Structure Development during Additive Manufacturing

0 downloads 0 Views 598KB Size Report
osition, 3D Printing, Stereolithography (STL), and. Selective Laser ... Ceramic. Metals. Composites. Chua et al, 2003. Melchels et al, 2010. 2 STRUCTURE IN ...
Structure Development during Additive Manufacturing A. Tojeira, S. Biscaia, T. Viana, P.J. Bártolo & G. R. Mitchell Centre for Rapid and Sustainable Product Development – Polytechnic Institute of Leiria, Marinha Grande, Portugal

ABSTRACT: Additive manufacturing involves the shaping of a product through the use of a liquid phase which is subsequently transformed to the solid state by cooling or through the use of chemical cross-linking reactions. Of particular note is the fused deposition modeling which utilizes semi-crystalline polymers such as poly(ε-caprolactone) or poly(lactic acid) and has been employed in CDRsp to prepare highly porous scaffolds for Tissue Engineering. We show that the crystallization process amplifies small levels of molecular anisotropy introduced in the additive writing process. We show that the level of anisotropy is significantly dependent on the process parameters such as temperature, write speed, and flow rate. The differences in the crystalline morphology introduced by changing these process parameters will have a marked impact on the mechanical properties. This in turn will alter the growth of tissue on such scaffold structures. As with other polymer processing procedures, tuning the process parameters provides a route to controlling and defining the structure and morphology of the scaffold and the properties exhibited by that scaffold. 1 INTRODUCTION Tissue Engineering is an emerging medical field dedicated to solve organs transplantation shortage by delivering implantable bio-fabricated tissue and organ constructs. The development of such implants relies on three main components: cells, growth factors and scaffolds. As major part of this complex process, scaffolds play important roles of form (given by the overall shape of the construct that fills in the tissue defect), function (as it provides temporary mechanical support to the growing tissue) and convey contact cues to the proliferating cells (Hollister, 2009). Nowadays, automated, complex and highly precise systems have been widely developed to respond society healthcare demands in shortest period of time possible. Therefore, Additive Manufacturing (AM) has been a reliable approach to obtain tridimensional scaffolds for Tissue Engineering by using medical imaging and computer-aided design (CAD) (Melchels et al 2012). However, the properties of the scaffold depend not only on the biomaterial itself but also on the technology used for its processing (Correlo et al, 2009). Therefore, scaffold functions are no longer just characterised by its macro-scale parameters such as scaffold geometries, porosity and degradation rate but for its microscopic relations between the materials of the scaffold and the processing procedures employed (Cui et al, 2008).

1.1 Additive Manufacturing in Tissue Engineering Additive Manufacturing systems consist in layer-bylayer construction of tri-dimensional objects as opposed to the well-known subtractive technologies. These systems operate in a more sustainable and reproducible ways since they minimize the waste of expensive biological materials (cells, growth factors and biomaterials) while presenting highly accurate features. Besides that, these technologies also offer customization conditions suited for patient-specific clinical treatment by providing high control over architectural characteristics of the implant (Melchels et al, 2012). 1.2 Additive Manufacturing technologies Scaffolds for Tissue Engineering may be produced by a wide number of technologies depending on the material and principle of operation. In the AM field, there are five main manufacturing processes: Fused Deposition Modelling (FDM) (3DF), 3D Fiber Deposition, 3D Printing, Stereolithography (STL), and Selective Laser Sintering (SLS). Nozzle-based technologies such as FDM process is based on polymers melts passing through a nozzle and a molten fibre is further deposited on the surface of a moving platform or merges with the previous layer where it cools down and solidifies. Geometrical and morphological parameters of scaffold are

