- Email: [email protected]
Photocrosslinking-based 3D printing of unsaturated polyesters from isosorbide: A new material for resorbable medical devices
Journal Pre-proof Photocrosslinking-based 3D printing of unsaturated polyesters from isosorbide: A new material for resorbable medical devices Jan Lammel-Lindemann, Isabela Autran Dourado, Johnathan Shanklin, Ciro A. Rodriguez, Luiz Henrique Catalani, David Dean PII:
S2405-8866(19)30021-1
DOI:
https://doi.org/10.1016/j.bprint.2019.e00062
Reference:
BPRINT 62
To appear in:
Bioprinting
Received Date: 5 April 2019 Accepted Date: 19 September 2019
Please cite this article as: J. Lammel-Lindemann, I.A. Dourado, J. Shanklin, C.A. Rodriguez, L.H. Catalani, D. Dean, Photocrosslinking-based 3D printing of unsaturated polyesters from isosorbide: A new material for resorbable medical devices, Bioprinting (2019), doi: https://doi.org/10.1016/ j.bprint.2019.e00062. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Photocrosslinking-Based 3D printing of Unsaturated Polyesters from Isosorbide: A New Material for Resorbable Medical Devices Jan Lammel-Lindemann1,2,3, Isabela Autran Dourado4, Johnathan Shanklin1, Ciro A. Rodriguez1,2,3, Luiz Henrique Catalani4 and David Dean1,* 1
Department of Plastic and Reconstructive Surgery, The Ohio State University, Columbus, OH 43210, United States. 2 Tecnologico de Monterrey, Escuela de Ingenieria y Ciencias, Monterrey, N.L. 64849, Mexico. 3 Laboratorio Nacional de Manufactura Aditiva y Digital (MADIT), Apodaca, N.L. 66629, Mexico. 4 Departamento de Química Fundamental, Instituto de Química, Universidade de São Paulo, São Paulo. 26077, Brazil.
*Corresponding Author: 460 West 12th Avenue, Rm. 388, Columbus, OH 43220 USA; E-mail: [email protected]; website: www.OsteoEngineering.com Abstract Words: 253 Manuscript without Abstract Words: 3192 Figures: 8 Tables: 0 Keywords: bone, tissue engineering, 3D printing (additive manufacturing), UV curing, digital light processing (DLP), porous scaffold
Abstract Clinically used tissue-engineered devices are few and far between. The search for biocompatible materials has been a great focus for the field for the past 30 years. The ones that exist are hard to work with when manufacturing patient-specific shapes with complex external and internal pore geometries because of their material and mechanical properties. Photo-cross-linkable polymers have arisen as a potential 3D printable solution for this problem. Here we present a new sugar-based polymer, isosorbide-derived polyester as a potential candidate to help fill this gap. Isosorbide-derived polyesters containing different amount of double bonds were synthesized to allow the material to be crosslinked or functionalized. Three resin formulations with different amounts of double bonds available (5, 12 and 17%) were used. After initial thermal and photo-crosslinking studies, Isosorbide-derived unsaturated polyesters were then mixed with a photoinitiator, co-crosslinker, and solvent to form a 3D printable resin formulation. Thin films of that resin were then cured using a Digital Light Projection (DLP) printer and analyzed. An attempt was made to modulate the cured thickness by adding HMB as a photo-attenuator. We assessed 3D printing success and tensile green strength for the 3 resin formulations. We were able to 3D print unsaturated polyesters from isosorbide with a DLP printer when enough double bonds were present to facilitate crosslinking and curing of the material. Finally, thin films were then seeded with fibroblasts following the ISO 10993-5 protocol. These coupons were found safe (i.e., non-cytotoxic and biocompatible), potentially marking the beginning of a new approach to preparing resorbable implants for tissue engineering uses. .
1. Introduction Despite at least 30 years of research, there are very few tissue-engineered devices used clinically. To this date, there is no “synthetic bone substitute”, and this is evident by the need for allografting bone from the patient’s own body to do major surgical reconstructions (1). Current implants tend to be strong metals that generate stress shielding and possibly resorption of the adjacent bone and preventing adaptive remodeling (2). The balance between providing structural support while minimizing stress shielding leaves a very tight window that material science is trying to fill. One of the main reasons for the search for new implant manufacturing strategies is the limited number of biocompatible resorbable materials, especially polymers, which can be used safely to prepare hydrogel and/or solid-cured scaffolds. Very few of these materials can be formed by 3D printing into porous, patient-specific shapes. In this study we present a novel, patent-pending material, isosorbide-derived polyester (3) which shows promise as a scaffold substrate bone tissue engineering, and likely regenerative uses for many other tissues as well. In this study, we show that this sugar molecule, isosorbide, can be prevented from degrading immediately in solution through light-induced crosslinking. Unlike other potential bone scaffolding materials have shown high compressive strength but are brittle (e.g., poly(propylene fumarate)) (4)., However crosslinked isosorbide-derived polyesters are expected to have high tensile strength. Bone regeneration and remodeling physiology require that resorbable devices ideally degrade in during a time span of between 3-6 months. This allows the bone to fully heals without obstructing this process (5). This is necessary because the initial bone healing response, resulting in a disorganized callus, needs to be remodeled by osteoclastic activity to become strong (6). Remodeling cannot occur if it encounters un-resorbed material. Unresorbed material would form
a barrier to this process (7,8). Producing devices for the regeneration of bone and other tissue types (e.g., skin, lung, liver, heart, muscle, etc.) will also different resorption rates in ranges that suit the tissue’s own regeneration (healing) and remodeling process (9–11). Polycondensation of dianhydrohexitols (isosorbide, isommanide, and isoidide) and saturated aliphatic diacids (adipic, gluconic and succinic) is a clever way to synthesize bioabsorbable polyesters. These polymers have unique properties; they are particularly suitable for applications in the medical device field since their degradation products are readily and locally metabolized. Additionally, the applications of these polymers may include surgical fixation devices for bone, tendon, and ligament reconstruction.
Isosorbide or 1,4:3,6-dianhydro-D-glucitol is a bioderived and biodegradable dianhydrohexitol compound qualified by US Food and Drug Administration as “generally recognized as safe” (GRAS). If derived into unsaturated polyester, then this sugar-based polymer with high tensile strength and flexibility is able to be photocured. This bicyclic diol shows thermal and chemical stabilities, it is non-toxic, biodegradable and can be used in copolymerization reactions (12).
Isosorbide can be obtained stereoselectively starting from starch. Enzymatic starch hydrolysis leads to D-glucose production. D-glucose hydrogenation process followed by sorbitol dehydration leads to isosorbide formation (13). Regarding its stereochemistry, isosorbide has a stereochemistry in wedge shape, composed of two dihydrofuranoid rings. There is a hydroxyl group in C2 in Exo configuration, while the hydroxyl in C5 is in endo configuration. Literature dating from the 1960s shows that when in Endo positions, hydroxyls are less reactive than those in the Exo position (14). Polycondensation
of dianhydrohexitols (isosorbide) and saturated or unsaturated aliphatic diacids is an interesting way to synthesize bioabsorbable polyesters (15,16).
Figure 1: Isosorbide structure
The exploration of isosorbide polyesters and the search for new materials was motivated by a desire to go beyond the use of petroleum-based materials. However, it was noticed that the applications of these isosorbide polymers can be diverse and interesting due to their biocompatible characteristics. Juais and co-workers showed the enzymatic production of isosorbide saturated polyesters (15). In that study, various synthesis conditions - such as solvents, temperatures, reaction times and monomers - were studied, so it was possible to produce isosorbide saturated polyesters with molar mass around 40,000 Da. In another study, Naves and co-workers showed the production of unsaturated polyesters of isosorbide and isomanide (one of the stereoisomers of isosorbide), using optimum conditions of the previous work. Copolymers in the order of 4,000-16,000 Da were produced with unsaturated monomers (16). As well as noted in Naves’ work, these double bonds provide the material the possibility of crosslinking or functionalization. In this article, it has been shown that unsaturated isosorbide polymers can be very promising in terms of biomedical applications because they can be 3D printed, giving biocompatible and biodegradable scaffolds.
In this work, photocurable polyesters were synthesized and printed. The C=C bonds form after light exposure, specific initiator starts radical reactions between polymer and crosslinkers, so that the initially liquid material becomes a solid film upon photocrosslinking. In addition to being biocompatible and biodegradable, these polyesters are photocurable. Since they are polyesters, it is known that they can easily undergo chemical hydrolysis inside the body (17,18). In terms of regenerative medicine, there are two types of typical cleavage in the biodegradation process: hydrolytic and enzymatic. In the human body, composed mostly of water, hydrolytic cleavages are strongly important and ester bonding is one of the most common that undergo hydrolysis. The hydrolysis of the ester bonds in the polymer chain leads to small molecules that can be excreted from the body (17). As regards to enzymatic cleavages, it is possible to affirm that it guarantees better control of biodegradation, since, by introducing them with correct peptide groups, one can control exactly which regions will be cleaved by specific enzymes (17). Therefore, the isosorbide polymers which are the subject of this work are biomedically interesting materials since their degradation byproducts are readily metabolized. The ability to photo-cure this material makes 3-D printing an option for our resin formulations. Development of additive manufacturing methods for this material may be able to fill a need in the medical implant industry by enabling specific parts to be manufactured for bone scaffolding, surgical fixation devices for bone, tendon, ligament reconstruction, vascular graft scaffold, surgical sutures, and carriers for controlled drug release. Using Digital Light Processing (DLP), we will project a mask that is computer-generated by slicing the desired object to print into 50 µm layers in software. The depth of these slices is varied to determine the best elevator movement in the Z direction. X and Y for 3D printing. Laser projection is controlled by a micromirror chip that reflects the UV light to the clear vase of the resin vat, where it cures the
resin crosslinking it. There’s some over penetration, that allows each new layer to attach to the previous one in what we call interlayer “stitching”. After the new layer has been stitched to the previously layer, the elevator will move again and the next mask will be projected. Most products used for these purposes currently are made from non-resorbable materials such as ceramics, polymers, and metals. Our isosorbide material would be able to fill a much-needed void for bio-implantable materials that are degradable, softer, more flexible and able to be used for manufacturing via 3D printing.
