Elsevier

Additive Manufacturing

Volume 34, August 2020, 101192
Additive Manufacturing

Printing direction dependent microstructures in direct ink writing

https://doi.org/10.1016/j.addma.2020.101192Get rights and content

Highlights

  • Direct ink writing with acoustophoresis is used to write tailored composite filaments.

  • An analytical model to diagnose sources of direction dependence is proposed.

  • Nozzle rotational asymmetry and printer calibration influence direction dependence.

  • Yield stress fluid support geometry influences direction dependence.

Abstract

Direct ink writing enables deposition of multiphase filaments with designed microstructures. Using acoustophoresis, we establish a narrow distribution of microparticles at the center of a direct-write nozzle. The distribution shifts and widens after deposition depending on the printing direction. We use particle image velocimetry and digital image analysis to identify flows transverse to the printing direction and characterize particle distributions in the printed filament. Sources of direction-dependent effects include square nozzles, co-deposition of support material, a rotationally asymmetric microstructure established in the nozzle, and speed inaccuracies that occur in 3-axis gantries. We propose an analytical model for predicting print direction-dependent flows and particle distributions as a function of anisotropy of the particle distribution in the nozzle, a disturbed zone near the nozzle, fluid reshaping of the print bead, uniform rotation of the print bead, calibration of the ink and support nozzle positions, and 3D printer motor error. Using the model, we propose strategies for controlling direction dependent microstructures in direct ink writing. The analytical model can be easily adapted to similar direct-write applications to diagnose sources of direction dependent microstructures.

Introduction

Direct ink writing (DIW) is an additive manufacturing technique that can produce complex geometries across a broad range of compositions and microstructures, including ceramics [1], [2], hydrogel cell scaffolds [3], [4], [5], conductive alloys [6], [7], and polymer-ceramic composites [8], [9]. Especially for two-phase or composite materials, spatial gradients in structural and functional properties can be integrated with geometric features via techniques which control microstructures “on-the-fly” during printing such as acoustophoresis [10], [11], [12], [13], [14], magnetic fields [15], and multimaterial mixing [16], [17], [18]. Advancing microstructural control requires a fundamental understanding of processing-structure relationships in DIW of multi-phase materials. One of the most critical processing-structure relationships is between printing direction and the resultant filament microstructure.

One of the central challenges of controlling microstructure during DIW is the presence of complex flow fields that connect the particle distribution in the nozzle to the distribution in the deposited line. Here, acoustic focusing is used to establish a narrow, anisotropic particle distribution in the nozzle (Fig. 1C). Critically, flows transverse to the printing direction on the substrate may disrupt this “target” microstructure and influence the final distribution of particles in the printed filament. These flows can cause narrow particle distributions established inside the nozzle to shift within the print path or widen as the filament exits the nozzle (Fig. 1A and B). Thus, if these flows depend on print direction, they could cause variations in properties throughout a printed part. For example, when using DIW with acoustophoresis to write carbon fibers in an insulating polymer matrix [12], direction dependent flows could cause variations in conductivity throughout the part. Alternatively, direction dependence could cause different sides of a printed SiC fiber-epoxy polygonal prism to have different stiffnesses [10]. In this work, we measure transverse flow velocities on the substrate experimentally, using inks with suspended particles and in situ optical imaging to track fluid movement. By writing single-layer, three-pass polygons, we probe various boundary conditions and print directions.

Several effects lead to print direction dependent microstructures. In this study, square nozzles are used because they enable effective acoustic control over microstructures in the nozzle [10], [11], [12]. While conceptually such anisotropic nozzles can be rotated to preserve alignment with the print path, doing so is a complicated controls problem. By investigating the mechanisms through which this misalignment influences filament microstructure, we highlight more elegant strategies for achieving microstructural control in DIW. A yield stress support material, implemented in a layer-by-layer or bath geometry, introduces a second source of direction dependent microstructures (Fig. 1D). Additionally, calibration of the stage introduces direction dependence. Understanding all of these sources is critical for controlling the microstructure of printed parts.

To diagnose sources of print direction dependent flows and particle distributions, we postulate several physical mechanisms and propose corresponding idealized analytical models of the flow patterns as a function of printing direction. To illustrate how to use the models and identify strategies for controlling the mechanisms, we fit these models to experimental observations. This identifies dominant mechanisms within our printing system and more general guidelines for selecting printing parameters which mitigate or accentuate direction dependence. Briefly, the mechanisms considered here are: (1) misalignment of the acoustic focusing direction and print direction, which alters the orientation of particle packing, (2) flows in a zone around the nozzle that disturb flows parallel to the printing direction (i.e. the disturbed zone), (3) reshaping of the fluid column exiting the nozzle as it is driven to lie flat on the substrate with a symmetrical fluid surface, (4) overlaps between the written line and existing features, and (5) motor errors which misalign designed and written paths. The models that describe these effects assume idealized flows and assume that changes in particle distributions result directly from transverse flows within the filament.

This work is part of a group of three papers which consider different aspects of the same data set. Ref. [19] considers printing direction-independent effects which manifest in straight lines. This paper considers direction-dependent effects in straight lines. Ref. [20] considers changes in microstructure at printed corners.

Section snippets

Hypotheses

We propose six effects which contribute to printing direction dependent microstructures (Table 1). For each effect, we formulate a dependence of either particle distribution position and transverse flow velocity or particle distribution width on the printing direction ϕ. A linear superposition of these models is used to fit experimental data and thus diagnose sources of print direction dependent microstructures and flows. Microstructure anisotropy involves a rotationally asymmetric distribution

Experimental approach

This paper draws from the same data set that is used in Refs. [19], [20], but this work focuses specifically on direction-dependent behaviors. Data and code can be found at Ref. [22].

Inks consisted of diurethane dimethacrylate (UDMA) (Sigma Aldrich, mixture of isomers with topanol inhibitor), triethylene glycol dimethacrylate (TEGDMA) (Sigma Aldrich, with MEHQ inhibitor), fumed silica (Evonik Aerosil R106), camphorquinone (CQ) (Sigma Aldrich), and 2-(Dimethylaminoethylmethacrylate) (DMAEMA)

Particle distribution width

Digital image analysis of in situ videos of the printed line just behind the nozzle indicates that the width of the distribution of particles in the printed line varies as a function of the print direction. Focusing anisotropy predicts that the distribution width should vary as a function of print direction because the distribution of particles in the nozzle is rectangular, and the width of the distribution in the line will be the projection of the rectangle onto the cross-section of the print

Discussion

The experimental particle distribution width in layer-by-layer support has a stronger dependence on the printing direction ϕ than the distribution in bath support. Comparing the experimental distribution width to the theoretical width, the shape of the printing direction-width profile indicates that the distribution of particles in the nozzle is rectangular rather than square, with a large aspect ratio. Bath support has a smaller dependence on printing direction than layer-by-layer support

Conclusions

Herein we proposed an analytical model that describes printing direction-dependent effects on the flow field near the nozzle and the microstructure of the printed line. Using acoustophoresis, we established a narrow distribution of particles in the nozzle which changes upon exiting the nozzle. By tracking fluid flows transverse to the printing direction and measuring the width and peak position of the particle distribution in the printed line, we determined that the proposed model predicts many

Authors’ contribution

Leanne Friedrich: conceptualization, methodology, software, formal analysis, investigation, data curation, writing – original draft, visualization. Matthew Begley: conceptualization, writing – review & editing, supervision.

Conflict of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Institute for Collaborative Biotechnologies through contract no. W911NF-09-D-0001 from the U.S. Army Research Office and a UCSB Chancellor's Fellowship. The work used the Microfluidics Laboratory at the California Nanosystems Institute and the Polymer Characterization Facility supported by the MRSEC Program of the National Science Foundation under award NSF DMR 1121053 at UCSB.

References (29)

  • R.A. Barry et al.

    Direct-write assembly of 3D hydrogel scaffolds for guided cell growth

    Adv. Mater.

    (2009)
  • Y. Jin et al.

    Functional nanoclay suspension for printing-then-solidification of liquid materials

    ACS Appl. Mater. Interfaces

    (2017)
  • T.J. Hinton et al.

    Three-dimensional printing of complex biological structures by freeform reversible embedding of suspended hydrogels

    Sci. Adv.

    (2015)
  • J.W. Boley et al.

    Direct writing of gallium-indium alloy for stretchable electronics

    Adv. Funct. Mater.

    (2014)
  • Cited by (16)

    • The emerging frontiers in materials for functional three-dimensional printing

      2022, Digital Manufacturing: The Industrialization of "Art to Part" 3D Additive Printing
    • A review on voids of 3D printed parts by fused filament fabrication

      2021, Journal of Materials Research and Technology
      Citation Excerpt :

      As the name implies, ME builds sliced 3D models layer-by-layer with viscous material (in paste or semi-liquid form) extruded through a nozzle or orifice [7]. Applications of ME include direct ink writing (DIW) [8,9], 3D concrete printing [10] and fused filament fabrication (FFF) [11]—also known as fused deposition modeling (FDM™) [12,13]. FFF is arguably the most widespread ME-based AM technique [5,6].

    • Additive manufacturing of polymeric composites from material processing to structural design

      2021, Composites Part B: Engineering
      Citation Excerpt :

      The printing nozzle allows a high deposition rate of 25–50 kg/h, and exhibit a large coarse and fine resolution in 7.6 mm and 5.1 mm [57]. DIW and reactive extrusion are newly developed AM processes to additively deposit a reactive liquid or slurry on a platform and then it will be immediately cured by thermal or ultraviolet (UV) sources to form solid structures as shown in Fig. 2 [35,60,61]. Usually, fibers or other additives are added to thermoset resin to modify the viscosity of feedstock materials so as to become sufficient viscous to hold their shape during printing.

    • Corner accuracy in direct ink writing with support material

      2020, Bioprinting
      Citation Excerpt :

      Ref. [18] considers printing direction-independent effects which manifest in straight lines. Ref. [25] considers direction-dependent effects in straight lines. This paper considers changes in microstructure at printed corners.

    View all citing articles on Scopus
    View full text