Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2019 Jul 12;4(7):12088–12097. doi: 10.1021/acsomega.9b00090

Effect of Polymer Binder on the Synthesis and Properties of 3D-Printable Particle-Based Liquid Materials and Resulting Structures

Nicholas R Geisendorfer †,‡, Ramille N Shah ‡,§,∥,*
PMCID: PMC6682019  PMID: 31460322

Abstract

graphic file with name ao-2019-00090u_0007.jpg

Recent advances have demonstrated the ability to 3D-print, via extrusion, solvent-based liquid materials (previously named 3D-Paints) which solidify nearly instantaneously upon deposition and contain a majority by volume of solid particulate material. In prior work, the dissolved polymer binder which enables this process is a high molecular weight biocompatible elastomer, poly(lactic-co-glycolic) acid (PLGA). We demonstrate in this study an expansion of this solvent-based 3D-Paint system to two additional, less-expensive, and less-specialized polymers, polystyrene (PS) and polyethylene oxide (PEO). The polymer binder used within the 3D-Paint was shown to significantly affect the as-printed and thermal postprocessing behavior of printed structures. This development enables users to select one of several polymers to impart the most desirable properties for a given application. Additionally, 3D-Paints based on these new binders are not adversely affected by classes of particles that can chemically degrade PLGA, notably particles containing large quantities of alkali ions. This study demonstrates the ability to successfully use PS and PEO as binders in the 3D-Paint system and compares the rheological, mechanical, microstructural, and thermal properties of the modified 3D-Paints and resulting as-printed and thermally post-processed objects. These objects include, for the first time, structures resulting from 3D-Painting which mostly contain soda-lime glass and 45S5 bioactive glass.

Introduction

The term additive manufacturing is used to describe a broad class of processes which involve the building of a part from a 3D model in an additive manner, often in a layer-by-layer process. Among the numerous advantages of additive manufacturing techniques is the ability to fabricate user-defined parts, including those with complex internal architecture or porosity, without the need for tooling, masks, or dies and with little material waste. The subclass of additive manufacturing processes which involve the deposition of feedstock materials through a nozzle or printhead are collectively termed 3D-printing,1 which includes (but is not limited to) direct ink writing, fused-deposition modelling, 3D-inkjet printing, and numerous others. The details of printing methods, compatible materials, and applications have been thoroughly reviewed elsewhere.2,3 Collectively, these techniques have been used to additively manufacture many different materials, including metals and alloys,4−7 ceramics8−10 and glasses,11−16 thermoplastics,17−19 and hydrogels and biomaterials.20−22

However, a single technique has been seldom used to additively manufacture hundreds of separate materials across many material classes. Recent progress by Shah, et al. has focused on developing a material-agnostic approach to 3D-printing which extrudes filaments of material containing a majority of embedded, functional particles within a polymeric matrix. This technique, named 3D-Painting, which has so far been used to successfully 3D-print metals and metal oxides,23−27 bioceramics,28 biologically-derived materials,29 2D nanomaterials,30,31 planetary regoliths,32,33 water-soluble salts,23,34 and mixed material systems,35 is capable of processing such a broad range of particulate materials because of the purely physical interaction between the particles, polymer, and solvents. The interaction between the components is physical, in that particles are simply embedded in a matrix of polylactic-co-glycolic acid (PLGA) which precipitates out of dichloromethane (DCM) solution upon extrusion because of the volatility of the DCM under atmospheric conditions. Although the ability to 3D-print a wide range of particulate materials using only the PLGA binder has historically been viewed as advantageous in its simplicity, there remain limitations to the use of a PLGA binder. For instance, PLGA is not chemically stable in the presence of all particles, especially those containing significant amounts of alkali elements/ions such as Li and Na; previous work has thoroughly established the fact that PLGA (as well as other polymers in the α-hydroxyester family) is degraded by alkali hydrolysis.36−43 This chemical instability precludes the use of the 3D-Painting technique with high-alkali glasses, such as soda lime glass and bioactive glasses. The usefulness of glasses in the fields of optics,44,45 fluidics,46 and medicine47 is well demonstrated. While additively manufactured inorganic and bioactive glasses have been demonstrated previously,11−16 these efforts have primarily relied on stereolithographic techniques which depend on the interaction of light with a photocurable polymer to maintain shape.11−13 Other previously demonstrated processes require the use of high-temperature 3D-printing equipment16 or rely on the extrusion of slurries which require extended (>24 h) post-print drying times.14,15 3D-Painting offers a relatively facile, highly scalable, room-temperature methodology to 3D-print materials requiring no drying time after printing and containing a majority by volume of the functional material in the resulting printed structure.

For these reasons, the 3D-Painting system must be expanded to enable the use of binding polymers other than PLGA and more generally to those outside of the polyester family, which possess a high level of stability in the presence of alkali cations, to enable the processing of materials such as high-alkali glasses. Alternative binding polymers also enhance the versatility of the 3D-Painting system, by potentially imparting a different combination of properties than those imparted by PLGA, even for systems containing particles that do not affect the stability of PLGA. To this end, we present two alternatives to PLGA in the form of polystyrene (PS) and polyethylene oxide (PEO). While both PS and PEO have been used in 3D-printing applications in the past,48,49 neither of the polymers has been previously utilized as a binder for the 3D-Painting system. Newly developed 3D-Paints (also more generally referred to as “3D-inks” or simply “inks”) in this work are based on the well-established PLGA formulation but incorporate not only a different polymer binder (PS or PEO) but also different quantities of ethylene glycol butyl ether (EGBE) and dibutyl phthalate (DBP), low-volatility solvents which ensure both extrudability of these new inks and that the printed filaments will be self-supporting. 3D-Paints based on these new polymers are not only 3D-printable but are also capable of incorporating 45S5 bioactive glass and soda-lime glass. Because properties of composites containing particles are significantly influenced by the interaction between the particles and the matrix, as well as the properties of the surrounding matrix,32,35 it is not surprising that a substitution of PLGA for PS or PEO leads to a marked change in properties of both the 3D-Paints and the 3D-printed materials. This study characterizes and compares the rheological properties of the liquid 3D-Paints, the microstructural, mechanical, and thermal properties of 3D-printed materials, and briefly, the microstructure of debound and sintered 3D-printed objects using a range of both previously demonstrated particles and newly demonstrated alkali-containing glasses.