2. Materials and Methods 2.1 Isosorbide Purification The isosorbide was recrystallized twice from ethyl acetate, using an external oil bath at 50°C, and dried under vacuum in the presence of phosphorus pentoxide until reaching constant mass. Their final purity was determined by Gas Chromatography-Mass Spectrometry (GC-MS QP2020 – Shimadzu with the column BPX5 ((5%-Phenyl)-methylpolysiloxane) - 30m.) Before use, all solvents used (Merck) were dried and distilled. Isosorbide (Sigma-Aldrich) was recrystallized twice from ethyl acetate, dried under vacuum for 72 hours and weighed in a Glove Box under N2 atmosphere (it is a very hygroscopic compound). Adipic diethyl and Fumaric acid (Sigma-Aldrich) were used as received. CAL-B (Novozyme; specific activity 10,500 PLU/g) is the Lipase B from Candida antarctica (NZ 435) immobilized in microporous resin Lewatit 1600. Hydroquinone (BHerzog) was used as received.
2.2 Synthesis of diethyl fumarate 23.83 g of fumaric acid was added in a round-bottomed flask with 120 mL of anhydrous ethanol. The mixture was heated under reflux over 15 minutes. Then, 5 mL of sulfuric acid was added dropwise, followed by 100 mL of toluene and 15 minutes of heating under reflux. Another 15 mL of sulfuric acid was added dropwise gently and the mixture was heated under reflux during more 30 minutes. The system was cooled at room temperature and neutralized with NaHCO3 saturated solution. Organic phases were extracted and the aqueous phases were washed with toluene. After combining organic phases, the mixture was dried with Na2SO4 over 24 hours. The diethyl fumarate was concentrated by rotary evaporation and then purified by distillation under reduced pressure. 2.3 Synthesis of unsaturated polyesters Polycondensation reactions of isosorbide with diethyl adipate and fractions of diethyl fumarate were carried out according to previous group development (16) from the following copolymerization reaction:
Figure 2: Schematic representation of co-polymerization reaction producing poly(isosorbide adipate-co-isosorbide fumarate).
To synthesize the unsaturated isosorbide polymer, isosorbide (IS), diethyl adipate (AD) and diethyl fumarate (FU) - molar ration IS:AD:FU (10:9.5:0.5) - were added into a flask containing cyclohexane:toluene (6:1) mixture. CAL-B (10% relative to total monomer mass) was employed as the catalyst and hydroquinone (1% w/w) was used to avoid radical addition reactions in double bonds during the copolycondensation. To produce 30 g of polymer, the amount of reactants were calculated in relation to 70 mmol of isosorbide and 270 ml of the same solvent mixture. The flask was attached to a Dean Stark apparatus that contained 4A molecular sieves for byproduct removal. The Dean Stark was attached to a condenser. The molecular sieves were replaced with fresh dried molecular sieves every 24 h and the solvent level was kept constant. The reaction was magnetically stirred and heated at 90°C sing an external oil bath. After 7 days of reaction, the solvent was removed, while chloroform was added and the enzyme was filtered. The product was concentrated, by rotary evaporation, removing chloroform. Three resin formulations were obtained this way, corresponding to different double bonds inserted 5% (A), 12% (B), and 17% (C). The unsaturation content in polyester chains was evaluated by 1H NMR by relation integral values of methylene groups from adipate and alkenyl groups from fumarate. 2.4 Initial Thermal and photocrosslinking studies Isosorbide polymer was mixed with chloroform as a solvent and stirred overnight. The polymer was mixed with a resin made from 3% BASF Irgacure 819 (BAPO) and 35% (w/w) crosslinker (1-vinylimidazole, N-vinylpyrrolidone or N-isopropylacrylamide). All solutions were mixed with a vortex, its contents were transferred to cylindric (2.5 cm of diameter and 1.5 cm of height) Teflon molds, which were subjected to photocrosslinking by UV light. Below, it is possible to see a schematization of the process.
Figure 3: Schematic representation of photocrosslinking tests The material was then tested at 30, 45 and 60 minutes in standard power UV light to induce crosslinking. It is worth noting that the UV light (LuzChem®) operates with radiance near to 4.6mW/cm² in the peak of wavelength 254 nm.
2.5 Photocurable Resin Formulation Three resins of varying molecular weights were used with 5% (A), 12% (B), and 17% (C) double bonds inserted respectively, using 58% w/w of the resin by weight. VIM (1vinylimidazole), (Sigma-Aldrich, MO) was used as co-crosslinker at a 32% w/w; as well as 3% Irgacure 819 (BAPO) (BASF, NJ) as photoinitiator and dissolved in Ethyl Acetate as the solvent at 7% w/w. Isosorbide polymer was mixed with the ethyl acetate/photoinitiator pack and stirred overnight in a sealed container.
2.6 Thin Film Curing and 3D Printing Methods Both thin film curing and 3D printing of these resins using a cDLP 3D printer are performed according to the work of Hernan Lara Padilla published in his doctoral thesis using an EnvisonTEC® µicro® printer (Dearborn, MI). That method allows the measurement of tensile
green strength. This device utilizes 405 nm light and a 450 mW/cm² intensity measured with a RM 12 radiometer with a VIS-B sensor (Dr. Grobel, Ettlingen, GER), much different than the initial photo-crosslinking light source in the Catalani lab. Thin-film measurements were done the same way as we have characterized other printable resins with as explained below. Microscope slides were measured previously and recorded. About 7 drops of resins were set on top of the base slide and then cured in the printer for the assigned time using 1.0 x 1.0 cm projected squares. Two separate time points were investigated for these thin film cure tests: 60 and 150 seconds. Once the curing was finished, the slide was gently wiped to take uncured material off, the top cover slide was placed over the cured part and this sandwich was measured with a caliper (Mitutoyo Absolute Digimatic, Mitutoyo, IL). After subtracting the initial slides thicknesses, we get the cure depth for the thin films. A total of 8 samples per time point were recorded.
2.7 Film Thickness Control
HMB (Oxybenzophenone) (Sigma, St. Louis, MO) was added to act as well as a light attenuator (reducing thin film thickness). Cure tests were repeated at 30, 60, 90 and 120 seconds, measured utilizing the same method as described above for the thin film curing. This was tested only with resin Formulation A.
2.8 Mechanical Testing In our experience resin formulations that exhibit thin films thickness of 150-250 µm can be 3D printed without an attenuator under our conditions in the EnvisonTEC® µicro® printer;
however, some resins were well above that thickness. The green strength of these 3D printed resin mixtures was investigated by using a load sensor that holds the build platform (after Lara Padilla, H, 2018, unpublished and patent-pending doctoral thesis) (19). A very small build plate was first coated with a “burn-in” patch of isosorbide. That burn-in patch acts as an anchor and support for the next 20 masks (layers) of 50 microns each to create a 1 mm diameter post. Then 3 consecutive layers were projected with increasing diameters of up to 8 mm to test the cured resin’s tensile strength. This also shows that the resin can be properly 3D printed by joining the layers during the cDLP 3D printing process (19). The tensile green strength sensor set up and the print visualization is shown below.
Figure 4: Schematic of the separation force and green strength in an EnvisionTEC (Dearborn, MI) µicro 3D printer. The load sensor and montage can be seen in (a). The build platform as it’s getting printed can be seen in (b). The build job schematic can be seen in (c), where after a burnin plate consisting of a wide base curing with 90 s for 3 layers and then 2 more with 60 s for
anchorage, followed by 20 layers of a 1 mm diameter are printed at a 60 s exposure time, and then followed by increasing diameter layers that become greater after every 3 layers printed in a matter of 0.5 mm as well at 60 s exposure time. The separation force is being tracked in real-time until the part fails by breaking from the 1 mm post when the separation force surpasses the material’s mechanical properties. 2.9 Cytotoxicity Per the ISO Standard 10993-5, 8 samples of 10 mm2 thin films were printed (4 for resin B and 4 for resin C, employing the same samples obtained from the printability tests transferred to PBS and left soaking for 5 minutes. This step was repeated 3 times. Then they were autoclaved in the liquid cycle for 30 minutes at a temperature of 121ºC. A total of 25,000 murine L929 American Type Culture Collection (ATCC, Rockville, MD, USA) fibroblasts were counted manually and seeded on an Isosorbide thin film. The seeded thin films were then placed in a low attachment well for culture in Dulbecco's Modified Eagle's Medium (DMEM; Gibco, Walthman, MA, USA) supplemented with 10% horse serum, 1% L-glutamine, 1% sodium pyruvate, 50 U/mL penicillin and 50 ug/mL streptomycin (HyClone, Logan, UT, USA). Samples were kept at 37 ºC and 5% CO2 atmosphere for three days. After 72 hours, cell proliferation was recorded using the PrestoBlue® (Invitrogen, MA, USA) assay and the number of cells present was calculated using a standard curve. Control groups included negative controls, 4 wells grown with complete DMEM for 72 hours as positive controls, and 4 samples per resin formulation. Effluent samples were collected in triplicate and measured as per the assay's instructions. Results were reported as average total cells per well. 2.10 Statistical Analysis
Quantitative data is shown as average ± standard deviation. Statistical analysis was performed by using GraphPad® Prism® 7.o, using two-way ANOVA with Turkey’s multiple comparisons for figure 5, ordinary one-way ANOVA with multiple comparisons for figure 6. Statistically significant values were defined as a p-value of <0.05. Cure tests were done with an n=8, cytotoxicity with an n=4. Legends are as follows ns= non-significant, *= p<0.05, **= <0.01, ***=<0.001 and ****=<0.0001.
