Dimitra Manatou, Arizona State University

Abstract

A Comprehensive and Automated Python-Based Pipeline for Microjet Characterization: Pushing the
Boundaries of GDVN Performance

Dimitra Manatou1, Konstantinos Karpos1, Sahba Zaare1, Vivek Krishnan1,
Roberto Alvarez1, Adil Ansari1, Reza Nazari1, Richard A. Kirian1

1 Department of Physics, Arizona State University, Tempe, Arizona 85287, USA

Gas Dynamic Virtual Nozzles (GDVNs) play a crucial role in XFEL crystallography and solution scattering experiments, yet their comprehensive characterization has been limited due to the tedium and complexity of both data collection and analysis procedures. To address this issue, we have developed a versatile pipeline, written in Python, that streamlines the characterization of GDVN micro- and nanojets. The pipeline offers multiple usage examples, including image cleaning with standard Python image
processing libraries, jet speed measurement, jet stability assessments, and the capability to determine jet diameters with precision that is better than the microscope resolution.

In addition to image analysis code, the pipeline includes a simple high-level Python interface to popular hardware such as Photron high-speed cameras, Sensirion liquid flowmeters, Bronkhorst gas flowmeters, Shimadzu HPLC pumps, and other devices. This allows a user to easily script automated scans of liquid and gas flow rates with the full flexibility of the Python language. A PyQt5 graphical interface is also provided for convenience. Such automation of both data collection and image processing allows for the
rapid production of diagrams that map gas/liquid flow conditions to jet properties (speed, diameter, angular variance, jetting/dripping status, etc.) and dimensionless numbers such as Weber, Reynolds, and Capillary numbers.

This comprehensive toolbox (measurement, control, analysis) equips GDVN users with the necessary tools to robustly characterize the “jetability” of their samples before conducting experiments. One can also conduct robust systematic characterizations of 3D-printed GDVN designs according to geometric design parameters along with liquid properties such as density, viscosity, surface tension, and non-Newtonian perturbations such as protein crystals. We are now in the process of applying our pipeline to various nozzles and liquids with the aim to fully optimize GDVN design. We also present our efforts to identify the smallest measurable jet, pushing the boundaries for nanojet production. We expect that users can eventually consult the relevant phase diagrams to identify the most suitable GDVN design for their specific sample characteristics, and thereby mitigate costly GDVN failures during XFEL beamtimes.

Acknowledgements:
This work is funded by the NSF BioXFEL STC (Award 1231306), NSF DBI Award 1943448, and NSF MCB Award 1817862.