the structure during the drying step creating fragile areas.

3. Chemical analysis – methods The first examined chemical feature of the mixture is the evaluation of pH, searching for the right combination of soil and concrete that could create a neutral environment for plants. To do this, we followed the steps specified in the UNI EN 13037 entitled ‘‘Soil improvers and growing media – Determination of pH’’ [R2]. Then, we evaluated the electrical conductivity and the presence of organic substances on the AAC sample, following, respectively, the UNI EN 13038, ‘‘Soil improvers and growing media – Determination of electrical conductivity’’ [R3] and the UNI EN 13039, ‘‘Soil improvers and growing media – Determination of organic matter and ashes’’ [R4]. 3.1. Determination of pH 3.1.1. Sample identification and methods As required by the standard [R2], the sample has a mass of 25 g and this should be added to 300 ml of distilled water. We prepared five different samples: – 25 g (60 ml) of granular AAC, with a size included between 3 and 10 mm; – 25 g of brown sphagnum peat; – 70% peat + 30% AAC: about 7.5 g of peat and 17.5 g of AAC, 25 g in total; – 60% peat + 40% AAC: about 10 g of peat and 15 g of AAC, 25 g in total; – 80% peat + 20% AAC: about 5 g of peat and 20 g of AAC, 25 g in total.

For all this causes, the last step of the process creates some wasted bricks, with fractures and irregularity. All these wastes are crushed into granules and collected into two sizes: – Particle (waste AAC) with a size included in the range 0.5–1 mm (Fig. 2). – Particle (waste AAC) with size included in the range 3–10 mm (Fig. 3). Peat is composed by organic matter, derived from vegetable matters decomposition. The first goal was to create a mixture (soil)–(waste AAC) with a neutral pH, so the first step was to find a right percentage of AAC in the substrate. For a green roof design, the European Committee of Standardisation published the UNI 11235 standard [R1]. To check chemical and physical characteristics, all the tests were performed according to the world regulations.

The sample, transferred into a plastic container, is added to 300 ml of distilled water. The container is hermetically closed with a stopper and it is put in a shaker for 1 h, maintaining a constant temperature around the (22 ± 3)°C. This procedure ensures that

Table 1 Summary of the main technical characteristics of brick with density 350 kg/m3. Technical characteristics Dimensions

Unit of measure Lenght High Thickness

L H T

q

Dry density Dry thermal conductivity Design thermal conductivity Mass Acoustic insulation Fire reaction Fire resistance Curtain wall Bearing wall

AIRBETON 350 mm

k10,dry ku m Rw Euroclass EI REI

600 250 240 kg/m3 350 W/m K 0.084 W/m K 0.105 kg/m2 116 dB 44 A1 (incombustible) min 240 min /

300

365

400

450

480

137 47

160 49

172 50

190 51

200 52

240 /

240 /

240 /

240 /

240 /

Table 2 Summary of the main technical characteristics of brick with density 500 kg/m3. Technical characteristics

Unit of measure

AIRBETON 500

Dimensions

L H T

mm

q

kg/m3 W/mK W/mK kg/m2 dB

625 250 50 80 500 0.12 0.145 51 66 33 36 A1 (incombustible) / 120 / /

Lenght High Thickness

Dry density Dry thermal conductivity Design thermal conductivity Mass Acoustic insulation Fire reaction Fire resistance Curtain wall Bearing wall

k10,dry ku m Rw Euroclass EI REI

min min

100

120

150

200

76 38

86 40

101 42

126 45

600 250 240 450 0.108 0.13 140 47

240 /

240 /

240 /

240 120

240 180

300

365

400

167 49

198 51

212 52

240 240

240 240

240 240

354

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

availability [15]: they create an opposition to the water permeability in the direction of the deep layers, holding it in a reversible way and contributing to the aggregation of mineral particles. Organic compounds intervene on ion exchange, making a more difficult solubilization and elements leaching. The experiment, described in this study, is based on placing the sample in an oven at a temperature gradually increasing and weighing its specimen; a weight loss should be noted due to the organic part that degrades before all other substances of which the sample is made. A fundamental aspect that should not be overlooked in this test is that the standard [R4] prescribes the study of organic matter to design a growing substrates without inert cementitious material. For this reason, we performed a Thermo Gravimetric Analysis, TGA, to deepen the study of our sample.

Fig. 2. Available AAC with size included between 0.5 mm and 1.00 mm.