3. Results Thin films were cured to measure the cure depth of the different resins without a light attenuator. We found that resins B and C were fully photo-curing at both exposure times. Resin A
ns ns
C 2
B 2
**
A 2
Cure Thickness ( m)
formulation did show a difference, but the photocrosslinking was found to be insufficient as the
parts were not fully cured this was then assessed by observing the resin A thin films immediately dissolve in phosphate-buffered saline solution. These results can be seen in figure 5.
Figure 5: Graph showing the cured thickness of resins A, B, and C. (5%, 12%, and 17% double bonds respectively) without photoattenuator (HMB). Image showcasing a cured film after being measured for illustration purposes. After noting that the droplets were fully cured, the decision to add HMB as a photoattenuator to control the cure depth for every layer was made. In previous work from our laboratory with other printable polymers, this was found as a key element in controlling the resolution for the printed parts and can aid in developing an optimized 3D printing process. Resin A was used for the graph shown in figure 6.
Figure 6: Influence of photoattenuator (HMB) on resin formulation type A (5% double bonds). Knowing that the resin formulations exhibit appropriate curing thickness, the mechanical properties of the materials were tested. While no light attenuator was used here, resins B and C were able to produce a 3D printed piece, that correlated quite well with the double bonds available to the resins. It was clear that increasing the percentage of double bonds improved the mechanical properties. Resin A did complete a burn-in patch at 90s, as witnessed by the force spikes seen in the first part of the curve (burn-in patch at 90 seconds), but the part completely failed to print at this point once the layers were printed at 60 seconds per layer. It is important to note that resin B started to fail during the print, as seen in both the graph showing positive force and the printed part with poor resolution. Resin C printed fully and the part was recovered as shown in panel C2 in figure 7. Resin B shows a green tensile strength of 123.79 Pa although the geometry was already partially failing while resin C formulation showed a tensile strength of
108.59 Pa with much improved geometric accuracy. The recovered parts and their respective force plots can be seen in figure 7.
Figure 7: Force plots taken during the cDLP printing of the resins and parts recovered. A1 corresponds to the resin A, and the part failed to print after the burn-in plate. B2 corresponds to resin B, and the part recovered is shown in B2. C1 corresponds to resin C, where the full part recovered is shown in C2 below the force graph. The force plots for all the resins show that there’s an increased separation negative force at the beginning where the anchoring and support burn-in masks were constructed, corresponding to the force it takes to separate the printed layer from the basement where the mask was projected. These were done in 3 masks of 90 seconds and 2 masks of 60 seconds each. Resin A failed to 3D print once the 1 mm diameter masks were projected. The force plot remains stable after this as
there’s no part being printed that exerts any force. For resins B and C, the higher amount of double bonds available correlates with better attachment (stitching) from one mask projection (i.e., 3D printing one layer by cDLP) to the next. This translates into a 3D printed part. The negative forces can be seen increasing with the increasing diameter of the masks in sets of 3, as expected. The positive spikes in the graphs can be explained as the force recorded once the build platform elevates enough to allow for the resin to flow back in and then compresses this uncured fluid. After the parts are printed, the plot returns to baseline as no more masks are being projected. The thin films from the cure depth studies were then recovered by scraping them off the microscope slides that held them., then washed 3 times in PBS. These were autoclaved and kept sterile until the seeding with murine fibroblast cells in low attachment well plates. The viability assay was then run on days 0 and 3 and graphed against a standard curve showing viability at day 3. The drop in cell count from day 0 to day 3 can be accounted by the low attachment well plates where cells that didn’t attach to the isosorbide proceeded to die, but the ones that attached to the films survived. The graph regarding this is shown in figure 8 below.
Figure 8: Cytotoxicity of the printed thin films was tested by presto blue. Cells were shown to survive after being seeded on the thin films after 3 days.
Discussion & Conclusions As we search for resorbable polymers that can be used for bone tissue engineering and regeneraion of other tissues, the field of biomanufacturing has been mostly dominated by polymers with rapid resorption times, such as polylactides (within 3 months) or slow resorption rates (e.g., polycaprolactones) which can take years if they resorb at all (20). Isosorbide provides a material that has appropriate green tensile strength, and this characteristic can aid in providing new uses for the material as scaffolding and other medical devices to be implanted. Isosorbide polyesters can designed to be resorbed in vivo in a period of 3 to 9 months and can fulfill the
need for a biomedical-grade, resorbable, 3D printable polymer for use in FDA-approved medical devices. It is important to note that we mention 3 month minimum as this is a critical span biologically for the scaffold to serve as a guide the infusion of new tissue through a 3D printed pore geometry. It is also expected to facilitate remodeling, as it is expected to degrade in the key window of 3-9 months. We report here the first time that this material has been 3D printed to the best of our knowledge. For that purpose, we used continuous Digital Light Processing (cDLP) projection of UV light. 3D printing, also referred to as additive manufacturing, opens the possibility of controlling both mechanical properties and geometry. The addition of isosorbide as a photocrosslinkable, resorbable, and biocompatible material will be valuable if it can be demonstrated to be 3D printed with a high level of accuracy while being biocompatible and, hypothetically, resorbable in the window expected of a tissue engineering scaffold.
Acknowledgments The authors are grateful for partial support received from a Global Gateway Mobility Grant, #2015/5041-0, from the FAPESP (Sao Paulo, Brazil) Foundation and the Ohio State University. Glossary
DLP: Digital light processing
BAPO: Irgacure 819 HMB: Oxybenzophenone DEF: Diethyl Fumarate EA: Ethyl Acetate
Declaration of interest
Drs. Dean and Catalani have submitted a patent application on the use of isosorbide for tissue engineering applications as presented in this manuscript.
References 1. Kinaci A, Neuhaus V, Ring DC. Trends in bone graft use in the United States. Orthopedics. 2014 Sep;37(9):e783-788. 2. Huiskes R, Weinans H, van Rietbergen B. The relationship between stress shielding and bone resorption around total hip stems and the effects of flexible materials. Clin Orthop Relat Res. 1992 Jan;(274):124–34. 3. David Dean, Luiz Henrique Catalani. Polyesters, Polymer Compositions, and Methods of Using Thereof. US Provisional Patent Application # 62/750,665., 2018. 4. Wilson JA, Luong D, Kleinfehn AP, Sallam S, Wesdemiotis C, Becker ML. Magnesium Catalyzed Polymerization of End Functionalized Poly(propylene maleate) and Poly(propylene fumarate) for 3D Printing of Bioactive Scaffolds. J Am Chem Soc. 2018 10;140(1):277–84. 5. DeLacure MD. Physiology of bone healing and bone grafts. Otolaryngol Clin North Am. 1994 Oct;27(5):859–74. 6. Frost HM. From Wolff’s law to the Utah paradigm: Insights about bone physiology and its clinical applications. The Anatomical Record. 2001;262(4):398–419. 7. Wang MO, Vorwald CE, Dreher ML, Mott EJ, Cheng M-H, Cinar A, et al. Evaluating 3Dprinted biomaterials as scaffolds for vascularized bone tissue engineering. Adv Mater Weinheim. 2015 Jan 7;27(1):138–44. 8. Willie BM, Petersen A, Schmidt-Bleek K, Cipitria A, Mehta M, Strube P, et al. Designing biomimetic scaffolds for bone regeneration: why aim for a copy of mature tissue properties if nature uses a different approach? Soft Matter. 2010 Oct 5;6(20):4976–87. 9. Atala A. Tissue Engineering and Regenerative Medicine: Concepts for Clinical Application. Rejuvenation Research. 2004 May 1;7(1):15–31. 10. Tur K. Biomaterials and Tissue Engineering for Regenerative Repair of Articular Cartilage Defects. Arch Rheumatol. 2009;24(4):206–17. 11. Materials for Biomedical Engineering: Absorbable Polymers - 1st Edition [Internet]. [cited 2019 Aug 2]. Available from: https://www.elsevier.com/books/materials-for-biomedicalengineering-absorbable-polymers/grumezescu/978-0-12-818415-8 12. Studies in Diuresis with Isomannide. Sugar Alcohols XVIII - John C. Krantz, C. Jelleff Carr, 1938 [Internet]. [cited 2019 Apr 1]. Available from: https://journals.sagepub.com/doi/abs/10.3181/00379727-39-10279 13. Beldi M, Medimagh R, Chatti S, Marque S, Prim D, Loupy A, et al. Characterization of cyclic and non-cyclic poly-(ether-urethane)s bio-based sugar diols by a combination of MALDI-TOF and NMR. European Polymer Journal. 2007 Aug 1;43(8):3415–33.