Results and Discussion

Ink Synthesis, Rheological Property Characterization, and 3D-Printing

A total of nine 3D-Paints were prepared for this study, seven of which were ultimately 3D-printed. 3D-Paints are prepared by dissolving the polymer in DCM and then adding the mixture of particles, EGBE, DBP, and excess DCM, to the dissolved polymer solution. These inks are formulated so that each 3D-printed material is 70% particles by volume (i.e., 30 vol % polymer) and included 45S5, SLG, and Fe metal particles with a PS binder, referred henceforth as 45S5:PS, SLG:PS, and Fe:PS, respectively, as well as (designated in the same way) 45S5:PEO, SLG:PEO, and Fe:PEO, and 45S5:PLGA, SLG:PLGA, and Fe:PLGA for particle inks made with PEO and PLGA binders, respectively. For reasons described above, the 45S5:PLGA and SLG:PLGA inks were not 3D-printable. The Fe-containing inks were synthesized in order to enable a direct comparison between the properties of 3D-printed objects containing PLGA, PS, or PEO (because Fe particles have no adverse effect on PLGA and have been demonstrated previously23,24). A schematic of the synthesis and processing steps involved in ink preparation, 3D-printing, and postprocessing/application are shown in Figure 1. Like the previously demonstrated PLGA-based materials, PEO and PS-based materials can either be used immediately or stored in air-tight, glass containers at approximately 4 °C until use. The combined pre-ink is allowed to thicken by DCM evaporation at ambient conditions, or under gentle sonication and gentle heating (which accelerates the process), until the inks reach the appropriate viscosity for 3D-printing. The 3D-printable viscosity for PLGA-based inks prepared in this manner has been previously reported to fall within the range of 30–35 Pa·s23,32,34 (rheological measurements performed for this study reveal the “appropriate viscosity” for the PS-based inks to fall within a range of 1–5 Pa·s (see Figure 2) but do not directly identify viscosity for PEO-based inks).

Figure 1.

Figure 1

Open in a new tab

Polymers (PLGA, PS, or PEO) are dissolved in excess DCM while particles are separately mixed with EGBE, DBP, and excess DCM. Once polymer dissolution is complete, the two components are combined and homogenized. The 3D-Paints are thickened via evaporation of DCM to appropriate printing viscosity and then extruded layer-by-layer. The resulting 3D-printed objects can be handled and used as printed, or undergo additional processing, such as pyrolysis and sintering, depending on the particle content and end application.

Figure 2.

Figure 2

Open in a new tab

Viscosity curves obtained from rotational tests (A,B) and storage and loss moduli curves obtained from oscillatory measurements (C–E). Viscosity of (degraded) 45S5:PLGA and SLG:PLGA was such that stresses higher than several Pa could not be generated by the instrument (A). All materials tested as ready-to-print; in this state, PEO materials could not be correctly tested in rotation without sample slipping.

All inks were thickened to their appropriate 3D-printing viscosity and the flow properties measured with a rheometer using both rotational shear and frequency sweep measurements (Figure 2). Differences in the flow behavior are apparent to the naked eye; PS-based inks possess a lower viscosity than (nondegraded) PLGA-based inks, while PEO-based inks have a dramatically higher viscosity (Video S1). In fact, the large viscosity of printable PEO-based inks, which better resembles a modeling putty or clay than the liquid-like PLGA- and PS-based inks, meant that they could not be successfully tested in rotation on the rheometer without sample slipping, thus violating the criterion for laminar flow in the material. Therefore, a 3D-printing viscosity for the PEO-based inks is not reported. In substitution of the rotational tests, frequency sweeps were conducted on the as-printable PEO materials (and also conducted on PLGA and PS-containing inks, for completeness). Additionally, in order to elucidate the effect of the presence of particles on 3D-printable materials, entirely polymeric inks were prepared (PLGA, PS, PEO dissolved in DCM and with added EGBE, DBP, but without particles) and measured. In order to mitigate the effects of drying (via DCM evaporation) during the measurements, a solvent trap was used. PLGA and Fe:PLGA materials exhibit shear-thinning behavior as expected, with Fe:PLGA possessing low-shear viscosity similar to the previously reported ideal range of 30–35 Pa·s (Figure 2A). SLG:PLGA and 45S5:PLGA, on the other hand, possess substantially lower low-shear viscosities, presumably because of the previously investigated degradation of the PLGA polymer in the presence of alkali ions,36−43 such as those present in SLG and 45S5. In fact, the viscosities are so low in these inks that the rheometer was unable to generate shear stress in the material across the entire test range, hence the apparently truncated curves in Figure 2A. Low shear viscosities of the as-printable PS-based inks are lower that than for PLGA and fall within a range of 1–5 Pa·s. The frequency sweeps (Figure 2C–E) generally reveal that the storage modulus (G′) of the inks is larger than loss modulus (G″), indicating dominant solid-like behavior, although it must be noted that conducting the measurement at all requires exposing the materials to air, albeit briefly and despite the use of the solvent trap, such that there may be some solidification that occurs as a result. More interesting is that in the case of PEO-based materials, G′ exceeds G″ at larger angular frequencies by nearly an order of magnitude, whereas the difference is significantly smaller in the same range for the PS- and PLGA-based inks. This result correlates with the qualitative, visual observation that the PEO-based materials are more “putty-like”, and the PLGA and PS-based materials possess a more-obvious liquid character. Otherwise, the results of the frequency sweeps are consistent with the data obtained from the rotational tests. The viscosity of the inks containing 45S5 and SLG is always higher than for those containing Fe (barring the case of PLGA, again because of polymer degradation). This effect is likely a result of the different particle size regime of the Fe (with a range of approximately 1–3 μm) versus SLG and 45S5 which are approximately 44 μm (corresponding to a 325 mesh)50 as seen in Figure 3.

Figure 3.

Figure 3

Open in a new tab

SEM micrographs of 3D-printed strut cross-sections of 3D-printed materials, except 45S5:PLGA and SLG:PLGA (A,B) which could not be printed. In these two cases, degraded ink was allowed to dry and subsequently imaged. Micrographs capture different morphologies and sizes of 45S5, SLG, and Fe particles (note different scale bar for Fe-containing samples). Unlike for 45S5:PLGA and SLG:PLGA, in the Fe:PLGA ink, the PLGA polymer is clearly visible coating the particles (C). PS and PEO polymers are not noticeably different in morphology for inks containing 45S5 and SLG particles than for Fe (D–I).

For each ink, 3D-printing occurs via extrusion followed by rapid solidification of the material because of precipitation of the binding polymer resulting from DCM vaporization. The results are 3D-printed filaments which are self-supporting, maintain shape, and fuse to previously printed layers because of residual solvent content (Videos S2–S4). Samples for tensile and compression testing (“dogbones” and cylinders) as well as fibers for thermal measurements and cylinders for thermal processing and sintering were 3D-printed (Figure S1). All 3D-printing was performed on a single, commercial 3D-printing instrument, although print parameters (linear speed, extrusion pressure, and so forth.) were tailored for each ink (Figure S2). See the Supporting Information for details on exact size and geometry of 3D-printed objects and on 3D-printing process parameters.

