Lapilli cement mortar lightened with ­treated expanded polystyrene beads

1 Introduction

The Canary Islands present a special lithology – different from the rest of the Spanish mainland – as it is mainly characterized by volcanic materials and structures, forming a landscape dominated mostly by lavas of different nature and pyroclastic deposits, with a wide range of different compositions [1]. One of these pyroclastic deposits is the “lapilli”. This rock is caused by volcanic eruptions and consists of fragments ranging in size from 2 to 64 mm generally with irregular shape, vitreous and porous. It has a basaltic composition and it is characterized by its black color which changes to reddish due to an oxidation phenomena [2].

Moreover, the Canary Islands, due to their volcanic origin, have a lack of adequate clay for manufacturing conventional Spanish concrete materials (bricks, hollow bricks, etc.). This issue has significantly influenced the construction activity of the Canary Islands, as local resources (i.e. volcanic materials) have been constantly used as raw materials for manufacturing local construction products likewise the ones used on the Spanish mainland. This local building cement is lightened with volcanic rocks, mainly lapilli, instead of adding expanded clay as used on the Spanish mainland.

Since the Technical Building Code regulation entered into force in 2007 [3], studies of airborne noise and heat resistance are performed on interior partitions and vertical building envelopes built with vibrated lapilli lightened concrete blocks currently placed on the market [4]. The result of these studies showed that partitions and vertical building envelopes built with these elements, without any other material, do not meet the minimum requirements set in the basic document “DB-HR Protection against noise” and “energy demand limitation” of the Technical Building Code regulation. Therefore it is necessary to include a second wall – block or drywall – placed in the interior in order to fulfill with the current regulation [5, 6].

Numerous research publications have been found analyzing lightweight concretes and mortars using different lightweight aggregates, such as expanded clay [7, 8]; expanded polystyrene [9]; rigid polyurethane foam [10] or even wood waste [11] and recycled plastics [12]. In general, these composites have lower densities and better thermal and acoustic behavior than traditional concretes or mortars. However, these reductions in density directly involve lower mechanical resistances [13].

In addition, from the previous literature, many works analyzing the characterization of mortars and concretes lightened with lapilli were found [14, 15, 16]. However, none of them analyzed the inclusion of other lightweight aggregates in addition to the lapilli.

Therefore, this research aims to determine the feasibility of a new building material based on lapilli concrete and cement mortar with treated expanded polystyrene beads (EPS_t) in order to enhance their thermal and acoustic performance. This new material will be used to manufacture concrete blocks.

2 Experimental plan and method

The experimental plan unfolds in three phases:

1^st Phase – Materials characterization

2^nd Phase – Reference mortar samples preparation (series I)

3^rd Phase – Preparation of two series – seven samples each – using different proportions according to the percentage of treated expanded polystyrene (EPS_t) added (series II and III)

Finally, standardized tests were conducted on all the above samples in order to measure:

1) Weight variation

2) Sound insulation

3) Thermal resistance

4) Mechanical resistance (bending and compression).

5) Water vapor transmission

2.1 Materials characterization

In a first phase, materials used in this study are characterized:

Drinking water

Cement

Coarse and fine aggregate: lapilli

Treated expanded polystyrene (EPS_t)

The results obtained are shown in Tables 1, 2 and 3. Table 3 shows a geotechnical characterization of weakly cemented pyroclast from the Canary Islands (Canary Islands Government Technical Report).

Virgin EPS beads sizing between 2 to 8 mm, gray colored and treated with a specific additive, allowing a better mixture of the polystyrene with water/binder, removing the floating phenomenon and ensuring a uniform distribution of the materials. The beads were manufactured by Politer [17].

2.2. Samples preparation

Samples were made at the Prefabricated Socas Company [18], following the same procedure as used to manufacture traditional vibrated concrete blocks. The samples preparation consisted of three phases:

Phase 1: Series I (PI) samples (Table 4) were performed using the same mortar as used in the factory for the production of traditional vibrated concrete blocks, so that their results serve as a reference for mortar samples made with lapilli and EPS_t. One third of the total mortar was used in each sample, vibrating it for 30 seconds and when reaching the end the mortar was also compacted (phase 2). For this, a metal cover was placed on the upper surface of the compound and fixed with two screws to the mold, allowing at last (phase 3) vibrating while compressing the mixture.

Phase 2: Test specimens of series II (P II) (Table 5) were prepared with the same mortar as PI, but once the mortar left the mixer, EPS_t (36.85 % of the total volume of the mixture) water (0.30 l) was added to a total of 126 l of mortar. The final mixture was used to fill the seven samples (62.91 % of lapilli mortar, 36.85 % EPS_t and 0.24 % of water). When adding EPS_t, the manufacturer’s instructions for 200 kg/m³ of concrete density were followed (17), subtracting from the total EPS_t the amount of lapilli used in the mortar proportions.