14. Hopton FJ, Thomas GHS. Conformations of some dianhydrohexitols. Can J Chem. 1969 Jul 1;47(13):2395–401. 15. Juais D, Naves AF, Li C, Gross RA, Catalani LH. Isosorbide Polyesters from Enzymatic Catalysis. Macromolecules. 2010 Dec 28;43(24):10315–9. 16. Naves AF, Fernandes HTC, Immich APS, Catalani LH. Enzymatic syntheses of unsaturated polyesters based on isosorbide and isomannide. Journal of Polymer Science Part A: Polymer Chemistry. 2013;51(18):3881–91. 17. Mondschein RJ, Kanitkar A, Williams CB, Verbridge SS, Long TE. Polymer structureproperty requirements for stereolithographic 3D printing of soft tissue engineering scaffolds. Biomaterials. 2017 Sep;140:170–88. 18. Gonçalves FAMM, Fonseca AC, Domingos M, Gloria A, Serra AC, Coelho JFJ. The potential of unsaturated polyesters in biomedicine and tissue engineering: Synthesis, structure-properties relationships and additive manufacturing. Progress in Polymer Science. 2017 May 1;68:1–34. 19. Hernan Lara-Padilla, Ciro A Rodriguez, David Dean. Fabrication of porous scaffolds using additive manufacturing with potential applications in bone tissue engineering. US Provisional Patent Application #62/715596, 2018. 20. Fisher JP, Holland TA, Dean D, Mikos AG. Photoinitiated cross-linking of the biodegradable polyester poly(propylene fumarate). Part II. In vitro degradation. Biomacromolecules. 2003 Oct;4(5):1335–42. 21. Hopton; F. J.Thomas GHS. Conformations of some dianhydrohexitols. Canadian Journal of Chemistry. 1969;47:2395–401. 22. Juais D, Naves AF, Li C, Gross RA, Catalani LH. Isosorbide polyesters from enzymatic catalysis. Macromolecules. 2010;43(24):10315–9.
Supplementary data: NMR spectra data of monomers and polymers (1) Isosorbide (1,4:3,6-dianhydro-D-glucitol)
1
H NMR (300MHz, D2O, δ, ppm): 4.62 (t, 1H, C4); 4,45 (d, 1H, C3); 4.4-4.33 (m, 1H, C2);
4.29-4.31 (dt, 1H, C5); 3.82-3.93 (m, 3H, C1’, C1” and C6”); 3.43-3.49 (m, 1H, C6’).
isosorbide.esp Water
DEUTERIUM OXIDE 1.0 0.9
Normalized Intensity
0.8 0.7 0.6 0.5
0.1
3.82
3.93 3.92
0.2
4.47 4.45 4.40 4.37 4.35 4.33 4.31 4.29
4.63 4.62 4.60
0.3
3.49 3.46 3.43
3.90 3.86 3.85
0.4
0 0.98 4.7
4.6
0.98 4.5
1.00 0.95 4.4
4.3
2.98 4.2
4.1 4.0 3.9 3.8 Chemical Shift (ppm)
1.00 3.7
3.6
3.5
3.4
3.3
3.2
It is worth to note that we have two diastereotopic protons in C1 and two in C6 – represented by C1’/C1” and C6/ C6’- related to hydrogens that are above or below the bicyclic ring.(21)
(2) Diethyl adipate:
1
H NMR (300MHz, CDCl3, δ, ppm): 1.26 (t, 6H, C1); 1.64-1.72 (m, 4H, C4); 2.30-2.34 (m, 4H,
C3); 4.13 (q, 4H, C2).
1.26
Diethyl Adipate.esp 1.0 0.9 0.8
1.28 1.23
0.6 0.5 4.14 4.12
0.4
1.66
Normalized Intensity
0.7
CHLOROFORM-d 0.1
1.72 1.64
0.2
2.34 2.32 2.30
4.16 4.09
0.3
3.99
4.46
TMS
0 4.00 7.0
6.5
6.0
5.5
5.0
4.5
4.0 3.5 3.0 Chemical Shift (ppm)
2.5
2.0
1.5
6.06 1.0
0.5
0
(3) Diethyl fumarate
1
H NMR (300MHz, CDCl3, δ, ppm): 1.32 (t, 6H, C1); 4.26 (q, 4H, C2); 6.85 (s, 2H, C4).
1.32
Diethyl Fumarate.esp CHLOROFORM-d 6.85
1.0 0.9 0.8
4.27 4.25
1.35 1.30
0.6 0.5 0.4 0.3 4.30 4.23
Normalized Intensity
0.7
0.2
TMS 0.1 0 1.88 7.0
3.95 6.5
6.0
5.5
5.0
4.5
4.0 3.5 3.0 Chemical Shift (ppm)
6.00 2.5
2.0
1.5
1.0
0.5
0
(4) Poly(isosorbide adipate-co-isosorbide fumarate) 4a) SAMPLE 12%
1
H RMN (300MHz, CDCl3, δ, ppm): 1.24-1.28 (m, from hydroxyl end groups); 1.61-1.68 (m,
4H, C9); 2.33-2.41 (m, 4H, C8, C13, C16); 3.53 (m, hydroxyl from C14); 3.78-3.83 (m, 1H, C6);
3.91-3.97 (m, 2H, C1); 4.00-4.02 (methyl and ethyl end groups); 4.12-4.14 (dd, C14, C17); 4.464.48 (d, 1H, C3); 4.83 (t, 1H, C4); 5.13-5.20 (m, 2H, C2 and C5); 7.16 e 7.19 (m, endo/exo C11).
7.27 7.27
201711291031_23446#2_1HFID.100.esp CHLOROFORM-d
0.7
TMS 0.01 2.36
0.5
1.68 1.26
1.70 1.69 1.67 1.65 1.62
2.41 2.40 2.35 2.35
3.97 3.94 3.92
4.83
3.83 3.81 3.80
0.1
4.82
5.20 5.16 5.13 5.15
0.2
4.48 4.47
0.3
0.08
3.96
0.4
7.26 7.24 7.19 7.17
Normalized Intensity
0.6
0 0.56 7.5
7.0
1.40 0.67 0.690.13 2.13 0.67 6.5
6.0
5.5
5.0
4.5
4.0 3.5 3.0 Chemical Shift (ppm)
3.35 2.5
4.00 2.0
1.5
1.0
0.5
0
Q: How is the percentage of double bonds determined? A: By relating protons integral values of C9 (methylene groups) and C11 (alkenyl groups). The unsaturation content is always in relation to saturated portion of the polymer. If we have 4 as proton integral value for C9 and 0,56 for C11, we have ~12% of fumarate residue (unsaturated monomer) inserted – in relation to adipate (saturated monomer). It is worth noting that we have, in NMR, an indication of Michael’s Addition in some double bonds, which is represented by some small peaks in 4.12-4.14 ppm (dd for C14 and C17). It’s
also represented in the structure purposed. The unsaturation ratio calculated take these side reactions in account.(22) 4b) SAMPLE 17% NMR attribution for Poly-(isosorbide adipate-co-isosorbide fumarate) 1H RMN (300MHz, CDCl3, δ, ppm): 1.24-1.28 (m, from hydroxyl end groups); 1.61-1.68 (m, 4H, C9); 2.34-2.40 (m, 4H, C8, C13, C16); 3.51-3.53 (m, hydroxyl from C14); 3.79-3.83 (m, 1H, C6); 3.92-3.97 (m, 2H, C1); 4.00-4.02 (methyl and ethyl end groups); 4.12-4.14 (dd, C14, C17); 4.47 (d, 1H, C3); 4.83 (t, 1H, C4); 5.14-5.20 (m, 2H, C2 and C5); 7.16-7.19 (d, endo/exo C11).
7.27
201711291028_23446#3_1HFID.100.esp CHLOROFORM-d
0.55
0.45 TMS 0.00
0.40 0.35
0.05
1.70 1.69 1.67 1.66 1.66 1.26 1.65 1.43
2.40 2.40 2.35 2.34
0.10
5.19 5.19 5.16 5.14 4.83
0.15
4.48 4.46
0.20
3.97 3.94 3.92 3.83 3.79
3.96
0.25
2.36
0.30
7.19 7.16 6.95 6.90 6.71
Normalized Intensity
0.50
0 0.68 0.07 0.03 7.0
6.5
1.96 0.96 6.0
5.5
5.0
0.89 0.26 2.90 0.76 4.5
4.0 3.5 3.0 Chemical Shift (ppm)
4.37 2.5
4.00 2.0
1.5
1.0
0.5
0
The structure is the same appointed for previous sample. The unsaturated content determined is ~17%. The calculation and other considerations follow the procedure explained for previous sample). 4c) SAMPLE 5% NMR attribution for Poly-(isosorbide adipate-co-isosorbide fumarate) 1H RMN (300MHz, CDCl3, δ, ppm): 1.24-1.28 (m, from hydroxyl end groups); 1.65-1.70 (m, 4H, C9); 2.34-2.40 (m, 4H, C8, C13, C16); 3.56-3.61 (m, hydroxyl from C14); 3.81-3.83 (m, 1H, C6); 3.92-3.97 (m, 2H, C1); 4.02 (methyl and ethyl end groups); 4.31-4.33 (dd, C14, C17); 4.46 (d, 1H, C3); 4.83 (t, 1H, C4); 5.14-5.20 (m, 2H, C2 and C5); 7.16-7.19 (d, endo/exo C11).
7.27
sample 5%.esp CHLOROFORM-d
1.0 0.9
TMS 0.00
0.7 0.6 0.5
2.36 2.35 2.34 1.70 1.69 1.66 1.67 1.65 1.64 1.43 1.28
2.40 2.40
4.02
3.94 3.92 3.83 3.81 3.61 3.56
0.1
4.48 4.33 4.46
0.2
4.62
5.20 5.16 5.14 4.85 4.83
0.3
3.97 3.96
0.4
7.19 7.16
Normalized Intensity
0.8
0 0.20 7.0
1.75 0.80 0.97 6.5
6.0
5.5
5.0
4.5
2.64 0.83 4.0 3.5 3.0 Chemical Shift (ppm)
4.28 2.5
4.00 0.25 2.0
1.5
1.0
0.5
0
-0.5
The structure is the same appointed for previous samples. The unsaturated content determined is ~5% . The calculation and other considerations follow the procedure explained for other samples.
