- Email: [email protected]
Drawn gelatin films with improved mechanical properties
Biomaterials 19 (1998) 2335 — 2340
Drawn gelatin films with improved mechanical properties A. Bigi!,*, B. Bracci!, G. Cojazzi", S. Panzavolta!, N. Roveri! !Department of Chemistry G. Ciamician, University of Bologna, via Selmi 2, 40126 Bologna, Italy "Centre for the Study of Macromolecular Physics (CNR), c/o Department of Chemistry G. Ciamician, University of Bologna, Italy Received 17 November 1997; accepted 2 June 1998
Abstract Chain anisotropic distribution in gelatin films has been obtained by uniaxial stretching at constant relative humidity, followed by air drying and successive cross-linking with glutaraldehyde. The drawn samples have been characterized by mechanical tests, differential scanning calorimetry and scanning electron microscopy. The Young’s modulus, E, and the stress at break, p , increase " linearly with the draw ratio and reach values which are about five times those characteristic of undrawn samples. Furthermore, on stretching the alignment of the gelatin strands along the direction of deformation increases while the thickness of the layers decreases significantly. The renaturation level, that is the fraction of gelatin in a collagen-like structure, has been calculated as the ratio between the melting enthalpy of gelatin samples and that of tendon collagen. The results indicate that the improvement of mechanical properties achieved by drawn gelatin is closely related to the renaturation level. The experimental approach utilized to induce segmental orientation in gelatin films, allows to obtain anisotropic materials with improved mechanical properties in the direction of deformation, and can be usefully applied in the preparation of biomaterials. ( 1998 Published by Elsevier Science Ltd. All rights reserved Keywords: Gelatin; Chain orientation; Mechanical properties; Renaturation; DSC; SEM
1. Introduction Gelatin is obtained by thermal denaturation or physical and chemical degradation of collagen. The process involves the disruption of noncovalent bonds and it is partially reversible, in agreement with the gelling properties of gelatin [1]. In fact gel formation, which is obtained by cooling gelatin aqueous solutions, is accompanied by some characteristic changes which have been ascribed to a partial regain of collagen triple-helix structure [2]. As a biomaterial, gelatin displays several attractivness: it is a natural polymer which has not shown antigenity, it is completely resorbable in vivo and its physicochemical properties can be suitably modulated [3—6]. Furthermore, due to the large number of functional side groups it contains, gelatin readily undergoes chemical cross-linking, which is very important for its possible use as a biomaterial. In fact, as collagen-based biomaterials are rapidly degraded in vivo, their structure must be reinforced so that
* Corresponding author. Fax: 0039 51 259456.
they will not significantly alter in the body for the required period. This is usually achieved through the use of cross-linking agents. Physical methods include dehydrothermal treatment and ultraviolet and gamma irradiation. Chemical cross-linking typically utilizes bifunctional reagents like glutaraldehyde (GTA) and diisocyanates, as well as carbodiimides, polyepoxy compounds and acyl azide methods [7—10]. Glutaraldehyde is by far the most widely used agent, due to its efficiency to stabilize collagen-based biomaterials and in spite of some local cytotoxicity and calcification of long-term implants [8, 9]. One of the main disadvantage of gelatin as a material is its poor mechanical properties. A successful method to improve the mechanical properties of materials consisting of semirigid chains is that of inducing segmental orientation. Due to the great interest towards anisotropic materials because of their improved mechanical properties in the direction of orientation, several methods have been utilized to induce anisotropic distribution of chain segments orientation in polymeric materials [4, 11]. The disadvantage of some of these methods is the loss of some or all of the anisotropy when the force which has been applied to induce orientation is removed. An approach which implies cross-linking of the materials
0142-9612/98/$—See front matter ( 1998 Published by Elsevier Science Ltd. All rights reserved. PII S 0 1 4 2 - 9 6 1 2 ( 9 8 ) 0 0 1 4 9 - 5
2336
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
before inducing uniaxial or biaxial deformation in order to maintain the chains anisotropic distribution during air drying has been successfully applied to several natural polymers [4, 12—14]. In the case of collagenous tissues, the anisotropic orientation of the collagen fibrils induced by deforming the material in native conditions, can be maintained by simply air drying the sample under deformation [15, 16]. Dried gelatin is barely extensible, while at over 75% relative humidity, its length can be extended over several percents, with the relative amount of triple-helix structure increasing on increasing deformation [6]. In order to obtain materials with controlled anisotropy, gelatin films have been uniaxially stretched at constant relative humidity, air dried under constant elongation and successively cross-linked with GTA. The results of the physicochemical investigation carried out on these samples indicate that the improved mechanical properties acquired by the drawn films are related to their renaturation levels determined by differential scanning calorimetry.
2. Materials and methods Type A gelatin (Italgelatine S.p.A.) from pig skin was used. Gelatin films were prepared from a 5% aqueous gelatin solution. Films were obtained on the bottom of Petri dishes (diameter"6 cm) after water evaporation at room temperature from 10 ml of gelatin solution. Strip shaped (3]30 mm, thickness around 0.12 mm) air-dried films were immersed in a mixture of water and ethanol in the ratio 2 : 3 for 72 h and stretched in the mixture using a Instrom testing machine with a crosshead speed of 5 mm min~1, as reported below. The use of water/ethanol mixture allows to maintain the samples at constant relative humidity (75%) without causing dissolution of the gelatin films. The samples were air dried under constant elongation, with a draw ratio, j"l/l 0 where l"final length and l "initial length, from 1.2 to 0 3.0. After air drying, the samples were cross-linked in a 2.5% (w/w) GTA solution in phosphate buffer at pH 7.4 for 24 h at room temperature, while kept at constant elongation. The cross-linked samples were then repeatedly washed with bidistilled water and air dried at room temperature. 2.1. Mechanical tests Stress—strain curves were recorded using an INSTRON Testing Machine 4465 and the Series IX software package. The Young’s modulus E, the stress at break p and the strain at break e of the cross-linked " " drawn strips were measured in a static mode. The tests were carried out in a mixture of water and ethanol in the ratio 2 : 3 on samples equilibrated in the same mixture for
72 h, in order to test the samples at the same relative humidity as that utilized for stretching. However, very similar mechanical parameters have been obtained on cross-linked drawn samples which have been tested in saline solution for comparison. Samples about 30 mm long and 0.2 mm thick, were examined using a crosshead speed of 5 mm min~1. The thickness of the samples was determined using a Leitz SM-LUX-POL microscope. For each draw ratio, mechanical tests were performed on at least 10 samples. 2.2. Differential scanning calorimetry Calorimetric measurements were performed using a Perkin Elmer DSC-7. The samples weights were in the range 3—4 mg. Thermograms were obtained using sealed aluminium pans to prevent any loss of water during heating at a rate of 5°C min~1. 2.3. Thermogravimetric analysis Thermogravimetric analysis was carried out on dried samples using a Perkin Elmer TGA-7. Heating was performed in a platinum crucible in an air flow (20 cm3 min~1) at a rate of 5° min~1 up to 900°C. The samples weights were in the range of 5—10 mg. 2.4. Scanning electron microscopy Morphological investigation of fracture surfaces of the dried films was performed using a Philips XL-20 scanning electron microscope. The samples were sputtercoated with gold prior to examination.
3. Results Typical stress—strain curves from samples air dried at different draw ratios and cross-linked with glutaraldehyde are reported in Fig. 1. The undrawn samples are much less extensible and display a higher stress at break than un-cross-linked gelatin films examined in the same conditions [17]. The stress at break improves significantly as the draw ratio of the samples increases, while drawing does not affect appreciably the value of strain at break, which averages around the value of 10%. At the highest draw ratio, the stress at break increases more than five-fold (Fig. 2a). A similar linear improvement can be appreciated in the values of the Young’s modulus E calculated tangentially to the first regions of the stress strain curves (Fig. 2b). The renaturation level of the gelatin films was determined by differential scanning calorimetry (DSC). A typical thermogram recorded from an unstretched crosslinked gelatin sample is reported in Fig. 3. It displays an endothermic peak at about 91.5°C associated to the
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
2337
Fig. 1. Typical stress—strain curves recorded from gelatin films air dried at different draw ratio and cross-linked with glutaraldehyde.
Fig. 2. Initial Young’s modulus, E (a) and stress at break, p (b) as a function of the draw ratio, j. The lines have been calculated by means of linear " regression analysis.
Fig. 3. Typical DSC thermogram of an undrawn gelatin film.
helix—coil transition of gelatin with a melting enthalpy *H"9.6 J g~1 The renaturation level was calculated as *H X(%)" 100 *H T where *H "27.8 J g~1 is the melting enthalpy of tenT don collagen which was examined in the same conditions
as gelatin samples, that is air dried after storage in a mixture of water and ethanol in the ratio 2 : 3 for 72 h and successive cross-linking in a 2.5% (w/w) GTA solution in phosphate buffer at pH 7.4 for 24 h at room temperature. The values of melting temperature, melting enthalpy and renaturation level obtained for the drawn gelatin samples are reported in Table 1. As water content greatly affects both the temperature and the enthalpy of the helix-coil transition [18], Table 1 reports also the values of water content of the samples as determined by thermogravimetric analysis. All the dried samples contain a relative amount of water corresponding to about 14 wt%, thus their thermal parameters can be directly compared. An increase in the renaturation level on increasing the draw ratio, up to about 75%, is evident. At the scanning electron microscope, undrawn crosslinked gelatin films exhibit a quite regular layered structure. The thickness of the gelatin layers and that of the interlayer space decrease dramatically as the draw ratio increases, as it can be observed in Fig. 4, which reports a comparison among the images recorded from samples with a draw ratio of 1 (a), 1.5 (b) and 2.5 (c).
2338
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
Table 1 Melting temperature, ¹ , melting enthalpy, *H, and renaturation level, . X, of cross-linked gelatin films as a function of the draw ratio, j. Each value is the mean of 10 determinations and is reported with its standard deviation. Water content, calculated by thermogravimetric analysis is also reported j
Water content ¹ (°C) . (wt %)
*H (J g~1)
X (%)
1.0 1.2 1.5 1.8 2.0 2.5 3.0 Tendon collagen
14.0$0.2 14.5$0.3 13.7$0.4 14.4$0.2 14.5$0.2 14.7$0.4 13.7$0.3 14.5$0.3
9.6$0.5 13.2$0.2 14.3$0.6 15.2$0.4 17.5$0.2 20.0$0.3 20.6$0.3 27.8$0.3
34.5 47.5 51.4 54.7 62.9 71.9 74.1
91.5$0.5 92.5$0.1 97.0$0.3 97.2$0.4 97.6$0.1 97.5$0.1 97.7$0.3 103$0.2
4. Discussion The results of this study indicate that the experimental approach we have utilized to induce segmental orientation in gelatin films, that is cross-linking the samples after air drying under uniaxial deformation, allows to obtain materials with improved mechanical properties in the direction of deformation. Both stress at break and initial Young’s modulus increase linearly with the draw ratio and reach values which are about five times those characteristic of undrawn samples. Although the strain at break does not seem to vary appreciably, the toughness of the material, that is the area under the stress—strain curve, increases with the extension ratio. Since toughness represents the energy required to break a material, it is an important parameter for the possible structural applications of the material itself. The improvements in mechanical properties achieved in this study are lower than those obtained by segmental orientation in other polymeric systems [12, 13] due to the peculiar characteristics of gelatin which account for its poor orientability [4]. In fact, gelation implies a partial reorganization of gelatin chains into collagen folds and intermolecular association of helices to form junctions. These physical cross-links can severely limit the extensibility of the material. However, the increase in Young’s modulus and stress at break reached in this investigation is undoubtedly higher than that obtained for gelatin submitted to chemical crosslinking before drawing and air drying [4]. This is not surprising since the introduction of chemical cross-links results in a further limitation of the controlled deformation and orientation of the chain segments. In the case of glutaraldehyde, the extensibility of the undrawn samples is reduced by a factor greater than 20 with respect to un-cross-linked samples [17]. Furthermore, while uncross-linked films display a coarse, porous network with a layered structure barely distinguishable [17], gelatin films cross-linked with GTA after air drying from
Fig. 4. Scanning electron micrographs of the fracture surfaces of gelatin films (a) undrawn and with a draw ratio of 1.5 (b) and 2.5 (c). Bar"10 lm.
water/ethanol solution exhibit an ordered laminated structure even when undrawn (Fig. 4a). The alignment of the gelatin strands along the direction of deformation increases and the thickness of the layers, as well as the interlayer space, decreases significantly as a function of stretching, as shown in Fig. 4b and c. It has been suggested that the mechanical properties of gelatin films are
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
2339
Fig. 5. Variations of Young’s modulus, E (a) and stress at break, p (b) as a function of the renaturation level, X. The lines have been calculated by " means of linear regression analysis.
