3D Printed PEG-Based Hybrid Nanocomposites Obtained by Sol–Gel

Research Article www.acsami.org

3D Printed PEG-Based Hybrid Nanocomposites Obtained by Sol−Gel Technique Annalisa Chiappone,*,† Erika Fantino,‡ Ignazio Roppolo,† Massimo Lorusso,† Diego Manfredi,† Paolo Fino,‡ Candido Fabrizio Pirri,†,‡ and Flaviana Calignano† †

Center for Space Human Robotics@polito, Istituto Italiano di Tecnologia, Torino 10129, Italy Department of Applied Science and Technology, Politecnico di Torino, Torino 10129, Italy

‡

Downloaded via DURHAM UNIV on July 5, 2018 at 09:36:25 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: In this work, three-dimensional (3D) structured hybrid materials were fabricated combining 3D printing technology with in situ generation of inorganic nanoparticles by sol−gel technique. Those materials, consisting of silica nanodomains covalently interconnected with organic polymers, were 3D printed in complex multilayered architectures, incorporating liquid silica precursors into a photocurable oligomer in the presence of suitable photoinitiators and exposing them to a digital light system. A post sol−gel treatment in acidic vapors allowed the in situ generation of the inorganic phase in a dedicated step. This method allows to build hybrid structures operating with a full liquid formulation without meeting the drawbacks of incorporating inorganic powders into 3D printable formulations. The influence of the generated silica nanoparticle on the printed objects was deeply investigated at macroand nanoscale; the resulting light hybrid structures show improved mechanical properties and, thus, have a huge potential for applications in a variety of advanced technologies. KEYWORDS: 3D printing, digital light processing (DLP), hybrid nanocomposite, sol−gel, mechanical properties

■

INTRODUCTION In recent years, three-dimensional (3D) printing technology has rapidly grown, showing great potentialities that make it suitable for different application fields, from bioengineering1,2 and microfluidics3,4 to electronics.5−7 A term often used synonymously with additive manufacturing (AM), “3D printing” indicates the fabrication process that enables the layer-by-layer construction of 3D objects from computer-aided design (CAD) data.8,9 Among the 3D printing technologies, Stereolithography (SL) was the first patented and commercialized process. This technology involves the curing of a liquid photosensitive polymer through the use of a light source, which supplies the energy needed to induce a chemical reaction, bonding large numbers of small molecules, and forming a highly cross-linked polymer.10−12 The manufacturing of 3D objects is based on the spatially controlled solidification of a liquid resin by photopolymerization, the spatial resolutions can approach 25−50 μm. In the last years, a modification of the SL process, digital light processing (DLP), has been recognized as one of the most powerful and versatile processes thanks to its high fabrication accuracy and to an increasing number of materials that can be processed.2,10 The main differences between SL and DLP are the building orientation and the method of illumination: in a DLP system light is projected from the bottom of the vat that contains the resin and the building platform is dipped into the © 2016 American Chemical Society

resin from above, while the irradiation system is equipped with a digital mirror device and an LED lamp. The DLP setup has some advantages: the production of high pieces (according to the printer setup) is allowed with the consumption of low volumes of resins and, above all, reduced printing times.9,12 Several papers report the use of SL and DLP for the fabrication complex Poly(ethylene glycol) (PEG)-based 3D structures.13−18 PEG is a synthetic hydrogel that has been widely used because of its hydrophilicity, biocompatibility, and ability to be chemically tailored.19 For these reasons, PEGbased photocurable (meth)acrylic oligomers20,21 are largely used and investigated in 3D printing, mainly for biomedical and biotechnology applications.16,22,23 Although these materials present interesting characteristics, their mechanical properties can largely be improved.24 The addition of nano fillers could surely help in reaching this aim but, at the same time, it could strongly modify the printing process introducing new issues: increased solution viscosity, limited light penetration depth, nanoparticles dispersion and stability. Considering these drawbacks, the in situ generation of nanoparticles after the printing step could be an alternative and better solution. Silica nanodomains can be directly generated in Received: December 24, 2015 Accepted: February 12, 2016 Published: February 12, 2016 5627