Microstructure

The microstructure of the 3D-printed objects was imaged using scanning electron microscopy (SEM). Micrographs showing cross-sections of 3D-printed struts are given in Figure 3. For the cases of 45S5:PLGA and SLG:PLGA inks, which could not be printed, degraded ink was allowed to dry and was subsequently imaged (shown in Figure 3A,B). Images capture different morphologies, sizes, and size distributions of 45S5, SLG, and Fe particles, demonstrating that the method is sufficiently robust to enable the 3D-printing of particles of various shapes and size regimes. Notably, the binding polymer is not evident in the microstructure of 45S5:PLGA and does not uniformly coat particles in SLG:PLGA, as it does in the case of Fe:PLGA (Figure 3C), likely because of degradation via alkali hydrolysis of PLGA by the alkali content in 45S5 and SLG. The PS- and PEO-based materials, on the other hand, are not noticeably different in morphology for inks containing 45S5 and SLG particles than for those containing Fe (Figure 3D–I), further supporting the hypothesis that these polymers are not chemically affected by the alkali content of the glasses. Apparent texture in the polymer in SLG:PEO (Figure 3H) and Fe:PEO (Figure 3I) is likely introduced by the sectioning processes prior to imaging.

Tension and Compression Properties

The mechanical properties of 3D-printed objects are influenced by the properties of the polymer binder; therefore, the choice of the binder has a marked effect on the properties of the 3D-printed objects. Samples printed from each 3D-printable ink were tested in both tension and compression. Representative stress versus strain curves for tensile specimens 3D-printed from each material are presented in Figure 4A. Compared to previously established PLGA-based materials, PS-based materials are stronger and stiffer, whereas PEO-based materials are generally more compliant and fail at much larger strains (Figure 4B,C). Thicker polymer bridges between particles, like those observed under the SEM in the case of PEO-based materials, enable much larger tensile strain-to-failure observed in the PEO-based materials compared to PLGA and PS (although the mechanical properties of the polymer itself are still a major contributor). For both PS- and PEO-based materials, the Fe-containing samples are stiffer and fail at lower strains than for the SLG- and 45S5-containing samples. This result is again likely attributable to the smaller particle size distribution of the Fe particles, which leads to smaller polymer bridges between particles (because volume fraction is the same in all cases), which strain harden more quickly and lead to failure at lower strains.

Figure 4.

Figure 4

Open in a new tab

(A) Representative stress vs strain curves from tensile tests on 3D-printed tensile specimens (note the break in the abscissa). (B) Tensile moduli calculated from the elastic portion of each stress vs strain curves and averaged over each sample. (C) Average strain to failure for each 3D-printed material.

The mechanical properties in compression are highly dependent on the 3D-printed geometry because the cylinders are not fully dense but rather have regular, macroscopic 3D-printed architecture (Figure S1).24 Representative compressive load versus strain curves are presented in Figure S3A. The PLGA- and PEO-based cylinders appear to compress uniformly and densify with increasing strain, while the PS-based cylinders tend to fail catastrophically and then pulverize in sections (Figure S3B, Video S5). The PLGA-based cylinders are elastic until approximately 5% strain and reach a compressive load of nearly 200 N, at which point the load (and therefore engineering stress) plateaus as the cylinder densifies until reaching approximately 42% strain. Beyond this point, the load required to continue compression increases at an accelerating rate. PEO-based cylinders exhibit a qualitatively similar behavior to PLGA-based cylinders, but instead strain under smaller loads plateauing between 50 and 75 N before beginning a sharp increase. PS-based cylinders are brittle and fail catastrophically in sections; after the failure of a particular section of the cylinders, the load naturally drops until the test apparatus encounters an intact section of the cylinder, at which point the load rises again until catastrophic failure of the underlying section. This process repeats until the entire cylinder is pulverized, usually after the failure of 3–4 separate sections. In both cases of PS and PEO, 45S5-containing cylinders begin to yield at lower loading than Fe or SLG-containing counterparts. This observation is a likely result of the different particle geometry; the 45S5 particles are highly angular and introduce stress concentrations into the polymer binders which lead to yield at lower loadings. For the spherical and relatively monodisperse Fe and SLG particles, stress throughout the loaded binder is more homogeneous, therefore yielding occurs at higher loading. In general, given these observations, PS would prove itself useful in an application where strength and rigidity, rather than ductility, are valued properties, whereas PEO would better serve in an application requiring large plasticity without failure.

Thermal Property Characterization

The rheological measurements and examination of the microstructure under SEM both support the idea of PLGA degradation by 45S5 and SLG. In order to empirically confirm the degradation of PLGA, as well as the stability of PS and PEO in the presence of 45S5 and SLG, fibers of each 3D-printed material were prepared for both thermogravimetry (TGA) and differential scanning calorimetry (DSC). In the cases of 45S5:PLGA and SLG:PLGA, samples of the ink were taken and allowed to dry overnight in ambient air before testing. For points of comparison, the as-received polymers and polymers dissolved in DCM and reprecipitated via solvent evaporation were also tested. As-received PLGA experiences glass transition at approximately 65 °C and melting at approximately 166 °C, as similarly reported in previous studies on the properties of PLA dissolved in DCM and reprecipitated51 (Figure 5A). Thermal decomposition of PLGA begins at approximately 275 °C with an endothermic peak at approximately 370 °C (Figure 5D), with complete decomposition by 400 °C (Figure 5A). A standard enthalpy of fusion (ΔHm*) between 93.0 and 93.6 J/g has historically been used for PLGA polymers possessing a majority of lactide monomers,51−53 based on the melting enthalpy of polylactic acid (PLA). Standard enthalpies of 93.0 and 197 J/g54 are used for estimations of PLGA and PEO crystallinity, respectively, according to eq 1, after the endotherms have been properly scaled to the mass of the polymer in the sample.

graphic file with name ao-2019-00090u_m001.jpg 1

Figure 5.

Figure 5

Open in a new tab

Thermogravimetric (A–C) and calorimetric (D–F) measurements on as-received polymer (orange), polymer dissolved in DCM and re-precipitated (green) and Fe- (gray), SLG- (red), and 45S5- (blue) containing 3D-printed fibers (all normalized to the mass of the overall sample). Because particle-containing inks were prepared as 70 vol % particles, weight fraction of pyrolizable material varies with particle density. (A) 45S5:PLGA and SLG:PLGA exhibit significant mass loss associated with PLGA-constituents prior to PLGA pyrolysis in Fe:PLGA or in as-received and re-precipitated PLGA. This difference is not observed in PS- and PEO-containing fibers (B,C).