3.3.1. Sample identification and methods As required by the standard [R4], the AAC sample has a mass of 5 g and, within a crucible of known mass (m0), it is placed in an oven at 103 ± 2 °C for 4 h. Removed from the oven, it is weighed again to assess the volatilization of organic matter. The crucible and its content are placed in the oven at the same temperature for another hour, and then the specimen is re-weighed. The procedure is repeated until the difference between two successive weigh measurements is less than 0.01 g. This value is called m1. Then, the sample is placed in the oven at 450 ± 25 °C for 6 h, to obtain, after the extraction, the value of the weight. The value is compared with a set of successive weightings, related to 1-h heating cycles at the same temperature, until the difference between two successive weigh measurements is less than 0.01 g. At this point, this new mass value is called m2. 3.4. Thermal Gravimetric Analysis (TGA)

Fig. 3. Available AAC with size included between 3 mm and 10 mm.

the water absorbs the solid sample characteristics dispersed inside it. A pH-meter values pH from the suspension inside the plastic container. To ensure the correctness of the test, this is repeated on three specimens of the same material. 3.2. Determination of electrical conductivity 3.2.1. Sample identification and methods As required by the standard [R3], the sample of AAC, with a grain size between 3 and 10 mm, has to occupy a 60 ml volume. Defined the right amount, the sample is transferred into a plastic container and it is added to 300 ml of distilled water, with an electrical conductivity value of 0.2 mS/m. The container is hermetically closed with a cap and it is put into a shaker for 1 h, maintaining a constant temperature around the (22 ± 3)°C. Following the instructions of the UNI standard, the test is repeated for three different samples of the same material, thus to identify and mediate any error. 3.3. Determination of organic matter The organic matter has an important role in the soil structure, because organic compounds indirectly act on the elements

A TGA (Thermo Gravimetric Analysis) [2,16] is a test of progressive loss weight measurement in relation to temperature variations. Weight variations are controlled by a steady increase in temperature (5 °C per minute). From the results, we can check when the sample loses its weight. If the lost is shown at low temperature, it is due to hydration water evaporation, if it is shown at high temperature, the presence of substance organic is confirmed. 3.5. X-Ray Diffraction (XRD) To complete our chemical analysis, we performed an X-ray diffraction study ([2,4,16,17]) to identify the nature of the chemical species contained in our waste AAC sample with a grain size between 0.5 mm and 1 mm. An X-ray test is based on the application of Bragg’s law: 2dsin h = nk, in other words, on the relation that links d parameter, representing the distance between two adjacent planes of a crystal structure of a chemical species, and the wavelength of a diffracted ray (k) directed onto the sample with an angle of incidence called h. If a species is polycrystalline diffracts the beam in the direction of a detector that will output the value of perceived intensity. 4. Chemical analysis – results and discussion 4.1. Determination of pH The results, averaged on tests carried out on the same samples, are summarized in Table 3. The results show an alkaline pH of the concrete, as expected. For this reason AAC is mixed with a large amounts of peat; peat

355

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361 Table 3 Results of pH analysis made on five different samples. Sample

Peat weight (g)

AAC weight (g)

Total weight (g)

pH

AAC Peat 20%AAC 80%peat 30%AAC 70% peat 40%AAC 60% peat

0 25.59 20.25 17.83 14.90

24.42 0 4.96 7.45 10.15

24.42 25.59 25.21 25.28 25.05

10.08 6.67 6.97 7.23 7.59

amounts and pH value are connected with an inversely proportional function. To perform all the other tests that require the mixture as a sample, we used the combination of 30% AAC and 70% soil, with a neutral pH: 7.23.

Table 5 The table is the summary of the test of evaluation of the organic substance within a AAC sample, performing the weighing step in an external environment and in a sealed environment. Measurements

Weight (g) Sample 1 External environment

Weight (g) Sample 2 Glove box

m0 m1 m2

56.890 61.079 60.908

56.893 61.213 60.958

Wsos.org.

4.082%

5.903%

4.2. Determination of electrical conductivity Table 4 contains sample names and their corresponding value of electrical conductivity. These results highlight that AAC is composed by many ionic particles as a consequence of the high conductivity in the fluid phase. High ionic particles concentration could hinder the absorption of water and nutrients by the roots. 4.3. Determination of organic matter

Fig. 4. TGA profile.