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633

Research Article

ACS Applied Materials & Interfaces

measurements were carried out in the linear viscoelastic region (strain amplitude 1%). Differential scanning calorimetry (DSC) measurements were performed with a DSC1 STARe System apparatus of TA Instruments equipped with a low temperature probe. The experiments were carried out between −80 and 60 °C with a heating rate of 10 °C/min. Thermogravimetric analysis (TGA) was performed using a TGA/ SDTA 851e instrument in the range between 25 and 700 °C, with a heating rate of 10 °C/min in air. DMTA measurements were performed on 3D printed flat samples with a Triton Technology TTDMA. All the experiments were conducted with a temperature ramp of 3 °C/min, applying a force with frequency of 1 Hz and with 10 μm of displacement. To compare the built parts with the CAD model, we digitized the samples using a 3D optical scanner, Atos Compact Scan 2 M by GOM Gmbh, and Geomagic Studio was used for the inspection. Before the comparative analysis, the compared objects have to be aligned one to another. After the alignment, the scan data and the reference model can be compared for deviations. The results of the comparative analysis are often displayed as colored 3D maps. In this study, the CAD model and scan data were aligned applying the “Best Fit” method which uses the least-squares principle, that is, the deviations between the scan and the models are mathematically segmented. The morphological characterization of the cured materials after 3D printing was carried out by field emission scanning electron microscopy (FESEM, Zeiss Supra 40) equipped with an energy dispersive X-ray spectrometer (EDX, Oxford INCA Energy 450) for compositional analyses. The samples were prepared by fracturing the obtained 3D structures (honeycomb and films) in liquid nitrogen, and both surfaces and cross sections of the cured materials were analyzed. Nanohardness and elastic modulus measurement were carried out using the NanoIndenter TI950 (Hysitron) equipped with in situ scanning probe microscopy (SPM) imaging capability. The most common method to analyze the hardness and modulus values from an indentation curve is the Oliver−Pharr method.35 The Oliver−Pharr method requires no imaging of the indentation, instead it is based on contact mechanics solutions developed by Sneddon.36 The tests are performed by applying and removing a controlled load to the specimens using a geometrically well-defined probe, producing traditional force versus displacement curves. The analysis of these curves provides information regarding the mechanical properties of the samples. The used indenter was a cono-spherical probe, that is a conical shaped probe with a spherical end with an included angle of 90° and a radius of 1 μm. Ten indentations, spaced 5 μm from each other, were realized for every sample. The load of 100 μN was applied with a rate of 20 μN/s, with a permanence time at the maximum load of 5 s. Printed flat specimens were used for this analysis, at least 10 different testing locations were analyzed for each sample. Tensile tests were carried out using a Instron 3366 dynamometer equipped with a load cell of 500 N. Printed flat specimens (10 × 80 × 0.7 mm) were prepared, and at least five specimens for each sample were tested. Compression tests were carried out using a Zwick Roell Z050 dynamometer equipped with a load cell of 1 kN. Alveolar structures were tested, and at least five specimens for each sample were tested.

a photocured matrix involving a series of hydrolysis and condensation reactions of metal alkoxide precursor dispersed in the initial formulation.25−27 The use of a coupling agent allows the formation of chemical bonds between the polymer matrix and the inorganic component, forming strictly interconnected organic−inorganic hybrid networks.24,25 This process allows to generate a reinforcing phase in a polymeric medium without meeting the above-mentioned drawbacks since all the precursor are processed in liquid form. It was already used both in thermoplastics28−30 and thermosettings31,32 to modify many properties such as hardness, Young’s modulus, or transparency. The resulting hybrid materials present a combination of the properties of ceramics with those of the organic polymer matrices. They have a huge potential for applications in a variety of advanced technologies, both as structural and functional materials.33 This study shows that by coupling the 3D printing fabrication with the sol−gel process it is possible to obtain highly reliable and precise 3D printed parts having a nanophasic morphology where the organic phase is strictly interconnected with the inorganic one (the so-called hybrid materials). The 3D structures were fabricated by incorporating methacryloyloxypropyl-trimethoxysilane (MEMO), as an organic−inorganic bridging monomer, and tetraethoxysilane (TEOS), as inorganic precursor, into the PEGDA oligomer in the presence of suitable photoinitiators and exposing them to the DLP system. The 3D fabrication is followed by hydrolysis and condensation reactions of silica precursors, in acidic humid atmosphere.34 Comprehensive studies were systematically performed on the morphology, the thermo-mechanical characteristics, and the mechanical properties of the 3D printed hybrid nanocomposites.