Phase 3: Test specimens of series III (PIII) (Table 6) were prepared with the same mortar as PI, but once the mortar left the mixer, EPS_t (73.70 % of the total volume of the mixture) water (0.50 l) was added to a total of 126 l of mortar. The final mixture was used to fill the seven samples (lapilli mortar 25.93 %, EPS_t 73.70 % and water 0.37 %). The quantity of EPS_t added in these specimens, exactly doubles the amount added in Series II specimens.

2.3 Tests

After 28 days, the dried samples were taken to the Quality Construction Laboratory of the Canary ­Islands, where the following tests were performed:

1) Thermal resistance

2) Mechanical resistances (bending and compression)

3) Water vapor transmission

2.3.1 Weight variation

In order to calculate the average weight of each series, the highest and lowest values were rejected and the mean weight of the remaining values was achieved.

2.3.2 Sound insulation

This test was performed as specified in the European Standards EN 140-4:1999 “Field measurements of airborne sound insulation between rooms”. The tested building enclosure is located between two adjacent rooms in the basement of the University of La Laguna, on the central campus. These rooms meet the requirements specified in the standard (Table 7). The equipment used was “OmniPower” B&K 4296 Serial No. 2485239, consisting of an output power amplifier of around 300 W.

Calculation:

The measurements were performed on 1/3-octave bands. The parameter used to measure the airborne sound insulation is the standardized level difference D_nT, (Equation 1):

D_nT = L₁ - L₂ + 10 log (T/T₀)⇥(1)

(L₁ = Sound pressure level in emitting room (room 1); L₂ = Sound pressure level in receiving room (room 2); T = Reverberation time in receiving room; T₀= Reference reverberation time (0.5 s))

Calculations for D_nT,w (weighted standardized level difference) and for the different frequency ranges were performed according to UNE EN 140-4: 1999 and UNE.EN ISO 717-1: 1997 standards, respectively.

Procedure:

The measurements of airborne sound insulation are performed by placing the sound source (emitting pink noise) in two different positions inside the emitting room and performing a total average of ten random positions both on the emitter and receiver rooms. The reverberation time was obtained using two positions of the sound source and six microphone positions. Previously, the background noise and reverberation had been measured.

Five tests were performed in the interior partition wall separating two offices. This partition wall was modified throughout the tests, as follows:

1. Interior partition built with traditional vibrated concrete blocks (15 cm thick) and coated on both sides with plaster (17.40 cm of total thickness)

Moreover, an opening of 180 x 120 cm was executed in the partition and was further filled in four different ways:

2. Using panels made with Series I, received with plaster

3. Using panels made with Series II, received with plaster

4. Using panels made with Series III, received with plaster, and plaster coatings on both sides (total thickness 7 cm)

2.3.3 Thermal resistance

The test was performed according to the UNE-EN 12664: 2002 and UNE-EN 1745: 2002 Standards. The objective of the test was to determine the heat flow of the samples under equilibrium conditions. The density of the heat flow ratio is measured using sensors located on the samples. A flow meter heat HFM 436/3/0 was used. The sample is placed between two plates kept at different temperatures (0 to 20 ºC) during the test. Finally, the thermal conductivity is determined when the thermal equilibrium between the two faces of the sample is reached and there is a uniform temperature gradient in the whole sample.

Four specimens of 30 cm x 30 cm x 5 cm were tested for each series and polished on both sides until they were perfectly parallel. In order to obtain a constant mass of the samples, they were dried with an oven at 50 °C until the weight difference between two successive days was 0.1 %. The time taken to achieve this weight stability was fourteen days.

From the samples the following parameters are obtained:

Density: The density ρ₀ and ρc of the dried sample tested was calculated using equation 2:

ρ₀ = m₂/V⇥(2)

(ρ₀ is the density of dry material; m₂ is the mass of the material after drying; V is the volume occupied by the material after drying)

Thermal resistance, R, was calculated using equation 3:

R = T1 - T2⇥(3)

         f eh

(f is the calibration factor of the heat flow meter; eh is the heat flow meter output; T1 is the sample temperature measurement on the hot side; T2 is the sample temperature measurement on the cold side)

And thermal conductivity, λ was calculated, using equation 4:

λ =   f eh.d⇥(4)

      T1 - T2

2.3.4 Flexural and compressive strength

Three specimens of 160 mm x 40 mm x 40 mm were cut from the sample panels (60 cm x 60 cm x 5 cm). The flexural strength was determined following the UNE-EN 1015-11: 2001 standard, i.e. applying a load at three points (using rollers) until failure. The contact points of the rollers and the sides of the specimens were cleaned with a cloth in order to remove particles or other materials. The smoothest side of the specimen –the one in contact with the mold – was placed on the rollers. The load was applied at a uniform rate between 10 N/s and 50 N/s, so that failure occurred between 30 s and 90 s.

The two halves of sample specimens obtained from the flexural strength test were used to test the compressive strength, following the UNE-EN 1015-11: 2001 standard.

The load was applied gradually on a rate of between 50 N/s and 500 N/s such that the failure occurred between 30 s and 90 s. The maximum load applied during the test was recorded – in N. The resistance of each specimen was recorded to the nearest 0.05 N/mm². The mean resistance was then calculated with an accuracy of 0.1 N/mm².