Crystallinity of the polymer for various samples is displayed in Figure S4. In general, dissolving the polymer in DCM and reprecipitating or 3D-printing leads to lower crystallinities compared to the as-received polymer. Similar levels of crystallinity would be expected for all 3D-printed PEO-based materials, given the apparent process-dependence of crystallinity, but higher crystallinity measured in SLG:PEO and 45S5:PEO may additionally be due to the presence of Na ions in the ink, which has been observed to increase crystallinity in PEO.55 Because the atactic PS is completely amorphous, no melting endotherm is observed in the PS-based materials (Figure 5E) and crystallinity is not calculated. The only visible endotherm for PS-based materials peaks at approximately 420 °C (Figure 5E), which is associated with substantial mass loss from the TGA measurement, and therefore corresponds to thermal degradation of PS (Figure 5B). The exception is for the PS decomposition endotherm of Fe:PS which peaks at approximately 450 °C. The mass loss curve from the independent TGA measurement is correspondingly offset. The PEO-based materials exhibit melting endotherms peaking at approximately 72 °C and thermal degradation (Figure 5F) at 350 °C (with corresponding mass loss, Figure 5C), consistent with previous observations.56 Thermal degradation of PEO also leaves behind a larger proportion of nonvolatile carbon residue, as seen in the larger mass fraction remaining in PEO-based materials after thermal degradation. Previous reports indicate that approximately 10% of the weight of PEO fails to become volatile, which is consistent with our results.56,57

Notably, we observe that the samples 45S5:PLGA and SLG:PLGA lack the characteristic mass loss associated with the thermal degradation of PLGA observed in the as-received, reprecipitated, and Fe-containing PLGA materials and also observe that this difference is not found in either the PS- or PEO-based materials. The evaporation and degradation temperatures for lactide and glycolide monomers have been previously reported and are indeed lower than those of the PLGA copolymer.58 Because of the 45S5 powder’s demonstrated tendency to adsorb water from the air, we expect that some mass loss occurring in 45S5-containing materials can be attributable to the evaporation of water at and above 100 °C.59 In addition, it has been observed that 45S5 begins shedding surface hydroxide (−OH) groups at approximately 400 °C.59 Desorption of surface hydroxides likely accounts for the slow mass loss in 45S5-containing materials at or above this temperature. The mass loss is most obvious in the 45S5:PLGA (Figure 5A) curve because PLGA decomposes at the lowest temperature of the three polymers employed in this study, exacerbated by the fact that degradation of polymer shifts mass loss associated with the vaporization of the polymer or its constituents to lower temperatures. Nevertheless, the loss associated with evaporation of surface hydroxides is presumably present in the 45S5:PS and 45S5:PEO materials as well. Limited mass loss associated with the evaporation of residual DBP and EGBE is also expected (at approximately 340 and 171 °C, respectively).

Sintering and Porosity

To focus on the effect of polymer binders on the thermal processing and sintered microstructure, we limit measurements in this study to the already-established metallic Fe system.24 3D-printed cylinders of Fe:PLGA, Fe:PS, and Fe:PEO were thermally processed under flowing forming gas (95% N2/5% H2) by ramping from room temperature to 450 °C and holding for 1 h to allow for binder burnout, followed by a ramp to 900 °C and holding for 2 h before ramping down to room temperature all at a rate of 5 °C/min. Cylinders before and after sintering are shown in Figure 6A. The nitrogen/hydrogen gas mixture was selected as the processing atmosphere in order to prevent oxidation and thermochemically reduce any surface oxides already present on the Fe particles. 3D-printed cylinders undergoing thermal processing experience homogeneous shrinkage associated with binder burnout and particle sintering and also preserve 3D-printed architecture without warping, cracking, or the delamination of 3D-printed layers. No significant differences in the radial shrinkage of the cylinders were observed (Figure 6B); as expected, the mass losses measured in Fe cylinders from PLGA and PS are both narrowly distributed and significantly different from one another, resulting from the different densities of the polymers (Figure 6C). More surprising is that the Fe cylinders from PEO exhibited a much wider distribution in mass loss between samples, and in aggregate are not significantly different from the cylinders resulting from PLGA or PS (based on the polymer density, the average mass loss is expected to be the greatest of the three). The reason for this wider distribution is unclear, and it is possible that distribution would narrow upon repeated measurement with a larger sample size. Notably, different binders employed led to significantly different porosities in the sintered Fe otherwise subjected to identical thermal treatments. The average porosity (density) exhibited by Fe resulting from Fe:PLGA was 27.2 ± 0.7% (72.8 ± 0.7%), whereas Fe:PEO and Fe:PS exhibited lower porosities of 15.0 ± 4.5% (85.0 ± 4.5%) and 5.0 ± 1.4% (95.0 ± 1.4%), respectively, as seen in Figure 6D. Representative micrographs demonstrating the porosity of these samples are shown in Figure 6E–G. Combustion fusion analysis performed on sintered samples indicates similarly limited dissolved N content across tested samples (Figure S5A). Similar levels of dissolved N indicate that N is likely not responsible for differences in sintering, and therefore porosity, between measured samples, and the relatively low levels of measured nitrogen generally indicate that Fe samples are not adversely affected by sintering in a N2/H2 mixed atmosphere. Combustion fusion analysis also indicates that carbon content in Fe trends qualitatively with the contribution, by weight, of carbon in the polymer, which is expected assuming that the polymer binder is the primary source of residual carbon remaining after firing (Figure S5B). Of note is that both residual carbon contents (presumably from polymer and solvents) and nitrogen contents in sintered Fe samples are at levels approximately consistent with those in medium carbon steels (0.31–0.60% C, and <130 ppm N), indicating that compositions relevant for engineering applications are achievable without necessarily requiring carbon or nitrogen additions or removal. Although the reason for the observed difference in porosity is unclear and the subject of additional work, any impact of the binding polymer on the sintered porosity of metallic Fe may provide future users with the ability to select a binder for 3D-printing which results in the most desirable microstructure; for instance, nearly dense for a structural part, or possibly more porous for an energy or catalytic application. Whether these observations are true for other sintered metals, or ceramics and glasses is the subject of future work.

Figure 6.

Figure 6

Open in a new tab

Fe-containing 3D-printed cylinders are thermally processed to debind and sinter Fe particles into a fully metallic structure. (A) Cylinders undergoing thermal processing experience homogeneous shrinkage associated with binder burnout and particle sintering, but also preserve 3D-printed architecture without warping or cracking. (B,C) Radial shrinkage and mass loss, respectively, experienced by 3D-printed Fe-containing cylinders undergoing thermal processing. (D) Porosity remaining within 3D-printed Fe struts measured from polished cross-sections. (E–G) Representative micrographs of polished strut cross-sections demonstrating difference in remaining porosity within sintered Fe cylinders.

Summary

We have demonstrated that 3D-Paints can be successfully synthesized based on the incorporation of polystyrenes and polyethers, evidenced by the use of PS and PEO in this study, in addition to the polyesters demonstrated previously. The ability to incorporate binding polymers with chemical stability in the presence of alkali cations has enabled the 3D-printing of majority-by-volume 45S5 bioactive glass and soda-lime glass, which chemically degrade the more-established PLGA, rendering the material unprintable. More generally, the availability of binders coming from different classes of polymers (polyesters, polyethers, and polystyrenes demonstrated thus far) enables the user to select a polymer binder which imparts the most desirable properties for the application, which might range from serving as a simple mechanical support for metal or ceramic greenbodies prior to thermal processing to a bioresorbable implant for regenerative medicine. Included in the selection criteria are price considerations, which are strengths of the demonstrated PS and PEO compared to the better-established and pricier PLGA (Millipore Sigma lists PLGA at $126.00/g, PS at $0.26/g and PEO at $54.10/g at the time of this writing). Although not demonstrated in this study, blends of polymers or polymer classes in the binding contents of the 3D-printable materials may also enable a range of properties tunable at the discretion of the user. In addition, multimaterial 3D-printing (either with inks made from the same particles but different binders or with different particles as well) is possible and would enable the fabrication of more complex structures, for example, using PEO as a water-soluble support material. The inherent material flexibility of 3D-Paints paired with the now-broadening selection of polymers available as binders further expands the horizon for the use of 3D-Painting in manufacturing for medical, structural, and energy applications.