■

EXPERIMENTAL SECTION

Materials. Poly(ethylenglycol)diacrylate, PEGDA, Mw 600 g mol was kindly provided by Allnex, tetraethyl orthosilicate, TEOS and 3(trimethoxysilyl) propyl methacrylate, MEMO, were purchased from Sigma−Aldrich and used as received. Bis-(2,4,6-trimethylbenzoyl) phenylphosphineoxide (Irgacure 819, BASF) and 2-hydroxy-2-methyl1- phenyl-propan-1-one (Irgacure 1173, BASF) were selected as initiating systems. Sample Preparation. 3D Printing. Mixtures containing PEGDA, MEMO 5 phr and TEOS (0, 20, 30, 40 phr) with the two photoinitiators were prepared, formulations percentage composition is reported in Table S1 in the Supporting Information. A 3DLPrinterHD 2.0 (Robot Factory), equipped with a projector with a resolution of 50 μm (1920 × 480 1080 pixels) was used for the printing. The build area is 100 × 56.25 × 150 mm with a layer thickness adjustable from 10 to 100 μm. The printing time ranged from 1 to 1.4 s/layer for increasing amounts of silica precursor. The printing process was followed by a post curing process performed with a medium−pressure mercury lamp also provided by Robot Factory. Sol−Gel Post Treatment. Hydrolysis and condensation reactions of silica precursors were carried out storing the printed samples overnight at the temperature of 70 °C in acidic humid atmosphere (1 wt % HCl in water) using the acidic vapors as catalyst for the reactions. PEGDA samples were also subjected to this treatment. Characterization. Real-time rheological measurements were performed using an Anton Paar rheometer (Physica MCR 302) in parallel plate mode with an Hamamatsu LC8 lamp with visible bulb and a cutoff filter below 400 nm equipped with 8 mm light guide. The gap between the two plates was set to 0.2 mm and the sample was kept at a constant temperature (25 °C) and under constant shear frequency of 1 rad/s, light was turned on after 1 min in order to stabilize the system. Concomitant changes in viscoelastic material moduli during polymerization were measured as a function of exposure time. The −1

■

RESULTS The aim of this work was to implement a method for creating 3D printed complex hybrid objects with enhanced mechanical properties exploiting a low cost unmodified DLP machine coupled with a sol−gel technique. Different formulations were prepared employing PEGDA as fundamental photocurable oligomer; to improve the mechanical properties of the threedimensional structures, we carried out in situ generation of inorganic nanoparticles. In comparison with more common printing formulations based on the NPs dispersion in the liquid oligomers, the possibility to generate silica nanodomains inside the printed structures after the object building opens new 5628

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633

Research Article

ACS Applied Materials & Interfaces

Each formulation and structure required an accurate study of the printing parameters. After the realization of the 3D objects, these were submitted to the thermal step for the sol/gel process during which the acidic vapors caused the condensation of silica nanoparticles inside the printed structures. The final resolution and fidelity to the CAD files was evaluated by 3D scanning. Figure 2C shows the results. Once a part is scanned and a point cloud is created for 3D inspection, this latter is virtually overlaid with the CAD model. The color chart indicates that green is nominal, red is extra material or warp in the positive direction, and blue is less material or warp in the negative direction. From the comparison, it has been determined that the average distance, calculating the medium value between positive values, negative values, and the medium between the two model, is about −0.035 mm. DSC measurements were performed on treated 3D samples and compared to those of the pristine PEGDA structures aiming to evaluate the stiffening effect induced by the presence of the silica domains. When only MEMO was added to the PEGDA resin, no variation of the Tg value was observable while, increasing the amount of silica precursor in the photocurable formulation, an increase of Tg values (Table 1)

