In this research work, processed granulate of corn cob (PCC, Figure 1c) is considered as a possible lightweight aggregate in the manufacturing process of lightweight concrete masonry units (CMU). During the PCC preparation study, expanded clay (EC, Figure 1a) was used as a reference lightweight aggregate because it is currently applied in the context of CMU.

PGCC is based on raw corn cob particles (Figure 1b) which are covered with a cement paste prepared with the ratio 1:1 (Portland cement 32.5 N: water). PCC was prepared in order to have a grade similar to EC. However, the shape of these two aggregates is quite different, Figure 1. EC has a spherical shape type (Figure 1a) and PCC has a random irregular shape (Figure 1c). The density and the water absorption coefficient of the PCC have been experimentally assessed and the respective values are 454.5 kg/m³ and 57.9%. This density seems acceptable because it is within the density range of the expanded clay lightweight aggregates (i.e. 60–850 kg/m³), (8). On the other hand, raw corn cob particles are covered with cement paste resulting in a waterproofing improvement of the aggregate.

Medium sand (MS; 0.0–4.0 mm), coarse sand (LS; 0.8–3.0 mm), gravel (G; 2.0–6.0 mm), Portland cement 32.5 N (C), lightweight aggregate (LWA) and water (W) are the constituents considered in this research for the manufacturing of lightweight concrete masonry units. Table 1 presents the amount of each constituent necessary to manufacture one CMU according to Mixtures 1 and 2, and considering a generalised LWA. Meanwhile, two different lightweight aggregates were studied: expanded clay (EC) and processed granulate of corn cob (PCC). Therefore, three cases (Cases 1, 2 and 3) were considered. Case 1 consists of manufacturing CMU according to Mixture 1 and using EC as lightweight aggregate (CMU-EC). Case 1 is to be used as a reference in this study. On the other hand, Case 2 corresponds to manufacturing CMU also according to Mixture 1 but using PCC (CMU-PCC) as lightweight aggregate instead of EC. This case intends to be a novelty in the light-weight concrete masonry units context, taking into account that an alternative organic lightweight aggregate has been considered (PCC), instead of an industrialized one (such as EC). At the same time, Case 3 is related to a manufacturing process scenario in which CMU is manufactured with Mixture 2 and it is also based on PCC (CMU-PCC). This last case intends to be a complement of Case 2 in which the amount of cement was slightly increased, Table 1.

The CMUs were manufactured in an industrialized context, Figures 2 and 3. The constituents were mixed automatically. After this stage, the mixture was introduced in moulds and under compaction the CMUs were moulded, Figure 2a. A set of five CMU were moulded automatically, Figure 2b. Then, the CMU were transported to a chamber room for the curing/drying process, Figure 3b, where the units were protected from the direct exposure of climate conditions (e.g. rain and sun) and the thermo-hygrometric conditions were the environmental ones.

In this research work, a six hollow block was manufactured in a standard shape, Figure 4. The size of the adopted lightweight concrete masonry unit (CMU) is 500 mm×200 mm×200 mm (length (L)×width (W)×height (H)) with a +3/−5 mm dimensional tolerance. Fifteen CMU related to each case study were manufactured, Figure 4.

The size, the dry mass (m_dry,s) and the bulk density (ρ) of these LWCMU were then assessed. The bulk density was quantified accordingly (15). The average (AVG), the standard deviation (SD) and the coefficient of variation (CoV) of these measures are presented in Table 2 for the three cases analysed. None of the CMU have shown the exact dimensions indicated above including the reference one (Case 1, Table 2). Meanwhile, all of them have satisfied the dimensional tolerance. They also proved to have a uniform size.

In terms of bulk density, the reference CMU-EC (Case 1, ρ=1364 kg/m³, Table 2) are lighter than CMU-PCC (Case 2 and Case 3, ρ=1681 kg/m³ and ρ=1748 kg/m³, respectively). Taking into account that the adopted manufacturing process and mixture were similar in Case 1 and 2, the mass of the lightweight aggregates considered (EC and PCC, respectively) may be the main justification for this discrepancy. In this research work, the bulk density of PGCC was estimated in ρ=454.494 kg/m³. At the same time, the density of EC ranges between 275 and 430 kg/m³ which is smaller than those of PCC. In addition, considering that the amount of cement was increased in Mixture 2, an increasing of the bulk density of CMU-PCC of Case 3 is also expected.

ASTM C 90-06a (16) defines three bulk density classes for concrete masonry units as follows: Lightweight – units having an average bulk density of less than 1680 kg/m³; Medium Weight – units having an average bulk density of 1680 kg/m³ or more, but less than 2000 kg/m³; Normal Weight – units have an average bulk density of 2000 kg/m³ or more. Therefore, according to ASTM C 90-06a (16) the reference CMU-EC (manufactured with Mixture 1) is clearly a lightweight concrete masonry unit. The CMU-PCC Case 2 (manufactured with Mixture 2) may be considered as medium weight concrete masonry units. On the other hand, the average bulk density obtained for CMU-PCC Case 2 (manufactured with Mixture 2) exceeds significantly the value of 1680 kg/m³. In this case, the concrete masonry units may be characterized as medium weight. These results indicate that using PCC as an alternative lightweight aggregate in the manufacturing process of CMU may require adjustments of a typical mixture used in this context in order to ensure the production of a lightweight concrete masonry unit.

Five CMU specimens of each case (C1, C2 and C3) were prepared and tested in terms of water absorption due to capillarity action at the age of 44 days and following the recommendations prescribed in the Portuguese NP EN 772-11 standard (17). The duration of this test was 14 days (i.e. 1209600 seconds). During this period of time, the CMU specimens were soaked in a 5 mm water layer. Table 3 presents the water absorption coefficient by capillarity, C_w,s, which was assessed by Expression [1]. C w , s = m s o , s - m d r y , s A s t s o 10 6

Where: C_w,s is the water absorption coefficient by capillarity (g/(m²s^0.5)); m_so,s is the mass of the CMU after immersion (g); m_dry,s is the dry mass of the CMU (g); A_s is the area of the face of the CMU which is soaked in water (mm²); t_so is the time of immersion in water (s - seconds).

Based on the water absorption coefficients by capillarity presented in Table 3 it is concluded that the CMU-PCC (CMU C2 and C3, Table 3) are more susceptible to absorb water due to capillarity than the CMU-EC (CMU C1, Table 3). In addition, this tendency is reduced by increasing the amount of cement in the mixture by increasing the cement paste which covers the organic particles. This last situation occurred for Cases 2 (CMU C2, prepared according Mixture 1, Tables 1 and 3) and 3 (CMU C3, prepared according Mixture 2, Tables 1 and 3) in which the respective coefficient decreased from C_w,s=7.7 g/(m²s^0.5) to 7.4 g/(m²s^0.5), in terms of average values, in Table 3.

The obtained value of coefficient of variation (CoV) of the water absorption coefficient by capillarity of the different studied cases is small, which means that there was an acceptable specimen variation and these experimental results may be considered consistent.

All the tested CMU kept their material integrity after being in contact with water during 14 days. This aspect is very important in terms of the suitability of the CMU as building materials. In fact, some building tasks may be compromised if the CMU are vulnerable to water. Exterior application, block laying, plastering and placing the units near the ground may be some of these building tasks.

A twenty-four hour cycle of aggressive thermo-hygrometric conditions was experimentally simulated (18–19). The aggressive thermo-hygrometric conditions of the cycle corresponds to having the specimen in the climate testing chamber device at a constant temperature of 60 °C during a period of time of 7 hours (first stage) followed by a second period of time of 1 hour in which the samples are soaked in water at the normal temperature of the laboratory of 22 °C approximately (second stage), and finalizing by having the specimens placed again in the climate testing chamber cell device at the constant temperature of −15 °C during an additional period of time of 16 hours (third stage). These extreme temperatures (60 °C and −15 °C), the dramatic change in temperature and the alternate dried/wet/frozen conditions allow simulating experimentally an accelerated aging process of the material.

The aging sensitivity of the CMU under study was assessed by testing the material during ten consecutive cycles of the described thermo-hygrometric conditions and following the suggested by (18–19). Two intact LWCMU specimens of each case (C1, C2 and C3) were tested. The aging effect signals of the samples were visually monitored at the end of each cycle. Figure 5 presents the CMU after being tested (C1′, C2′ and C3′). It is concluded that there was no expressive degradation of the tested materials. In fact, the tested CMU kept their integrity after being under such environmental aggressiveness. This technical aspect is relevant considering exterior applications. The tested specimens only suffered minor degradation, such as a slight change of colour (e.g. a brownish tendency) and a certain erosion of the edges of the specimens, Figure 5.

Partition walls seem to be the main building domain of application of CMU. Therefore, lightness, material integrity, adequate durability, good thermal and acoustic insulation ability, affordable and sustainable, are some expected material properties. CMU is a non-structural element. However, it is important that this type of masonry unit presents a certain compressive strength. For instance, it has to be able to support the weight of the underlying portion of the wall. Additionally, an adequate compressive strength also indicates that the CMU presents an acceptable material integrity and therefore it can be shipped, stored and applied in the building site. For instance, BS EN 771-3:2011 (20) indicates the range of the compressive strength of aggregate concrete masonry units (dense and lightweight aggregates) from 2.9 MPa to 10.4 MPa.