Experimental Section

3D-Ink Synthesis

3D-inks were synthesized in ratios of 7:3 particles:polymer (by volume after solidification) by combining particles (Fe (spheres, APS 1–3 μm, 98+% carbonyl iron, Alfa Aesar), SLG (spheres, APS <25 μm, Mo-Sci), and 45S5 Bioactive Powder (APS <25 μm, Mo-Sci)) and EGBE, DBP solvents (Sigma-Aldrich) with polymers first dissolved in excess DCM: PLGA, (Resomer LG 824 S, 82:18 lactide:glycolide; density = 1.15 g/cm3, Evonik Industries, GmbH), PS (Mw = 350 000, density = 1.04 g/cm3, Sigma-Aldrich), and PEO (Mw = 1 000 000, density = 1.21 g/cm3 Sigma-Aldrich) in a BioPlotter cartridge (Nordson EFD). PLGA-based inks contain 0.9 g of EGBE, 0.45 g of DBP per cm3 powder, PS-based inks contain 0.6 g of EGBE, 0.1 g of DBP per cm3 powder, and PEO-based inks contain 0.6 g of EGBE, 0.03 g DBP per cm3 powder. The particle slurry is then added to the fully-dissolved polymer solution and the ink is homogenized by vortex mixing for 2–3 min. After combination, the inks were thickened (via DCM evaporation) to appropriate viscosity in a sonic water bath at elevated temperature (40–50 °C).

3D-Printing

All 3D-printing was performed at room temperature using a 3D-BioPlotter (EnvisionTEC, GmbH) with extrusion pressures ranging between 0.5 and 5.0 bar and linear speeds of 3–6 mm/s for PEO-containing inks, and 8–15 mm/s for PLGA and PS-based inks (see Figure S2). Inks were extruded from a 410 μm diameter nozzle (Nordson EFD).

Rheology

Rheometry was performed using an MCR Rheometer (Anton Paar) with a couette fixture (double-gap cylinder attachment, gap distance = 0 mm) on PLGA-, PS-based inks, and the 20 mm parallel plate attachment for PEO-based inks (gap distance = 1 mm). Rotational tests were stress-controlled with a stress range of 100–0.01 Pa. Frequency sweeps were performed at a strain amplitude of 1%. All tests were performed at 25 °C with the aid of a solvent trap, and all tests had a measuring point duration of 3 s.

Mechanical Testing

In order to eliminate the effect of residual solvents on polymer mechanical properties (especially DBP, the plasticizer), PLGA and PS-containing samples were washed and lyophilized according to the method established in prior work.28 PEO-containing samples were washed in 100% ethanol for 1 h and lyophilized until dry. Tensile measurements were performed using an LF Plus Mechanical Tester (Lloyd Instruments, USA) with a rate of extension of 2.0 mm/s. Six samples of each material were tested. Compression tests were performed on compression cylinders (1 cm diameter, 2 cm tall) using an MTS Sintech 20G at a compression rate of 2.0 mm/s.

Thermal Measurements

TGA measurements were performed using a SDTA851e Thermogravimetric Analyzer (Mettler Toledo), and DSC measurements were performed using a DSC822e Differential Scanning Calorimeter (Mettler Toledo), both under flowing N2 atmosphere between 30 and 500 °C at heating rates of 10 °C/min.

Imaging

SEM imaging was performed using a Hitachi S-4800 under an accelerating voltage of 2 kV. Optical microscopy/metallography was performed using an Olympus PMG3 inverted metallurgical microscope.

Sintering and Shrinkage/Porosity Analysis

Sintering of 3D-printed Fe samples occurred in an alumina furnace boat inside a tube furnace (Carbolite, Type CTF 17/300) under a flowing forming gas atmosphere (5% H2/95% N2, ∼5 scfm, Airgas). The furnace protocol involved a ramp at 5 °C/min to 450 °C and a subsequent hold lasting 1 h for complete polymer pyrolysis, followed by another ramp at the same rate to 900 °C for 2 h for Fe particle sintering and finally ramping down at 5 °C/min to room temperature. Linear reduction measurements were performed using a caliper and reported reduction was taken as the average radius (of three measurements) before and after thermal processing. Mass reduction was measured by weighing shrinkage cylinders before and after thermal processing. Porosity was measured by a stereological method using image analysis software ImageJ on polished sample cross sections. In ImageJ, the “binarize” function is used after setting a threshold to turn pixels either white or black (black for pores, in this case). After binarizing, the pixels are counted, thereby providing the fraction associated with pores. Images are captured of six different regions of a single sample, and the average porosity and standard deviation between regions are reported.

Elemental Analysis

Analysis of trace elemental compositions was performed using combustion fusion analysis, performed externally by Genitest, Inc. (Montreal, Canada).

Statistical Methods

Error bars represent one standard deviation from the mean, and significance tests were two-tailed Student’s t-tests with two sample unequal variance with a p-value of 0.05.

Acknowledgments

The authors gratefully acknowledge Prof. David C. Dunand and Dr. Christoph Kenel, for recommendations on interpreting data from sintered Fe samples, and Dr. Adam Jakus for helpful comments on the manuscript. This work made use of the MatCI Facility which receives support from the MRSEC Program (NSF DMR-1720139) of the Materials Research Center at Northwestern University, the Central Laboratory for Materials Mechanical Properties supported by the MRSEC program of the National Science Foundation (DMR-1720139), the EPIC facility of Northwestern University’s NUANCE Center, which has received support from the Soft and Hybrid Nanotechnology Experimental (SHyNE) Resource (NSF ECCS-1542205); the MRSEC program (NSF DMR-1720139) at the Materials Research Center; the International Institute for Nanotechnology (IIN); the Keck Foundation; and the State of Illinois, through the IIN, and the Simpson Querry Institute (SQI) at Northwestern University, which received funding from the Army Research Office, the U.S. Army Medical Research and Material Command, and Northwestern University. N.R.G. was supported by a NASA Space Technology Research Fellowship (80NSSC17K0192). Additional funding support was provided through a gift from Google (R.N.S.).

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.9b00090.

