Additive Manufacturing of Poly(phenylene Sulfide) Aerogels via Simultaneous Material Extrusion and Thermally Induced Phase Separation

Abstract

Additive manufacturing (AM) of aerogels increases the achievable geometric complexity, and affords fabrication of hierarchically porous structures. In this work, a custom heated material extrusion (MEX) device prints aerogels of poly(phenylene sulfide) (PPS), an engineering thermoplastic, via in situ thermally induced phase separation (TIPS). First, pre‐prepared solid gel inks are dissolved at high temperatures in the heated extruder barrel to form a homogeneous polymer solution. Solutions are then extruded onto a room‐temperature substrate, where printed roads maintain their bead shape and rapidly solidify via TIPS, thus enabling layer‐wise MEX AM. Printed gels are converted to aerogels via postprocessing solvent exchange and freeze‐drying. This work explores the effect of ink composition on printed aerogel morphology and thermomechanical properties. Scanning electron microscopy micrographs reveal complex hierarchical microstructures that are compositionally dependent. Printed aerogels demonstrate tailorable porosities (50.0–74.8%) and densities (0.345–0.684 g cm⁻³), which align well with cast aerogel analogs. Differential scanning calorimetry thermograms indicate printed aerogels are highly crystalline (≈43%), suggesting that printing does not inhibit the solidification process occurring during TIPS (polymer crystallization). Uniaxial compression testing reveals that compositionally dependent microstructure governs aerogel mechanical behavior, with compressive moduli ranging from 33.0 to 106.5 MPa.

Full text

RESEARCH ARTICLE
www.advmat.de
Additive Manufacturing of Poly(phenylene Sulfide) Aerogels
via Simultaneous Material Extrusion and Thermally Induced
Phase Separation
Garrett F. Godshall, Daniel A. Rau, Christopher B. Williams,* and Robert B. Moore*
Additive manufacturing (AM) of aerogels increases the achievable geometric
complexity, and affords fabrication of hierarchically porous structures. In this
work, a custom heated material extrusion (MEX) device prints aerogels of
poly(phenylene sulfide) (PPS), an engineering thermoplastic, via in situ
thermally induced phase separation (TIPS). First, pre-prepared solid gel inks
are dissolved at high temperatures in the heated extruder barrel to form a
homogeneous polymer solution. Solutions are then extruded onto a
room-temperature substrate, where printed roads maintain their bead shape
and rapidly solidify via TIPS, thus enabling layer-wise MEX AM. Printed gels
are converted to aerogels via postprocessing solvent exchange and
freeze-drying. This work explores the effect of ink composition on printed
aerogel morphology and thermomechanical properties. Scanning electron
microscopy micrographs reveal complex hierarchical microstructures that are
compositionally dependent. Printed aerogels demonstrate tailorable porosities
(50.0–74.8%) and densities (0.345–0.684 g cm−3), which align well with cast
aerogel analogs. Differential scanning calorimetry thermograms indicate
printed aerogels are highly crystalline (≈43%), suggesting that printing does
not inhibit the solidification process occurring during TIPS (polymer
crystallization). Uniaxial compression testing reveals that compositionally
dependent microstructure governs aerogel mechanical behavior, with
compressive moduli ranging from 33.0 to 106.5 MPa.
G. F. Godshall, R. B. Moore
Department of Chemistry
Macromolecules Innovation Institute
Virginia Tech
Blacksburg, VA 24061, USA
E-mail: rbmoore3@vt.edu
D. A. Rau, C. B. Williams
Department of Mechanical Engineering
Macromolecules Innovation Institute
Virginia Tech
Blacksburg, VA 24061, USA
E-mail: cbwill@vt.edu
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202307881
© 2023 The Authors. Advanced Materials published by Wiley-VCH
GmbH. This is an open access article under the terms of the Creative
Commons Attribution-NonCommercial-NoDerivs License, which permits
use and distribution in any medium, provided the original work is
properly cited, the use is non-commercial and no modifications or
adaptations are made.
DOI: 10.1002/adma.202307881
1. Introduction
Aerogels are a unique class of materials
comprised of a solid network phase and
a porous air-filled phase. These materi-
als boast high porosities and surface ar-
eas, combined with low densities and typ-
ically low thermal conductivities.[1,2]Aero-
gels are attractive materials for various
applications including mass reduction of
parts, thermal insulation, passive solar in-
sulation, absorption, filtration, drug de-
livery, and biomedical scaffolding, among
many others.[1]Aerogels have been fab-
ricated from silica and other inorganic
materials,[3]as well as synthetic[1,2]and bio-
based polymers,[4]metals,[5]and carbon[6]
using a variety of processing methods
including sol–gel processing,[7]chemical
cross-linking,[8]freeze-templating,[6]and
phase inversion.[9,10]
Traditionally, aerogels are processed via
casting routes, which limit the processed
geometries to monolithic primitive shapes
that may then be machined to the desired
final form via subtractive machining. A
recent evolution of aerogel processing is
the use of additive manufacturing (AM),
which allows for geometrical customization and the expansion of
structural complexity of the inherently lightweight microporous
bodies. Specifically, AM of aerogels affords the realization of parts
with hierarchical porosity, wherein the aerogel structure provides
stochastic nano/microporosity, and the printing process enables
tailoring of the meso/macroscale porosity via part design and lay-
ered processing. In addition, AM of aerogels could afford the op-
portunity to directly and conformally deposit the aerogel onto a
conventionally manufactured substrate.
Generally, fabrication of aerogels via AM has been accom-
plished through one of three manufacturing modalities: ma-
terial extrusion (MEX) of gels and pastes (also referred to as
direct ink write, DIW), material jetting (MJT), and vat pho-
topolymerization (VP).[6]Due to strict viscosity constraints im-
posed by MJT and VP, which specifically require low vis-
cosities (<10 Pa s),[11]DIW is a common route for printing
aerogel precursor resins. In DIW, liquidous inks are selec-
tively extruded through a nozzle moving on a defined tool-
path, and three-dimensional parts are constructed in a layer-
wise process.[11]One major benefit of DIW is its ability to
Adv. Mater. 2024,36, 2307881 2307881 (1 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
process a wide range of ink viscosities (>10 000 Pa s in
shear-thinning fluids),[12,13]including those loaded with various
polymers,[14,15]metals,[16 ]ceramics,[17]eutectic materials,[18 ]and
biomaterials.[19]To enable successful printing, the solid–liquid
transitions of the ink must be well-understood. During extrusion,
the ink must flow as a liquid through the nozzle onto the sub-
strate. Once deposited, the ink should undergo a rapid solidifica-
tion process whereby the extrudate displays a modulus increase
which allows it to maintain its shape and support subsequent
layers.[20]Solidification may occur via photopolymerization,[14,15 ]
yield-stress recovery,[21,22]chemical reactivity,[23]or solvent
evaporation.[24,25]
Broadly, DIW of a range of aerogels has been achieved by
printing inks containing solvent to produce solvated green parts.
Printed aerogels are formed once the solvent has been removed
from the green parts through common, nondestructive postpro-
cess solvent evacuation techniques such as supercritical CO2ex-
traction or freeze-drying. Recent work on the DIW printing of
aerogels reveals a variety of materials and methods utilized to
fabricate porous bodies. Chandrasekaran et al. fabricated carbon
aerogels via a sol–gel solidification technique using an aqueous
resorcinol-formaldehyde (RF) ink.[26]Solutions were extruded
into a solvent bath to prevent ink evaporation and subsequent
structural collapse; solidification of printed roads was enabled by
yield-stress reformation of the solids in the ink, enabling DIW
printing. A curing process initiated the sol–gel cross-linking step
of the RF, after which a solvent exchange process and super-
critical CO2drying formed the porous aerogel. Finally, two-step
carbonization and etching yielded the activated carbon aerogels,
which boasted high surface areas (1894 m2g−1)andhighca-
pacitance as compared to nonactivated DIW and bulk carbon
aerogels. Yao et al. also fabricated carbon aerogels via DIW, us-
ing cellulose nanocrystals (CNCs) as the carbon precursor.[27]
Briefly, they created aqueous CNC dispersions and also included
silica nanospheres to act as porogens. Parts were printed in
air and then freeze dried to create aerogels. The printed parts
were then carbonized, followed by the chemical removal of
SiO2to create macropores, and finished with a KOH treatment
to create nanoscale pores. This yielded a hierarchical porous
structure for improved performance as a supercapacitor at low
temperatures.