The results obtained from this test are summarized in Table 5 that shows the values of masses noted during the experiment and the percentage of the volatilized substance. The measurement step is made in the external environment and in a sealed glove-box environment to avoid the samples to absorb moisture. The results show that the value of lost mass in an open environment is less because AAC has a high hygroscopic degree. The standard [R4], however, indicates that this percentage is entirely made up of organic components of the sample. In our case, it is possible that part of this percentage corresponds to a weight loss corresponding to the evaporation of hydration water. 4.4. Thermal Gravimetric Analysis (TGA) To check if part of the weight loss observed during the determination of the organic matter was due to water loss, we performed a TGA analysis. The result (Fig. 4) shows the trend of the weight loss (%) of an AAC sample compared to a temperature increase from 20 °C to 500 °C. At the beginning, the sample has a mass of 5.6 mg. The weight loss as a function of the temperature is linear within three ranges. A first range between 20 °C and 60 °C with a total lost weight of 14.53%; a second range between 60 °C and 250 °C with a total lost weight of about 2.5%; a last range between 250 °C and 500 °C with a total lost weight of 1.2%. AAC is mainly composed of tobermorite [3,17] whose chemical formula is Ca5Si6O16(OH)24H2O that is a calcium silicate hydrate, with a 9.86% by molecular weight of water joined with hydrogen bonds. Looking at the graph resulting from the TGA, it is possible to demonstrate that the initial weight loss, about 14.53% by weight Table 4 Electrical conductivity. Granular size of AAC falls in a range between 3 and 10 mm. Sample no.

Content

Weight (g)

Conductivity (lS/cm)

1 2 3 4

Distilled water AAC + distilled water AAC + distilled water AAC + distilled water

/ 24.73 25.33 23.61

0.054 808 872 785

of the sample, can be ascribed to the evaporation of moisture and pore water. Until 250 °C the loss of weight (5.02%) depends to hydrogarnet groups. In the total 20% of lost of sample initial weight there are both loss of free water, present in the sample, and water joined with hydrogen bonds. The final 2% of lost weight (in a temperature range between 250 °C and 550 °C) could be the Portlandite Ca(OH)2 dehydration. The organic substance is contained just in the surface level due to carbonation processes that take place with the reaction of calcium with carbon dioxide. Organic content starts to decompose at higher temperatures. The result justifies the initial assumption that AAC should not contain dissolved organic matter and, according to the standard [R4], the loss of weight has not an organic nature. 4.5. X-Ray Diffraction (XRD) The test result, shown in Fig. 5, represents the intensity of the crystal diffracted ray variation relating to the h variation. The large number of visible peaks shows that the sample presents a crystalline structure. With a more detailed analysis, from the quantitative point of view it is possible to say, as expected, that the tobermorite is the most important chemical species. However, there are other crystalline structures composed of calcium oxides or silicon and they are not tied together to form the complex structure of tobermorite. In a final analysis, it is possible to verify that at the beginning of the production process, aluminum does not react with the complex structure of tobermorite, but with single oxides and hydroxides, justifying the basic nature of the sample.

5. Physical analysis – methods To make a physical description of the mixture, composed of 70% of peat and 30% of AAC, we have carried out the following laboratory tests:

356

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

– two samples of AAC with a volume of 275 ml; – two samples of the mixture 30%AAC and 70%peat: 250 ml of the mixture, about 27.85 g of dry soil and 43.62 g of AAC, for a total of 71.47 g.

θ Fig. 5. X-ray test on a sample of AAC. The graph presents the relation between intensity registered by the detector and h variation. The peaks with the star represent the tobermorite phase while peaks with circle represent the silica phase and peaks with triangle the portlandite phase.

– – – –

Water content measurement. Particle density and water absorption determination. Compaction experiment (Proctor test). Permeability determination.

following, for each of them, a specific reference standard (respectively [R5–R8]). The relationship existing between soil and water is the key feature in our study and analysis. The soil retains water due to two chemical–physical phenomena: imbibition and capillarity. The first is due to the particles properties having an electric charge to attract the polar water and, due of this characteristic, the water tends to form a veil; capillarity is a physical process resulting from gravitational forces, from adhesion and cohesion forces. The importance of absorption for a cultivation soil is related to the ability to retain water that will be released to the plants when needed. 5.1. Water content measurement This test consists in the measurement of the wet weight and the dry weight of a specimen made dry in the oven at about 100 °C and on the determination of the lost weight, given by water evaporation. 5.1.1. Sample identification and methods The specimen is placed in a container with known mass (mc) and the total mass is noted as m1. The sample is placed in oven at 105 °C for a minimum of 16 h to be dried to constant mass. The variation of the mass must be less than 0.1% of the specimen mass. The specimen removed from the oven must be placed inside the dryer to cool down to environment temperature and weighed, recording the value of the container and the dry specimen (m2). 5.2. Particle density and water absorption determination 5.2.1. Sample identification According to the standard [R6], the sample must have a particle size included in a range between 4 mm and 31.5 mm. For this reason, the material is sieved. In the present case, we used samples with a grain size between 4.75 mm and 10. The volume of material have to be included in a range between 0.5 and 0.6 l. Due to the pycnometer capacity of 500 ml, our samples occupy a volume between 250 ml and 300 ml. We prepared two different experiment:

5.2.2. Methods The dry sample is placed inside the pycnometer (tare noted as m1). Then the pycnometer and its contents are weighed, noting the value obtained as m2. The pycnometer is filled with water, up to the reference mark, and the counting time starts. After 5 min, we shake the pycnometer to eliminate air bubbles that have formed between the grains. The container is filled again, up to the reference mark, and carefully the outer surface is dried. The specimen is weight and the new mass value is M5. All the operations described above are repeated after 24 h and it is assigned to the last value of the mass the symbol M24. After the final measurement, the pycnometer is emptied just from the water and the aggregate are moved on a dry tissue to remove surface water, rolling it gently. Wet aggregate mass is called Mw. 5.3. Compaction test (Proctor test) Normally, soil consists of solid particles and intergranular voids [18] and its volume can be reduced. This action increases the number of contact points between solid particles, improving the strength of the soil. If the compaction effort is kept constant, the dry density will vary with the variation of water content. Density of soils with low water content rises rapidly with the increasing of moisture because particles slide over each other due to the lubricating effect of the water. However, increasing the water content, the value of dry density decreases because the water tends to keep separate the solid components, increasing the intergranular spaces. 5.3.1. Sample identification The sample is the mixture (peat + AAC) in the proportion of 70% peat + 30% AAC. The quantity of the sample is defined by a standard [R7] and it is in relation to the size of the cylindrical mold and to the material to be tested. The standard requires 2.5 kg of sample, but, in this case, with lightweight material, to ensure the appropriate volume of evidence, 600 g of material are enough. Therefore: – 70% by volume of dry soil = 381 g; – 30% by volume of dry AAC = 219 g; 5.3.2. Methods A minimum of five specimens with various moisture contents to be able to create the compaction curve are needed. After the drying step, the sample is mixed with water, creating inside a uniform distribution of the water content. The mold that will contain the sample to be compacted is weighed (tare). The moist mixture is divided into 3 equal quantities that will be neatly placed inside the mold and they will be progressively compacted with 25 strokes of a steel pestle. Hits are seated in a decentralized point of the specimen that, rotating, will result uniformly compacted over the entire surface. At this point, the specimen must be shaped on the surface thus to have the same volume of the mold. The specimen is weighed and the value of the wet gross weight (WGW) is obtained.

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

Wet net weight (WNW) is obtained subtracting the mass of the mold (tare) to the WGW. CEN ISO/TS 17892-1 [R5] describes the method to calculate the moisture. All this operations must be repeated until the mass of the last specimen will be equal or less than the one of the previous specimen. 5.3.3. Source of error Sources of error of a Proctor test are essentially linked to the real moisture value loaded into in the sample and to the uniformity degree within the structure. To measure the value of the water content for a sample compacted with the Proctor test, it is better to select a small portion of the volume pulling it out from the central area of the specimen, assuming that it will be representative of the total. The uniformity degree influences the test, because successive layers of soil with different water content will be compacted in a different way. 5.4. Porosity determination The term ‘‘porosity’’ (n) means the ratio between the total voids volume (Vv) and the total volume of the structure (Vs), and it is expressed by the percentage of volume occupied by the void [19], that can be filled with air or water.

n ¼ V v =V s The soil porosity ranges between 0 and 1. The 0 value corresponds to a ground completely devoid of pores, formed just from a solid fraction; the value 1 corresponds to the purely hypothetic case of a soil structure consisting just of pores.

357

5.5.1. Sample identification and methods This test requires a permeameter (Fig. 6), a tool with a mold for the specimen, with filters both on the base and on the cover, connected to a charging system of water that permeates the specimen and emerges from a nozzle placed in the base of support. The lowering of the water can be directly viewed reading the notches at the top of the burette. Samples must be compacted with the Proctor equipment and sealed inside the mold of the permeameter, making sure to order: base, filter, specimen, filter, cover. To determine the permeability, the load of water must be measured at defined intervals reading the notch corresponding to the reached level. 5.5.2. Error sources A permeability test has several limitations, related to the sample disturbance. It is difficult to recreate in laboratory a sample that represents in a realistic way the degree of resistance to the crossing of a flow, both in the vertical direction and in the horizontal one. In the present case, the compaction degree inside the sample depends to the moisture content created with a Proctor test and it will be higher than the one produced in a real case, where, maybe the soil has a manual compaction. Another source of error is related to the fact that the small size of the sample does not consider the macro-structure of internal storage. Finally, it should be considered, as a source of error the presence of air bubbles in the sample or in the circuits and the use of a sample not completely water saturated. All these sources of error can eventually be resolved by carrying out the test for a large number of specimens, maintaining as uniform as possible the characteristics of the initial constitution.

