 
          2821
        
        
          Technical Committee 212 /
        
        
          
            Comité technique 212
          
        
        
          Table 2. Elastic modulus at compressive stress range from 1.0 N/mm
        
        
          2
        
        
          to
        
        
          2.0 N/mm
        
        
          2
        
        
          .
        
        
          
            2
          
        
        
          
            4
          
        
        
          
            5
          
        
        
          E (N/mm
        
        
          2
        
        
          )
        
        
          254
        
        
          304
        
        
          274
        
        
          3 GEOTECHNICAL CAPACITY
        
        
          So far there have only been two load tests for the cohesive
        
        
          polymer pillar. Both of the load tests have taken place in
        
        
          Southwest of Finland. The first load test was in June 2010 in
        
        
          city of Turku. The second load test was in November 2012 in
        
        
          city of Salo.
        
        
          Both of the load tests were made in co-operation with
        
        
          Tampere University of Technology. The pillars were tested until
        
        
          they could not hold the load any more.
        
        
          The test pillars are named with a letter T or S depending on
        
        
          the city and a number.
        
        
          3.1
        
        
          
            Load test in Turku, 2010
          
        
        
          The load test was accomplished at the front yard of a test
        
        
          project. The polymer pillars for the load test were installed into
        
        
          a soft clay layer. Undrained shear strength of the clay was
        
        
          measured with vane test and the result was 15 kPa.
        
        
          Total of five polymer pillars were tested. The top of each
        
        
          pillar was excavated to sight and they were cut to achieve
        
        
          smooth surface for loading. Therefore the lengths of pillars are
        
        
          not equal. The lengths of pillars and amounts of injected
        
        
          geopolymer are shown in Table 3.
        
        
          Table 3. Tested polymer pillars in Turku 2010.
        
        
          
            T2
          
        
        
          
            T3
          
        
        
          
            T4
          
        
        
          
            T6
          
        
        
          
            T7
          
        
        
          Original length (m)
        
        
          4.0
        
        
          4.0
        
        
          4.0
        
        
          4.0
        
        
          8.0
        
        
          Tested length (m)
        
        
          3.65
        
        
          3.53
        
        
          3.54
        
        
          3.00
        
        
          6.21
        
        
          Total injection (kg) 210.5 165.8 196.0 142.5 300.8
        
        
          Injection (kg/m)
        
        
          52.6
        
        
          41.4
        
        
          49.0
        
        
          35.6
        
        
          37.6
        
        
          The arrangement of the load test is shown in Picture 3. The
        
        
          loading was achieved with hydraulic jack against steel beam
        
        
          and counterweights. Applied load was measured with load
        
        
          sensor and the movement was measured as relative to normal
        
        
          ground surface.
        
        
          Picture 3. Arrangement of the load test in Turku 2010.
        
        
          The load-movement curve was drawn from the results as
        
        
          shown in Figure 2. Because of different lengths of pillars, the
        
        
          load-movement curve does not give the information needed.
        
        
          Therefore was needed to calculate the shear stress between shaft
        
        
          of pillar and clay. The shear stress-movement curve is shown on
        
        
          Figure 3.
        
        
          T6
        
        
          T7
        
        
          T2
        
        
          T3
        
        
          T4
        
        
          Figure 2. Load-movement curves of the load test in Turku 2010.
        
        
          T6
        
        
          T4
        
        
          T7
        
        
          T2
        
        
          T3
        
        
          Figure 3. Shear stress-movement curve of the load test in Turku 2010.
        
        
          The measured C
        
        
          u
        
        
          value of the clay is shown with dashed line.
        
        
          The installation on test pillar T7 was not succeeded. There
        
        
          had been some problems with the injection pump during the
        
        
          injection. Nevertheless it was decided to be tested. The result is
        
        
          significantly worse than the result of the shorter pillars.
        
        
          With other pillars the test succeeded. The test shows the
        
        
          cohesive grip between pillar and clay is good. The shear stress
        
        
          between the shaft of the pillar and the clay is bigger than the
        
        
          undrained shear strength of the clay. Partly this can be
        
        
          explained with the speed of the load test. Each load step was 30
        
        
          minutes and the entire loading took about 4 hours per pillar.
        
        
          3.2
        
        
          
            Load test in Salo, 2012
          
        
        
          This load test was also accomplished at the front yard of a test
        
        
          project. The polymer pillar for the load test was installed into a
        
        
          soft clay layer. Undrained shear strength of the clay was
        
        
          measured with vane test and the result was 10 kPa.
        
        
          Only one pillar was tested at this site. The preparing
        
        
          procedure for pillar was similar to the test pillars in Turku. The
        
        
          length of pillar and the amount of injected geopolymer is shown
        
        
          in Table 4.
        
        
          Table 4. Tested polymer pillar in Salo 2012.
        
        
          
            S1
          
        
        
          Original length (m)
        
        
          3.0
        
        
          Tested length (m)
        
        
          2.36
        
        
          Total injection (kg)
        
        
          90.0
        
        
          Injection (kg/m)
        
        
          30.0