Synthetic polymers have also been fabricated into aerogels
via DIW, often using a phase inversion process to yield parts
with porous microstructures and complex geometries. Liu et al.
utilized a nonsolvent induced phase separation (NIPS) pro-
cess to create porosity in printed polyimide (PI) aerogels.[28]PI
powder was mixed with a good solvent for the polymer (N,N-
dimethylformamide, DMF) and a poor solvent for the poly-
mer (poly(ethylene glycol), PEG) to form a homogeneous so-
lution. As the ink was deposited, the good solvent volatilized,
decreasing the thermodynamic stability of the PI and PEG,
eventually leading to microphase separation and solidification
of PI domains. After printing, the PEG was removed and the
parts were dried, yielding porous PI aerogels boasting high
glass transition temperatures (Tg=315 °C) and high ten-
sile strengths. Interestingly, Di Luca et al. printed a porous
poly(ethylene oxide-terephthalate)/poly(butylene terephthalate)
(PEOT/PBT) copolymer ink via a thermally induced phase sep-
aration (TIPS) process.[29]The copolymer was dissolved in N-
methyl-2-pyrrolidone (NMP) at 70 °C inside a commercial bio-
plotter and selectively extruded in a layer by layer fashion.
Upon deposition, the solution cools and the system undergoes
phase separation, yielding a two-phase PEOT/PBT and NMP mi-
crostructure. After printing, NMP was exchanged with a volatile
alcohol or water, and the parts were ambiently dried, forming
PEOT/PBT aerogels. It was demonstrated that the initial polymer
concentration in the ink as well as the drying conditions affected
the microstructure.
In this work, the authors present an expansion of available
AM-processable aerogel materials. Specifically, the first reported
printing of poly(phenylene sulfide) (PPS) aerogels is realized.
PPS is an engineering thermoplastic with excellent thermal sta-
bility (Tg=90 °C, Tm=280 °C, UL 94 Flame Rating =V-0), me-
chanical properties (E=3.4 GPa, 𝜎y=80 MPa), and resistance
to many organic solvents.[30,31]PPS is often used as a substitute
for metal parts in aerospace and automotive applications where
weight reduction and chemical resistance is required.[32,33]Addi-
tive manufacturing of PPS has been demonstrated through the
use of MEX of molten filament (also referred to as fused filament
fabrication, FFF)[34,35]and powder bed fusion (PBF).[36,37 ]While
these approaches have shown the ability to produce dense PPS
parts with complex geometries, the inherent melt processing pre-
vents the production of complex PPS aerogel morphologies with
intentional microscale porosity.
To enable printing of PPS aerogel, the authors leverage prior
research in fabricating PPS[38]and other polymer[39–43 ]aerogels
via a TIPS phase inversion process. In this process, a poly-
mer is premixed with a solvent and brought to high tempera-
tures where polymer–solvent interactions become favorable, and
a homogenous polymer solution is formed. Gelation proceeds if
an appropriate gelation solvent is selected, where temperature-
dependent polymer-solvent interactions promote microphase
separation.[44,45]During cooling, the solution becomes thermo-
dynamically unstable, leading to microphase separation into
discrete polymer-rich and polymer-poor domains. This process
yields a continuous polymeric network imbibed with the gela-
tion solvent, forming a solvated gel.[44–46]Extraction of the gela-
tion solvent leaves behind an air-filled phase, creating a poly-
meric aerogel. Recently, we have demonstrated that PPS aero-
gels may be produced using an environmentally friendly, toxically
benign gelation solvent, 1,3-diphenylacetone (DPA), yielding ro-
bust fibrillar morphologies, low densities, and good mechanical
properties.[38]
While previous works have demonstrated simultaneous DIW-
TIPS printing of a variety of material systems, this work fea-
tures printing aerogels from a high-performance polymer with
high processing temperatures (>230 °C) that necessitate a cus-
tom DIW printing apparatus. The overall aims of this study are
to (1) create a generalizable protocol for printing PPS aerogels
via DIW, (2) to understand the solidification process dictating the
printability of the PPS/DPA system, and (3) to evaluate the effect
of polymer composition on the resulting morphological, physi-
cal, and mechanical properties of printed PPS aerogels. Further-
more, comparisons between printed PPS aerogels and cast aero-
gels are also drawn to interpret any influence of the printing pro-
cess on the physical and mechanical nature of PPS aerogels.
Adv. Mater. 2024,36, 2307881 2307881 (2 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
Figure 1. A diagram of the heated DIW process. a) PPS/DPA gel is created in a laboratory via TIPS, then broken up and added to the heated-DIW barrel.
b) The solvated gel dissolves at the hot end of the barrel, whereby pneumatically pressurized argon gas is used to extrude the polymer solution through a
nozzle. c) The printed road cools rapidly and solidifies via TIPS, and parts are then built in a layer-wise fashion. d) The printed gels are solvent exchanged
and freeze-dried to create PPS aerogels with a porous, semicrystalline morphology.
2. Results and Discussion
2.1. Printing PPS Aerogels
PPS solvated gels were successfully printed via selective extru-
sion of a PPS/DPA gel ink through a custom-designed, heated
DIW extrusion system. An annotated diagram of the heated DIW
extrusion process is shown in Figure 1. To make the ink, PPS
is first dissolved into a suitable high boiling point solvent. For
this work DPA was selected because it possessed a high boiling
point (Tb=330 °C) and it is relatively benign as an FDA-approved
food grade additive.[38]Upon forming a homogenous solution
at 270 °C, the system was cooled to room temperature, during
which it underwent a TIPS process to form a phase separated
PPS/DPA solvated gel. This organogel is then easily broken into
pellet-sized pieces constituting the gel ink feedstock (Figure 1a)
and loaded into the custom-built heated extruder (heated-DIW).
Within the lower heating zone, the temperature is raised to form
a liquid ink (Figure 1b). With application of pneumatic pressure,
the viscous PPS/DPA solution is extruded through the nozzle and
deposited upon the print substrate. The extrudate rapidly cools
once it contacts the room temperature substrate, at which point
it once again phase separates via TIPS and quickly solidifies by
crystallization of the PPS component[38](Figure 1c). Due to the
high viscosity of the extrudate and the rapid solidification, the ex-
truded road maintains its shape and is capable of supporting the
deposition of subsequent layers as the part is built. After printing,
the part is solvent exchanged and freeze-dried to remove the gela-
tion solvent, leaving behind a hierarchically structured, porous
PPS aerogel (Figure 1d).
A visual demonstration of the heated-DIW/TIPS printing pro-
cess is presented for the deposition of a linear bead (Figure 2a)
and a multilayer cylindrical build (Figure 2d). Initially, the hot
PPS/DPA liquid is clear and dark brown in color, but quickly
transforms to an opaque and tan solid appearance within sec-
onds after the heated nozzle moves past the deposited bead.
This change in visual appearance is attributed to the crystalliza-
tion/solidification process occurring during the TIPS thermore-
versible gelation taking place as the liquid cools and gels. Using
an infrared (IR) camera to monitor the temperature of deposited
bead during the deposition/build, it is clear that temperatures
above 200 °C (i.e., the crystallization temperature for a 30 wt%
PPS/DPA system, Figure S1, Supporting Information) are ob-
served as bright yellow/orange pixels only in regions very near
the moving nozzle. Moreover, the rapid cooling needed for so-
lidification and structural stabilization during the deposition of
multiple layers is apparent in the IR image of the cylindrical part
build (Figure 2e).