5.4.1. Sample identification and methods ‘‘n’’ can be determined with measurements of the average specific weight of the solid sample (cs) and the weight of dry volume of the sample (cd), according to the relation

n ¼ 1  cd =cs with – cd (dry specific weight) obtained as result from the relation between Proctor test and calculation of the water content; – cs (average specific weight of the sample) obtained as a result of the absorption test. The value of porosity can be calculated for all samples prepared for the Proctor test. 5.5. Permeability determination Permeability is the measure of the speed of a fluid that flows through a porous structure. The physical law that allows to connect the speed of the flow and the hydraulic gradient is Darcy’s law:

V ¼ K  dH=dL with dH that represents the loss of hydraulic load; dL that represents the stretch of the porous structure where the flow occurs. The flow is reported to the whole section of the sample, not only to the intergranular voids, and for this reason V represents the apparent speed of the flow, lower than the effective one. The proportionality constant K is the hydraulic conductivity, also called coefficient of permeability, that it should be directly calculated for each sample.

Fig. 6. Equipment to realize a test to determinate the value of permeability. In the picture is shown a water reservoir that provides the water to all the samples; there are four graduated burettes where the water level can be directly read; there are three molds connected with plastic tubes with the water load, containing the samples.

358

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

6. Physical analysis – results and discussions

volume of water to reach 250 ml of the pycnometer. At this point, the equation derives from a ratio of the mass of the sample and its density, unknown, and the mass of water that would occupy more volume and its mass density. The results for the bulk density of the samples is reported in Table 8. Density values are slightly larger than that of water (0.9973 Mg/ m3), according to the fact that part of the material floats inside the pycnometer. Water absorption is calculated with the following equation:

6.1. Water content measurement

WF ¼

At first, the test is performed for a sample composed just of peat, to get a preliminary idea of the natural water content in the soil used. To calculate the moisture of the sample there is the following equation:

Table 9 reports the results, as a percentage of the mass of water absorbed respect to the dry mass. Apparently, the result of water absorption by the mixture is very high, but it is justified by the result of the test carried out for the water content just for the specimen containing peat. In fact, the soil has an absorption capacity of about 192% of its dry weight in normal conditions, therefore, the value of 222% relative to the absorption of water by the mixture is an acceptable value.

Table 6 Results of water content measurements. Capsule mass (mc) Capsule + wet soil mass (m1) Capsule + dry soil mass (m2) Moisture ‘‘W’’ (peat)

W¼

(g) (g) (g) %

17.60 36.96 24.23 192.01

ðm1  m2 Þ  100 ðm2  mc Þ

obtaining values of water content expressed as a percentage of the total weight of the dry sample. Results are given in Table 6 that shows all the values of the masses noted during the test, expressed in grams. 6.2. Particle density and water absorption determination The results of these tests are summarized in Table 7, where there are all the mass values recorded during the test. The particle density of the granules of the lightweight aggregate according to the equation:

qa ¼

ðm2  m1 Þ  qw m1 þ ðV p  qw Þ þ m  M

6.3. Compaction test (Proctor test) The results of a Proctor test are contained in Table 10. To draw the compaction curve we calculated the density of the dry sample cd, with the following equation:

cd ¼

WNW cW
¼ W W V  1 þ 100 1 þ 100

where the moisture content ‘‘w’’ was obtained with the following equation:

w¼

is calculated for each sample. The denominator [(m1 + (Vpqw) + m  M)/qw] corresponds to the volume occupied by the sample, corresponding to a lost

M W  ðm2  m1 Þ  100 ðm2  m1 Þ

m1  m2  100 m2  mc

Summing up, we obtained the data shown in Table 11 and in Fig. 7 that represents the compaction curve for the mixture of 70% of soil and 30% of AAC.