Obs: All NMR data were recorded in a Bruker 300MHz equipment. The data processing was made in ACD Lab/NMR Processor Academic Edition software.
cDLP figure ilustrating the printing process and photocuring.
S2405-8866(19)30021-1
DOI:
https://doi.org/10.1016/j.bprint.2019.e00062
Reference:
BPRINT 62
To appear in:
Bioprinting
Received Date: 5 April 2019 Accepted Date: 19 September 2019
Please cite this article as: J. Lammel-Lindemann, I.A. Dourado, J. Shanklin, C.A. Rodriguez, L.H. Catalani, D. Dean, Photocrosslinking-based 3D printing of unsaturated polyesters from isosorbide: A new material for resorbable medical devices, Bioprinting (2019), doi: https://doi.org/10.1016/ j.bprint.2019.e00062. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Photocrosslinking-Based 3D printing of Unsaturated Polyesters from Isosorbide: A New Material for Resorbable Medical Devices Jan Lammel-Lindemann1,2,3, Isabela Autran Dourado4, Johnathan Shanklin1, Ciro A. Rodriguez1,2,3, Luiz Henrique Catalani4 and David Dean1,* 1
Department of Plastic and Reconstructive Surgery, The Ohio State University, Columbus, OH 43210, United States. 2 Tecnologico de Monterrey, Escuela de Ingenieria y Ciencias, Monterrey, N.L. 64849, Mexico. 3 Laboratorio Nacional de Manufactura Aditiva y Digital (MADIT), Apodaca, N.L. 66629, Mexico. 4 Departamento de Química Fundamental, Instituto de Química, Universidade de São Paulo, São Paulo. 26077, Brazil.
*Corresponding Author: 460 West 12th Avenue, Rm. 388, Columbus, OH 43220 USA; E-mail: [email protected]; website: www.OsteoEngineering.com Abstract Words: 253 Manuscript without Abstract Words: 3192 Figures: 8 Tables: 0 Keywords: bone, tissue engineering, 3D printing (additive manufacturing), UV curing, digital light processing (DLP), porous scaffold
Abstract Clinically used tissue-engineered devices are few and far between. The search for biocompatible materials has been a great focus for the field for the past 30 years. The ones that exist are hard to work with when manufacturing patient-specific shapes with complex external and internal pore geometries because of their material and mechanical properties. Photo-cross-linkable polymers have arisen as a potential 3D printable solution for this problem. Here we present a new sugar-based polymer, isosorbide-derived polyester as a potential candidate to help fill this gap. Isosorbide-derived polyesters containing different amount of double bonds were synthesized to allow the material to be crosslinked or functionalized. Three resin formulations with different amounts of double bonds available (5, 12 and 17%) were used. After initial thermal and photo-crosslinking studies, Isosorbide-derived unsaturated polyesters were then mixed with a photoinitiator, co-crosslinker, and solvent to form a 3D printable resin formulation. Thin films of that resin were then cured using a Digital Light Projection (DLP) printer and analyzed. An attempt was made to modulate the cured thickness by adding HMB as a photo-attenuator. We assessed 3D printing success and tensile green strength for the 3 resin formulations. We were able to 3D print unsaturated polyesters from isosorbide with a DLP printer when enough double bonds were present to facilitate crosslinking and curing of the material. Finally, thin films were then seeded with fibroblasts following the ISO 10993-5 protocol. These coupons were found safe (i.e., non-cytotoxic and biocompatible), potentially marking the beginning of a new approach to preparing resorbable implants for tissue engineering uses. .
1. Introduction Despite at least 30 years of research, there are very few tissue-engineered devices used clinically. To this date, there is no “synthetic bone substitute”, and this is evident by the need for allografting bone from the patient’s own body to do major surgical reconstructions (1). Current implants tend to be strong metals that generate stress shielding and possibly resorption of the adjacent bone and preventing adaptive remodeling (2). The balance between providing structural support while minimizing stress shielding leaves a very tight window that material science is trying to fill. One of the main reasons for the search for new implant manufacturing strategies is the limited number of biocompatible resorbable materials, especially polymers, which can be used safely to prepare hydrogel and/or solid-cured scaffolds. Very few of these materials can be formed by 3D printing into porous, patient-specific shapes. In this study we present a novel, patent-pending material, isosorbide-derived polyester (3) which shows promise as a scaffold substrate bone tissue engineering, and likely regenerative uses for many other tissues as well. In this study, we show that this sugar molecule, isosorbide, can be prevented from degrading immediately in solution through light-induced crosslinking. Unlike other potential bone scaffolding materials have shown high compressive strength but are brittle (e.g., poly(propylene fumarate)) (4)., However crosslinked isosorbide-derived polyesters are expected to have high tensile strength. Bone regeneration and remodeling physiology require that resorbable devices ideally degrade in during a time span of between 3-6 months. This allows the bone to fully heals without obstructing this process (5). This is necessary because the initial bone healing response, resulting in a disorganized callus, needs to be remodeled by osteoclastic activity to become strong (6). Remodeling cannot occur if it encounters un-resorbed material. Unresorbed material would form
a barrier to this process (7,8). Producing devices for the regeneration of bone and other tissue types (e.g., skin, lung, liver, heart, muscle, etc.) will also different resorption rates in ranges that suit the tissue’s own regeneration (healing) and remodeling process (9–11). Polycondensation of dianhydrohexitols (isosorbide, isommanide, and isoidide) and saturated aliphatic diacids (adipic, gluconic and succinic) is a clever way to synthesize bioabsorbable polyesters. These polymers have unique properties; they are particularly suitable for applications in the medical device field since their degradation products are readily and locally metabolized. Additionally, the applications of these polymers may include surgical fixation devices for bone, tendon, and ligament reconstruction.
Isosorbide or 1,4:3,6-dianhydro-D-glucitol is a bioderived and biodegradable dianhydrohexitol compound qualified by US Food and Drug Administration as “generally recognized as safe” (GRAS). If derived into unsaturated polyester, then this sugar-based polymer with high tensile strength and flexibility is able to be photocured. This bicyclic diol shows thermal and chemical stabilities, it is non-toxic, biodegradable and can be used in copolymerization reactions (12).
Isosorbide can be obtained stereoselectively starting from starch. Enzymatic starch hydrolysis leads to D-glucose production. D-glucose hydrogenation process followed by sorbitol dehydration leads to isosorbide formation (13). Regarding its stereochemistry, isosorbide has a stereochemistry in wedge shape, composed of two dihydrofuranoid rings. There is a hydroxyl group in C2 in Exo configuration, while the hydroxyl in C5 is in endo configuration. Literature dating from the 1960s shows that when in Endo positions, hydroxyls are less reactive than those in the Exo position (14). Polycondensation
of dianhydrohexitols (isosorbide) and saturated or unsaturated aliphatic diacids is an interesting way to synthesize bioabsorbable polyesters (15,16).
Figure 1: Isosorbide structure
The exploration of isosorbide polyesters and the search for new materials was motivated by a desire to go beyond the use of petroleum-based materials. However, it was noticed that the applications of these isosorbide polymers can be diverse and interesting due to their biocompatible characteristics. Juais and co-workers showed the enzymatic production of isosorbide saturated polyesters (15). In that study, various synthesis conditions - such as solvents, temperatures, reaction times and monomers - were studied, so it was possible to produce isosorbide saturated polyesters with molar mass around 40,000 Da. In another study, Naves and co-workers showed the production of unsaturated polyesters of isosorbide and isomanide (one of the stereoisomers of isosorbide), using optimum conditions of the previous work. Copolymers in the order of 4,000-16,000 Da were produced with unsaturated monomers (16). As well as noted in Naves’ work, these double bonds provide the material the possibility of crosslinking or functionalization. In this article, it has been shown that unsaturated isosorbide polymers can be very promising in terms of biomedical applications because they can be 3D printed, giving biocompatible and biodegradable scaffolds.
In this work, photocurable polyesters were synthesized and printed. The C=C bonds form after light exposure, specific initiator starts radical reactions between polymer and crosslinkers, so that the initially liquid material becomes a solid film upon photocrosslinking. In addition to being biocompatible and biodegradable, these polyesters are photocurable. Since they are polyesters, it is known that they can easily undergo chemical hydrolysis inside the body (17,18). In terms of regenerative medicine, there are two types of typical cleavage in the biodegradation process: hydrolytic and enzymatic. In the human body, composed mostly of water, hydrolytic cleavages are strongly important and ester bonding is one of the most common that undergo hydrolysis. The hydrolysis of the ester bonds in the polymer chain leads to small molecules that can be excreted from the body (17). As regards to enzymatic cleavages, it is possible to affirm that it guarantees better control of biodegradation, since, by introducing them with correct peptide groups, one can control exactly which regions will be cleaved by specific enzymes (17). Therefore, the isosorbide polymers which are the subject of this work are biomedically interesting materials since their degradation byproducts are readily metabolized. The ability to photo-cure this material makes 3-D printing an option for our resin formulations. Development of additive manufacturing methods for this material may be able to fill a need in the medical implant industry by enabling specific parts to be manufactured for bone scaffolding, surgical fixation devices for bone, tendon, ligament reconstruction, vascular graft scaffold, surgical sutures, and carriers for controlled drug release. Using Digital Light Processing (DLP), we will project a mask that is computer-generated by slicing the desired object to print into 50 µm layers in software. The depth of these slices is varied to determine the best elevator movement in the Z direction. X and Y for 3D printing. Laser projection is controlled by a micromirror chip that reflects the UV light to the clear vase of the resin vat, where it cures the
resin crosslinking it. There’s some over penetration, that allows each new layer to attach to the previous one in what we call interlayer “stitching”. After the new layer has been stitched to the previously layer, the elevator will move again and the next mask will be projected. Most products used for these purposes currently are made from non-resorbable materials such as ceramics, polymers, and metals. Our isosorbide material would be able to fill a much-needed void for bio-implantable materials that are degradable, softer, more flexible and able to be used for manufacturing via 3D printing.
2. Materials and Methods 2.1 Isosorbide Purification The isosorbide was recrystallized twice from ethyl acetate, using an external oil bath at 50°C, and dried under vacuum in the presence of phosphorus pentoxide until reaching constant mass. Their final purity was determined by Gas Chromatography-Mass Spectrometry (GC-MS QP2020 – Shimadzu with the column BPX5 ((5%-Phenyl)-methylpolysiloxane) - 30m.) Before use, all solvents used (Merck) were dried and distilled. Isosorbide (Sigma-Aldrich) was recrystallized twice from ethyl acetate, dried under vacuum for 72 hours and weighed in a Glove Box under N2 atmosphere (it is a very hygroscopic compound). Adipic diethyl and Fumaric acid (Sigma-Aldrich) were used as received. CAL-B (Novozyme; specific activity 10,500 PLU/g) is the Lipase B from Candida antarctica (NZ 435) immobilized in microporous resin Lewatit 1600. Hydroquinone (BHerzog) was used as received.
2.2 Synthesis of diethyl fumarate 23.83 g of fumaric acid was added in a round-bottomed flask with 120 mL of anhydrous ethanol. The mixture was heated under reflux over 15 minutes. Then, 5 mL of sulfuric acid was added dropwise, followed by 100 mL of toluene and 15 minutes of heating under reflux. Another 15 mL of sulfuric acid was added dropwise gently and the mixture was heated under reflux during more 30 minutes. The system was cooled at room temperature and neutralized with NaHCO3 saturated solution. Organic phases were extracted and the aqueous phases were washed with toluene. After combining organic phases, the mixture was dried with Na2SO4 over 24 hours. The diethyl fumarate was concentrated by rotary evaporation and then purified by distillation under reduced pressure. 2.3 Synthesis of unsaturated polyesters Polycondensation reactions of isosorbide with diethyl adipate and fractions of diethyl fumarate were carried out according to previous group development (16) from the following copolymerization reaction:
Figure 2: Schematic representation of co-polymerization reaction producing poly(isosorbide adipate-co-isosorbide fumarate).
To synthesize the unsaturated isosorbide polymer, isosorbide (IS), diethyl adipate (AD) and diethyl fumarate (FU) - molar ration IS:AD:FU (10:9.5:0.5) - were added into a flask containing cyclohexane:toluene (6:1) mixture. CAL-B (10% relative to total monomer mass) was employed as the catalyst and hydroquinone (1% w/w) was used to avoid radical addition reactions in double bonds during the copolycondensation. To produce 30 g of polymer, the amount of reactants were calculated in relation to 70 mmol of isosorbide and 270 ml of the same solvent mixture. The flask was attached to a Dean Stark apparatus that contained 4A molecular sieves for byproduct removal. The Dean Stark was attached to a condenser. The molecular sieves were replaced with fresh dried molecular sieves every 24 h and the solvent level was kept constant. The reaction was magnetically stirred and heated at 90°C sing an external oil bath. After 7 days of reaction, the solvent was removed, while chloroform was added and the enzyme was filtered. The product was concentrated, by rotary evaporation, removing chloroform. Three resin formulations were obtained this way, corresponding to different double bonds inserted 5% (A), 12% (B), and 17% (C). The unsaturation content in polyester chains was evaluated by 1H NMR by relation integral values of methylene groups from adipate and alkenyl groups from fumarate. 2.4 Initial Thermal and photocrosslinking studies Isosorbide polymer was mixed with chloroform as a solvent and stirred overnight. The polymer was mixed with a resin made from 3% BASF Irgacure 819 (BAPO) and 35% (w/w) crosslinker (1-vinylimidazole, N-vinylpyrrolidone or N-isopropylacrylamide). All solutions were mixed with a vortex, its contents were transferred to cylindric (2.5 cm of diameter and 1.5 cm of height) Teflon molds, which were subjected to photocrosslinking by UV light. Below, it is possible to see a schematization of the process.
Figure 3: Schematic representation of photocrosslinking tests The material was then tested at 30, 45 and 60 minutes in standard power UV light to induce crosslinking. It is worth noting that the UV light (LuzChem®) operates with radiance near to 4.6mW/cm² in the peak of wavelength 254 nm.
2.5 Photocurable Resin Formulation Three resins of varying molecular weights were used with 5% (A), 12% (B), and 17% (C) double bonds inserted respectively, using 58% w/w of the resin by weight. VIM (1vinylimidazole), (Sigma-Aldrich, MO) was used as co-crosslinker at a 32% w/w; as well as 3% Irgacure 819 (BAPO) (BASF, NJ) as photoinitiator and dissolved in Ethyl Acetate as the solvent at 7% w/w. Isosorbide polymer was mixed with the ethyl acetate/photoinitiator pack and stirred overnight in a sealed container.
2.6 Thin Film Curing and 3D Printing Methods Both thin film curing and 3D printing of these resins using a cDLP 3D printer are performed according to the work of Hernan Lara Padilla published in his doctoral thesis using an EnvisonTEC® µicro® printer (Dearborn, MI). That method allows the measurement of tensile
green strength. This device utilizes 405 nm light and a 450 mW/cm² intensity measured with a RM 12 radiometer with a VIS-B sensor (Dr. Grobel, Ettlingen, GER), much different than the initial photo-crosslinking light source in the Catalani lab. Thin-film measurements were done the same way as we have characterized other printable resins with as explained below. Microscope slides were measured previously and recorded. About 7 drops of resins were set on top of the base slide and then cured in the printer for the assigned time using 1.0 x 1.0 cm projected squares. Two separate time points were investigated for these thin film cure tests: 60 and 150 seconds. Once the curing was finished, the slide was gently wiped to take uncured material off, the top cover slide was placed over the cured part and this sandwich was measured with a caliper (Mitutoyo Absolute Digimatic, Mitutoyo, IL). After subtracting the initial slides thicknesses, we get the cure depth for the thin films. A total of 8 samples per time point were recorded.
2.7 Film Thickness Control
HMB (Oxybenzophenone) (Sigma, St. Louis, MO) was added to act as well as a light attenuator (reducing thin film thickness). Cure tests were repeated at 30, 60, 90 and 120 seconds, measured utilizing the same method as described above for the thin film curing. This was tested only with resin Formulation A.
2.8 Mechanical Testing In our experience resin formulations that exhibit thin films thickness of 150-250 µm can be 3D printed without an attenuator under our conditions in the EnvisonTEC® µicro® printer;
however, some resins were well above that thickness. The green strength of these 3D printed resin mixtures was investigated by using a load sensor that holds the build platform (after Lara Padilla, H, 2018, unpublished and patent-pending doctoral thesis) (19). A very small build plate was first coated with a “burn-in” patch of isosorbide. That burn-in patch acts as an anchor and support for the next 20 masks (layers) of 50 microns each to create a 1 mm diameter post. Then 3 consecutive layers were projected with increasing diameters of up to 8 mm to test the cured resin’s tensile strength. This also shows that the resin can be properly 3D printed by joining the layers during the cDLP 3D printing process (19). The tensile green strength sensor set up and the print visualization is shown below.
Figure 4: Schematic of the separation force and green strength in an EnvisionTEC (Dearborn, MI) µicro 3D printer. The load sensor and montage can be seen in (a). The build platform as it’s getting printed can be seen in (b). The build job schematic can be seen in (c), where after a burnin plate consisting of a wide base curing with 90 s for 3 layers and then 2 more with 60 s for
anchorage, followed by 20 layers of a 1 mm diameter are printed at a 60 s exposure time, and then followed by increasing diameter layers that become greater after every 3 layers printed in a matter of 0.5 mm as well at 60 s exposure time. The separation force is being tracked in real-time until the part fails by breaking from the 1 mm post when the separation force surpasses the material’s mechanical properties. 2.9 Cytotoxicity Per the ISO Standard 10993-5, 8 samples of 10 mm2 thin films were printed (4 for resin B and 4 for resin C, employing the same samples obtained from the printability tests transferred to PBS and left soaking for 5 minutes. This step was repeated 3 times. Then they were autoclaved in the liquid cycle for 30 minutes at a temperature of 121ºC. A total of 25,000 murine L929 American Type Culture Collection (ATCC, Rockville, MD, USA) fibroblasts were counted manually and seeded on an Isosorbide thin film. The seeded thin films were then placed in a low attachment well for culture in Dulbecco's Modified Eagle's Medium (DMEM; Gibco, Walthman, MA, USA) supplemented with 10% horse serum, 1% L-glutamine, 1% sodium pyruvate, 50 U/mL penicillin and 50 ug/mL streptomycin (HyClone, Logan, UT, USA). Samples were kept at 37 ºC and 5% CO2 atmosphere for three days. After 72 hours, cell proliferation was recorded using the PrestoBlue® (Invitrogen, MA, USA) assay and the number of cells present was calculated using a standard curve. Control groups included negative controls, 4 wells grown with complete DMEM for 72 hours as positive controls, and 4 samples per resin formulation. Effluent samples were collected in triplicate and measured as per the assay's instructions. Results were reported as average total cells per well. 2.10 Statistical Analysis
Quantitative data is shown as average ± standard deviation. Statistical analysis was performed by using GraphPad® Prism® 7.o, using two-way ANOVA with Turkey’s multiple comparisons for figure 5, ordinary one-way ANOVA with multiple comparisons for figure 6. Statistically significant values were defined as a p-value of <0.05. Cure tests were done with an n=8, cytotoxicity with an n=4. Legends are as follows ns= non-significant, *= p<0.05, **= <0.01, ***=<0.001 and ****=<0.0001.