2.3.5 Determining the properties of water vapor transmission

For this test, conducted according to the UNE-EN ISO 12572: 2001, method “C” was used as it is the least favorable and the most suitable to meet the requirements set by the CTE regulation, 11 cm x 11 cm x 5 cm specimens were performed. This sizing was chosen because the lab test plates had this section. Moreover, the thickness (5 cm) complies with paragraph (6.2.3) of the Standard, which specifies that the section of concrete with aggregates should be at least three times the largest particle size.

The samples were sealed to the open end of the test plate containing a water saturated solution and then they were placed in a test environment with controlled temperature and humidity.

Due to the difference in water vapor pressures between the test set and the chamber, a water vapor flow passes through the test specimens. The test set was weighed periodically to determine the resistance factor to water vapor when the steady state was reached.

The test was performed once the weight of the samples was stabilized. The lapilli mortar specimens with EPS_t adhered to the bassel and were laterally waterproofed with an adhesive sealant “Sika­flex 11 FC”.

The set of specimens was introduced into the test chamber DYCOMETAL Model: CCM-0/19380. Specimens were weighted every 48 hours until five successive weightings variation in mass per unit time was constant within 5 % of the mean value.

The test results were calculated as follows:

For each set of successive weightings of the test specimens, the mass change (∆m₁₂) was calculated using equation 5:

∆m₁₂ = m 2 - m1⇥(5)

               t2 - t1

(∆m₁₂ mass change per time for one determination [kg/s]; m1 mass of the test set at time t1 [kg]; m2 mass of the test set at time t2 [kg]; t1 and t2 are successive weighting times [s])

The regression line between mass and time was calculated – excluding the previous test period (non-linear). The slope of this line is G, kg/s.­Furthermore, the density of water vapor flow, i.e. g, is calculated using equation 6:

g =  G⇥(6)

      A

(A is the is the exposed area of both sides)

Finally, water vapor retention, resistance to water vapor, the water vapor permeability and the resistance factor to water vapor were all calculated following UNE-EN ISO 12572: 2001 Standard.

3 Results and discussion

3.1 Samples weight variation

When EPS_t is added to traditional lapilli mortar, the weight of the samples decreases depending on the percentage added to the original proportions up to 41.84 % (Table 8).

3.2 Sound insulation

The acoustic insulation of the interior partition placed between the rooms improved when using samples containing EPS_t. The interior partitions built with Series II samples (PII) showed an improvement of 2 dBA over Series I (PI), as shown in Table 6.

Furthermore, when the interior partition was modified, the best results were obtained with specimens of the Series 2 (PII), despite having less percentage of EPS_t than Series 3 (PIII) samples. This could be justified because the last series contained more interconnected pores. Furthermore, it was found that in some areas the expanded polystyrene beads were not surrounded by mortar, were bonded in small groups, letting the noise pass without difficulty through the plate section.

Finally, P (III) samples with plaster coating on both sides obtained the best result, nearly reaching the value of the 1^st test (Table 9).

3.3 Thermal resistance analysis

As shown in Table 10, adding EPS_t improves thermal resistance, making this material more thermal resistant than the original one. In addition, this phenomenon rises as the percentage of EPS_t increases.

3.4 Determining flexural and compressive strength

The resistance of the samples containing EPS_t, decreases in relation to EPS_t content (Table 11). The greater mechanical resistance reduction is reached in compression rather than flexural strength, achieving reductions of up to 85 %.

3.5 Determining the properties of water vapor transmission

Results from Table 12, show that P (II) proportions have better resistance than the reference sample P (I). Therefore, the water vapor resistance of the material improves when EPS_t is included in a determined quantity. This suggests that porosity and water absorption of the material decreases when adding EPS_t.

Moreover, P (lll) samples exceed the optimal EPS_t proportions for ensuring a water vapor resistance, because the water vapor resistance decreases and consequently the water vapor permeability increases. However, this depends on the place where the material will be located and the requirements needed.

5 Conclusions

From the results obtained from the tests the following conclusions can be drawn:

When high percentage of lapilli is replaced by EPS_t significant improvements on sound insulation and thermal conductivity are obtained. However, adding high quantities of EPS_t is not the solution to increase the thermal resistance of the material, as high quantities of EPS_t weaken other properties. Therefore, the original materials composition needs to be optimized.

It is feasible to replace lapilli by treated expanded polystyrene (EPS_t) – up to 73.70 % of the total volume – in traditional lapilli mortars, maintaining an optimal workability of the blocks. Furthermore, a homogeneous mixture is achieved eliminating the floating phenomenon of the polystyrene beads. Adding EPS_t reduces the weight of the test samples up to 41.84 %. Nevertheless, strength properties are also reduced.

Therefore, this new material can be used for manufacturing vibrated concrete blocks meeting the requirements of the Spanish CTE regulation without having to place an inner second block wall or drywall and replacing the blocks currently used in the Canary Islands. This procedure has led to a patent which is in the process of commercialization (ref: P201132108).