Table 7 Samples 1 and 2 are just of granular AAC, while samples number 3 and 4 are made with 30% of AAC and 70% of peat. Measurements

Symbol

Weight (g) Sample 1 (AAC)

Weight (g) Sample 2 (AAC)

Weight (g) Sample 3 (mixture)

Weight (g) Sample 4 (mixture)

Tare pycnometer Dry gross weight (pycnometer and sample) Dry net weight Wet gross weight (pycnometer, sample and water) Wet gross weight (after 5 min) Wet gross weight (after 24 h) Wet net weight

m1 m2 m M M5 M24 Mw

203.4 296.4 93.0 768.3 770.2 773.2 147.2

207.3 300.0 92.7 775.8 780.4 782.9 147.7

203.38 274.84 71.5 700.90 757.9 791.9 239.0

207.30 270.73 71.4 727.80 764.7 798.2 222.0

Table 8 Values of particle density of AAC and the mixture. Sample 1 (AAC)

Sample 2 (AAC)

Sample 3 (mixture)

Sample 4 (mixture)

ca

3.470 Mg/m3

5.066 Mg/m3

0.982 Mg/m3

1.437 Mg/m3

ca (averege)

4.268 Mg/m3

1.209 Mg/m3

Table 9 Percentage of mass of absorbed water respect to dry mass. Samples 1 and 2 are made of AAC; samples 3 and 4 are made of mixture (30% of AAC and 70% of peat). Sample 1 (AAC)

Sample 2 (AAC)

Sample 3 (mixture)

Sample 4 (mixture)

WF

58.280%

59.331%

234.452%

210.793%

WF (average)

58.805%

222.623%

359

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

The curve shows a maximum whose ordinate identifies the maximum dry density. In other words, the maximum compaction degree achieved, while the x-axis represents the optimal moisture content. This curve is built from the approximation of all the obtained data with a second order polynomial trend line. From the graph it is possible to derive the value of the optimal moisture content and its corresponding value of density (Table 12). The value of optimal moisture content of 71.429% is relative to the total dry mass necessary to achieve the best compaction of the sample, in other words the maximum density value.

Table 11 Summary table regarding moisture content and density values necessary to verify the compaction curve. SAMPLES Moisture content (w) Dry specific volume (cd)

2

3

4

5

60.1 0.460

81.3 0.461

100.1 0.459

119.5 0.45

COMPACTION CURVE y = -7·10 -6 x 2 + 0.001x + 0.4235 R² = 0.9951

6.5. Permeability determination

Dry density (g/cm3)

0.465

Table 13 summarizes the results of the test. The value of maximum compaction corresponds to a dry specific weight of 0.460 g/cm3 with a moisture content of 71%, the value of porosity will be included between 61.932% and 61.848%.

0.460 0.455 0.450 0.445 0.440 0.435

Following the Darcy equation, who studies the motion of the water within the porous structure with a variable load of water:

K Ah L

0.430 20

30

40

50

60

70

80

90

100

110

120

Moisture content (%) Fig. 7. Compaction curve obtained by representing all the results of the Proctor test.

where: Q represents the flow rate crossing the sample; K is the proportionality constant, the hydraulic conductivity, permeability; A is the cross sectional area; H is the load of water applied over the sample; L is the length of the sample. A very long tube with very small section (a) is mounted above the specimen and filled until the level h0 at time t0. The equation that describes the motion follows:

dh Q ¼ a  dt Thus:

dh K  A dt ¼  dt a L And finally:

ln

1 23.1 0.44

0.470

6.4. Porosity determination

Q¼

(%) (g/cm3)

h KA ¼  ðt  t 0 Þ h0 aL

K is obtained performing a linear regression of the measurements made during the test with the help of a graph relating the logarithm of the load difference and Dt. Four permeability tests are carried on, on four different specimens with comparable initial conditions.

Table 12 Optimal moisture content and its relative density value, obtained by deriving the equation of the compaction curve. Optimal moisture content (%) Maximum density value (g/cm3)

71.429 0.46

The results of the values of K are shown in Table 14, where is reported the value of the permeability obtained from a test carried out immediately after the assembly of the specimen and the value of the permeability evaluated after 24 h from the assembly. The results are obtained with a linear regression on a pack of readings noted with a scanning time of 1 min (and 3000 in some cases). For the permeability determination, a value obtained on a saturated sample must be used, choosing a value among those calculated for t24. Two time ranges are chosen to perform the test (t0 and t24), as reported in the literature [18]. This, to allow the sample, connected to a water tank placed above of the entire plant, to be fully saturated (Fig. 6). The value of h0 used for the test is of 139.5 cm, equivalent to the height from the beginning of a graduated burette to the nozzle of the mold.