Using a temperature calibrated IR camera, in situ measure-
ments of the cooling rate of a printed PPS/DPA road may be
quantified to evaluate the solidification process during the DIW
print. Correlation between the observed cooling rate and ther-
mal transitions of the PPS/DPA solution measured via differen-
tial scanning calorimetry (DSC) (Figure S1, Supporting Informa-
tion) allows for a deeper understanding of the time-dependent
processes responsible for the solidification that enables success-
ful DIW printing. Figure 2c presents a plot of temperature versus
time of the highlighted pixel of the 30 wt% printed road (Inset of
Figure 2c). The cooling rate profile demonstrates that immedi-
ately upon deposition, the ink temperature drops from 216 °Cto
183 °C within just 0.97 s after extrusion, yielding a cooling rate
of −34.0 °C. According to the DSC thermograms, the polymer
in the 30 wt% PPS/DPA solution crystallizes at approximately
200 °C at the fastest possible cooling rate available to the DSC
used (≈60 °Cmin
−1)(FigureS1, Supporting Information). To-
gether, these cooling rate and DSC data indicate that the extru-
date spends less than a second in the solution state after deposi-
tion, prior to polymer crystallization (≈0.84 s). Consequently, at
the extrusion speed of 4 mm s−1, the bead is solid at distances
of less than 1 mm from the moving print nozzle. After the part
cools below ≈183 °C, the cooling rate gradually decreases until
the part equilibrates near room temperature, approximately 35 s
after deposition. Figure 2d,e demonstrates the ability of the PPS
gel ink to solidify fast enough to form stable beads that are needed
to create complex parts in a layer-wise fashion. The macro- and
microscale origins of this rapid solidification can be explained
Adv. Mater. 2024,36, 2307881 2307881 (3 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
Figure 2. a) Digital camera image of in situ DIW printed 30 wt% PPS gel in a single road deposition, with b) the corresponding in situ IR image of
the cooling bead, immediately after deposition. c) A temperature versus time profile of the highlighted pixel (pixel area =17 ×17 μm) of the single
road deposition (inset, quantified using the IR camera). d) Digital camera image of DIW printed 30 wt% PPS gel of the multilayer build of a cylindrical
part, with e) the corresponding in situ IR image of the cooling layers, during the cylindrical deposition. Extrusion speeds for all depositions were set at
4mms
−1.
through multiple thermal and material factors. A rapid cooling
rate of the extrudate containing fast-crystallizing PPS leads to
quick solidification upon extrusion. The initial precipitous drop
in temperature upon deposition, observed in Figure 2c is related
to the low thermal mass of the extrudate and the high temper-
ature gradient between the nozzle and the environment. PPS is
a fast crystallizing polymer, with reported isothermal crystalliza-
tion half times of 3–10 s at temperatures between 160 °Cand
200 °C.[47]Therefore, the almost-instantaneous solidification of
printed PPS/DPA roads can be explained by the fast drop in tem-
perature, which brings about rapid PPS crystallization.
Tall single-wall cylindrical parts (height =10–12 mm, OD
=10–12 mm) were also printed, as highlighted in Figure 2d.
Figure 2e reveals an in situ IR thermography image capturing
the printing of a hollow cylinder used for further physical and
mechanical testing. From the IR image, it is clear that heat is
retained in the upper two or three layers (approximately 100–
140 °C) for 30 to 60 s after being deposited, while the lower lay-
ers appear cool to about 60 °C in a few minutes after printing.
The deposition speed was set to 4 mm s−1for these experiments,
resulting in layer print times of ≈10 s. Residual heat is impor-
tant for remelting previously deposited layers, which improves
interlayer bonding in FFF printing.[48,49]A similar influence of
retained heat is thought to occur in PPS/DPA gel printing as well,
since the mechanisms of layer solidification (polymer crystalliza-
tion) and interlayer adhesion (interlayer entanglement) are pre-
sumed to be the same as in FFF. Figure 2d,e again reveals that the
printed bead begins to solidify so rapidly (<1 s) that the layer tran-
sitions to a solid prior to the completion of that layer (layer time
estimate =10 s, translation speed =4mms
−1). This is unique, as
typical DIW inks feature solidification mechanisms with slow so-
lidification rates (i.e., yield stress formation) where lower layers
may sag and spread during printing of tall structures. PPS/DPA
ink begins to opacify rapidly, indicating solidification (based on
previous work on PPS/DPA correlating PPS crystallization and
cloud point temperatures[38]). This allows for tall parts of many
layers to be printed from the PPS polymer solution.
2.2. Printed Aerogel Geometries
A benefit of additive manufacturing aerogels is the ability to
maintain porosity on the microscale while tailoring part poros-
ity on the mesoscale and part geometry on the macroscale. This
affords the opportunity to tailor aerogel architecture to satisfy spe-
cific design requirements and applications. Figure 3a,b displays
multiple printed parts from both 30 wt% (a) and 50 wt% (b) inks.
Printed parts include tensile bars, hexagonal lattices, and cylin-
ders. The heated-DIW system demonstrated the ability to form
parts with sharp corners, filled or hollow layers, and curvature.
Images of printed cylinders in Figure 3a,b shows high resolu-
tion layers (≈0.4 mm layer thickness) with good shape retention
afforded by rapid road solidification. Additionally, printed cylin-
ders highlight the ability of heated-DIW to print tall structures
of over 25 layers 0.5 mm each in height, indicating that both
30 and 50 wt% PPS/DPA inks possess sufficient gel strength for
supporting subsequent layer deposition. Parts do not contain ob-
servable macro-scale voids and there is no evidence of delami-
nation between layers, which suggests sufficient interlayer adhe-
sion, likely due to some degree of remelting (as confirmed by the
in situ IR thermography) as subsequent layers are deposited. Al-
though rapid polymer crystallization is associated with warping
of FFF parts,[50]the presence of solvent within the PPS/DPA ex-
trudate tends to decrease the volume change upon crystallization
when compared to a neat polymer, which may explain the lack of
shrinkage-induced defects in the gel parts. Some inconsistencies
in layer geometry and layer thickness were caused by tempera-
ture variations within the barrel, leading to viscosity changes and
slight over/under extrusion during a given print. These issues
can be remedied through further hardware development.
Adv. Mater. 2024,36, 2307881 2307881 (4 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
Figure 3. Digital photographs of DIW printed PPS gels; a) 30 and b) 50 wt%. c) Scanning electron microscope (SEM) micrographs of printed 30 wt%
PPS aerogels ((i), (iii), and (v)), and printed 50 wt% PPS aerogels ((ii), (iv), and (vi)). Magnification increases from left to right. d) SEM micrographs
of the cross sections of 50 wt% DIW printed PPS aerogel walls, which were printed with (i) no pause between layer deposition, and (ii) a 15 s pause
between layer deposition. Interlayer welds are highlighted between arrows.
After solvent exchange and freeze-drying, printed parts clearly
retain their as-printed size and shape. Interestingly, changes in
color are observed once the solvated gel is solvent extracted to
form the aerogel. The darker color of the solvated gels is due to
slight degradation of the solvent at the high extrusion temper-
atures. Precautions were taken to reduce the residence time of
solutions to avoid excess degradation.
2.3. Printed Aerogel Morphology
Understanding the morphology of aerogels allows for the de-
velopment of key process–structure–property relationships that
can be used to design microstructures yielding desired proper-
ties. Figure 3c highlights SEM micrographs that reveal the mor-
phologies of printed PPS aerogel cylinders. Figure 3c(i),(ii) dis-
plays four layer cross-sections of printed cylinders of 30 and
50 wt% inks, respectively. Both samples display smooth outer
skins, likely products of rapid cooling and solvent evaporation at
the exterior surface once phase separation begins upon ink depo-
sition, as seen in other TIPS systems.[51]The fast cooling rate of
the polymer solution in air leads to a concentration gradient at the
surface as solvent evaporates; as the quantity of solvent decreases,
the porosity of the surface layer decreases as well, forming a skin
at the extrudate surface.[51]
Figure 3c(iii)–(vi) reveals the finer textures of the aerogels fab-
ricated from precursor inks featuring 30 and 50 wt% solids, re-
spectively. Aerogels of both concentrations appear highly porous,
with visible micron-scale pores created upon the sublimation of
ice during the freeze drying process. Nitrogen sorption analy-
sis reveals Brunauer-Emmett-Teller (BET) surface areas of the 30
and 50 wt% printed aerogels to be 178.6 and 109.1 m2g−1,respec-
tively, as presented in Figure S2 (Supporting Information). These
values are high, indicating the presence of nanoscale porosity
in both 30 and 50 wt% aerogels which scales inversely with the
polymer composition. The surface areas of the printed parts are
somewhat lower than those of the 30 and 50 wt% aerogels cast in
a laboratory setting (284.5 and 178.6 m2g−1, respectively; Figure
S2, Supporting Information), indicating that the printing process
has a modest effect on the nitrogen-accessible surface area of PPS
aerogels. The differences between the cast and printed surface ar-
eas at each composition are potentially linked to the thick layer in-
terfaces present within the printed aerogels, which are discussed
in more detail below. A representative image of a cast PPS aerogel
is presented in Figure S3 (Supporting Information).