interesting perspective. In fact this method enables a fast and easy printing step operating always with liquid compounds without viscosity or stability issues. For this reason TEOS, a liquid silica precursor, was added in different amounts (20, 30, and 40 phr) in the formulations together with MEMO that was used as coupling agent (5 phr), enabling a direct chemical interconnection between the organic matrix and the silica nanoparticles. Being the DLP process based on a light source mainly emitting in the visible range and the post curing process based on a UV source, two different light sensitive molecules were also added to the formulations: BAPO was chosen for its absorption in the deep blue to near UV, while Irgacure 1173 was chosen as UV−photoactive compound. Prior to the printing step, the influence of the different amounts of TEOS onto the photopolymerization kinetics was investigated by means of photorheology. The test measures the variation of G′ modulus during visible light irradiation under constant oscillation frequency. Figure 1 reports the curve

Table 1. Thermal Properties of the Composite 3D Printed Materials sample

Tg (°C)

PEGDA MEMO TEOS 20 TEOS 30 TEOS 40

−41 −41 −34 −33 −33

a

Tg (°C)

fwhm

TGA residue (%)

−36 −33 −30 −27 −28

10.4 12.7 17.5 17.1 19.1

2.4 7.2 9 11.5

b

theoretical inorganic amount (%) 1.2 6.5 8.9 11.3

a

Figure 1. Photorheology characterization of the different formulations.

obtained from DSC measurements. bobtained from DMTA measurements.

relative to the pure PEGDA formulation compared to the one containing PEGDA and MEMO (namely, MEMO) and to those containing different amounts of TEOS; the plots show fast reaction times for all the formulations with a slight delay on the beginning of the reaction when TEOS is added, indicating the need of longer exposition time during printing. Once we investigated the behavior of the formulations under irradiation, different computer-aided design (CAD) files were produced aiming to print different 3D objects, ranging from tensile test specimens to honeycomb structures (Figure 2).

was observed. This increase is due to the presence of the silica phase generated after the sol−gel process that hinders the mobility of the polymer chains, increasing the stiffness and therefore the glass transition temperature. Moreover, the use of MEMO as coupling agent leads to the formation of direct chemical bonds between organic and inorganic domains; this causes a higher hindering effect that reaches a plateau for higher amount of precursor (40 phr). The trend observed for the variation of the Tg was also confirmed by DMTA (Table 1). Figure 3 show the curves obtained for printed flat samples with a thickness of 0.7 mm. It is evident that a higher amount of

Figure 2. (A) 3D printed structures. (B) Section of the CAD image of the alveolar structure with dimensions. (C) Map of the final resolution and fidelity to the CAD files evaluated by 3D scanning. 5629

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633

Research Article

ACS Applied Materials & Interfaces

Figure 3. (A) Tan(δ) curves and (B) E′ curves obtained from DMTA measurements for the hybrid samples containing different amounts of TEOS.

