Multi-Objective Optimization Design through Machine Learning for Drop-on-Demand Bioprinting

Drop-on-demand (DOD) bioprinting has been widely used in tissue engineering due to its highthroughput efficiency and cost effectiveness. However, this type of bioprinting involves challenges such as satellite generation, too-large droplet generation, and too-low droplet speed. These challenges reduce the stability and precision of DOD printing, disorder cell arrays, and hence generate further structural errors. In this paper, a multi-objective optimization (MOO) design method for DOD printing parameters through fully connected neural networks (FCNNs) is proposed in order to solve these challenges. The MOO problem comprises two objective functions: to develop the satellite formation model with FCNNs;and to decrease droplet diameter and increase droplet speed. A hybrid multi-subgradient descent bundle method with an adaptive learning rate algorithm (HMSGDBA), which combines the multisubgradient descent bundle (MSGDB) method with Adam algorithm, is introduced in order to search for the Pareto-optimal set for the MOO problem. The superiority of HMSGDBA is demonstrated through comparative studies with the MSGDB method. The experimental results show that a single droplet can be printed stably and the droplet speed can be increased from 0.88 to 2.08 m·s^-1 after optimization with the proposed method. The proposed method can improve both printing precision and stability, and is useful in realizing precise cell arrays and complex biological functions. Furthermore, it can be used to obtain guidelines for the setup of cell-printing experimental platforms.

Effects of nozzle and fluid properties on the drop formation dynamics in a drop-on-demand inkjet printing

The droplet formation dynamics of a Newtonian liquid in a drop-on-demand (DOD) inkjet process is numerically investigated by using a volume-of-fluid (VOF) method. We focus on the nozzle geometry, wettability of the interior surface, and the fluid properties to achieve the stable droplet formation with higher velocity. It is found that a nozzle with contracting angle of 45° generates the most stable and fastest single droplet, which is beneficial for the enhanced printing quality and high-throughput printing rate. For this nozzle with the optimal geometry, we systematically change the wettability of the interior surface, i.e., different contact angles. As the contact angle increases, pinch-off time increases and the droplet speed reduces. Finally, fluids with different properties are investigated to identify the printability range.

Experimental Analysis of a Pneumatic Drop-on-Demand(DOD)Injection Technology for 3D Printing Using a Gallium-Indium Alloy

Many liquid metals have a high boiling point,strong electrical conductivity,high thermal conductivity,and nontoxic properties,which make them ideal targets for applications in different fields such as optics,microcircuits,electronic switches,micro-electromechanical System(MEMS)devices and 3D printing manufacturing.However,owing to the generally high surface tension of these liquids,achieving uniform micro-droplets is often a challenge due to the inherent difficulties in controlling their size and shape.In this study,a gallium indium alloy(GaIn24.5)has been used in combination with a pneumatic drop-on-demand(DOD)injection technology to carry out a series of experiments.The micro-droplet forming process has been explored for different pressure and pulse width conditions.Uniform metal droplets(diameter 1080μm)have been obtained with a 1.5 kPa jet pressure,100 ms pulse width,and 50%duty ratio.The standard deviation of the measured metal droplets diameter has been found to be approximately 20μm.

Effects of the actuation waveform on the drop size reduction in drop-on-demand inkjet printing

In this study the effects of the actuation waveforms on the droplet generation in a drop-on-demand inkjet printing are studied systematically by numerical simulations.Two different types of waveforms,namely the unipolar and bipolar actuations,are investigated for three fluids with different physical properties.We focus on two key parameters,which are the dwell time and the velocity amplitude.For the unipolar driving,the ejection velocity and the ejected liquid volume are both increased as the velocity amplitude becomes larger.The dwell time only has minor effects on both the ejection velocity and the ejected liquid volume.The ejection velocity decreases significantly for large liquid viscosity,while the influences of viscosity on the ejected liquid volume are much weaker.Four different droplet morphologies and the corresponding parameter ranges are identified.The droplet radius can be successfully reduced to about 40%e of the nozzle exit radius.For the bipolar waveforms,same droplet morphologies are observed but with shifted boundaries in the phase space.The minimal radius of stable droplet produced by the bipolar waveforms is even smaller compared to the unipolar ones.

Polyvinylpyrrolidone-based bioink:influence of bioink properties on printing performance and cell proliferation during inkjet-based bioprinting

Among the different bioprinting techniques,the drop-on-demand(DOD)jetting-based bioprinting approach facilitates contactless deposition of pico/nanoliter droplets ofmaterials and cells for optimal cell–matrix and cell–cell interactions.Although bioinks play a critical role in the bioprinting process,there is a poor understanding of the influence of bioink properties on printing performance(such as filament elongation,formation of satellite droplets,and droplet splashing)and cell health(cell viability and proliferation)during the DOD jetting-based bioprinting process.An inert polyvinylpyrrolidone(PVP360,molecular weight=360 kDa)polymerwas used in this study to manipulate the physical properties of the bioinks and investigate the influence of bioink properties on printing performance and cell health.Our experimental results showed that a higher bioink viscoelasticity helps to stabilize droplet filaments before rupturing from the nozzle orifice.The highly stretched droplet filament resulted in the formation of highly aligned“satellite droplets,”which minimized the displacement of the satellite droplets away from the predefined positions.Next,a significant increase in the bioink viscosity facilitated droplet deposition on the wetted substrate surface in the absence of splashing and significantly improved the accuracy of the deposited main droplet.Further analysis showed that cell-laden bioinks with higher viscosity exhibited higher measured average cell viability(%),as the presence of polymer within the printed droplets provides an additional cushioning effect(higher energy dissipation)for the encapsulated cells during droplet impact on the substrate surface,improves the measured average cell viability even at higher droplet impact velocity and retains the proliferation capability of the printed cells.Understanding the influence of bioink properties(e.g.,bioink viscoelasticity and viscosity)on printing performance and cell proliferation is important for the formulation of new bioinks,and we have demonstrated precise DOD deposition of living cells and fabrication of tunable cell spheroids(nL–μL range)using multiple types of cells in a facile manner.

Study on linear bio-structure print process based on alginate bio-ink in 3D bio-fabrication

Ink-jet printing is a non-impact printing technology in which drops jetted from an orifice onto a designated position.This technology can digitally transport fluids containing cells precisely onto desired substrates to construct three-dimensional organs.In order to obtain the stable uniform droplets,a stream is the key point of this technology.However,there are so many factors that affect the uniform droplet stream construction process:print parameters,material parameters,control method,etc.A good understanding of the various coupled transport processes that occur during bio-ink impact and spreading on bio-structure can improve the success of print-ability.This paper aims to obtain a good linear bio-structure with ink-jet printing technology.First,a typical droplet deposition process model is constructed;including droplet dynamics impact models and droplet diffusion cap models.Second,a model of successive droplet overlap,to form linear bio-structures,is constructed.Third,the finite element method is used to simulate the droplet impact,collision,and fusion process.Finally,the main influencing factors of the continuous injection printing process,namely the time interval between consecutive droplets and the droplet contact angles,are discussed.Sodium alginate is selected as bio-ink to verify the theory,and it is found that a good linear bio-structure could be obtained if the printing parameters are controlled optimally,i.e.,if the initial contact angle is set as 60 degrees and the trigger frequency is set as 150 kHz.With a proper printing speed and gel coating,a good survival rate of printed cells could be obtained.