Interestingly, the morphology of the printed aerogel is highly
dependent on the ink concentration. In Figure 3c(iii),(v),30wt%
PPS aerogels display an isotropic fibrillar morphology. In our
previous work,[38]these fibrils have been identified as axi-
alites, which are premature spherulites that form under specific
processing conditions, including systems in which the nucle-
ation density is high.[52,53]In contrast, Figure 3c(iv),(vi) reveal
spherulitic morphologies of the 50 wt% printed aerogels. The
porous surfaces of the spherulites likely originate from solvent
which was expelled radially outwards as lamellae crystallize and
grow. Upon the removal of such solvent, intraspherulitic porosity
was formed.
The influence of polymer composition on microstructure is
an important relationship to understand, as the resulting part
morphology has implications for its performance. One explana-
tion for the observed morphological difference of the aerogels
formed from the different ink concentrations is that a specific
Adv. Mater. 2024,36, 2307881 2307881 (5 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
processing parameter, such as printing temperature, is affecting
the nucleation density of the inks, whereby the 30 wt% parts
display high nucleation densities and thus axialitic morpholo-
gies, and the 50 wt% parts with lower nucleation densities are
comprised of spherulites. Controlling nucleation density through
processing without the addition of nucleation agents has been
demonstrated through self-nucleation.[54]
When a semicrystalline polymer is heated above its melting
temperature and below its equilibrium melting temperature (i.e.,
the theoretical melting point for a crystal of infinite thickness),
then the polymer melt should be fully amorphous and should not
contain any residual crystallinity. However, depending on tem-
perature, there may exist some melt memory, whereby chains
still maintain correlations to each other in solution. These corre-
lations, also termed self-nuclei,[54]do not exist in a purely amor-
phous state nor a purely crystalline state and act as nucleation
agents, increasing the nucleation density of the system. When
a polymer solution or melt containing these self-nuclei is re-
crystallized, increased nucleation density typically yields smaller
spherulites or even axialites. Inversely, when a solution is dis-
solved at high enough temperatures to remove self-nuclei, the
nucleation density is reduced and larger spherulites form upon
crystallization. The presence of self-nuclei has been observed in
many polymers[54]; it has specifically been observed in PPS crys-
tallized from both the melt[55,56]and from solution.[53 ]
Work by Chiang and Lloyd[53]has demonstrated the effect
of the dissolution temperature, or the temperature at which a
solvated gel is dissolved back into a homogenous solution, on
the resulting morphology of thin PPS membranes fabricated
via TIPS. Membranes fabricated at higher dissolution tempera-
tures displayed spherulitic morphologies, while those formed at
low dissolution temperatures displayed axialitic morphologies,
indicating that self-nucleation has a strong affect on the mi-
crostructures of PPS membranes. The resulting morphologies
of printed 30 and 50 wt% PPS aerogels in this work suggests that
self-nucleation may be occurring during the printing process of
PPS/DPA. Using the IR camera, the average extrudate temper-
ature of the 30 wt% parts is 208 °C±8°C, while the temper-
ature of the barrel must be raised to extrude the more viscous
50 wt% solutions, resulting in higher extrudate temperatures
(243 ±8°C). Apparently, the extrusion temperatures required
to print the more concentrated ink (50 wt%) are high enough to
destroy self-nuclei, producing a spherulitic microstructure upon
crystallization. Interestingly, the morphologies of 30 and 50 wt%
PPS aerogels formed by a laboratory-scale casting process at the
same dissolution temperature presented in Figure S4 (Support-
ing Information) both reveal axialitic morphologies, confirm-
ing that the change in polymer concentration is not responsi-
ble for the difference in morphology between the 30 and 50 wt%
printed aerogels. These results suggest self-nucleation is respon-
sible for the observed morphological differences between the
printed aerogels.
Understanding the effect of printing parameters on the mor-
phology of the interlayer welds of a polymer AM part guides
processing conditions to maximize part effectiveness and to cre-
ate porous microstructures with as few heterogeneities as pos-
sible. This understanding also helps to interpret the process
physics governing any partial remelting of a previously printed
road once the next layer is deposited. Additionally, layer inter-
faces in AM polymer parts dictate many mechanical properties;
increased chain entanglements between layers at the interfaces
generally lead to stronger parts.[48,57]Figure 3d highlights results
from a simple experiment aimed at probing the effect of pause
time taken between the deposition of each layer on the interfaces
of a 10 layer 50 wt% PPS gel. Two printed parts were formed from
a 1 cm long line toolpath, 10 layers tall, at the same print speed
(8 mm s−1), pressure (10 psi), and layer height (0.3 mm). How-
ever, in one part, no pause was taken between the deposition of
successive layers (the cross section for the no-pause part is shown
in Figure 3d(i)), while in the other printed specimen a 15 s pause
was taken between the printing of each layer (Figure 3d(ii)). This
pause time is in addition to the approximately 5 s each layer takes
to be deposited. Note, a faster print speed was used in this ex-
periment to isolate the effect of pause time on morphology with
less influence of print time. The key result of this comparison
is that the interlayer welds, identified between the arrows shown
in Figure 3d(i),(ii) are much thicker in the z-direction in the part
printed without a pause taken between layers. Image analysis in-
dicates that the average interface thickness in the part printed
without pause is 81 ±28 microns, versus 51 ±31 microns in the
part printed with 15 s pauses between layers.
The thicker interfaces for the no-pause part can be explained
by the increase in the retained heat (as demonstrated in Figure 2d
for a sample printed without pauses). As a layer is printed, it
rapidly begins to lose heat and approaches the environmental
temperature, as observed in Figure 2c. The next deposited layer
then re-melts the previous layer, where polymer chains diffuse at
the interface, forming the interlayer weld. Therefore, upon depo-
sition of a given layer, the previously deposited layer is hotter in
the part printed without pausing. This aids in remelting that pre-
viously printed layer, and thus the part printed without pausing
displays thicker interfaces.
In the case of the PPS gels highlighted in Figure 3d, the welds
are much less porous than the bulk morphology. Recent simu-
lations of TIPS solutions cooling in the presence of a solid sub-
strate that has good interactions with the polymer demonstrate
that polymer chains will preferentially migrate out of solution
and form a solid polymeric layer on the substrate.[58]If the dense
outer skin of the previously deposited layer acts as the substrate,
then upon extrusion of the next layer polymer chains in solution
may migrate to this substrate forming a layer of high PPS con-
centration. This phenomenon may explain the lack of porosity of
the interlayer welds between regions of porous aerogel.[58]Addi-
tionally, the interlayer regions are much thicker than the exterior
skin of both prints (9 ±3 microns), indicating that remelting is
the likely cause of these dense interfaces rather than just an evap-
oration process that occurs after each layer is deposited.
As further evidence of this heat retention and remelting phe-
nomenon, the interlayer welds of both prints in Figure 3d are
thicker towards the top of the part, indicating that the amount
of heat retained in the part increases with the introduction of
more layers. In situ monitoring of printed parts with an IR cam-
era showed that the minimum temperature each layer reached
prior to the deposition of the next layer tended to increase as part
height increased. It must be noted that this trend is complex, with
the first layers experiencing cooling primarily via conduction to
the room temperature build plate, and the upper layers experi-
encing heat loss through convection to the air. Future work must
Adv. Mater. 2024,36, 2307881 2307881 (6 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
Figure 4. a) Porosity versus density of cast and printed poly(phenylene sulfide) (PPS) aerogels. b) Percent crystallinity of cast and printed 30 and 50 wt%
PPS aerogels. c) Compressive stress versus strain profiles of printed PPS aerogels. d) Compressive modulus versus density plot of cast and printed PPS
aerogels. Open green symbols =series of low-density cast aerogels prepared at concentrations ranging from 9.1 to 23.1 wt%.[38]Closed green symbols
=30 and 50 wt% cast aerogel specimen.
be conducted to rigorously study the relationship between heat
transfer, number of layers, layer height, layer cross-sectional area
and related toolpath time, and the resulting part microstructure.