TEOS in the object results in a progressive shift of tan(δ) toward higher temperature. Moreover, an increase of fwhm was observed, indicating an increasing variability of the cross-link density (Table 1). This could be correlated to the formation of a hybrid network in which both C−C bonds coming from the opening of acrylic double bonds and polymer/NPs bonds could be considered as cross-linking points. TGA analyses were performed in order to evaluate the thermal stability of the matrix and to determine the amount of silica created in the thermal hydrolysis/condensation step. The presence of the filler does not cause significant variations in the stability of the matrix; the weight loss at 200 °C is lower than 2% for all the samples (Figure S1). Table 1 reports the inorganic residue measured at 700 °C compared with the theoretical amount of silica estimated from the quantity of precursor: the measured values are close to the theoretical ones. The size, quantity, and distribution of the silica domains inside the printed structures were investigated by FESEM analysis, arms of the honeycomb printed samples were cryofractured and both surface and core were observed. The collected pictures (Figure 4) show that silica nanoparticles are well dispersed in the matrices. In all the cases the particles observed on the surface result slightly larger than those seen in the cross section and this could be due to the method selected for the sol/gel process. The hydrolysis/condensation reaction was performed at 70 °C in acid vapors, in such conditions the concentration of the acid species is higher on the surface than in the bulk of the material, where diffusion needs to occur in order to activate the reaction. Moreover the formation of NPs on the surface could constitute a further barrier for the diffusion of acid species toward the core. This gradient of acidic conditions could induce a more effective sol/gel process on the surface than in the bulk of the material. Furthermore, it can be observed that by varying the quantity of silica precursor the shape of the generated nanoparticles also varies. When only MEMO is added to the PEGDA matrix, few spherical particles are visible with an average diameter of about 100 nm in the cross section and 300 nm on the top; for an amount of TEOS of 20 phr the quantity of spherical nanoparticles is increased both on the surface (av diameter, 200 nm) and in the bulk material (av diameter, 400 nm). When 30 phr of TEOS are added, some thin whisker-like nanoparticles (av length, 600 nm) are observable in the bulk material together with the spherical ones, for higher amounts of precursor only elongated nanoparticles, with relatively low aspect ratio, are visible (300 × 100 nm in the bulk, 700 × 300

Figure 4. SEM images of the (left) cross section and (right) top view of samples MEMO, TEOS 20, TEOS 30, and TEOS 40.

nm on the top). Such differences in the morphology of the inorganic phase could surely influence the mechanical properties of the structures. 5630

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633

Research Article

ACS Applied Materials & Interfaces

Figure 5. (A) Plot of load versus depth. (B) Hardness. (C) Reduced elastic modulus values.

Figure 6. (A) Tensile tests performed on flat samples. (B) Compression tests performed on alveolar structures.

influences the elastic modulus and the hardness; the best performances, considering the nanoscale behavior, are obtained with 40 phr of TEOS, this sample has also the highest isotropic linear elastic recovery. Nanoindentation tests were also performed on the cross section of the samples after cryofracture (Figure S2). PEGDA and MEMO specimens gave results similar to those obtained on the surface, while samples containing TEOS, in this case, did not show any improvement in H or Er with respect to MEMO one. This is in good agreement with FESEM measurements that showed a NPs surface enrichment. Flat specimens with a thickness of 0.7 mm were printed to be used for tensile tests. Figure 6A reports the stress/strain curves of the different treated samples: as visible the Young’s modulus is doubled with respect to that of neat PEGDA when TEOS is added to the matrix, as expected the quantity of the filler influences the material stiffness. MEMO sample, containing 2.4 wt % of silica according to TGA measurements, shows an increase of E of 50% with respect to neat PEGDA while samples containing the silica precursor, and thus more nanoparticles (more than 7 wt %), present a further increase between 100 and 120%. Sample TEOS 20 presents the highest Young’s modulus, while samples TEOS 30 and 40 gave slightly

NP surface enrichment could have also an influence on the mechanical properties of the 3D printed object. Sol−gel process for the in situ generation of silica nanoparticles has been largely used to increase the surface hardness of coatings;34 nano indentation tests were then performed on different samples aiming to observe the variation of the hardness and of the mechanical properties at the nanoscale,37,38 the results are shown in Figure 5. As seen in Figure 5A, the addition of TEOS in the organic matrix causes an increase in the normal load necessary to reach a 1 μm indentation depth. This indicates a higher hardness of the hybrid structures compared to the pure PEGDA ones. In addition, lower final penetration depth was observed for higher inorganic phase content. Hybrid samples also show better elastic recovery as visible from the lower hysteresis observable. Figure 5B,C reports the surface nanohardness (H) and elastic reduced modulus (Er): both increase when TEOS is added to the matrix. The Er of MEMO sample is doubled with respect to neat PEGDA, with a further small increase for samples containing 20 phr and 30 phr of TEOS. Sample with 40 phr of TEOS shows the highest value of Er (32 MPa). Hardness has a similar trend and the highest value is obtained, also in this case, for sample TEOS 40. Thus, the presence of the filler 5631