  • Images of 3D-printed samples used in this study; schematic showing the 3D-printing parameter space for inks used in this study; compressive properties of 3D-printed cylinders produced for this study; calculated polymer crystallinities in 3D-printed/dried inks based on DSC melting endotherms; nitrogen and carbon content of metallic Fe specimens produced for the study, determined by combustion fusion analysis (PDF)

  • Inks at appropriate viscosities for 3D-printing. 45S5:PLGA is degraded and therefore does not thicken appropriately (AVI)

  • 3D-printing of compression cylinders from PLGA-based inks (AVI)

  • 3D-printing of compression cylinders from PS-based inks (AVI)

  • 3D-printing of compression cylinders from PEO-based inks (AVI)

  • Compression of 3D-printed cylinders and qualitative differences in the mechanical behavior between cylinders containing different polymer binders are readily visible (AVI)

The authors declare the following competing financial interest(s): Authors NRG and RNS have submitted an invention disclosure to Northwestern University with intent to file a patent application on the subject of this work. RNS is a co-founder of and shareholder in Dimension Inx, LLC which develops and manufactures new advanced manufacturing compatible materials and devices for medical and non-medical applications, and could potentially benefit from the outcome of this research. As of November 2018, RNS serves part time as Chief Science Officer of Dimension Inx, LLC. Dimension Inx, LLC did not influence the conduct, description or interpretation of the findings in this study.

Supplementary Material

ao9b00090_si_001.pdf (712.5KB, pdf)
ao9b00090_si_002.avi (15.3MB, avi)
ao9b00090_si_003.avi (27.9MB, avi)
ao9b00090_si_004.avi (76.9MB, avi)
ao9b00090_si_005.avi (100.9MB, avi)
ao9b00090_si_006.avi (7.7MB, avi)

References

  1. ASTM Standard F2792 . Standard Terminology for Additive Manufacturing Technologies; West Conshocken: PA, 2015. [Google Scholar]
  2. Vaezi M.; Seitz H.; Yang S. A Review on 3D Micro-Additive Manufacturing Technologies. Int. J. Adv. Manuf. Technol. 2013, 67, 1721–1754. 10.1007/s00170-012-4605-2. [DOI] [Google Scholar]
  3. Ngo T. D.; Kashani A.; Imbalzano G.; Nguyen K. T. Q.; Hui D. Additive Manufacturing (3D Printing): A Review of Materials, Methods, Applications and Challenges. Compos. Part B Eng. 2018, 143, 172–196. 10.1016/j.compositesb.2018.02.012. [DOI] [Google Scholar]
  4. Xu C.; Wu Q.; L’Espérance G.; Lebel L. L.; Therriault D. Environment-Friendly and Reusable Ink for 3D Printing of Metallic Structures. Mater. Des. 2018, 160, 262–269. 10.1016/j.matdes.2018.09.024. [DOI] [Google Scholar]
  5. Xu C.; Bouchemit A.; L’Espérance G.; Laberge Lebel L.; Therriault D. Solvent-Cast Based Metal 3D Printing and Secondary Metallic Infiltration. J. Mater. Chem. C 2017, 5, 10448–10455. 10.1039/c7tc02884a. [DOI] [Google Scholar]
  6. Xu C.; Quinn B.; Lebel L. L.; Therriault D.; L’Espérance G. Multi-Material Direct Ink Writing (DIW) for Complex 3D Metallic Structures with Removable Supports. ACS Appl. Mater. Interfaces 2019, 11, 8499–8506. 10.1021/acsami.8b19986. [DOI] [PubMed] [Google Scholar]
  7. Ladd C.; So J.-H.; Muth J.; Dickey M. D. 3D Printing of Free Standing Liquid Metal Microstructures. Adv. Mater. 2013, 25, 5081–5085. 10.1002/adma.201301400. [DOI] [PubMed] [Google Scholar]
  8. Eckel Z. C.; Zhou C.; Martin J. H.; Jacobsen A. J.; Carter W. B.; Schaedler T. A. Additive Manufacturing of Polymer-Derived Ceramics. Science 2016, 351, 58–62. 10.1126/science.aad2688. [DOI] [PubMed] [Google Scholar]
  9. Zanchetta E.; Cattaldo M.; Franchin G.; Schwentenwein M.; Homa J.; Brusatin G.; Colombo P. Stereolithography of SiOC Ceramic Microcomponents. Adv. Mater. 2016, 28, 370–376. 10.1002/adma.201503470. [DOI] [PubMed] [Google Scholar]
  10. Peng E.; Wei X.; Herng T. S.; Garbe U.; Yu D.; Ding J. Ferrite-Based Soft and Hard Magnetic Structures by Extrusion Free-Forming. RSC Adv. 2017, 7, 27128–27138. 10.1039/c7ra03251j. [DOI] [Google Scholar]
  11. Kotz F.; Arnold K.; Bauer W.; Schild D.; Keller N.; Sachsenheimer K.; Nargang T. M.; Richter C.; Helmer D.; Rapp B. E. Three-Dimensional Printing of Transparent Fused Silica Glass. Nature 2017, 544, 337–339. 10.1038/nature22061. [DOI] [PubMed] [Google Scholar]
  12. Kotz F.; Plewa K.; Bauer W.; Schneider N.; Keller N.; Nargang T.; Helmer D.; Sachsenheimer K.; Schäfer M.; Worgull M.; et al. Liquid Glass: A Facile Soft Replication Method for Structuring Glass. Adv. Mater. 2016, 28, 4646–4650. 10.1002/adma.201506089. [DOI] [PubMed] [Google Scholar]
  13. Tesavibul P.; Felzmann R.; Gruber S.; Liska R.; Thompson I.; Boccaccini A. R.; Stampfl J. Processing of 45S5 Bioglass by lithography-based additive manufacturing. Mater. Lett. 2012, 74, 81–84. 10.1016/j.matlet.2012.01.019. [DOI] [Google Scholar]
  14. Nguyen D. T.; Meyers C.; Yee T. D.; Dudukovic N. A.; Destino J. F.; Zhu C.; Duoss E. B.; Baumann T. F.; Suratwala T.; Smay J. E.; et al. 3D-Printed Transparent Glass. Adv. Mater. 2017, 29, 1701181. 10.1002/adma.201701181. [DOI] [PubMed] [Google Scholar]
  15. Eqtesadi S.; Motealleh A.; Miranda P.; Lemos A.; Rebelo A.; Ferreira J. M. F. A simple recipe for direct writing complex 45S5 Bioglass 3D scaffolds. Mater. Lett. 2013, 93, 68–71. 10.1016/j.matlet.2012.11.043. [DOI] [Google Scholar]
  16. Luo J.; Gilbert L. J.; Qu C.; Landers R. G.; Bristow D. A.; Kinzel E. C. Additive Manufacturing of Transparent Soda-Lime Glass Using a Filament-Fed Process. J. Manuf. Sci. Eng. 2016, 139, 061006. 10.1115/1.4035182. [DOI] [Google Scholar]
  17. Ahn S. H.; Baek C.; Lee S.; Ahn I. S. Anisotropic Tensile Failure Model of Rapid Prototyping Parts - Fused Deposition Modeling (FDM). Int. J. Mod. Phys. B 2003, 17, 1510–1516. 10.1142/s0217979203019241. [DOI] [Google Scholar]
  18. Ahn S. H.; Montero M.; Odell D.; Roundy S.; Wright P. K. Anisotropic Material Properties of Fused Deposition Modeling ABS. Rapid Prototyp. J. 2002, 8, 248–257. 10.1108/13552540210441166. [DOI] [Google Scholar]
  19. McCullough E. J.; Yadavalli V. K. Surface Modification of Fused Deposition Modeling ABS to Enable Rapid Prototyping of Biomedical Microdevices. J. Mater. Process. Technol. 2013, 213, 947–954. 10.1016/j.jmatprotec.2012.12.015. [DOI] [Google Scholar]
  20. Rutz A. L.; Hyland K. E.; Jakus A. E.; Burghardt W. R.; Shah R. N. A Multimaterial Bioink Method for 3D Printing Tunable, Cell-Compatible Hydrogels. Adv. Mater. 2015, 27, 1607–1614. 10.1002/adma.201405076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Zhu J. Bioactive Modification of Poly(Ethylene Glycol) Hydrogels for Tissue Engineering. Biomaterials 2010, 31, 4639–4656. 10.1016/j.biomaterials.2010.02.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Emami S. H.; Salovey R. Crosslinked Poly(Ethylene Oxide) Hydrogels. J. Appl. Polym. Sci. 2003, 88, 1451–1455. 10.1002/app.11813. [DOI] [Google Scholar]
  23. Jakus A. E.; Taylor S. L.; Geisendorfer N. R.; Dunand D. C.; Shah R. N. Metallic Architectures from 3D-Printed Powder-Based Liquid Inks. Adv. Funct. Mater. 2015, 25, 6985–6995. 10.1002/adfm.201503921. [DOI] [Google Scholar]
  24. Taylor S. L.; Jakus A. E.; Shah R. N.; Dunand D. C. Iron and Nickel Cellular Structures by Sintering of 3D-Printed Oxide or Metallic Particle Inks. Adv. Eng. Mater. 2017, 19, 1600365. 10.1002/adem.201600365. [DOI] [Google Scholar]
  25. Taylor S. L.; Shah R. N.; Dunand D. C. Ni-Mn-Ga Micro-Trusses via Sintering of 3D-Printed Inks Containing Elemental Powders. Acta Mater. 2018, 143, 20–29. 10.1016/j.actamat.2017.10.002. [DOI] [Google Scholar]
  26. Taylor S. L.; Ibeh A. J.; Jakus A. E.; Shah R. N.; Dunand D. C. NiTi-Nb Micro-Trusses Fabricated via Extrusion-Based 3D-Printing of Powders and Transient-Liquid-Phase Sintering. Acta Biomater. 2018, 76, 359–370. 10.1016/j.actbio.2018.06.015. [DOI] [PubMed] [Google Scholar]
  27. Calvo M.; Jakus A. E.; Shah R. N.; Spolenak R.; Dunand D. C. Microstructure and Processing of 3D Printed Tungsten Microlattices and Infiltrated W-Cu Composites. Adv. Eng. Mater. 2018, 20, 1800354. 10.1002/adem.201800354. [DOI] [Google Scholar]
  28. Jakus A. E.; Rutz A. L.; Jordan S. W.; Kannan A.; Mitchell S. M.; Yun C.; Koube K. D.; Yoo S. C.; Whiteley H. E.; Richter C.-P.; et al. Hyperelastic ″bone″: A highly versatile, growth factor-free, osteoregenerative, scalable, and surgically friendly biomaterial. Sci. Transl. Med. 2016, 8, 358ra127. 10.1126/scitranslmed.aaf7704. [DOI] [PubMed] [Google Scholar]
  29. Jakus A. E.; Laronda M. M.; Rashedi A. S.; Robinson C. M.; Lee C.; Jordan S. W.; Orwig K. E.; Woodruff T. K.; Shah R. N. ″Tissue Papers″ from Organ-Specific Decellularized Extracellular Matrices. Adv. Funct. Mater. 2017, 27, 1700992. 10.1002/adfm.201700992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jakus A. E.; Secor E. B.; Rutz A. L.; Jordan S. W.; Hersam M. C.; Shah R. N. Three-Dimensional Printing of High-Content Graphene Scaffolds for Electronic and Biomedical Applications. ACS Nano 2015, 9, 4636–4648. 10.1021/acsnano.5b01179. [DOI] [PubMed] [Google Scholar]