2.4. Comparison of Cast and Printed PPS Aerogel Physical
Properties
The physical properties of the printed aerogels were compared to
cast aerogels, which act as control samples, to determine any ef-
fect of the printing process on density, porosity, and crystallinity
of DIW aerogels. Data of 30 and 50 wt% cast PPS aerogels, in ad-
dition to data from the authors’ previous study, was used to elu-
cidate the effects of printing on aerogel properties.[38]Figure 4
compares the physical properties of the printed and cast PPS
aerogels. Figure 4a shows that, for all aerogels, porosity decreased
linearly as polymer concentration of the aerogel (and thus aero-
gel density) increased, as expected. A linear trendline through
all samples, both cast and printed, displays a R2value of 0.9925,
indicating that the printing process has little effect on the den-
sity versus porosity relationship of the PPS aerogels. The vari-
ability of the density and porosity at each concentration is larger
in the printed parts than in the cast parts, likely due to overextru-
sion/underextrusion deviations, which created imperfect cylin-
ders and slight variations in the volume measurements (and thus
the density calculations) of some of the printed parts.
Figure 4b displays the percent crystallinity of the printed aero-
gels, and compares them to cast analogues at the same composi-
tions. The crystallinity of all four sets of aerogels is high, around
40–44%, suggesting that the TIPS process is useful for creating
highly crystalline parts. The difference in crystallinity between
cast and printed parts is minimal, suggesting that the printing
process does not disrupt nor does it enhance the crystallinity
of the aerogel. PPS is not known to be polymorphic, and wide-
angle x-ray scattering (WAXS) patterns (Figure S5, Supporting
Information) of cast and printed 30 and 50 wt% PPS aerogels
display similar features at similar scattering vectors, indicating
that printing does not affect the PPS crystal structure. The vari-
ation of the crystallinity of the printed parts is larger than those
in the cast parts; there are at least two possible explanations for
this variation. One explanation is that temperature oscillations in
the heated-DIW barrel alters the crystallization window once the
solution is extruded, creating additional variability in the crys-
tallinity of the AM parts. Another possibility is that, due to the
inconsistent cooling across each layer, some variation in crys-
tallinity is introduced throughout the part. Overall, Figure 4a,b in-
dicates that critical physical properties are generally unperturbed
by the printing process, suggesting that heated-DIW with in situ
TIPS is a viable processing option for creating complex aerogel
parts without sacrificing important properties inherent to the cast
aerogels.
2.5. Mechanical Behavior of Cast and Printed PPS Aerogels
Mechanical properties of printed PPS aerogels may be tuned
through a change in ink formulation. Figure 4c,d reveals the
compressive mechanical behavior of printed PPS aerogels. In
Figure 4c, compressive stress versus strain curves for each of
the individual printed 30 and 50 wt% aerogels are plotted. The
Adv. Mater. 2024,36, 2307881 2307881 (7 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
polymer composition, and thus the morphology, of the aerogels
greatly impacts their compressive response. The printed 30 wt%
aerogels display a typical compressive response for a porous poly-
meric sample: an initial linear-elastic zone, followed by a yield
and a plateau of deformation as members buckle and bend, and
finally densification of the porous structure.[59]Interestingly, the
printed 50 wt% aerogels show a deformation profile that is very
different from a cast part (Figure S5, Supporting Information),
but is instead more similar to that of a solid semiductile poly-
mer: an initial linear-elastic region, then a yield point, followed by
eventual failure at relatively low strains.[60,61]Figure S6 (Support-
ing Information) reveals that the stress versus strain profiles of
cast 30 and 50 wt% aerogels display very similar cellular solid-like
responses. Both 30 and 50 wt% cast PPS aerogels contain axialitic
morphologies, which suggests that the origins of the difference in
stress versus strain profiles between 30 and 50 wt% DIW aerogels
arise from morphological differences rather than compositional
differences. The deviation between compressive strengths in the
four 50 wt% aerogels (approximately 8.25 MPa between highest
and lowest compressive strength) is related to the specific density
of each aerogel and not a unique defect generated via printing, as
reported in Figure S7 (Supporting Information).
The results in Figure 4c suggest that the difference in mor-
phology between the two sets of aerogels leads to different com-
pressive failure modes. The axialitic morphology of the 30 wt%
parts, which forms a robust interconnected network, responds
in a fashion similar to traditional porous polymeric bodies,[59]
while the spherulitic morphology of the 50 wt% parts fail in a
more brittle manner, as there is less intercrystalline connectivity
compared to the axialitic morphologies, and the spherulitic inter-
stitial boundaries act as cracks and thus concentrate stress.[61,62]
This reveals the importance of control over the nucleation density
when applying TIPS technology to a DIW process, as it affects the
aerogel microstructure and thus the final part properties.
Figure 4d highlights the modulus versus density relationship
of the printed PPS aerogels, and compares those to the cast aero-
gels. In the printed parts, the moduli increases with composition
and trends towards the modulus of bulk PPS (3.4 GPa),[30]in-
dicating that heated-DIW printed PPS parts can be constructed
with tunable mechanical properties. The 50 wt% parts display
much higher moduli (E30 DIW =33.0 ±7.93 MPa; E50 DIW =
106.5 ±24.0 MPa), but much lower ductility (𝜖break,30 DIW =0.725
±0.026 mm mm−1;𝜖break,50 DIW =0.323 ±0.123 mm mm−1)
than the 30 wt% parts. The compressive moduli of the printed
PPS aerogels are high compared to other aerogels fabricated
via AM,[63–65]and approach the moduli of their cast analogues
(E30 Cast =106.8 ±4.9 MPa; E50 Cast =360.5 ±28.4 MPa).
The moduli of the printed parts display more variability than
those in the cast aerogels at the same compositions (30 and
50 wt%). Variations in density and crystallinity, in addition to the
presence of defects and layer interfaces could explain this vari-
ability. A scaling law, n, of 2.78 was fit through the cast aerogel
samples indicating good strut interconnectivity, between that of
an ideal cellular solid (n=2) and a poorly connected mass fractal
system (n=3.6).[38]While the modulus increase for the printed
30 and 50 wt% aerogels parallels the scaling observed for the
cast aerogels, the morphological differences between the printed
aerogels diminishes the physical significance of a power law rela-
tionship of modulus versus density for these samples. In general,
the ability of the printed gels to maintain a high percentage of the
mechanical behavior of the cast aerogels indicates the efficacy of
printing PPS TIPS solutions via heated-DIW.
3. Conclusion
In this work, the printing of PPS aerogels was realized for the
first time via a unique integration of heated MEX and a TIPS
process to rapidly solidify the printed roads. Fabrication of a
heated extruder allowed for printing of PPS/DPA solutions into
solvated gels which were subsequently processed into PPS aero-
gels. These new structures are otherwise unprintable via com-
mercially available AM methods. In this approach, PPS was first
gelled in DPA to create two concentrations of inks; those feed-
stocks were dissolved in the heated extruder forming a solution,
which was then extruded using pneumatic pressure into discrete
roads. Upon deposition, printed roads solidified rapidly via the
TIPS process. Solvated gel parts were built in a layer-wise fashion,
and upon solvent extraction the printed aerogels were formed.
This work presents a pathway for printing aerogels from PPS and
other high performance thermoplastics which can undergo TIPS.