influence by the processing conditions, namely, nozzle diameter, writing speed, extrusion rate and polymer physics (de Mulder et al 2009). In case of 3DF, it does not require temperature enabling the use of cell-laden materials to be processed. 3D printing creates tri-dimensional objects by depositing liquid binder onto a powdered bed. This liquid may act as glue or promote a reaction causing the particles to bind together. After each layer, a new powder bed in spread and the object is created by n cycles of this procedure. In this case, the unbounded particles aid the construction of the following layers as it provides temporary support to the object (Butscher, 2011). Light-based technologies such as stereolithography and selective laser sintering are promising techniques to fabricate tissue engineering constructs. STL relies on the photo-curing process of a material by incidence of UV and/or IR radiation, whereas SLS uses IR lasers to heat up the powder beyond its melting point causing neighbouring particles to fuse forming a solid structure. When each layer if finished, a new powder layer is placed with a mechanical roller and the exceeding material from the precious layer provides in place to support the continuance of the process (Bártolo, 2009). Inherently to each SFF process, material physical properties are important characteristics to be considered. In Table 1, it is summarized a list of materials for each type of scaffold fabrication technology for Tissue Engineering. Table 1. Materials used in SFF techniques for Tissue Engineering. AM Technology FDM

3DF

3D printing STL

SLS

Material

Ref.

Amorphous and semi-crystalline thermoplastics Ceramics Natural and synthetic polymers Ceramics Powder bulk polymers Powder ceramics Amorphous reactive resin Ceramics Composites Hydrogels Bulk polymers Ceramic Metals Composites

Chua et al, 2003 Landers et al, 2002

Landers et al, 2002 Corbel et al, 2011 Melchels et al, 2010

Chua et al, 2003 Melchels et al, 2010

2 STRUCTURE IN POLYMERS 2.1 Shaping polymers Any process which produces a polymer-based part involves shaping in the liquid state and a subsequent transformation to the solid state to preserve that state. Additive manufacturing technologies are no different. For some the transformation to a solid takes place via cross-linking reactions as in stereolithography (Davis & Mitchell, 2011) and for others, cooling below the glass transition to form a glass. In some cases the cooling process initiates crystallization which can be easily determined using x-ray diffraction as shown in Figure 1. The glassy amorphous phase is characterised by broad diffuse peaks indicative of the limited spatial correlations. In contrast the crystalline phase shows sharp diffraction peaks.

Figure 1. The wide-angle scattering patterns for a crystalline sample of isotactic polystyrene (solid line) and glassy isotactic polystyrene (broken line) formed by rapidly cooling from the melt at 160º C (G.R.Mitchell, 2001).

2.2 Cross-linked polymer networks Suitable polymers have the possibility of forming cross-links between neighbouring chains as shown schematically in Figure 2. As a consequent of this cross-linking, the chains have restricted translational mobility and form a ‘rubber’ like solid. It may be that the increase in cross-link density leads to rise in the glass transition which is above the ambient temperature and so the material forms a glass. From the additive manufacturing perspective of the applications described here, this cross-linking provides 3dimensional stability, i.e. forms a solid; additionally such materials cannot melt (for crystalline polymers) or dissolve. This is particularly important in the development of 3-D devices using stereolithography (Davis and Mitchell 2011) since it can ensure that the final material can be separated from monomer or pre-polymer. In some cases the cross-linking reactions lead to a phase separation as is common in the formation of polyurethanes (Mateus et al, 2013).

Figure 2. Schematic representation of cross-linking in a polymer (the broken line represents short-length cross-linking chains) (Davis & Mitchell, 2008).

2.3 Glassy polymers Polymer glasses may be viewed as frozen macromolecular fluids in which the structure on both the long and short length scales is essentially that found in the molten state (Mitchell, 2001). The short range order reflects the particular chemical structure of the polymer. Polymers display this disordered structure either as a consequence of the thermal history which inhibits crystallization via rapid cooling or through the presence of disorder along the chain such as is present for a random copolymer or atactic systems; atactic polystyrene is an example of the latter. In cases where the material is based on 2 or more polymers, there is a strong probability that these are immiscible both in the melt and in the glass leading to a phase separated structure which will exhibit a characteristic length scale. Such behaviour is similar to that displayed by block copolymers, although in such cases the length scale is limited by the chemical connectivity.

rected by row nuclei formed from extended chains (Pople et al, 1999). This directing process may lead to relatively low levels of anisotropy in the melt being greatly amplified by the crystallization process. The presence of the row nuclei at the time of crystallisation to direct the subsequent crystallization depends on the relaxation time of the polymer chains involved. If the chains are able to relax before crystallization takes place, then there will be no memory of the flow alignment of the chains which form the row nuclei. Such relaxation times are strongly molecular weight dependent. 3 CASE STUDY 3.1 PCL-based scaffolds Tissue engineered PCL-based scaffolds were obtained by fused-deposition method using the BIOEXTRUDER equipment developed at Institute Polytechnic of Leiria (Figure 4).

a)

2.4 Semi-crystalline polymers Stereo regular polymers are able to crystallize. The basic element is the chain folded lamellar as shown in Figure 3. Lamella crystal

b)

Figure 4. a) Bioextruder system and b) deposition trajectory of tri-dimensional scaffolds (Domingos et al 2009).

Tri-dimensional constructs were obtained by deposition of fibbers with diameter of 300µm of previously molten mixtures describing a 0º/90º configuration in 30mm x 6mm x 0.56mm scaffolds as presented in Figure 5 and processing parameters were evaluated as shown in Table 2. Amorphous region Figure 3. A schematic of the molecular configuration in a chain folded lamellar. The view shown is a cross-section through the lamellar thickness. Typically the chain folded thickness is of the order of 10-20nm but the lateral extent may exceed 1-2µm.

Typically for a quiescent melt these chain folded lamellae form spherulites, but in the case of crystallization from a sheared melt the lamellar may be di-

Figure 5. PCL scaffold with pore size of 900 µm.

Table 2. Variables analyzed in scaffolds produced by the Bioextruder. Processing conditions Case studies Temperature (ºC) 80 100 120 140 160 Writing speed (mm s-1) 10 15 20 25 30 Screw Rotation 40 50 60 70 80 Velocity (rpm)

the thickness of the chain folded lamellae and the degree of crystallisation. The azimuthal distribution of intensity at that value of Q0 reveals the levels of preferred orientation of the lamellar crystals. The evaluation of the preferential orientation of the polymer lamella crystals was determined using a second order Legendre polynomial (equation 1) (Mitchell 2013):

3.2 Synchrotron Experimental Setup Small-angle X-ray scattering (SAXS) time-resolved experiments were performed using X-ray synchrotron radiation in the beamline BM26B at the European Synchrotron Radiation Facility (ESFR) (Grenoble, France). The experimental apparatus of the BM26B beamline consisted in a Pilatus 1M detector (169mm x 179mm active area) for 2D SAXS patterns acquisition, a sample-to-detector distance of about 7 m and X-ray wavelength of 1.54 Å (Figure 6). The data are reported as a function of |Q| where Q = 4πsinθ/λ where 2θ is the scattering angle and λ is the wavelength of the incident x-rays.

= 2 Q

2 < P >m 2

(1)

where, is the orientation parameter of the measured intensity distribution and m is the orientation parameter of a perfectly aligned system (Mitchell 2013). Equation 1 may be written is a more directly useful way as shown in equation 2: π 2

< P2 >= ∫ 0

(

)

I Q , α sin αP (cos α )dα 2 I Q , α sin αdα

(

)

(2)

where, I ( Q , α ) is the azimuthal intensity at a specific Q and α is the azimuthal angle. If Q ≈ 1 the orientation is near perfect, whereas Q ≈ 0 the distribution is isotropic. 4 RESULTS AND DISCUSSION Figure 6. Scheme of the SAXS experiment impinging on two different strands (one vertical and one horizontal).

3.3 Analysis of the 2D SAXS patterns After acquiring the patterns, 2D SAXS images were analysed in the FIT2D software (from the European Synchrotron Radiation Facility – ESRF) were azimuthal integrations were obtained as exemplified in Figure 7.

a)

Azimuthal angle (α)

b)

Figure 7. a) 2D SAXS pattern from a PCL+0.5% Graphene, b) azimuthal integration of Intensity I(α) at |Q| = 0,0395 Å-1.

The scattering patterns exhibited by the scaffolds are typical of a semi-crystalline polymer and arised from the presence of lamellar crystals sandwiched by uncrystallised material. The peak position Q0 reflects

4.1 Scaffolds processed at different temperatures PCL scaffolds were design and manufacturing using five different melting temperatures. In Figure 8 is possible to observe a clear alteration of the structure. Scaffolds processed at high temperatures are less oriented due to extended time of relaxation of the molecules after extrusion. The key feature here is the time taken for crystallization to occur. The maximum crystallization rate for the PCL used in this work is ~ 35°C. The higher the temperature of the polymer in the extruder barrel, the longer it will take to cool to that temperature. If at that point there is a memory of the extrusion process in terms of extended chains, then this will direct the growth direction of the lamellar crystals (Figure 8a). If the time is sufficiently long for the melt to have relaxed then an isotropic distribution will result (Figure 8c). It is also possible that if the writing speed i.e. movement of the platen is faster than the feed rate from the extruder then the filament about to be deposited could be extended. Analysis of the SAXS patterns from each scaffold shows that the lamellar scattering peak position (Q0) is constant with varying temperature conditions

whereas the fraction of crystals which have an isotropic distribution increased.

b) c) a) Figure 8. 2D SAXS pattern from a PCL scaffold processed at a) 100ºC, b) 140ºC and c) 160ºC.

4.2 Scaffolds processed at different writing speeds The write speed refers to the speed of movement of the support platform. Increasing this has two consequences; if this is faster than the flow rate of the polymer from the extruder, the filament will be drawn down. Movement of the platform also affects the cooling rate of the deposited polymer and this will change the time taken to reach the crystallisation temperature. A combination of these two effects leads to an increase in the fraction of the material which is isotropic. Under all conditions the position of the peak in the small-angle x-ray scattering remains constant emphasising that changing process affects the level of anisotropy not the basic semicrystalline morphology.

Figure 10. Peak position (Q0), orientation parameter (P2) and fraction of isotropic material (Fiso) of PCL scaffolds manufactured at different screw rotation velocities.

5 CONCLUSIONS Crystalline morphology in scaffold for Tissue Engineering is strongly dependent on the processing condition which will impact the properties of the scaffold, such as mechanical strength and degradation rate. SAXS techniques using the synchrotron beam-line provide a powerful approach to evaluate the morphology of materials on a localized scale. By varying the processing parameters, crystalline peak position remain unchanged. However the degree of orientation of crystals alignment and the fraction of aligned crystals were highly impact by the processing conditions. ACKLOGDEMENTS The authors would like to thank the European Synchrotron Radiation Facility (ESFR) in Grenoble, France, with special regards to the Beamline BM26B scientists, Guiseppe Portale and Daniel Hermida Merino.

Figure 9. Peak position (Q0), orientation parameter (P2) and fraction of isotropic material (Fiso) of PCL scaffolds manufactured with different writing speeds.

4.3 Scaffolds processed at different Screw Rotation velocity Figure 10 shows that an increase in the flow rate through the extruder decreases the level of isotropic material, in otherwords the level of anisotropy increases. We attributed this to the increase level of anisotropy induced in the melt phase in the extruder.

REFERENCES Bártolo, P.J., Chua, C.K., Almeida, H.A., Chou, S.M. & Lim, A.S.C. 2009. Biomanufactuirng for tissue engineering: Present and future trends. Virtual and Physical Prototyping 4(4): 203-216. Butscher, A., Bohner, M., Hofmman, S., Gauckler, L. & Müller, R. 2011. Structural and material approaches to bone tissue engineering in powder-based three-dimensional printing. Acta Biomaterialia 7(3):907-920. Chua, C.K., Leong, K.F., Cheah, C.M. & Chua, S.W. 2003. Development of a Tissue Engineering Scaffold Structure Library for Rapid Prototyping. Part 1: Investigation and Classification. International Journal of Advanced Manufacturing Technology 21(4): 291-301. Corbel, S., Dufaud, O. & Roques-Carmes, T. 2011. Materials for Stereolithography. In Bártolo, P.J. (ed) Stereolithogra-

phy: Materials, Processes and Applications: 141-160, Springer. Correlo, V.M., Boesel, L.F., Pinho, E., Costa-Pinto, A.R. Alves da Silva, M.L., Bhattacharya, M., Mano, J.F., Neves, N.M., & Reis, R.L. 2009. Melt-based compression-molded scaffolds from chitosan-polyester blends and composites: Morphology and mechanical properties. Journal of Biomedical Materials Research Part A 91(2): 489-504. Cui, C., Jorgensen, S. M., Eaker, D. R. & Ritman, E. L. 2008. Coherent X-ray Scattering for Discriminating Biocompatible Materials in Tissue Scaffolds. In S.R. Stock, Proceedings of SPIE Volume 7078, Developments in XRay Tomography VI: 1-10. Davis, F.J. & Mitchell, G.R. 2011. Polymeric materials for rapid manufacturing in Rapid Manufacturing. In P.J. Bártolo(ed.), Stereolithography: Materials, Processes and Applications, 113-139, Springer. de Mulder, E.L.W., Buma, P. & Hannik, G. 2009. Anisotropic Porous Biodegradable Scaffolds for Musculoskeletal Tissue Engineering. Materials 2(4): 1674-1696. Domingos, M., Dinucci, D., Cometa, S., Alderighi, M., Bártolo & P.J., Chiellini, F., 2009. Polycaprolactone Scaffolds Fabricated via Bioextrusion for Tissue Engineering Applications. International Journal of Biomaterials, Article ID 239643, 1-9. Hollister, S.J., 2009. Scaffold Design and Manufacturing: From Concept to Clinic. Advanced Materials 21(32-33): 33303342. Landers, R., Pfister, A., Hübner, U., John, H., Schmelzeisen, R. & Mülhaupt, R. 2002. Fabrication of soft tissue engineering scaffolds by means of rapid prototyping techniques. Journal of Materials Science 37(15): 3107-3116. Mateus, A., Bartolo, P. & Mitchell, G.R. 2013. Modelling Reaction Injection Moulding. In G. R. Mitchell (ed.) Rheology: Theory, Properties and Practical Applications, Novapress Melchels, F.P.W., Domingos, M.A.N., Klein, T.J., Malda, J., Bártolo, P.J. & Hutmacher, D.W., 2012. Additive manufacturing of tissue and organs. Progress in Polymer Science 37(8): 1079-1104. Melchels, F.P.W., Feijen, J. & Grijpma, D.W. 2010. A review on stereolithography and its applications in biomedical engineering. Biomaterials 31(24): 6161-6130. Mitchell G.R. 2001. Structure of polymer glasses: short range order. In K.H.J. Buschow, R.W. Cahn, M.C. Flemings, B. Ilschner, E.J. Kramer, S. Mahajan & P. Veyssière, Encyclopedia of Materials: Science and Technology, 8926-8932, Elsevier, Mitchell G.R. 2013 Orientation in Polymers, Springer: Berlin Pople, J.A., Mitchell, G.R., Sutton, S.J., Vaughan, A.S. & Chai, C. 1999. The development of organised structures in polyethylene crystallised from a sheared melt, analysed by WAXS and TEM. Polymer 40(10): 2769-2777.