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633

ACS Applied Materials & Interfaces

■

lower values indicating that intermediate quantities of silica particles with lower dimensions lead to the best performances under tensile stresses. We can also supposed that the different shapes observed in the SEM analyses can influence the obtained values. Once we evaluated the positive influence of the generation of silica nanoparticles on the behavior of the bulky samples under tensile stress, honeycomb structures were printed and tested in compression tests aiming to observe the response of empty and light architectures under this kind of solicitation. Figure 6B reports the stress−strain curves under compression. In this case, the obtained values are more similar to those reported from the nanoindentation test; in fact, even if at the nanoscale, this latter also imposes a compression stress to the sample. The measurements performed on the alveolar structures showed that in the presence of MEMO the Young’s modulus is reduced with respect to PEGDA sample, this can be explained considering that in the former sample few nanoparticles are generated, and thus, these are probably not sufficient to make the material stiffer. Furthermore, the use of the coupling agent reduces the cross-linking density of the polymer causing a decrease of the mechanical properties of the entire structure as observed. When TEOS is also added to the matrix, the presence of the generated nanoparticles implies an evident increase of the Young’s modulus which reaches values of 35 MPa for the sample TEOS 40.

■

CONCLUSIONS

■

ASSOCIATED CONTENT

Research Article

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Gou, M.; Qu, X.; Zhu, W.; Xiang, M.; Yang, J.; Zhang, K.; Wei, Y.; Chen, S. Bio-inspired Detoxification Using 3D-Printed Hydrogel Nanocomposites. Nat. Commun. 2014, 5, 3774. (2) Derby, B. Printing and Prototyping of Tissues and Scaffolds. Science 2012, 338, 921−926. (3) Anderson, K. B.; Lockwood, S. Y.; Martin, R. S.; Spence, D. M. A 3D Printed Fluidic Device that Enables Integrated Features. Anal. Chem. 2013, 85, 5622−5626. (4) Shallan, A. I.; Smejkal, P.; Corban, M.; Guijt, R. M.; Breadmore, M. C. Cost-Effective Three-Dimensional Printing of Visibly Transparent Microchips within Minutes. Anal. Chem. 2014, 86, 3124−3130. (5) Lewis, J. A.; Ahn, B. Y. Device Fabrication: Three-Dimensional Printed Electronics. Nature 2015, 518, 42−43. (6) Espalin, D.; Muse, D.; MacDonald, E.; Wicker, R. 3D Printing Multifunctionality: Structures with Electronics. Int. J. Adv. Manuf Technol. 2014, 72, 963−978. (7) Kong, Y. L.; Tamargo, I. A.; Kim, H.; Johnson, B. N.; Gupta, M. K.; Koh, T.-W.; Chin, H.-A.; Steingart, D. A.; Rand, B. P.; McAlpine, M. C. 3D Printed Quantum Dot Light-Emitting Diodes. Nano Lett. 2014, 14, 7017−7023. (8) Hofmann, M. 3D Printing Gets a Boost and Opportunities with Polymer Materials. ACS Macro Lett. 2014, 3, 382−386. (9) Gross, B. C.; Erkal, J. L.; Lockwood, S. Y.; Chen, C.; Spence, D. M. Evaluation of 3D Printing and Its Potential Impact on Biotechnology and the Chemical Sciences. Anal. Chem. 2014, 86, 3240−3253. (10) Lee, M. P.; Cooper, G. J. T.; Hinkley, T.; Gibson, G. M.; Padgett, M. J.; Cronin, L. Development of a 3D Printer Using Scanning Projection Stereolithography. Sci. Rep. 2015, 5, 9875. (11) Bártolo, P. J. Stereolithography: Materials, Processes, and Applications. Springer: New York, 2011. (12) Vaezi, M.; Seitz, H.; Yang, S. A review on 3D Micro-Additive Manufacturing Technologies. Int. J. Adv. Manuf Technol. 2013, 67, 1721−1754. (13) Chan, V.; Zorlutuna, P.; Jeong, J. H.; Kong, H.; Bashir, R. Three-dimensional Photopatterning of Hydrogels Using Stereolithography for Long-Term Cell Encapsulation. Lab Chip 2010, 10, 2062−2070. (14) Chan, V.; Jeong, J. H.; Bajaj, P.; Collens, M.; Saif, T.; Kong, H.; Bashir, R. Multi-material Bio-Fabrication of Hydrogel Cantilevers and Actuators with Stereolithography. Lab Chip 2012, 12, 88−98. (15) Arcaute, K.; Ochoa, L.; Medina, F.; Elkins, C.; Mann, B.; Wicker, R. Three-Dimensional PEG Hydrogel Construct Fabrication using Stereolithography. MRS Online Proceedings Library 2005, 874. (16) Skardal, A.; Atala, A. Biomaterials for Integration with 3-D Bioprinting. Ann. Biomed. Eng. 2015, 43, 730−746. (17) Zhu, W.; Li, J.; Leong, Y. J.; Rozen, I.; Qu, X.; Dong, R.; Wu, Z.; Gao, W.; Chung, P. H.; Wang, J.; Chen, S. 3D-Printed Artificial Microfish. Adv. Mater. 2015, 27, 4411−4417. (18) Soman, P.; Fozdar, D. Y.; Lee, J. W.; Phadke, A.; Varghese, S.; Chen, S. A Three-dimensional Polymer Scaffolding Material Exhibiting a Zero Poisson’s Ratio. Soft Matter 2012, 8, 4946−4951. (19) Harris, J. M. Poly (ethylene glycol) Chemistry: Biotechnical and Biomedical Applications. Springer Science & Business Media: Berlin, Germany, 2013. (20) Nguyen, K. T.; West, J. L. Photopolymerizable Hydrogels for Tissue Engineering Applications. Biomaterials 2002, 23, 4307−4314. (21) Mann, B. K.; Gobin, A. S.; Tsai, A. T.; Schmedlen, R. H.; West, J. L. Smooth Muscle Cell Growth in Photopolymerized Hydrogels with Cell Adhesive and Proteolytically Degradable Domains: Synthetic