Large thermal gradients between the nozzle and the print en-
vironment lead to fast PPS crystallization and thus rapid printed
road solidification, allowing for time-efficient printing of multi-
layer solvated gel parts. Cooling and solidification behavior was
confirmed through correlation between in situ IR camera video
and the thermal transitions of PPS/DPA solutions via DSC. Mul-
tiple complex architectures were printed involving sharp corners
and rounded curvature, infill, and many stacked layers. The gel
microstructure was observed to change with polymer composi-
tion, likely due to the self-nucleation behavior of PPS. Through
the printing parameters used in this study, the 30 wt% PPS aero-
gels displayed axialitic morphologies, while 50 wt% aerogels dis-
played spherulitic morphologies. Porosity, density, and the crys-
tallinity of the aerogels followed the same trend as cast aero-
gels, indicating that the printing process does not greatly impart
unique defects on the physical nature of the aerogels as compared
to their cast analogues. Compressive properties of the printed
aerogels changed with composition and also with aerogel mor-
phology, suggesting that control of printing conditions can have
a significant impact on final part properties. Importantly, printed
PPS aerogels display improved compressive moduli over previ-
ously reported DIW-printed aerogels,[63–65]providing a new al-
ternative when robust, geometrically complex, solvent-resistant,
and thermally stable aerogels are desired.
Given the compositionally dependent morphology of the
printed parts (e.g., axialitic versus spherulitic), it is anticipated
that control of this behavior could be leveraged in the future to
create parts with gradient microstructures. Parts requiring ar-
eas of designed failure (such as a crumple zone in an automo-
bile) could be printed at higher temperatures, yielding spherulitic
morphologies with brittle mechanical response (see Figure 4c),
without sacrificing density, porosity, crystallinity, or surface area.
These zones could be reinforced by aerogel made of an axialitic
microstructure featuring increased ductility and strain-at-break.
Future studies investigating the interrelationships between print-
ing temperature, ink composition, morphology, and mechanical
response could allow greater, and even in situ, control over these
designed failure and reinforcement zones. Investigation of the
Adv. Mater. 2024,36, 2307881 2307881 (8 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
heat transfer within a printed part as a function of the number
of layers, layer height, and pause time between layer prints will
also allow for a deeper understanding of the complex formation
of layer interfaces. Probing the effects of PPS gel microstructure
on additional properties such as thermal conductivity, absorption
or filtration of specific pollutants, and acoustic behavior will allow
for expanded applications of printed aerogels. Also, use of a pres-
sure cell attachment to a parallel plate rheometer could allow for
the rheology of PPS/DPA inks to be measured in a sealed envi-
ronment, where vital relationships between temperature, shear
rate, and flow behavior could be studied. Overall, printed PPS
aerogels are mechanically robust and highly porous, and feature
tailorable properties based on the initial ink concentration and
printing parameters.
4. Experimental Section
Materials:PPS (Ryton QA 200N) was provided by Solvay Specialty
Polymers (Alpharetta, GA), and DPA)(99.0%) was purchased from Oak-
wood Chemical (Estill, SC) and Sigma-Aldrich Inc. (St. Louis, MO).
Ethanol (200 proof, 100%, USP, Decon Labs) was purchased from Fisher
Scientific Company (Pittsburgh, PA). All polymers and chemicals were
used as-received.
PPS/DPA Gel Ink Formulation:Ink feedstocks (PPS/DPA gels) were
fabricated in a similar fashion as reported by Godshall et al.[38]Briefly, PPS
and solid DPA (Tm=32–34 °C) were premixed at the proper mass ratios
(either 30 or 50 wt% PPS for this study) inside of large test tubes, sealed,
and purged with Argon gas for at least 15 min to remove air and decrease
potential oxidation of the solution, which may occur to PPS and/or DPA at
elevated temperatures.[66]The mixture was then heated to 270 °Cinsideof
a metal bath whereby the DPA dissolved the PPS after 20–60 min depend-
ing on the concentration. Once the polymer was dissolved, the mixture
was stirred vigorously for 5–10 min to homogenize the solution. After ho-
mogenization, the solution was removed from the metal bath and allowed
to cool to room temperature under ambient conditions. Upon cooling, the
solution phase separated and formed an opaque, solidified gel in 2–3 min.
The room temperature gel was removed from the test tube and cut into
smaller pellet-sized pieces, ready for loading into the heated-DIW extruder
barrel.
Heated-DIW Printing:To enable simultaneous DIW-TIPS printing, a 3-
axis gantry was constructed and outfitted with a customized heated DIW
barrel, displayed in Figure S8 (Supporting Information). This custom ex-
truder was designed to meet the need for high temperatures (>215 °C)
while also being resistant to solvents. As opposed to a plunger based
DIW extruder that would require chemically and thermally stable seals,
a simple pneumatically actuated extruder was chosen and designed with
Schedule 80 stainless steel fittings capable of resisting the elevated pres-
sures at temperatures well above 200 °C. The Schedule 80 stainless steel
barrel contains two heating zones. An upper heating zone, powered by
an insulated band heater, prevented excess heat loss from the hot end,
and reduced clogging in the upper portion of the extruder. The lower heat-
ing zone, which dissolved the gel ink into a homogenous solution, was
powered via a second insulated band heater. The temperatures of both
heating zones were controlled by independent J-KEM Model 210 temper-
ature controllers; thermocouples were attached onto the upper barrel and
onto the brass extrusion nozzle to monitor the temperature in each zone.
Pneumatic pressure was supplied to the extruder via a Nordson Ultimus
V system, using argon gas instead of compressed air to reduce poten-
tial oxidative degradation of the polymer solution at high temperatures. A
wooden build plate was used as the printing substrate, as the deposited
PPS/DPA ink adhered well to its surface, yet the completed part was still
easily removed after printing.
For determination of proper printing temperatures, the upper heating
zone was set to 125 °C during all printing. Temperature and pressure pa-
rameters were selected that produced a consistent extrudate that remelted
previous layers and did not lead to degradation of the polymer solution.
The lower heating band controlled the temperature needed to enable a
consistent flow (i.e., extrudate temperature was measured by IR camera
to be 208 ±8°C and 243 ±8°C for the 30 and 50 wt% solutions, re-
spectively). Full printing parameters are presented in Table S1 (Supporting
Information).
For both ink formulations, a 0.4 mm diameter nozzle was used to de-
posit layer heights of 0.5 mm at an extrusion speed of 4.0 mm s−1.Dueto
imperfect barrel insulation and the nature of PID controlled heating, slight
oscillations in temperature during printing lead to transiently varying vis-
cosity and thus a need for minor pressure modulation during extrusion.
Existing DIW analytical models for predicting optimal extrusion pressures
of PPS/DPA inks were not sufficient since the experimental set up could
not allow for the exact non-Newtonian viscosity of the ink to be measured
in its solution state. For the 30 wt% ink, an average extrusion pressure of
2 psi was used; for the 50 wt% ink, an average extrusion pressure of 6 psi
was used.
PPS Aerogel Postprocessing:Printed aerogels were allowed to cool to
room temperature, then were submerged in 50 °C ethanol for 4 days to
extract DPA. Next, the ethanol-containing organogels were submerged
in deionized water at room temperature for 4 days to create hydrogels.
During these solvent exchange processes, the ethanol and water were re-
freshed daily. The hydrogels were then frozen at −18 °Covernightand
freeze dried using a LabConco lyophilizer to yield printed PPS aerogels.
Printed Part Characterization:In situ IR camera videos of DIW print-
ing were taken using a Micro-Epsilon thermoIMAGER TIM 640 infrared
camera. The TIMConnect Software was used to record video and collect
simultaneous spatial, temporal, and thermal data.
The density of printed hollow cylindrical aerogel samples was mea-
sured by dividing the mass of each part by its measured volume. True part
volume was calculated directly using the measured dimensions of each
printed aerogel cylinder. Nominal cylinder dimensions were: a height of
10 mm, an outer radius of 5 mm and an inner radius of 4 mm (yielding an
approximate wall thickness of 1 mm). Four replicate parts of each printed
polymer composition were measured.
Porosity of printed hollow cylindrical aerogel samples was measured
using a Micromeritics AccuPyc II 1340 helium pycnometer, according to
ASTM D6226-21.[67]A3.5cm
3insert and an equilibration rate of 0.03 psig
min−1were used. To calculate porosity, the following equation was used:
P=V−Vspec
V×100 (1)
where Pis the volumetric porosity (%), Vis the hollow cylinder volume,
and Vspec is the skeletal density. Four replicate parts of each printed poly-
mer composition were measured. In addition, nitrogen porosimetry was
carried out on an Anton Paar Autosorb iQ C-XR adsorption analyzer to
calculate (BET) surface areas of printed and cast PPS aerogels.
