DEVELOPMENT OF 3D-PRINTED MICROFLUIDIC DEVICES FOR SYNCHROTRON X-RAY BEAMLINES

ENABLING TECHNOLOGIES

180 ESRF

The quality of 3D-printed microfluidic devices was improved by filtering the LED light of a desktop 3D printer in combination with optimisation of the printing resin. Single-day turnaround from design to use, growth of crystals in micro-droplets carried by an oil phase, and in-device diffraction render the method suitable for X-ray investigations.

A. de Maria, S. Malbet-Monaco and J. McCarthy. ESRF

The authors wish to acknowledge the many people involved in the development, implementation and testing of the sample tracking module, including, in particular, the ICAT, PaNOSC and SMIS developers, members of the Experiment Safety team, and the User Office and Business Development Office staff. This project has received funding from the European Union s Horizon2020 research and innovation programme under grant agreements No. 870313 and No. 823852.

AUTHORS

ACKNOWLEDGEMENTS

The ICAT sample tracking tool was launched in time for the first run (25/08/2020 to 14/10/2020) of the ESRF restart following the EBS shutdown. During this run, 132 parcels were sent for experiments on non-MX beamlines (sample tracking for MX beamlines using ISPyB is not included in the scope of this report). More detailed information on the use of the sample tracking tool in that period is given in Table 2.

These developments have paved the way for many mail-in experiments to be carried out. This does not mean that the task of implementing a mail-in service is now complete. With the experience gained from the first implementation, the next phase will be to improve and adapt the tool based on feedback from users and analysis of the workflows. Features for providing links that make it easy to follow the sample from its creation in ICAT to data acquisition, its entry in the electronic logbook, and to the raw and processed data produced from the sample will be implemented during the next development cycle as part of STREAMLINE.

In the Run 2/2020 time period (25/08 to 14/10):

Parcels sent to the ESRF 132 # experiments 89 # samples 733 # equipment and tools 49

Table 2: Data showing use of the ICAT sample

tracking tool on non- MX beamlines during the first run of 2020.

The use of microfluidic devices as a sample delivery and characterisation platform on X-ray beamlines has been steadily growing over the past years [1,2]. In comparison with traditional sample-handling approaches, microfluidics provide access to shorter length and timescales while using smaller sample quantities. Examples in the areas of soft matter and life sciences include nanoparticle and crystal growth, protein folding and fibre alignment. However, the construction of microfluidic devices for use on X-ray beamlines is often a time-consuming and complex multi-step process [3,4]. Possible benefits of 3D printing are access to the third dimension for manufacturing and significant reduction of the time needed to turn a new concept into a working device. The current method of choice for 3D printing of microfluidic devices is the solidification of a resin under UV light irradiation using Digital Light Processing (Figure 156a). With this method, the channel sizes, until now, mainly remained in the range of several hundreds of micrometres because the UV light penetrates through previously printed

layers, resulting in unwanted solidification of the resin inside the microfluidic channels during the printing process. By filtering the long wavelengths of the LED source and carefully choosing the resin components, the UV penetration depth was efficiently reduced such that smaller channels and fine in-channel detail can now be printed (Figures 156b-c).

Using a standard desktop DLP printer modified by the addition of a bandpass filter, different microfluidic components were printed and tested. The microfluidic components, comprising particle filters, droplet generators and droplet traps, were combined into devices used for the crystallisation of biominerals (e.g., gypsum) and proteins (e.g., lysozyme and thaumatin). The device for gypsum crystallisation consisted of a droplet generator with three aqueous phase inputs and one oil input followed by a droplet trapping channel featuring a density of 224 traps within an area of 1.4 cm2. After filling the device, the flow was stopped, and crystals grew in all trapped droplets within 12 hours.