In order to evaluate the compressive behaviour of the CMU (based on EC and PCC aggregates), in particular the CMU-PCC, five CMU samples of each case were prepared and tested in terms of uniaxial compression test. Specimens were tested at the age of 50 days because of logistic aspects specifically concerned with this research work. This test was performed according to NP EN 772-1 (21). A 300 kN load bearing capacity servo-hydraulic actuator was used. The test was carried out in force closed-loop displacement control with a load displacement rate of 1.2 mm/minute.

In Figure 6, the stress vs strain curves of the CMU tested under uniaxial compression are presented. In terms of compression, it is undoubtful that CMU-EC (C1) is stronger than CMU-PCC (C2 and C3). However, all the tested CMU seem to have a similar behaviour in compression. An approximate initial straight-line portion of the diagram followed by a well-defined Yield Point are two aspects that characterize the stress vs strain curves of the specimens.

Additionally, the AVG, the SD and the CoV of the ultimate compressive strength (fc_max) of the CMU tested are presented in Table 4. Two aged CMU specimens of each case (C1′, C2′ and C3′) were also tested in compression. The respective estimated ultimate compressive strength is also presented in Table 4. Complementarily, the failure mode under compression faced by the tested CMU is shown in Figure 7.

The CMU-EC (C1) showed high compressive strength than CMU-PCC (Cases C2 and C3), Figure 7 and Table 4. In fact, in terms of average, the compressive strength of the reference CMU (CMU-EC, C1) is 2.14 and 1.90 times higher than the compressive strength of the proposed CMU-PCC C2 and C3, respectively. The fact that an organic product was used as aggregate may justify this compressive strength discrepancy. At the same time, the compressive strength showed by the CMU-PCC is lower than 2.9 MPa which is the minimum compressive strength admissible by (20). On the other hand, the obtained compressive strength of CMU-PCC may still be acceptable in the context of CMU. For instance, in a 3 m height wall built with CMU-PCC, the masonry units placed at the first layer (critical ones) will be under an approximate 1.9 kN compressive force (FCd) corresponding to the dead load related to the weight of the overlying wall, Figure 8. On the other hand, the obtained experimental ultimate compressive force (F_CRd) for CMU-PCC related to Case 2 was 84.5 kN approximately which is significantly higher than 1.9 kN. Therefore, the proposed CMU-PCC may have an acceptable mechanical behaviour in the context of lightweight concrete masonry units for non-structural purposes. This material achievement is even more interesting considering that an agricultural waste product is proposed as a lightweight aggregate. The scatter of the results is highlighted by the CoV. The verified dispersion may be acceptable taking into account the obtained small values of the CoV and considering that an organic raw material was employed.

In Table 4, the fact that an approximate 12% mechanical behaviour improvement of CMU-PCC achieved by increasing the ratio of C:LWA (in terms of weight) from 1 (Mixture 1 in Table 1) to 1.308 (Mixture 2 in Table 1) is also featured. This technical aspect is in accordance with the expected results in the lightweight concrete for non-structural purposes domain.

In order to complement the aging sensitivity analysis done in the previous section of this paper, the specimens that were tested under the aggressive environmental conditions were then also tested in terms of uniaxial compression test (C1′, C2′ and C3′, Table 4). Comparing the ultimate compressive strength of the intact CMU specimens (C1, C2 and C3, Figure 6 and Table 4) with the respective one of the aged CMU specimens (C1′, C2′ and C3′, Table 4), it is noticed again that both CMU-EC and CMU-PCC shown an adequate durability. In fact, the compressive behaviour of the CMU-PCC was not affected significantly during the aging test because there was no reduction of the compressive strength. In contrast, the CMU-EC was slightly affected after being tested under repeated cycles of aggressive environmental conditions because there was a reduction of its compressive strength from 4.55 MPa to 3.75 MPa (C1 and C1′, Table 4). This fact may indicate that the material has faced a certain level of degradation after being tested.

In addition, the CMU-PCC (C2 and C3, Figure 7) specimens tested under uniaxial compression showed failure modes similar to the expected one of a regular conventional lightweight concrete masonry unit such as CMU-EC (C1, Figure 7). This experimental analogy is another interesting output that enhances the practicability of the proposed technical solution of manufacturing lightweight concrete masonry units with processed granulate of corn cob as an aggregate.