The morphology of printed aerogels was analyzed via a Zeiss LEO 1550
field-emission SEM (FE-SEM) with in-lens detection. Parts were freeze-
fractured and mounted on carbon tape, then sputter coated with 5 nm of
Iridium using a Leica EM ACE600 sputter coater. Image analysis of SEM
micrographs was performed using the ImageJ image processing software.
The degree of crystallinity of the solvent extracted PPS aerogels was
measured using first-heat scans (10 °Cmin
−1from 40 °C to 330 °C) from
a TA Instruments Q2000 DSC. Percent crystallinity (%Xc) was calculated
via:
%Xc=ΔHm
ΔHo
m×100 (2)
where ΔHmis the measured heat of fusion, taken as the integrated area
under the melting peak, and ΔHmois the theoretical heat of fusion of 100%
crystalline PPS, equal to 111.7 J g−1.[68]
Thermogravimetric analysis (TGA) using a TA Instruments Discov-
ery 550 TGA was performed on printed PPS/DPA samples to measure
Adv. Mater. 2024,36, 2307881 2307881 (9 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
DPA content and to confirm complete removal of DPA in the solvent ex-
tracted aerogels. Heating ramps were performed from room temperature
to 850 °Cat20°Cmin
−1in a nitrogen environment.
WAXS experiments were conducted at beamline 9-ID-C at the Advanced
Photon Source (APS) at Argonne National Labs (Lemont, IL).[69,70]The
Nika macro for Igor Pro was utilized to develop WAXS data.[71]
To measure the compressive properties of the printed cylindrical aero-
gels, an Instron 3340 Universal Testing System was equipped with com-
pressive platens, a 5 kN load cell, and a crosshead speed of 1 mm min−1.
Four replicate parts of each polymer composition were measured.
Cast aerogels were fabricated according to the process outlined in God-
shall et al.[38]30 and 50 wt% cast aerogels were tested according to the
same methods as outlined above (density, porosity, SEM, BET, DSC, and
uniaxial compression).
In order to tune printing parameters and optimize the printing pro-
cess, the rheological properties of DIW inks are often studied. Due to the
close proximity of the melting temperature of the PPS/DPA solvated gel
inks (Tm, 30 wt% =243.9 °C, Tm, 50 wt% =253.3 °C, Figure S9, Supporting
Information) and the pure DPA boiling point (Tb, DPA =330 °C), solvent
evaporation was inevitable during rheological testing. While the ink in the
extruder barrel is ideally sealed off from the environment, the material be-
tween the parallel plates is not; excess evaporation in the rheometer pre-
vented accurate characterization of the shear thinning behavior and solid-
ification/dissolution transition temperatures of the inks that occur during
the printing process. Qualitatively, however, the 50 wt% ink was noticeably
more viscous than the 30 wt% ink, and Figure S9 (Supporting Information)
indicates that the 50 wt% ink also has a higher melting point. The increase
in viscosity with increasing polymer concentration was also reflected in the
extrusion pressure required to print each ink, with the 30 wt% ink (2 psi)
requiring a lower pressure than the 50 wt% ink (6 psi) (Table S1,Sup-
porting Information). Importantly, the quantity of solvent that may have
evaporated during the printing process should be determined, to confirm
that the ink concentration remains consistent during and after printing.
The nozzle is the only escape point for vapors during printing, and ideally
solvent loss should be kept to a minimum to maintain a constant ink com-
position. Figure S10 (Supporting Information) highlights TGA degrada-
tion curves that quantify the amount of DPA remaining in solvated printed
parts. In a 50 wt% printed part, 46.6% of the solvent is retained, indicat-
ing that the solvent content in the printed part was approximately equal to
that in the precursor ink, suggesting that solvent evaporation was not an
issue during heated-DIW of PPS/DPA inks.
Statistical Analysis:Data presented as mean ±SD, n=4 for printed
aerogels and n=5 for cast aerogels, unless otherwise stated.
Supporting Information
Supporting Information is available from the Wiley Online Library or from
the author.
Acknowledgements
G.F.G. and D.A.R. contributed equally to this work. This material is based
upon work supported by the National Science Foundation under Grant
Nos. DMR-1809291 and DMR-2104856. This research used resources of
the Advanced Photon Source, a U.S. Department of Energy (DOE) Office
of Science user facility operated for the DOE Office of Science by Argonne
National Laboratory under Contract No. DE-AC02-06CH11357. WAXS data
was collected on the 9-ID-C beamline at the APS, Argonne National Labo-
ratory.
Conflict of Interest
The authors declare no conflict of interest.
Data Availability Statement
The data that support the findings of this study are available from the cor-
responding author upon reasonable request.
Keywords
additive manufacturing, hierarchical porosity, material extrusion, polymer
aerogel, polyphenylene sulfide, thermally induced phase separation
Received: August 5, 2023
Revised: October 30, 2023
Published online: December 6, 2023
[1] C.I.Idumah,A.C.Ezika,V.U.Okpechi,Surf. Interfaces 2021,25,
101258.
[2] P. Paraskevopoulou, D. Chriti, G. Raptopoulos, G. C. Anyfantis, Ma-
terials 2019,12, 1543.
[3] S. S. Kistler, Nature 1931,127, 741.
[4] S. Zhao, W. J. Malfait, N. Guerrero-Alburquerque, M. M. Koebel, G.
Nystrom, Angew. Chem., Int. Ed. 2018,57, 7580.
[5] X. Jiang, R. Du, R. Hübner, Y. Hu, A. Eychmüller, Matter 2021,4, 54.
[6] H. Tetik, Y. Wang, X. Sun, D. Cao, N. Shah, H. Zhu, F. Qian, D. Lin,
Adv. Funct. Mater. 2021,31, 2103410.
[7] J. L. Gurav, I.-K. Jung, H.-H. Park, E. S. Kang, D. Y. Nadargi, J. Nano-
mater. 2010,2010, 409310.
[8] Y. Li, B. Ma, R. Zhang, X. Luo, Polymer 2022,253, 125035.
[9] X.Sun,T.Fujimoto,H.Uyama,Polym. J. 2013,45, 1101.
[10] Y. Guo, T. Zhang, M. Chen, C. Li, H. Wu, S. Guo, J. Membr. Sci. 2021,
619, 118778.
[11] I. Gibson, D. Rosen, B. Stucker, M. Khorasani, Additive Manufacturing
Technologies, Springer International Publishing, Cham 2021.
[12] B. G. Compton, J. A. Lewis, Adv. Mater. 2014,26, 5930.
[13] M. Vaezi, H. Seitz, S. Yang, Int. J. Adv. Manuf. Technol. 2013,67, 1721.
[14] D. A. Rau, J. Herzberger, T. E. Long, C. B. Williams, ACS Appl. Mater.
Interfaces 2018,10, 34828.
[15] C. B. Arrington, D. A. Rau, C. B. Williams, T. E. Long, Polymer 2021,
212, 123306.
[16] C. Xu, B. Quinn, L. L. Lebel, D. Therriault, G. L’Espérance, ACS Appl.
Mater. Interfaces 2019,11, 8499.
[17] A. Shahzad, I. Lazoglu, Composites, Part B 2021,225, 109249.
[18] J. W. Boley, K. Chaudhary, T. J. Ober, M. Khorasaninejad, W. T. Chen,
E. Hanson, A. Kulkarni, J. Oh, J. Kim, L. K. Aagesen, A. Y. Zhu, F.
Capasso, K. Thornton, P. V. Braun, J. A. Lewis, Adv. Mater. 2017,29,
1604778,.
[19] S. Ji, M. Guvendiren, Front. Bioeng. Biotechnol. 2017,5, 23.
[20] M. A. S. R. Saadi, A. Maguire, N. T. Pottackal, M. S. H. Thakur, M. M.
Ikram, A. J. Hart, P. M. Ajayan, M. M. Rahman, Adv. Mater. 2022,34,
2108855.
