Assessing Mortar Extrudability through the Jolting TestLESAGE Karei1 a', INGELBRECHT Thomas1b and DE SCHUTTER Geert1c

1Magnel Laboratory for Concrete Research, Ghent University, Ghent, Belgium

aKarel. Lesage@ugent.be, cGeert. DeSchutter@ugent.be

‘corresponding author

Keywords: extrusion, size ratio, 3D printing

Abstract.The main problem with concrete printing is that a proper concrete mix bas to be used which provides the right propertjes at the right time. The concrete should provide a sufficiënt amount of “extrudability” in order for it to be able to pass through the printer head and its nozzles. The layers which are printed on top of each other should be able to bond together to erect a homogeneous construction. Furthermore, when a layer is deposited on top of another layer, the latter one should retain its position and form, also determined by the extruding nozzle.

When starting extrusion experiments, it becomes clear that only a limited number of techniques are available to determine the extrusion potential of a certain mix and none of these techniques are 100% exclusive. Of course, the extrusion process itself can be executed but that often leads to time consuming and material wasting labor. Therefore, it could be interesting to develop less labor intensive tests to judge the extrudability of mixtures in advance of real scale extrusion experiments. For example, it was investigated in this study what would be the indicative potential of a small scale jolting table test.

The material combination of interest is cement mortar. However, in order to have a clearer influence of the size distribution of the aggregates, it is chosen to use several monodisperse fractions of glass spheres ranging from 50 to approximately 400pm. The OPC pastes are always prepared in advance with a water cement ratio of 0,33 and glass particles are added within ten minutes by machine mixing at low speed. No admixtures or other additions were used.

The extrudability is addressed with a direct and indirect technique: The direct technique is the extrusion of the paste sample through a square die with a 4 by 4 cm dimension. The paste pressure in front of the die is build up by accumulating paste, provided by a constantly rotating auger in a cylindrical tube.

The second and indirect technique is the measuring of the impact resistance by using a manual jolting table compatible with ASTM C230. Instead of a cone shaped sample volume, a constant volume of extruded beam sample was used and the number of shocks was determined for the sample to laterally flow for Icm.

For the left three mortars in Figure 1, it is clear that for an increase in partiele volume fraction the resistance to impact increases. On the other hand, the right two points are demonstrating a lower resistance to impact while their partiele volume fraction is significantly higher.

For the same mortar samples, Figure 2 demonstrates that the decrease in impact resistance is following an increase in size ratio. This means that for a higher difference in particles size, a lower resistance to impact is obtained. It is expected that this phenomenon is due to the lower viscosity created by the higher size ratio.

XIV DBMC - 14th International Conference on Durability of Building Materials andComponents, 29-31 May 2017, Ghent University, Belgium

261

XIV DBMC - 14th International Conference on Durability of Building Materials andComponents, 29-31 May 2017, Ghent University, Belgium

14

~ 12 llO

16

«

IS

14

12

11

£ 10

0 5 10 15 20

Volume fraction of particles {%)

Figure 1 Shocks vs. Volume fraction

25 0 5 10Size ratio (-)

Figure 2 Shocks vs. Size Ratio

15

Both above observations, suggest that by steering the size ratio of particles, a less viscous suspension with a higher partiele volume fraction can be obtained which is of interest for many applications. Moreover, for the purpose of extrusion, it would be interesting to see how an increase in size ratio reduces the extrusion time and how this relation relates to the number of shocks. Because both the partiele volume fraction and the size ratio are varied simultaneously in the above experiments, the partiele volume fraction will be fixed to 87.5 in the remaining.

co

Ëo»£

9:36

8:24

7:12

6:00

4:48

3:36

2:24

1:12

0:000.00

y = -0.0873X + 0.3981 Rz = 0.964

1.00 2.00 3.00 4.00Size Ratio (-)

21

CD 19co

JXLUo 17

2 13

110.00

y = -0.8149X + 20.703 '**T R2 = 0.9475

■•I T

2.00 4.00 6.00 8.00 10.00Size Ratio (-)

Figure 3 Extrusion time vs. Size Ratio Figure 4 Shocks vs. Size Ratio

From Figure 3 it can be seen that for an increase in size ratio, the extrusion time is lowered, meaning that the extrusion of fixed sample volume took less time. This can only be due to a lower viscosity of the samples with a higher size ratio. From Figure 4 it is clear that a similar trend is observed and that a lower viscosity also leads to a lower resistance to impact. However, the suitability of the jolting test to indicate extrudability or extrusion time is disowned, since a factor two increase in size ratio leads to almost a factor two decrease in extrusion time but only to approximately a factor 1.1 decrease in number of shocks.

It can be concluded that the jolting test, as proposed above, has only limited value in indicating the extrudability of a bidisperse partiele System. Although the flow spread range can be adapted to particular pastes etc., there is still a need to distinct the shock number within certain accuracy. For the experiments in this work, the accuracy of shock numbers was barely significant enough to predict the extrudability. Nevertheless, the qualitative correlation was indeed observed and could potentially be further refined towards significant ranges of shock numbers and flow spreads.

262