Table 10 Results of the test of water content determination and Proctor test. The specimen is the mixture of 70% of peat and 30% of AAC. All the specimen are prepared with an increasing water content. Mold weight (tare) Mold volume (V)

(g) (cm3)

SAMPLES Proctor test

Water content determination

Mold weight + wet compacted sample (WGW) Wet net weight (WNW) Wet specific volume (WNW/V) cw Capsule weight (mc) Capsule weight + wet sample (m1) Capsule weight + dry sample (m2)

(g) (g) (g/cm3) (g) (g) (g)

3528 947 1

2

3

4

5

4045 517 0.55 18.09 39.01 35.09

4226 698 0.74 17.34 40.64 31.89

4320 792 0.84 17.75 42.86 31.6

4397 869 0.92 17.84 51.56 34.69

4459 931 0.98 18.03 56.22 35.43

360

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361

Table 13 Summary table for the test to determine the porosity of the mixture prepared for the Proctor test (70% peat and 30% AAC). Specific weight of the solid sample (cs)

g/cm3

1.209 1

2

3

4

5

g/cm3 (%) %

0.443 23.1 63.308

0.460 60.1 61.932

0.461 81.3 61.848

0.459 1001 62.975

0.447 119.5 62.954

SAMPLES Dry specific weight (cd) Moisture content (w) Porosity (n)

Table 14 Results obtained by performing a permeability test on four different sample prepared with 70% of peat and 30% of AAC and with an optimal moisture content obtained with a Proctor test. SAMPLES k (t0) k (t24)

mm/min mm/min

1

2

3

4

0.127 0.061

0.058 0.045

0.122 0.054

0.043 0.030

 The value of the porosity should be greater than 75%, while the result obtained is 62%. These differences, however, are not ascribed to real deviations from the request, but ascribed to the limits inherent to the Proctor test. The found results allow us to state that granulated AAC waste can be used as lightning material within a structure for a green roof.

7. Conclusion All the analyzed physical and chemical characteristics of the AAC waste are comparable to those reported in literature for a natural lightening material such as lapillus or pumice rock. The UNI 11235 [R1], which lists the characteristics of a substrate of a green roof (Section 8.9 about materials and components of the cultural substrate): – Density values included between 350 g/l and 1000 g/l, as written in the UNI EN 13041 [R9]. – Permeability, as in the UNI EN 1097-6 [R6]:  For intensive green roofs > 0.3 mm/min.  For extensive green roofs > 0.6 mm/min.  Water retention, following the UNI EN 13041 [R9].  Total porosity P 75%, following the UNI EN 13041 [R9].  Air volume at 10 cm of water column P 18%, UNI EN 13041 [R9].  Water volume at 10 cm of water column, UNI EN 13041 [R9]: – for intensive green roofs P 40%; – for extensive green roofs P 30%; – pH value in according with the UNI EN 13037 [R2];

Organic content, in according with the UNI EN 13039 [R4]. The results, compared with the most important data of natural lightening material normally used, are the following:  pH value is of 7.23, neutral. This ensures the survival of Crassulaceae plants such as the aeonium, the Graptopetalum paraguayense or the rubrotinctum, selected for their strong resistance to extreme conditions, such as strong sunshine, and the scarcity of water.  The organic matter is less than 4.08%, ensuring a lack of interaction with the ground and with the water resources, that are not adsorbed by nutritive elements external to the soil.  The apparent density of 459.2 kg/m2 is included in the range between 350 kg/m2 < D < 1000 kg/m2 imposed by the legislation on soil improvers and growing media [R1].  The demand for high water retention capacity is completely satisfied by the value of 222.62% of the mass of water absorbed relative to the mass of the dry sample. Two values, however, do not meet the requirements:  The permeability value equal to 0.048 mm/min is lower than required, that should be > 0.3 mm/min.

Acknowledgements The authors acknowledge the University of Parma for financial support and Airbeton SpA for providing AAC. References [1] Kus H, Carlsson T. Microstructural investigations of naturally and artificially weathered autoclaved aerated concrete. Cem Concr Res 2003;33:1423–32. [2] Wongkeo W, Thongsanitgarn P, Pimraksa K, Chaipanich A. Compressive strength flexural strength and thermal conductivity of autoclaved concrete block made using bottom ash as cement replacement materials. Mater Des 2012;35:434–9. [3] Matsushita F, Aono Y, Shibata S. Carbonation degree of autoclaved aerated concrete. Cem Concr Res 2000;30:1741–5. [4] Matsui K, Kikuma J, Tsunashima M, Ishikawa T, Matsumo S, Ogawa A, et al. In situ time-resolved X-ray diffraction of tobermorite formation in AAC: influence of silica source reactivity and Al addition. Cem Concr Res 2011;41:510–9. [5] Houston JH, Maxwell RS, Carroll SA. Transformation of meta-stable calcium silicate hydrates to tobermorite: reaction kinetics and molecular structure from XRD and NMR spectroscopy. Geochem Trans 2009;10:510–9. [6] Zhanga X, Shen L, Tam VWY, Wing Yan Lee W. Barriers to implement extensive green roof systems: a Hong Kong study. Renew Sustain Energy Rev 2012;16:314–9. [7] Yang J, Wang Z. Physical parameterization and sensitivity of urban hydrological models: application to green roof systems. Build Environ 2014;75:250–63. [8] Nagase A, Dunnett N. The relationship between percentage of organic matter in substrate and plant growth in extensive green roofs. Landscape Urban Plan 2011;103:230–6. [9] Savi T, Andri S, Nardini A. Impact of different green roof layering on plant water status and drought survival. Ecol Eng 2013;57:188–96. [10] Molineaux CJ, Fentiman CH, Gange AC. Characterizing alternative recycled waste materials for use as green roof growing media in the UK. Ecol Eng 2009;35:507–1513. [11] Zinzi M, Agnoli S. Cool and green roofs. An energy and comfort comparison between passive cooling and mitigation urban heat island techniques for residential buildings in the Mediterranean region. Energy Build 2012;55:66–76. [12] Parizotto S, Lamberts R. Investigation of green roof thermal performance in temperate climate. A case study of an experimental building in Florianópolis city, Southern Brazil. Energy Build 2011;43:1712–22. [13] Gregoire BG, Clausen JC. Effect of a modular extensive green roof on stormwater runoff and water quality. Ecol Eng 2011;37:963–9. [14] Stovin V, Vesuviano G, Kasmin H. The hydrological performance of a green roof test bed under UK climatic conditions. J Hydrol 2012;414–415:148–61. [15] Nagase A, Dunnett N. The relationship between percentage of organic matter in substrate and plant growth in extensive green roofs. Landscape Urban Plan 2011;103:230–6. [16] Wongkeo W, Chaipanich A. Compressive strength microstructure and thermal analysis of autoclaved and air cured structural lightweight concrete made with coal bottom ash and silica fume. Mater Sci Eng, A 2010;527:3676–84. [17] Rios CA, Williams CD, Fullen MA. Hydrothermal synthesis of Hydrogarnet and tobermorite at 175 °C from kaolinite and metakaolinite in the CaO–Al2O3– SiO2–H2O system: a comparative study. Appl Clay Sci 2009;43:228–37. [18] K.H. Head, Manual of soil laboratory testing, second ed., MA (Cantab), C. Eng. FICE, FGS, John Wiley & Sons, Inc., NY, Toronto.

F. Bisceglie et al. / Construction and Building Materials 69 (2014) 351–361 [19] P.L. Raviolo, Il laboratorio geotecnico – procedure di prova, elaborazione e acquisizione dati’’ (geotechnical laboratory – tests methods, data acquisition and study), Controls Editor. [R1] UNI 11235, Criteria for design, execution, testing and maintenance of roof Garden, by UNI – Ente Nazionale Italiano di Unificazione (UNI), May 2007. [R2] UNI EN 13037, Soil improvers and growing media – Determination of pH, Ente Nazionale Italiano di Unificazione (UNI), January 2012. [R3] UNI EN 13038, Soil improvers and growing media – Determination of electrical conductivity, Ente Nazionale Italiano di Unificazione (UNI), January 2012. [R4] UNI EN 13039, Soil improvers and growing media – Determination of organic matter content and ash, Ente Nazionale Italiano di Unificazione (UNI), January 2012. [R5] CEN ISO/TS 17892-1, Geotechnical investigation and testing – Laboratory testing of soil – Part: Determination of water content, Ente Nazionale Italiano Unificazione (UNI), Committee European Normalisation (CEN), International Standard Organisation (ISO), January 2005.

361

[R6] UNI EN 1097-6, Tests for mechanical and physical properties of aggregates – Part 6: Determination of particle density and water absorption, Ente Nazionale Italiano Unificazione (UNI), February 2008. [R7] C.N.R. 69, Norme sui materiali stradali – prova di costipamento di una terra, (Rules on road materials – Compaction test of soils) November 1978, Official publication of National Research Council, nowadays UNI EN 13286-2:2005. [R8] UNI CEN ISO/TS 17892-11, Geotechnical investigation and testing – Laboratory testing of soil – Part 11: Determination of permeability by constant and falling head, Ente Nazionale Italiano Unificazione (UNI), Committee European Normalisation (CEN), International Standard Organisation (ISO), February 2005. [R9] UNI EN 13041, Ammendanti e substrati di coltivazione - Determinazione delle proprietà fisiche - Massa Volumica apparente secca, volume d’aria, volume d’acqua, coefficiente di restringimento e porosità totale, Ente Nazionale Italiano Unificazione (UNI), January 2012.