[21] S. S. L. Chan, R. M. Pennings, L. Edwards, G. V. Franks, Addit. Manuf.
2020,35, 101335.
[22] P. J. Scott, D. A. Rau, J. Wen, M. Nguyen, C. R. Kasprzak, C. B.
Williams, T. E. Long, Addit. Manuf. 2020,35, 101393.
[23] O. Rios, W. Carter, B. Post, P. Lloyd, D. Fenn, C. Kutchko, R. Rock, K.
Olson, B. Compton, Mater. Today Commun. 2018,15, 333.
[24] S.-Z. Guo, M.-C. Heuzey, D. Therriault, Langmuir 2014,30, 1142.
[25] S.-Z. Guo, F. Gosselin, N. Guerin, A.-M. Lanouette, M.-C. Heuzey, D.
Therriault, Small 2013,9, 4118.
[26] S. Chandrasekaran, B. Yao, T. Liu, W. Xiao, Y. Song, F. Qian, C. Zhu, E.
B. Duoss, C. M. Spadaccini, Y. Li, M. A. Worsley, Mater. Horiz. 2018,
5, 1166.
Adv. Mater. 2024,36, 2307881 2307881 (10 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
[27] B. Yao, H. Peng, H. Zhang, J. Kang, C. Zhu, G. Delgado, D. Byrne, S.
Faulkner, M. Freyman, X. Lu, M. A. Worsley, J. Q. Lu, Y. Li, Nano Lett.
2021,21, 3731.
[28] H. C. Liu, A. D. Auguste, J. O. Hardin, A. Sharits, J. D. Berrigan, ACS
Appl. Mater. Interfaces 2021,13, 27364.
[29] A. D. Luca, J. R. d. Wijn, C. A. v. Blitterswijk, S. Camarero-Espinosa,
L. Moroni, Macromol. Rapid Commun. 2017,38, 1700186.
[30] Ryton QA200N Polyphenylene Sulfide Technical Data Sheet,Solvay,
2020. [Online]. Available: https://www.matweb.com/search/
datasheet.aspx?matguid=fed1d60725244d9ba944875f7c0891dd&
ckck=1
[31] H. N. Beck, J. Appl. Polym. Sci. 1992,45, 1361.
[32] A. S. Rahate, K. R. Nemade, S. A. Waghuley, Rev. Chem. Eng. 2013,
29, 471.
[33] P. Zuo, A. Tcharkhtchi, M. Shirinbayan, J. Fitoussi, F. Bakir, Macromol.
Mater. Eng. 2019,304, 1800686.
[34] M. Q. Ansari, M. J. Bortner, D. G. Baird, Addit. Manuf. 2019,29,
100814.
[35] P. Yeole, A. A. Hassen, S. Kim, J. Lindahl, V. Kunc, A. Franc, U. Vaidya,
Addit. Manuf. 2020,34, 101255.
[36] C. A. Chatham, T. E. Long, C. B. Williams, Addit. Manuf. 2019,28, 506.
[37] C. A. Chatham, T. E. Long, C. B. Williams, MRS Commun. 2021,11,
179.
[38] G. F. Godshall, G. A. Spiering, E. R. Crater, R. B. Moore, ACS Appl.
Polym. Mater. 2023,5, 7994.
[39] S. J. Talley, X. Yuan, R. B. Moore, ACS Macro Lett. 2017,6, 262.
[40] S. J. Talley, C. L. AndersonSchoepe, C. J. Berger, K. A. Leary, S. A.
Snyder, R. B. Moore, Polymer 2017,126, 437.
[41] S. J. Talley, S. L. Vivod, B. A. Nguyen, M. B. Meador, A. Radulescu, R.
B. Moore, ACS Appl. Mater. Interfaces 2019,11, 31508.
[42] A. Elhaj, K. Irgum, ACS Appl. Mater. Interfaces 2014,6, 15653.
[43] R. Zeinali, L. J. d. Valle, J. Torras, J. Puiggali, Int. J. Mol. Sci. 2021,22,
3504.
[44] D. R. Lloyd, S. S. Kim, K. E. Kinzer, J. Membr. Sci. 1991,64,1.
[45] S. S. Kim, D. R. Lloyd, J. Membr. Sci. 1991,64, 13.
[46] D. R. Lloyd, K. E. Kinzer, H. S. Tseng, J. Membr. Sci. 1990,52, 239.
[47] Y. Furushima, M. Nakada, Y. Yoshida, K. Okada, Macromol. Chem.
Phys. 2018,219, 1700481.
[48] X. Gao, S. Qi, X. Kuang, Y. Su, J. Li, D. Wang, Addit. Manuf. 2021,37,
101658.
[49] M. Zhou, B. Chen, Y. Li, J. Jiang, W. Zhai, Adv. Eng. Mater. 2023,25,
2201468.
[50] C. A. Chatham, C. E. Zawaski, D. C. Bobbitt, R. B. Moore, T. E. Long,
C. B. Williams, Macromol. Mater. Eng. 2019,304, 1800764.
[51] H.-h. Lin, Y.-h. Tang, T.-y. Liu, H. Matsuyama, X.-l. Wang, J. Membr.
Sci. 2016,507, 143.
[52] M. Trujillo, M. L. Arnal, A. J. Müller, M. A. Mujica, C. Urbina de
Navarro, B. Ruelle, P. Dubois, Polymer 2012,53, 832.
[53] C.-Y. Chiang, D. R. Lloyd, J. Porous Mater. 1996,2, 273.
[54] L. Sangroniz, D. Cavallo, A. J. Muller, Macromolecules 2020,53,
4581.
[55] P. Yan, F. Yang, M. Xiang, T. Wu, Q. Fu, Polymer 2020,195, 122439.
[56] S. Hu, F. Yang, M. Xiang, Y. Cao, T. Wu, Q. Fu, Polymer 2022,262,
125427.
[57] N. P. Levenhagen, M. D. Dadmun, ACSAppl.Polym.Mater.2019,1,
876.
[58] S. Farzaneh Kalourazi, F. Wang, H. Zhang, M. Selzer, B. Nestler, J.
Phys.: Condens. Matter 2022,34, 444003.
[59] M. F. Ashby, Metall. Mater. Trans. A 1983,14A, 1755.
[60] I. M. Ward, J. Sweeney, Mechanical Properties of Solid Polymers,Wiley,
Chichester, West Sussex, UK 2012.
[61] J. L. Way, J. R. Atkinson, J. Nutting, J. Mater. Sci. 1974,9, 293.
[62] A. Lustiger, C. N. Marzinsky, R. R. Mueller, J. Polym. Sci., Part B:
Polym. Phys. 1998,36, 2047.
[63] S. Camarero-Espinosa, M. Carlos-Oliveira, H. Liu, J. F. Mano, N.
Bouvy, L. Moroni, Adv. Healthcare Mater. 2022,11, 2101415.
[64] V. C. F. Li, A. Mulyadi, C. K. Dunn, Y. Deng, H. J. Qi, ACS Sustainable
Chem. Eng. 2018,6, 2011.
[65] L. Huang, X. Du, S. Fan, G. Yang, H. Shao, D. Li, C. Cao, Y. Zhu, M.
Zhu, Y. Zhang, Carbohydr. Polym. 2019,221, 146.
[66] P.Yan, W. Peng, F. Yang, Y. Cao, M. Xiang, T. Wu, Q. Fu, Polym. Degrad.
Stab. 2022,197, 109863.
[67] D6226-21 Standard Test Method for Open Cell Content of Rigid Cellular
Plastics, ASTM International, PA, USA 2021.
[68 ] P. Huo , P. C ebe , Colloid Polym. Sci. 1992,270, 840.
[69] J. Ilavsky, P. R. Jemian, A. J. Allen, F. Zhang, L. E. Levine, G. G. Long,
J. Appl. Crystallogr. 2009,42, 469.
[70] J. Ilavsky, F. Zhang, A. J. Allen, L. E. Levine, P. R. Jemian, G. G. Long,
Metall. Mater. Trans. A 2013,44, 68.
[71] J. Ilavsky, J. Appl. Crystallogr. 2012,45, 324.
Adv. Mater. 2024,36, 2307881 2307881 (11 of 11) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH