The developed partition system presents an innovative shape and geometry and intends to be used in non-loadbearing partitions. The definition of the geometry was based on four main ideas, namely: 1) obtaining a partition system that results in an easy construction technology and thus decrease construction times; 2) to be possible to install infra-structures during construction of the walls and thus reduce the production of wastes; 3) obtaining a partition system that function in tensile and not in compression with lightweight materials; and 4) to permit portability and multiple assembling/disassembling.

The panel is composed by cubic modular components, equivalent to an outlet box, and is composed of two equal halves (Figure 1), that is connected by male/female fixings, and present vertical and horizontal perforations, which are adequate to an efficient integration of the infra-structures, as can be seen on Figure 1e).

Besides the conception of the geometry and shape of the partition, some details of materials are presented here. The process of selecting materials for the interior partition is a challenging task. In previous studies it was made a material selection to the sandwich panel configuration attending functional, economic and environmental demands for AdjustMembrane design. Within this project the functional performance (acoustic and thermal) and sustainable assessment profile were already described in detail on previous publications (9) (10) (11).

AdjustMembrane partition system constitutes a modular building solution of a non-load bearing wall, tensioned between the pavements and ceiling slabs that are used as anchoring elements. It is important to understand at which part of the partition is able to bear impact loads to assign it a certain use. Table 3 presents the material’s mechanical properties used in the interior walls.

AdjustMembrane core system is composed by grid and insulation modular components. The insulation is constituted by malleable fibrous material with grooves whose geometry allows the integration of pipes, as previously mentioned. The grid and the core are composed by modules that can be subtracted or added anytime to allow the integration of facilities such as water or electrical pipes and switches or outlets, without panel damaging (Figure 1e).

AdjustMembrane presents a coating in the external faces of the core that are assembled by fitting to the core grid (Figure 1b). The panel coating used on experimental test was polyester hot pressed bat it is possible to use polymers, textile materials, metal sheets or composite materials with reinforcing fibbers, or other types of rigid panels or films (fibbers, polymers or composites).

An over-coating may also be placed on site. The over-coatings to be applied can be fabrics or films. A tensioned PVC membrane attached to a frame system is available in the market, with specific profiles which make the adjustment of the membrane. Membranes are also available on the market. The most common are: PVC, PTFE, silicone, and vinyl with or without reinforcing structure (usually woven polyester or fibreglass). There may also be added OSB panels, MDF, plywood, etc., mechanically fixed by “Velcro” or other fastening system, like quick coupler type which does not compromise the disassembly. In spite of hamper the easy disassembly desired, it admits the possibility of using more conventional over-coatings such as plaster, mortar, particularly with synthetic linkers (e.g. acrylic).

The European guidelines for technical approval of interior partition kits (5) define two states in relation to which the resistance to impact should be verified, namely the state which foresee mechanical damage and the state that ensure adequate behaviour under service loads. For each state distinct levels of loading, which are associated to distinct categories of use, are defined from the loading area categories stated in EC1 (12). The categories of use are: Category A (areas of residential use); Category B (offices); Category C, which foresees areas with the possibility of people concentration; Category D (shopping areas); and Category E, where the concentration of goods is probable. In (5), the categories of use I e II include the previous categories A and B, the category of use III includes the zones C1 to C4, D and E and the category of use IV includes the zones C5, A, B, C1-C4, D and E. The definition of the categories I to IV relates to the need of protection of the walls, with the risk to occur accidents and to exist improper use. The present partition kit solution was mainly conceived to be used on Categories A and B (I and II, respectively).

At this point, it should be referred that the guidelines for interior partition kits (5) also recommends carrying out mechanical tests for eccentric vertical load and horizontal linear static load. Due to the nature of the partition, namely its flexibility or low bending stiffness, the two additional mechanical tests were not carried out, since the concept of the AdjustMembrane partition in the present version cannot be directly compared to traditional partitions in this parameter. Further research on prototypes using other materials on the coating and/or grid of this solution would be necessary to improve the response of this partition concept concerning these issues, which was out of the scope on the present work. Additionally, all the test procedures were adjusted to the available prototype size, which was built mainly to analyse the assemblage phases and the workability of the AdjustMembrane partition. At this moment there are no specific standards to carry out impact tests. The impact tests can be made following the recommendations of (13) and (14).

Impact tests were carried out on a prototype of 1.2 m length, 2.8 m height and 0.07 m thickness. The dimension of this prototype is smaller than the dimension preconized on the guidelines (5), due to the fact that it is considered that the prototype was built using a rigid steel frame in one side, representing the position for a door or other discontinuities in its surface, such as a window. The prototype wall to the left of this rigid frame was built according to the construction technology described previously. The connection to the steel structure has been defined by screwing the solution to the metal support. The impact simulation tests carried out at this phase considering the lightweight interior partition kit composed by a PVC grid (constituted by modular components with mechanical male/female fixings between them), polyurethane foam and PVC rigid cover layer. These were the materials with which was possible to build the prototype in this phase, and the main objective of the prototype in this phase was to test the assembling process. The use of other materials in a bigger wall size will be considered for future research, as soon as there will be companies interested in producing the technology in a bigger scale, what is in the present moment under negociation.

According to the recommendations (13), (15) and (16), two distinct types of impact resistances should be considered, namely the hard body and soft body impact tests, aiming at simulating the resistance of the wall to the projection of objects or the collision between users and partitions. The hard body impact is materialized with a steel sphere, whereas the soft impact is materialized by the collision of a bag with a specified geometry and mass. The steel spheres should have diameters of 50.0 mm and 63.5 mm and weights of 0.5 kg and 1.0 kg, respectively, to verify the safety in use and the partition serviceability, respectively (see Figure 2a). To obtain the behaviour of the wall to soft body impact, a bag with a specified geometry (13) should be used. The bag should be filled with glass spheres of 3 mm diameter in a total mass of 50 kg (see Figure 2b).

As already mentioned, the impact resistance of a partition kit consists of its ability to withstand impact loads after which it is possible to maintain its normal use. The resistance to soft body impact load is associated to certain impact energy, E_i, given by the expression:

The laboratorial prototype was used to evaluate the mechanical behaviour of the AdjustMembrane solution. The values of deformations corresponding to the impact and the permanent deformations are indicated in Table 5, for the serviceability tests carried out.

The visual inspection made at the level of the interfaces between the wall and the steel frame revealed the absence of any separation. The safety criteria of no collapse, no penetration and no projection were thus fully accomplished according to (5) and (13), being the partitions in perfect condition of use after the different phases of loading. Notice that due to the absence of damage, it was decided to make all tests using the same prototype.

It is observed that the partition presents a good behaviour under distinct soft body impact energies, which is confirmed by no permanent deformation and the verification of the criteria needed to be found and that correspond to the global integrity of the wall after impact tests (see Table 5). The deformation increased with the increase on the impact energy, but in all cases was lower than the limit mentioned in (5). No permanent deformation was recorded both for the safety in use and serviceability conditions. It was also observed that no increment on the deformation was recorded after the successive impacts.

During impact tests, no damage was observed, neither on the structure or the coating. After impact, the specimen returned to the original position without residual displacements, with the exception of the tensioned straps on the soft body impact test with 120 Nm energy. In this case, it was visible a maximum displacement (slipping) of 5 mm (top to bottom) of the 2^nd and 3^rd strap (from left to right), as can be seen in Figure 6.

Based on the results obtained for impact loads, it should be concluded that the partition solution proposed here is appropriated to resist impact loads, both corresponding to the projection of objects and collision of users.