3. Results Thin films were cured to measure the cure depth of the different resins without a light attenuator. We found that resins B and C were fully photo-curing at both exposure times. Resin A
ns ns
C 2
B 2
**
A 2
Cure Thickness ( m)
formulation did show a difference, but the photocrosslinking was found to be insufficient as the
parts were not fully cured this was then assessed by observing the resin A thin films immediately dissolve in phosphate-buffered saline solution. These results can be seen in figure 5.
Figure 5: Graph showing the cured thickness of resins A, B, and C. (5%, 12%, and 17% double bonds respectively) without photoattenuator (HMB). Image showcasing a cured film after being measured for illustration purposes. After noting that the droplets were fully cured, the decision to add HMB as a photoattenuator to control the cure depth for every layer was made. In previous work from our laboratory with other printable polymers, this was found as a key element in controlling the resolution for the printed parts and can aid in developing an optimized 3D printing process. Resin A was used for the graph shown in figure 6.
Figure 6: Influence of photoattenuator (HMB) on resin formulation type A (5% double bonds). Knowing that the resin formulations exhibit appropriate curing thickness, the mechanical properties of the materials were tested. While no light attenuator was used here, resins B and C were able to produce a 3D printed piece, that correlated quite well with the double bonds available to the resins. It was clear that increasing the percentage of double bonds improved the mechanical properties. Resin A did complete a burn-in patch at 90s, as witnessed by the force spikes seen in the first part of the curve (burn-in patch at 90 seconds), but the part completely failed to print at this point once the layers were printed at 60 seconds per layer. It is important to note that resin B started to fail during the print, as seen in both the graph showing positive force and the printed part with poor resolution. Resin C printed fully and the part was recovered as shown in panel C2 in figure 7. Resin B shows a green tensile strength of 123.79 Pa although the geometry was already partially failing while resin C formulation showed a tensile strength of
108.59 Pa with much improved geometric accuracy. The recovered parts and their respective force plots can be seen in figure 7.
Figure 7: Force plots taken during the cDLP printing of the resins and parts recovered. A1 corresponds to the resin A, and the part failed to print after the burn-in plate. B2 corresponds to resin B, and the part recovered is shown in B2. C1 corresponds to resin C, where the full part recovered is shown in C2 below the force graph. The force plots for all the resins show that there’s an increased separation negative force at the beginning where the anchoring and support burn-in masks were constructed, corresponding to the force it takes to separate the printed layer from the basement where the mask was projected. These were done in 3 masks of 90 seconds and 2 masks of 60 seconds each. Resin A failed to 3D print once the 1 mm diameter masks were projected. The force plot remains stable after this as
there’s no part being printed that exerts any force. For resins B and C, the higher amount of double bonds available correlates with better attachment (stitching) from one mask projection (i.e., 3D printing one layer by cDLP) to the next. This translates into a 3D printed part. The negative forces can be seen increasing with the increasing diameter of the masks in sets of 3, as expected. The positive spikes in the graphs can be explained as the force recorded once the build platform elevates enough to allow for the resin to flow back in and then compresses this uncured fluid. After the parts are printed, the plot returns to baseline as no more masks are being projected. The thin films from the cure depth studies were then recovered by scraping them off the microscope slides that held them., then washed 3 times in PBS. These were autoclaved and kept sterile until the seeding with murine fibroblast cells in low attachment well plates. The viability assay was then run on days 0 and 3 and graphed against a standard curve showing viability at day 3. The drop in cell count from day 0 to day 3 can be accounted by the low attachment well plates where cells that didn’t attach to the isosorbide proceeded to die, but the ones that attached to the films survived. The graph regarding this is shown in figure 8 below.
Figure 8: Cytotoxicity of the printed thin films was tested by presto blue. Cells were shown to survive after being seeded on the thin films after 3 days.
Discussion & Conclusions As we search for resorbable polymers that can be used for bone tissue engineering and regeneraion of other tissues, the field of biomanufacturing has been mostly dominated by polymers with rapid resorption times, such as polylactides (within 3 months) or slow resorption rates (e.g., polycaprolactones) which can take years if they resorb at all (20). Isosorbide provides a material that has appropriate green tensile strength, and this characteristic can aid in providing new uses for the material as scaffolding and other medical devices to be implanted. Isosorbide polyesters can designed to be resorbed in vivo in a period of 3 to 9 months and can fulfill the
need for a biomedical-grade, resorbable, 3D printable polymer for use in FDA-approved medical devices. It is important to note that we mention 3 month minimum as this is a critical span biologically for the scaffold to serve as a guide the infusion of new tissue through a 3D printed pore geometry. It is also expected to facilitate remodeling, as it is expected to degrade in the key window of 3-9 months. We report here the first time that this material has been 3D printed to the best of our knowledge. For that purpose, we used continuous Digital Light Processing (cDLP) projection of UV light. 3D printing, also referred to as additive manufacturing, opens the possibility of controlling both mechanical properties and geometry. The addition of isosorbide as a photocrosslinkable, resorbable, and biocompatible material will be valuable if it can be demonstrated to be 3D printed with a high level of accuracy while being biocompatible and, hypothetically, resorbable in the window expected of a tissue engineering scaffold.
Acknowledgments The authors are grateful for partial support received from a Global Gateway Mobility Grant, #2015/5041-0, from the FAPESP (Sao Paulo, Brazil) Foundation and the Ohio State University. Glossary
DLP: Digital light processing
BAPO: Irgacure 819 HMB: Oxybenzophenone DEF: Diethyl Fumarate EA: Ethyl Acetate
Declaration of interest
Drs. Dean and Catalani have submitted a patent application on the use of isosorbide for tissue engineering applications as presented in this manuscript.
References 1. Kinaci A, Neuhaus V, Ring DC. Trends in bone graft use in the United States. Orthopedics. 2014 Sep;37(9):e783-788. 2. Huiskes R, Weinans H, van Rietbergen B. The relationship between stress shielding and bone resorption around total hip stems and the effects of flexible materials. Clin Orthop Relat Res. 1992 Jan;(274):124–34. 3. David Dean, Luiz Henrique Catalani. Polyesters, Polymer Compositions, and Methods of Using Thereof. US Provisional Patent Application # 62/750,665., 2018. 4. Wilson JA, Luong D, Kleinfehn AP, Sallam S, Wesdemiotis C, Becker ML. Magnesium Catalyzed Polymerization of End Functionalized Poly(propylene maleate) and Poly(propylene fumarate) for 3D Printing of Bioactive Scaffolds. J Am Chem Soc. 2018 10;140(1):277–84. 5. DeLacure MD. Physiology of bone healing and bone grafts. Otolaryngol Clin North Am. 1994 Oct;27(5):859–74. 6. Frost HM. From Wolff’s law to the Utah paradigm: Insights about bone physiology and its clinical applications. The Anatomical Record. 2001;262(4):398–419. 7. Wang MO, Vorwald CE, Dreher ML, Mott EJ, Cheng M-H, Cinar A, et al. Evaluating 3Dprinted biomaterials as scaffolds for vascularized bone tissue engineering. Adv Mater Weinheim. 2015 Jan 7;27(1):138–44. 8. Willie BM, Petersen A, Schmidt-Bleek K, Cipitria A, Mehta M, Strube P, et al. Designing biomimetic scaffolds for bone regeneration: why aim for a copy of mature tissue properties if nature uses a different approach? Soft Matter. 2010 Oct 5;6(20):4976–87. 9. Atala A. Tissue Engineering and Regenerative Medicine: Concepts for Clinical Application. Rejuvenation Research. 2004 May 1;7(1):15–31. 10. Tur K. Biomaterials and Tissue Engineering for Regenerative Repair of Articular Cartilage Defects. Arch Rheumatol. 2009;24(4):206–17. 11. Materials for Biomedical Engineering: Absorbable Polymers - 1st Edition [Internet]. [cited 2019 Aug 2]. Available from: https://www.elsevier.com/books/materials-for-biomedicalengineering-absorbable-polymers/grumezescu/978-0-12-818415-8 12. Studies in Diuresis with Isomannide. Sugar Alcohols XVIII - John C. Krantz, C. Jelleff Carr, 1938 [Internet]. [cited 2019 Apr 1]. Available from: https://journals.sagepub.com/doi/abs/10.3181/00379727-39-10279 13. Beldi M, Medimagh R, Chatti S, Marque S, Prim D, Loupy A, et al. Characterization of cyclic and non-cyclic poly-(ether-urethane)s bio-based sugar diols by a combination of MALDI-TOF and NMR. European Polymer Journal. 2007 Aug 1;43(8):3415–33.
14. Hopton FJ, Thomas GHS. Conformations of some dianhydrohexitols. Can J Chem. 1969 Jul 1;47(13):2395–401. 15. Juais D, Naves AF, Li C, Gross RA, Catalani LH. Isosorbide Polyesters from Enzymatic Catalysis. Macromolecules. 2010 Dec 28;43(24):10315–9. 16. Naves AF, Fernandes HTC, Immich APS, Catalani LH. Enzymatic syntheses of unsaturated polyesters based on isosorbide and isomannide. Journal of Polymer Science Part A: Polymer Chemistry. 2013;51(18):3881–91. 17. Mondschein RJ, Kanitkar A, Williams CB, Verbridge SS, Long TE. Polymer structureproperty requirements for stereolithographic 3D printing of soft tissue engineering scaffolds. Biomaterials. 2017 Sep;140:170–88. 18. Gonçalves FAMM, Fonseca AC, Domingos M, Gloria A, Serra AC, Coelho JFJ. The potential of unsaturated polyesters in biomedicine and tissue engineering: Synthesis, structure-properties relationships and additive manufacturing. Progress in Polymer Science. 2017 May 1;68:1–34. 19. Hernan Lara-Padilla, Ciro A Rodriguez, David Dean. Fabrication of porous scaffolds using additive manufacturing with potential applications in bone tissue engineering. US Provisional Patent Application #62/715596, 2018. 20. Fisher JP, Holland TA, Dean D, Mikos AG. Photoinitiated cross-linking of the biodegradable polyester poly(propylene fumarate). Part II. In vitro degradation. Biomacromolecules. 2003 Oct;4(5):1335–42. 21. Hopton; F. J.Thomas GHS. Conformations of some dianhydrohexitols. Canadian Journal of Chemistry. 1969;47:2395–401. 22. Juais D, Naves AF, Li C, Gross RA, Catalani LH. Isosorbide polyesters from enzymatic catalysis. Macromolecules. 2010;43(24):10315–9.
Supplementary data: NMR spectra data of monomers and polymers (1) Isosorbide (1,4:3,6-dianhydro-D-glucitol)
1
H NMR (300MHz, D2O, δ, ppm): 4.62 (t, 1H, C4); 4,45 (d, 1H, C3); 4.4-4.33 (m, 1H, C2);
4.29-4.31 (dt, 1H, C5); 3.82-3.93 (m, 3H, C1’, C1” and C6”); 3.43-3.49 (m, 1H, C6’).
isosorbide.esp Water
DEUTERIUM OXIDE 1.0 0.9
Normalized Intensity
0.8 0.7 0.6 0.5
0.1
3.82
3.93 3.92
0.2
4.47 4.45 4.40 4.37 4.35 4.33 4.31 4.29
4.63 4.62 4.60
0.3
3.49 3.46 3.43
3.90 3.86 3.85
0.4
0 0.98 4.7
4.6
0.98 4.5
1.00 0.95 4.4
4.3
2.98 4.2
4.1 4.0 3.9 3.8 Chemical Shift (ppm)
1.00 3.7
3.6
3.5
3.4
3.3
3.2
It is worth to note that we have two diastereotopic protons in C1 and two in C6 – represented by C1’/C1” and C6/ C6’- related to hydrogens that are above or below the bicyclic ring.(21)
(2) Diethyl adipate:
1
H NMR (300MHz, CDCl3, δ, ppm): 1.26 (t, 6H, C1); 1.64-1.72 (m, 4H, C4); 2.30-2.34 (m, 4H,
C3); 4.13 (q, 4H, C2).
1.26
Diethyl Adipate.esp 1.0 0.9 0.8
1.28 1.23
0.6 0.5 4.14 4.12
0.4
1.66
Normalized Intensity
0.7
CHLOROFORM-d 0.1
1.72 1.64
0.2
2.34 2.32 2.30
4.16 4.09
0.3
3.99
4.46
TMS
0 4.00 7.0
6.5
6.0
5.5
5.0
4.5
4.0 3.5 3.0 Chemical Shift (ppm)
2.5
2.0
1.5
6.06 1.0
0.5
0
(3) Diethyl fumarate
1
H NMR (300MHz, CDCl3, δ, ppm): 1.32 (t, 6H, C1); 4.26 (q, 4H, C2); 6.85 (s, 2H, C4).
1.32
Diethyl Fumarate.esp CHLOROFORM-d 6.85
1.0 0.9 0.8
4.27 4.25
1.35 1.30
0.6 0.5 0.4 0.3 4.30 4.23
Normalized Intensity
0.7
0.2
TMS 0.1 0 1.88 7.0
3.95 6.5
6.0
5.5
5.0
4.5
4.0 3.5 3.0 Chemical Shift (ppm)
6.00 2.5
2.0
1.5
1.0
0.5
0
(4) Poly(isosorbide adipate-co-isosorbide fumarate) 4a) SAMPLE 12%
1
H RMN (300MHz, CDCl3, δ, ppm): 1.24-1.28 (m, from hydroxyl end groups); 1.61-1.68 (m,
4H, C9); 2.33-2.41 (m, 4H, C8, C13, C16); 3.53 (m, hydroxyl from C14); 3.78-3.83 (m, 1H, C6);
3.91-3.97 (m, 2H, C1); 4.00-4.02 (methyl and ethyl end groups); 4.12-4.14 (dd, C14, C17); 4.464.48 (d, 1H, C3); 4.83 (t, 1H, C4); 5.13-5.20 (m, 2H, C2 and C5); 7.16 e 7.19 (m, endo/exo C11).
7.27 7.27
201711291031_23446#2_1HFID.100.esp CHLOROFORM-d
0.7
TMS 0.01 2.36
0.5
1.68 1.26
1.70 1.69 1.67 1.65 1.62
2.41 2.40 2.35 2.35
3.97 3.94 3.92
4.83
3.83 3.81 3.80
0.1
4.82
5.20 5.16 5.13 5.15
0.2
4.48 4.47
0.3
0.08
3.96
0.4
7.26 7.24 7.19 7.17
Normalized Intensity
0.6
0 0.56 7.5
7.0
1.40 0.67 0.690.13 2.13 0.67 6.5
6.0
5.5
5.0
4.5
4.0 3.5 3.0 Chemical Shift (ppm)
3.35 2.5
4.00 2.0
1.5
1.0
0.5
0
Q: How is the percentage of double bonds determined? A: By relating protons integral values of C9 (methylene groups) and C11 (alkenyl groups). The unsaturation content is always in relation to saturated portion of the polymer. If we have 4 as proton integral value for C9 and 0,56 for C11, we have ~12% of fumarate residue (unsaturated monomer) inserted – in relation to adipate (saturated monomer). It is worth noting that we have, in NMR, an indication of Michael’s Addition in some double bonds, which is represented by some small peaks in 4.12-4.14 ppm (dd for C14 and C17). It’s
also represented in the structure purposed. The unsaturation ratio calculated take these side reactions in account.(22) 4b) SAMPLE 17% NMR attribution for Poly-(isosorbide adipate-co-isosorbide fumarate) 1H RMN (300MHz, CDCl3, δ, ppm): 1.24-1.28 (m, from hydroxyl end groups); 1.61-1.68 (m, 4H, C9); 2.34-2.40 (m, 4H, C8, C13, C16); 3.51-3.53 (m, hydroxyl from C14); 3.79-3.83 (m, 1H, C6); 3.92-3.97 (m, 2H, C1); 4.00-4.02 (methyl and ethyl end groups); 4.12-4.14 (dd, C14, C17); 4.47 (d, 1H, C3); 4.83 (t, 1H, C4); 5.14-5.20 (m, 2H, C2 and C5); 7.16-7.19 (d, endo/exo C11).
7.27
201711291028_23446#3_1HFID.100.esp CHLOROFORM-d
0.55
0.45 TMS 0.00
0.40 0.35
0.05
1.70 1.69 1.67 1.66 1.66 1.26 1.65 1.43
2.40 2.40 2.35 2.34
0.10
5.19 5.19 5.16 5.14 4.83
0.15
4.48 4.46
0.20
3.97 3.94 3.92 3.83 3.79
3.96
0.25
2.36
0.30
7.19 7.16 6.95 6.90 6.71
Normalized Intensity
0.50
0 0.68 0.07 0.03 7.0
6.5
1.96 0.96 6.0
5.5
5.0
0.89 0.26 2.90 0.76 4.5
4.0 3.5 3.0 Chemical Shift (ppm)
4.37 2.5
4.00 2.0
1.5
1.0
0.5
0
The structure is the same appointed for previous sample. The unsaturated content determined is ~17%. The calculation and other considerations follow the procedure explained for previous sample). 4c) SAMPLE 5% NMR attribution for Poly-(isosorbide adipate-co-isosorbide fumarate) 1H RMN (300MHz, CDCl3, δ, ppm): 1.24-1.28 (m, from hydroxyl end groups); 1.65-1.70 (m, 4H, C9); 2.34-2.40 (m, 4H, C8, C13, C16); 3.56-3.61 (m, hydroxyl from C14); 3.81-3.83 (m, 1H, C6); 3.92-3.97 (m, 2H, C1); 4.02 (methyl and ethyl end groups); 4.31-4.33 (dd, C14, C17); 4.46 (d, 1H, C3); 4.83 (t, 1H, C4); 5.14-5.20 (m, 2H, C2 and C5); 7.16-7.19 (d, endo/exo C11).
7.27
sample 5%.esp CHLOROFORM-d
1.0 0.9
TMS 0.00
0.7 0.6 0.5
2.36 2.35 2.34 1.70 1.69 1.66 1.67 1.65 1.64 1.43 1.28
2.40 2.40
4.02
3.94 3.92 3.83 3.81 3.61 3.56
0.1
4.48 4.33 4.46
0.2
4.62
5.20 5.16 5.14 4.85 4.83
0.3
3.97 3.96
0.4
7.19 7.16
Normalized Intensity
0.8
0 0.20 7.0
1.75 0.80 0.97 6.5
6.0
5.5
5.0
4.5
2.64 0.83 4.0 3.5 3.0 Chemical Shift (ppm)
4.28 2.5
4.00 0.25 2.0
1.5
1.0
0.5
0
-0.5
The structure is the same appointed for previous samples. The unsaturated content determined is ~5% . The calculation and other considerations follow the procedure explained for other samples.
Obs: All NMR data were recorded in a Bruker 300MHz equipment. The data processing was made in ACD Lab/NMR Processor Academic Edition software.
cDLP figure ilustrating the printing process and photocuring.