Hybrid organic−inorganic 3D structures were fabricated by incorporating silica precursors into PEGDA oligomer in the presence of suitable photoinitiators and exposing them to a digital light system. A post treatment in acidic vapors allowed a sol−gel reaction of the metal-alkoxysilane trapped in the polymeric network, and thus, the in situ generation of the inorganic phase covalently interconnected to the polymeric matrix through a bridging agent. This method allows to build 3D structures with improved mechanical properties operating with a full liquid formulation without the typical drawbacks of incorporating inorganic powders. The kinetics of the reaction was followed by photoreology evidencing that the presence of inorganic precursor does not influence the photopolymerization reaction. The generation of an inorganic phase remarkably changes the mechanical properties inducing an increase of Tg, an increase of Young’s modulus both in tensile and in compression tests, and an increase of surface hardeness. The latter could be related to a preferential growth of the silica NPs on the surfaces induced by the acid-vapors sol−gel treatment, as also evidenced by FESEM. In the synthesis in this study, we showed a novel and alternative procedure for improving the mechanical properties of DLP 3D printed objects without compromising printability of standard formulations.

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.5b12578. Composition of the formulations, TGA plots, hardness and reduced elastic modulus measured on the cross sections, tensile test strain/stess curves reporting the standard error and compression test strain/stess curves reporting the standard error. (PDF) 5632

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633

Research Article

ACS Applied Materials & Interfaces ECM Analogs for Tissue Engineering. Biomaterials 2001, 22, 3045− 3051. (22) Maher, P. S.; Keatch, R. P.; Donnelly, K.; Mackay, R. E.; Paxton, J. Z. Construction of 3D Biological Matrices Using Rapid Prototyping Technology. Rapid Prototyping Journal 2009, 15, 204−210. (23) Tibbitt, M. W.; Anseth, K. S. Hydrogels as Extracellular Matrix Mimics for 3D Cell Culture. Biotechnol. Bioeng. 2009, 103, 655−663. (24) Malucelli, G.; Priola, A.; Sangermano, M.; Amerio, E.; Zini, E.; Fabbri, E. Hybrid Nanocomposites Containing Silica and PEO Segments: Preparation through Dual-Curing Process and Characterization. Polymer 2005, 46, 2872−2879. (25) Amerio, E.; Fabbri, P.; Malucelli, G.; Messori, M.; Sangermano, M.; Taurino, R. Scratch Resistance of Nano-Silica Reinforced Acrylic Coatings. Prog. Org. Coat. 2008, 62, 129−133. (26) Sangermano, M.; Amerio, E.; Epicoco, P.; Priola, A.; Rizza, G.; Malucelli, G. Preparation and Characterization of Hybrid Nanocomposite Coatings by Cationic UV-Curing and the Sol-Gel Process of a Vinyl Ether Based System. Macromol. Mater. Eng. 2007, 292, 634− 640. (27) Bandyopadhyay, A.; Bhowmick, A. K.; De Sarkar, M. Synthesis and Characterization of Acrylic Rubber/Silica Hybrid Composites Prepared by Sol-Gel Technique. J. Appl. Polym. Sci. 2004, 93, 2579− 2589. (28) Toskas, G.; Cherif, C.; Hund, R.-D.; Laourine, E.; Mahltig, B.; Fahmi, A.; Heinemann, C.; Hanke, T. Chitosan(PEO)/Silica Hybrid Nanofibers as a Potential Biomaterial for Bone Regeneration. Carbohydr. Polym. 2013, 94, 713−722. (29) Georgieva, N.; Bryaskova, R.; Tzoneva, R. New Polyvinyl Alcohol-Based Hybrid Materials for Biomedical Application. Mater. Lett. 2012, 88, 19−22. (30) Wei, W.; Zhang, H.; Guan, S.; Jiang, Z.; Yue, X. Preparation and Characterization of Transparent Polyarylethers-Silica Hybrid Membranes with Covalently Connected Phases. Polymer 2012, 53, 5002− 5009. (31) Han, Y.-H.; Taylor, A.; Mantle, M.; Knowles, K. UV Curing of Organic−Inorganic Hybrid Coating Materials. J. Sol-Gel Sci. Technol. 2007, 43, 111−123. (32) Zou, K.; Soucek, M. D. UV-Curable Organic-Inorganic Hybrid Film Coatings Based on Epoxidized Cyclohexene Derivatized Linseed Oil. Macromol. Chem. Phys. 2004, 205, 2032−2039. (33) Ajayan, P. M.; Schadler, L. S.; Braun, P. V. Nanocomposite Science and Technology. John Wiley & Sons: New York, 2006. (34) Roppolo, I.; Messori, M.; Perruchas, S.; Gacoin, T.; Boilot, J.-P.; Sangermano, M. Multifunctional Luminescent Organic/Inorganic Hybrid Films. Macromol. Mater. Eng. 2012, 297, 680−688. (35) Oliver, W. C.; Pharr, G. M. An Improved Technique for Determining Hardness and Elastic Modulus Using Load and Displacement Sensing Indentation Experiments. J. Mater. Res. 1992, 7, 1564−1583. (36) Sneddon, I. N. The Relation Between Load and Penetration in the Axisymmetric Boussinesq Problem for a Punch of Arbitrary Profile. Int. J. Eng. Sci. 1965, 3, 47−57. (37) Belon, C.; Chemtob, A.; Croutxé-Barghorn, C.; Rigolet, S.; Le Houérou, V.; Gauthier, C. A. Simple Method for the Reinforcement of UV-Cured Coatings via Sol-Gel Photopolymerization. Macromol. Mater. Eng. 2011, 296, 506−516. (38) Soloukhin, V. A.; Posthumus, W.; Brokken-Zijp, J. C.; Loos, J.; de With, G. Mechanical Properties of Silica−(Meth) Acrylate Hybrid Coatings on Polycarbonate Substrate. Polymer 2002, 43, 6169−6181.

5633

DOI: 10.1021/acsami.5b12578 ACS Appl. Mater. Interfaces 2016, 8, 5627−5633