  31. Guiney L. M.; Mansukhani N. D.; Jakus A. E.; Wallace S. G.; Shah R. N.; Hersam M. C. Three-Dimensional Printing of Cytocompatible, Thermally Conductive Hexagonal Boron Nitride Nanocomposites. Nano Lett. 2018, 18, 3488–3493. 10.1021/acs.nanolett.8b00555. [DOI] [PubMed] [Google Scholar]
  32. Jakus A. E.; Koube K. D.; Geisendorfer N. R.; Shah R. N. Robust and Elastic Lunar and Martian Structures from 3D-Printed Regolith Inks. Sci. Rep. 2017, 7, 44931. 10.1038/srep44931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Taylor S. L.; Jakus A. E.; Koube K. D.; Ibeh A. J.; Geisendorfer N. R.; Shah R. N.; Dunand D. C. Sintering of Micro-Trusses Created by Extrusion-3D-Printing of Lunar Regolith Inks. Acta Astronaut. 2018, 143, 1–8. 10.1016/j.actaastro.2017.11.005. [DOI] [Google Scholar]
  34. Jakus A. E.; Geisendorfer N. R.; Lewis P. L.; Shah R. N. 3D-Printing Porosity: A New Approach to Creating Elevated Porosity Materials and Structures. Acta Biomater. 2018, 72, 94. 10.1016/j.actbio.2018.03.039. [DOI] [PubMed] [Google Scholar]
  35. Jakus A. E.; Shah R. N. Multi and Mixed 3D-Printing of Graphene-Hydroxyapatite Hybrid Materials for Complex Tissue Engineering. J. Biomed. Mater. Res. Part A 2017, 105, 274–283. 10.1002/jbm.a.35684. [DOI] [PubMed] [Google Scholar]
  36. Blaker J. J.; Bismarck A.; Boccaccini A. R.; Young A. M.; Nazhat S. N. Premature degradation of poly(α-hydroxyesters) during thermal processing of Bioglass-containing composites. Acta Biomater. 2010, 6, 756–762. 10.1016/j.actbio.2009.08.020. [DOI] [PubMed] [Google Scholar]
  37. Pardie S. P. The Effects Of Chemical And Environmental Conditions On The Degradation Of Poly Lactic Acids. Int. J. Eng. Appl. Sci. 2012, 4, 17–23. [Google Scholar]
  38. Tracy M.; Ward K. L.; Firouzabadian L.; Wang Y.; Dong N.; Qian R.; Zhang Y. Factors Affecting the Degradation Rate of Poly(Lactide-Co-Glycolide) Microspheres in Vivo and in Vitro. Biomaterials 1999, 20, 1057–1062. 10.1016/s0142-9612(99)00002-2. [DOI] [PubMed] [Google Scholar]
  39. Simpson R. L.; Nazhat S. N.; Blaker J. J.; Bismarck A.; Hill R.; Boccaccini A. R.; Hansen U. N.; Amis A. A. A Comparative Study of the Effects of Different Bioactive Fillers in PLGA Matrix Composites and Their Suitability as Bone Substitute Materials: A Thermo-Mechanical and in Vitro Investigation. J. Mech. Behav. Biomed. Mater. 2015, 50, 277–289. 10.1016/j.jmbbm.2015.06.008. [DOI] [PubMed] [Google Scholar]
  40. Stamboulis A.; Hench L. L.; Boccaccini A. R. Mechanical Properties of Biodegradable Polymer Sutures Coated with Bioactive Glass. J. Mater. Sci. Mater. Med. 2002, 13, 843–848. 10.1023/a:1016544211478. [DOI] [PubMed] [Google Scholar]
  41. Blaker J. J.; Maquet V.; Boccaccini A. R.; Jerome R.; Bismarck A. Wetting of Bioactive Glass Surfaces by Poly(Alpha-Hydroxyacid) Melts: Interaction between Bioglass and Biodegradable Polymers. e-Polymers 2005, 5, 1–13. 10.1515/epoly.2005.5.1.248. [DOI] [Google Scholar]
  42. Shah Mohammadi M.; Ahmed I.; Marelli B.; Rudd C.; Bureau M. N.; Nazhat S. N. Modulation of Polycaprolactone Composite Properties through Incorporation of Mixed Phosphate Glass Formulations. Acta Biomater. 2010, 6, 3157–3168. 10.1016/j.actbio.2010.03.002. [DOI] [PubMed] [Google Scholar]
  43. Maquet V.; Boccaccini A. R.; Pravata L.; Notingher I.; Jérôme R. Preparation, characterization, andin vitrodegradation of bioresorbable and bioactive composites based on Bioglass-filled polylactide foams. J. Biomed. Mater. Res. Part A 2003, 66A, 335–346. 10.1002/jbm.a.10587. [DOI] [PubMed] [Google Scholar]
  44. Luo J.; Gilbert L. J.; Bristow D. A.; Landers R. G.; Goldstein J. T.; Urbas A. M.; Kinzel E. C.. Additive Manufacturing of Glass for Optical Applications; Gu B., Helvajian H., Piqué A., Eds.; International Society for Optics and Photonics, 2016; Vol. 9738, p 97380Y. [Google Scholar]
  45. Choi H.-K.; Ahsan M. S.; Yoo D.; Sohn I.-B.; Noh Y.-C.; Kim J. T.; Jung D.; Kim J. H.. Formation of Cylindrical Micro-Lens Array in Fused Silica Glass Using Laser Irradiations. In Micro/Nano Materials, Devices, and Systems; Friend J., Tan H. H., Eds.; International Society for Optics and Photonics, 2013; Vol. 8923, p 89234T. [Google Scholar]
  46. Gervais L.; de Rooij N.; Delamarche E. Microfluidic Chips for Point-of-Care Immunodiagnostics. Adv. Mater. 2011, 23, H151–H176. 10.1002/adma.201100464. [DOI] [PubMed] [Google Scholar]
  47. Vallet-Regí M.; Ruiz-Hernández E. Bioceramics: From Bone Regeneration to Cancer Nanomedicine. Adv. Mater. 2011, 23, 5177–5218. 10.1002/adma.201101586. [DOI] [PubMed] [Google Scholar]
  48. Rymansaib Z.; Iravani P.; Emslie E.; Medvidović-Kosanović M.; Sak-Bosnar M.; Verdejo R.; Marken F. All-Polystyrene 3D-Printed Electrochemical Device with Embedded Carbon Nanofiber-Graphite-Polystyrene Composite Conductor. Electroanalysis 2016, 28, 1517–1523. 10.1002/elan.201600017. [DOI] [Google Scholar]
  49. Nesaei S.; Rock M.; Wang Y.; Kessler M. R.; Gozen A. Additive Manufacturing With Conductive, Viscoelastic Polymer Composites: Direct-Ink-Writing of Electrolytic and Anodic Poly(Ethylene Oxide) Composites. J. Manuf. Sci. Eng. 2017, 139, 111004. 10.1115/1.4037238. [DOI] [Google Scholar]
  50. Metzner A. B. Rheology of Suspensions in Polymeric Liquids. J. Rheol. 1985, 29, 739. 10.1122/1.549808. [DOI] [Google Scholar]
  51. Guo S.-Z.; Heuzey M.-C.; Therriault D. Properties of Polylactide Inks for Solvent-Cast Printing of Three-Dimensional Freeform Microstructures. Langmuir 2014, 30, 1142–1150. 10.1021/la4036425. [DOI] [PubMed] [Google Scholar]
  52. Harris A. M.; Lee E. C. Improving Mechanical Performance of Injection Molded PLA by Controlling Crystallinity. J. Appl. Polym. Sci. 2008, 107, 2246–2255. 10.1002/app.27261. [DOI] [Google Scholar]
  53. Melo L. P. d.; Salmoria G. V.; Fancello E. A.; Roesler C. R. D. M. Influence of Processing Conditions on the Mechanical Behavior and Morphology of Injection Molded Poly(lactic-co-glycolic acid) 85:15. Int. J. Biomater. 2017, 2017, 1–8. 10.1155/2017/6435076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Blaine R. L.THERMAL APPLICATIONS NOTE Polymer Heats of Fusion.
  55. Strawhecker K. E.; Manias E. Crystallization Behavior of Poly(ethylene oxide) in the Presence of Na+Montmorillonite Fillers. Chem. Mater. 2003, 15, 844–849. 10.1021/cm0212865. [DOI] [Google Scholar]
  56. Madorsicy S. L.; Straus S. Thermal Degradation of Polyethylene Oxide and Polypropylene Oxide. J. Polym. Sci. 1959, 36, 183–194. 10.1002/pol.1959.1203613015. [DOI] [Google Scholar]
  57. Lewis J. A. Binder Removal From Ceramics. Annu. Rev. Mater. Sci. 1997, 27, 147–173. 10.1146/annurev.matsci.27.1.147. [DOI] [Google Scholar]
  58. D’Avila Carvalho Erbetta C.; Alves R. J.; Resende J. M.; Freitas R. F. d. S.; Sousa R. G. d. Synthesis and Characterization of Poly(D,L-Lactide-Co-Glycolide) Copolymer. J. Biomater. Nanobiotechnol. 2012, 03, 208–225. 10.4236/jbnb.2012.32027. [DOI] [Google Scholar]
  59. Lefebvre L.; Chevalier J.; Gremillard L.; Zenati R.; Thollet G.; Bernache-Assolant D.; Govin A. Structural Transformations of Bioactive Glass 45S5 with Thermal Treatments. Acta Mater. 2007, 55, 3305–3313. 10.1016/j.actamat.2007.01.029. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao9b00090_si_001.pdf (712.5KB, pdf)
ao9b00090_si_002.avi (15.3MB, avi)
ao9b00090_si_003.avi (27.9MB, avi)
ao9b00090_si_004.avi (76.9MB, avi)
ao9b00090_si_005.avi (100.9MB, avi)
ao9b00090_si_006.avi (7.7MB, avi)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES