3D-printing of undisturbed soil imaged by X-ray

Abstract: An artificial soil pore network can help to analyse preferential macropore flow which is important for pollutant leaching and degradation in the environment. Reproducing soil macro pores in an artificial, durable material offers the opportunity of repeating experiments in contrast to real soil pore networks. Therefore potential and limitations of reproducing an undisturbed soil sample by 3Dprinting was evaluated. An undisturbed soil column of Ultuna clay soil with a diameter of 7 cm was scanned by micro X-ray computer tomography at a resolution of 51 micron. A subsample cube of 2.03 cm length with connected macropores was cut out from this 3D-image and printed in five different materials by a 3D-printing service provider. The materials were ABS, Alumide, High Detail Resin, Polyamide and Prime Grey. The five print-outs of the subsample have been tested on their hydraulic conductivity by using the falling head method and the hydrophobicity has been tested by an adapted sessile drop method. To determine the morphology of the print-outs and compare it to the real soil the print-outs have been scanned by X-ray. The images were analysed with the open source program ImageJ. The five 3D-image print-outs copied from the subsample of the soil column were compared by means of their macropore network connectivity, porosity, surface volume, tortuosity and skeleton. The comparison of pore morphology between the real soil and the print-outs showed that Polyamide was the most consistent print-out while Alumide was the least detailed. Only the largest macropore was represented throughout all materials. Bottlenecks or dead ends in the printed pores were caused by lacking detail or residual support-material from the printing process. The physical analysis confirmed all materials as non-dissoluble and the sessile drop method shows angles between 54 and 75 degrees, rather wettable to slightly hydrophobic. Prime grey, Polyamide and ABS had a connected macropore throughout the sample and a hydraulic conductivity decreasing in this order, while two materials were not conducting. If the blocking of the pore was caused by faulty printing or printingaid material couldn’t be determined. Comparing the macropores in the soil and in the 3D-print-outs, the level of detail in the print-outs was not correlated with the infiltration velocity, residual printing material was suspected to block the pores in some materials. The thesis showed that the each material has its limitations but Prime Grey and Polyamide are prospective materials, although those and ABS need further research for residual material blocking pores. Independent from the 3Dprinting material, the fine pore matrix cannot be printed. Therefore soil with connected macropores is required.