closely related to the renaturation of random gelatin strands to the triple-helix structure that exists in native collagen [19]. Furthermore, the relative amount of triplehelix structure has been found to increase on increasing deformation [6]. Among the different techniques used to determine gelatin renaturation, differential scanning calorimetry seems to be the more suitable [20]. In fact, the enthalpy of the helix—coil transition is considered to be due mainly to hydrogen bond interactions and therefore the renaturation level determined by DSC is a measure of the gelatin strand fraction in a real collagenlike molecular structure. In order to quantify the renaturation level of gelatin films as a function of the draw ratio, we have used tendon collagen as a reference sample. The enthalphy value obtained for tendon is lower than that previously reported [18] since it is stabilized by GTA cross-linking, as shown also by its high melting temperature [21, 22]. Undrawn gelatin displays just a 34.5% of renaturation level compared to tendon collagen. On stretching, the renaturation level increases almost linearly as a function of the draw ratio up to about 75%. Furthermore, stretching provokes an increase in the melting temperature of gelatin, in agreement with an increase in thermal stability, which is more significant at small draw ratios. The variation of the Young’s modulus, as well as of the stress at break, with the renaturation level of the drawn gelatin films is almost linear, as it can be appreciated in Fig. 5a and b. It can be concluded that the renaturation level plays a key role in the improvement of mechanical properties achieved by drawn gelatin, although the contribution of other factors, such as the degree of preferential orientation of collagen molecular portions along the direction of stretching [17], must be considered.
5. Conclusions Gelatin sponges are widely and successfully utilized in clinical practice, and cross-linked gelatin cups are used in veterinary surgery, whereas other applications, such as sutures prepared from gelatin, are hindered by the poor mechanical properties of the material [9, 23]. The
use of cross-linking-under-deformation method allows to prepare gelatin films which display a significant improvement of the mechanical properties in the direction of deformation. In particular, cross-linked drawn gelatin films exhibit mechanical parameters which are comprised in the range of work of several connective tissues, including arterial walls and skin [24]. Thus, the use of this approach can be exploited in the design and preparation of biomaterials which could cover a wider field of applications than that characteristic of undrawn gelatin and offer a valuable and cheaper alternative to other materials, such as collagen itself.
Acknowledgements We thank Prof A. Ripamonti for useful advice and stimulating discussion. This research was supported by MURST, CNR (PF MSTA II) and the University of Bologna (Funds for Selected Research Topics). One of the authors (BB) carried out this research activity, thanks to a fellowship awarded by the Italgelatine S.p.A.
References [1] Ross PI. Gelatin. In: Mark HF, Bikales NM, Overberger CG, Menges G, editors. Encyclopedia of polymer science and engineering, vol. 7. New York: Wiley, 1987:488. [2] Veis A. The macromolecular chemistry of gelatin. New York and London: Academic Press, 1964. [3] Yao CH, Sun JS, Lin FH, Liao CJ, Huang CW. Biological effects and cytotoxicity of tricalcium phosphate and formaldehyde cross-linked gelatin composite. Mater Chem Phys 1996;45:6—14. [4] Fakirov S, Sarac Z, Anbar T, Boz B, Bahar I, Evstatiev M, Apostolov AA, Mark JE, Kloczkowski A. Mechanical properties and transition temperatures of crosslinked oriented gelatin. 1. Static and dynamic mechanical properties of cross-linked gelatin. Colloid Polym Sci 1996;274:334—41. [5] Pezron I, Tarawa T, Miyasaka K, Ishikawa K. X-ray diffraction of gelatin fibers in the dry and swollen states. Biopolymers 1974;13:1823—39. [6] Tanioka A, Miyasaka K, Ishikawa K. Reconstitution of collagen fold structure with stretching gelatin film. Biopolymers 1976;15:1505—11.
2340
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
[7] Rault I, Frei V, Herbage D, Abdul-Malak N, Huc A. Evaluation of different chemical methods for cross-linking collagen gel, films and sponges. J Mater Sci Mater Med 1996;7:215—21. [8] Khor E. Methods for the treatment of collagenous tissues for bioprostheses. Biomaterials 1997;18;95—103. [9] Nimni ME, Cheung DT, Strates B, Kodama M, Sheikh K. Bioprosthesis derived from cross-linked and chemically modified collagenous tissues. In: Nimni ME, editor. Collagen, vol. 3. Boca Raton: CRC Press, 1988:1—38. [10] Olde Damink LHH, Dijkstra PJ, Van Luyn MJA, Van Wachem PB, Nieuwenhuis P, Feijen J. Glutaraldehyde as a crosslinking agent for collagen-based biomaterials. J Mater Sci Mater Med 1995;6:460—72. [11] Ciferri A, Ward IM. Ultra high modulus polymers. London: Applied Science Publications, 1979. [12] Yang Y, Kloczkowski A, Mark JE, Erman B, Bahar I. A novel orientation technique for semirigid polymers. 1. Preparation of cross-linked cellulose—acetate and hydroxypropylcellulose films having permanent anisotropy in the swollen state. Colloid Polym Sci 1994;272:284—92. [13] Mark JE, Yang Y, Kloczkowski JE, Erman B, Bahar I. A novel orientation technique for semirigid polymers. 2. Mechanical properties of cellulose—acetate and hydroxypropylcellulose films. Colloid Polym Sci 1994;272:393—9. [14] Fakirov S, Sarac Z, Anbar T, Boz B, Bahar I, Evstatiev M, Apostolov AA, Mark JE, Kloczkowski A. Mechanical properties and transition temperatures of cross-linked oriented gelatin. II. Effect of orientation and water content on transition temperatures. Colloid Polym Sci 1997;275:307—14.
[15] Bigi A, Jeronimidis G, Purslow PP, Ripamonti A, Roveri N. Collagen orientation by X-ray pole figures and mechanical properties of media carotid wall. J Mater Sci 1981;16:2557—62. [16] Purslow PP, Bigi A, Ripamonti A, Roveri N. Collagen fibre reorientation around a crack in biaxially stretched aortic media. Int J Biol Macromol 1984;6:21—5. [17] Bigi A, Panzavolta S, Ripamonti A, Roveri N. Hydroxyapatite—gelatin films: a structural and mechanical characterization. Biomaterials 1998;19:739—44. [18] Bigi A, Cojazzi G, Roveri N, Koch MHJ. Differential scanning calorimetry and X-ray diffraction study of tendon collagen thermal denaturation. Int J Biol Macromol 1987;9:363—7. [19] Achet D, He XW. Brittleness of gelatin films. Ely Lilly Capsule Research and Development Internal Report. September 1993. [20] Achet D, He XW. Determination of the renaturation level in gelatin films. Polymer 1995;36:787—91. [21] Privalov PL, Tiktopulo EI. Thermal conformational transformation of tropocollagen. 1. Calorimetric study. Biopolymers 1970;9:127—39. [22] McPherson JM, Ledger PW, Sawamura S, Conti A, Wade S, Reihanian H, Wallace DG. The preparation and physicochemical characterization of an injectable form of reconstituted, glutaraldehyde cross-linked, bovine corium collagen. J Biomed Mater Res 1986;20:79—92. [23] Hastings GW, Ducheyne P. Macromolecular biomaterials. Boca Raton: CRC Press, 1984. [24] Vincent J. Structural biomaterials. Princeton: Princeton University Press, 1990.
Drawn gelatin films with improved mechanical properties A. Bigi!,*, B. Bracci!, G. Cojazzi", S. Panzavolta!, N. Roveri! !Department of Chemistry G. Ciamician, University of Bologna, via Selmi 2, 40126 Bologna, Italy "Centre for the Study of Macromolecular Physics (CNR), c/o Department of Chemistry G. Ciamician, University of Bologna, Italy Received 17 November 1997; accepted 2 June 1998
Abstract Chain anisotropic distribution in gelatin films has been obtained by uniaxial stretching at constant relative humidity, followed by air drying and successive cross-linking with glutaraldehyde. The drawn samples have been characterized by mechanical tests, differential scanning calorimetry and scanning electron microscopy. The Young’s modulus, E, and the stress at break, p , increase " linearly with the draw ratio and reach values which are about five times those characteristic of undrawn samples. Furthermore, on stretching the alignment of the gelatin strands along the direction of deformation increases while the thickness of the layers decreases significantly. The renaturation level, that is the fraction of gelatin in a collagen-like structure, has been calculated as the ratio between the melting enthalpy of gelatin samples and that of tendon collagen. The results indicate that the improvement of mechanical properties achieved by drawn gelatin is closely related to the renaturation level. The experimental approach utilized to induce segmental orientation in gelatin films, allows to obtain anisotropic materials with improved mechanical properties in the direction of deformation, and can be usefully applied in the preparation of biomaterials. ( 1998 Published by Elsevier Science Ltd. All rights reserved Keywords: Gelatin; Chain orientation; Mechanical properties; Renaturation; DSC; SEM
1. Introduction Gelatin is obtained by thermal denaturation or physical and chemical degradation of collagen. The process involves the disruption of noncovalent bonds and it is partially reversible, in agreement with the gelling properties of gelatin [1]. In fact gel formation, which is obtained by cooling gelatin aqueous solutions, is accompanied by some characteristic changes which have been ascribed to a partial regain of collagen triple-helix structure [2]. As a biomaterial, gelatin displays several attractivness: it is a natural polymer which has not shown antigenity, it is completely resorbable in vivo and its physicochemical properties can be suitably modulated [3—6]. Furthermore, due to the large number of functional side groups it contains, gelatin readily undergoes chemical cross-linking, which is very important for its possible use as a biomaterial. In fact, as collagen-based biomaterials are rapidly degraded in vivo, their structure must be reinforced so that
* Corresponding author. Fax: 0039 51 259456.
they will not significantly alter in the body for the required period. This is usually achieved through the use of cross-linking agents. Physical methods include dehydrothermal treatment and ultraviolet and gamma irradiation. Chemical cross-linking typically utilizes bifunctional reagents like glutaraldehyde (GTA) and diisocyanates, as well as carbodiimides, polyepoxy compounds and acyl azide methods [7—10]. Glutaraldehyde is by far the most widely used agent, due to its efficiency to stabilize collagen-based biomaterials and in spite of some local cytotoxicity and calcification of long-term implants [8, 9]. One of the main disadvantage of gelatin as a material is its poor mechanical properties. A successful method to improve the mechanical properties of materials consisting of semirigid chains is that of inducing segmental orientation. Due to the great interest towards anisotropic materials because of their improved mechanical properties in the direction of orientation, several methods have been utilized to induce anisotropic distribution of chain segments orientation in polymeric materials [4, 11]. The disadvantage of some of these methods is the loss of some or all of the anisotropy when the force which has been applied to induce orientation is removed. An approach which implies cross-linking of the materials
0142-9612/98/$—See front matter ( 1998 Published by Elsevier Science Ltd. All rights reserved. PII S 0 1 4 2 - 9 6 1 2 ( 9 8 ) 0 0 1 4 9 - 5
2336
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
before inducing uniaxial or biaxial deformation in order to maintain the chains anisotropic distribution during air drying has been successfully applied to several natural polymers [4, 12—14]. In the case of collagenous tissues, the anisotropic orientation of the collagen fibrils induced by deforming the material in native conditions, can be maintained by simply air drying the sample under deformation [15, 16]. Dried gelatin is barely extensible, while at over 75% relative humidity, its length can be extended over several percents, with the relative amount of triple-helix structure increasing on increasing deformation [6]. In order to obtain materials with controlled anisotropy, gelatin films have been uniaxially stretched at constant relative humidity, air dried under constant elongation and successively cross-linked with GTA. The results of the physicochemical investigation carried out on these samples indicate that the improved mechanical properties acquired by the drawn films are related to their renaturation levels determined by differential scanning calorimetry.
2. Materials and methods Type A gelatin (Italgelatine S.p.A.) from pig skin was used. Gelatin films were prepared from a 5% aqueous gelatin solution. Films were obtained on the bottom of Petri dishes (diameter"6 cm) after water evaporation at room temperature from 10 ml of gelatin solution. Strip shaped (3]30 mm, thickness around 0.12 mm) air-dried films were immersed in a mixture of water and ethanol in the ratio 2 : 3 for 72 h and stretched in the mixture using a Instrom testing machine with a crosshead speed of 5 mm min~1, as reported below. The use of water/ethanol mixture allows to maintain the samples at constant relative humidity (75%) without causing dissolution of the gelatin films. The samples were air dried under constant elongation, with a draw ratio, j"l/l 0 where l"final length and l "initial length, from 1.2 to 0 3.0. After air drying, the samples were cross-linked in a 2.5% (w/w) GTA solution in phosphate buffer at pH 7.4 for 24 h at room temperature, while kept at constant elongation. The cross-linked samples were then repeatedly washed with bidistilled water and air dried at room temperature. 2.1. Mechanical tests Stress—strain curves were recorded using an INSTRON Testing Machine 4465 and the Series IX software package. The Young’s modulus E, the stress at break p and the strain at break e of the cross-linked " " drawn strips were measured in a static mode. The tests were carried out in a mixture of water and ethanol in the ratio 2 : 3 on samples equilibrated in the same mixture for
72 h, in order to test the samples at the same relative humidity as that utilized for stretching. However, very similar mechanical parameters have been obtained on cross-linked drawn samples which have been tested in saline solution for comparison. Samples about 30 mm long and 0.2 mm thick, were examined using a crosshead speed of 5 mm min~1. The thickness of the samples was determined using a Leitz SM-LUX-POL microscope. For each draw ratio, mechanical tests were performed on at least 10 samples. 2.2. Differential scanning calorimetry Calorimetric measurements were performed using a Perkin Elmer DSC-7. The samples weights were in the range 3—4 mg. Thermograms were obtained using sealed aluminium pans to prevent any loss of water during heating at a rate of 5°C min~1. 2.3. Thermogravimetric analysis Thermogravimetric analysis was carried out on dried samples using a Perkin Elmer TGA-7. Heating was performed in a platinum crucible in an air flow (20 cm3 min~1) at a rate of 5° min~1 up to 900°C. The samples weights were in the range of 5—10 mg. 2.4. Scanning electron microscopy Morphological investigation of fracture surfaces of the dried films was performed using a Philips XL-20 scanning electron microscope. The samples were sputtercoated with gold prior to examination.
3. Results Typical stress—strain curves from samples air dried at different draw ratios and cross-linked with glutaraldehyde are reported in Fig. 1. The undrawn samples are much less extensible and display a higher stress at break than un-cross-linked gelatin films examined in the same conditions [17]. The stress at break improves significantly as the draw ratio of the samples increases, while drawing does not affect appreciably the value of strain at break, which averages around the value of 10%. At the highest draw ratio, the stress at break increases more than five-fold (Fig. 2a). A similar linear improvement can be appreciated in the values of the Young’s modulus E calculated tangentially to the first regions of the stress strain curves (Fig. 2b). The renaturation level of the gelatin films was determined by differential scanning calorimetry (DSC). A typical thermogram recorded from an unstretched crosslinked gelatin sample is reported in Fig. 3. It displays an endothermic peak at about 91.5°C associated to the
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
2337
Fig. 1. Typical stress—strain curves recorded from gelatin films air dried at different draw ratio and cross-linked with glutaraldehyde.
Fig. 2. Initial Young’s modulus, E (a) and stress at break, p (b) as a function of the draw ratio, j. The lines have been calculated by means of linear " regression analysis.
Fig. 3. Typical DSC thermogram of an undrawn gelatin film.
helix—coil transition of gelatin with a melting enthalpy *H"9.6 J g~1 The renaturation level was calculated as *H X(%)" 100 *H T where *H "27.8 J g~1 is the melting enthalpy of tenT don collagen which was examined in the same conditions
as gelatin samples, that is air dried after storage in a mixture of water and ethanol in the ratio 2 : 3 for 72 h and successive cross-linking in a 2.5% (w/w) GTA solution in phosphate buffer at pH 7.4 for 24 h at room temperature. The values of melting temperature, melting enthalpy and renaturation level obtained for the drawn gelatin samples are reported in Table 1. As water content greatly affects both the temperature and the enthalpy of the helix-coil transition [18], Table 1 reports also the values of water content of the samples as determined by thermogravimetric analysis. All the dried samples contain a relative amount of water corresponding to about 14 wt%, thus their thermal parameters can be directly compared. An increase in the renaturation level on increasing the draw ratio, up to about 75%, is evident. At the scanning electron microscope, undrawn crosslinked gelatin films exhibit a quite regular layered structure. The thickness of the gelatin layers and that of the interlayer space decrease dramatically as the draw ratio increases, as it can be observed in Fig. 4, which reports a comparison among the images recorded from samples with a draw ratio of 1 (a), 1.5 (b) and 2.5 (c).
2338
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
Table 1 Melting temperature, ¹ , melting enthalpy, *H, and renaturation level, . X, of cross-linked gelatin films as a function of the draw ratio, j. Each value is the mean of 10 determinations and is reported with its standard deviation. Water content, calculated by thermogravimetric analysis is also reported j
Water content ¹ (°C) . (wt %)
*H (J g~1)
X (%)
1.0 1.2 1.5 1.8 2.0 2.5 3.0 Tendon collagen
14.0$0.2 14.5$0.3 13.7$0.4 14.4$0.2 14.5$0.2 14.7$0.4 13.7$0.3 14.5$0.3
9.6$0.5 13.2$0.2 14.3$0.6 15.2$0.4 17.5$0.2 20.0$0.3 20.6$0.3 27.8$0.3
34.5 47.5 51.4 54.7 62.9 71.9 74.1
91.5$0.5 92.5$0.1 97.0$0.3 97.2$0.4 97.6$0.1 97.5$0.1 97.7$0.3 103$0.2
4. Discussion The results of this study indicate that the experimental approach we have utilized to induce segmental orientation in gelatin films, that is cross-linking the samples after air drying under uniaxial deformation, allows to obtain materials with improved mechanical properties in the direction of deformation. Both stress at break and initial Young’s modulus increase linearly with the draw ratio and reach values which are about five times those characteristic of undrawn samples. Although the strain at break does not seem to vary appreciably, the toughness of the material, that is the area under the stress—strain curve, increases with the extension ratio. Since toughness represents the energy required to break a material, it is an important parameter for the possible structural applications of the material itself. The improvements in mechanical properties achieved in this study are lower than those obtained by segmental orientation in other polymeric systems [12, 13] due to the peculiar characteristics of gelatin which account for its poor orientability [4]. In fact, gelation implies a partial reorganization of gelatin chains into collagen folds and intermolecular association of helices to form junctions. These physical cross-links can severely limit the extensibility of the material. However, the increase in Young’s modulus and stress at break reached in this investigation is undoubtedly higher than that obtained for gelatin submitted to chemical crosslinking before drawing and air drying [4]. This is not surprising since the introduction of chemical cross-links results in a further limitation of the controlled deformation and orientation of the chain segments. In the case of glutaraldehyde, the extensibility of the undrawn samples is reduced by a factor greater than 20 with respect to un-cross-linked samples [17]. Furthermore, while uncross-linked films display a coarse, porous network with a layered structure barely distinguishable [17], gelatin films cross-linked with GTA after air drying from
Fig. 4. Scanning electron micrographs of the fracture surfaces of gelatin films (a) undrawn and with a draw ratio of 1.5 (b) and 2.5 (c). Bar"10 lm.
water/ethanol solution exhibit an ordered laminated structure even when undrawn (Fig. 4a). The alignment of the gelatin strands along the direction of deformation increases and the thickness of the layers, as well as the interlayer space, decreases significantly as a function of stretching, as shown in Fig. 4b and c. It has been suggested that the mechanical properties of gelatin films are
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
2339
Fig. 5. Variations of Young’s modulus, E (a) and stress at break, p (b) as a function of the renaturation level, X. The lines have been calculated by " means of linear regression analysis.
closely related to the renaturation of random gelatin strands to the triple-helix structure that exists in native collagen [19]. Furthermore, the relative amount of triplehelix structure has been found to increase on increasing deformation [6]. Among the different techniques used to determine gelatin renaturation, differential scanning calorimetry seems to be the more suitable [20]. In fact, the enthalpy of the helix—coil transition is considered to be due mainly to hydrogen bond interactions and therefore the renaturation level determined by DSC is a measure of the gelatin strand fraction in a real collagenlike molecular structure. In order to quantify the renaturation level of gelatin films as a function of the draw ratio, we have used tendon collagen as a reference sample. The enthalphy value obtained for tendon is lower than that previously reported [18] since it is stabilized by GTA cross-linking, as shown also by its high melting temperature [21, 22]. Undrawn gelatin displays just a 34.5% of renaturation level compared to tendon collagen. On stretching, the renaturation level increases almost linearly as a function of the draw ratio up to about 75%. Furthermore, stretching provokes an increase in the melting temperature of gelatin, in agreement with an increase in thermal stability, which is more significant at small draw ratios. The variation of the Young’s modulus, as well as of the stress at break, with the renaturation level of the drawn gelatin films is almost linear, as it can be appreciated in Fig. 5a and b. It can be concluded that the renaturation level plays a key role in the improvement of mechanical properties achieved by drawn gelatin, although the contribution of other factors, such as the degree of preferential orientation of collagen molecular portions along the direction of stretching [17], must be considered.
5. Conclusions Gelatin sponges are widely and successfully utilized in clinical practice, and cross-linked gelatin cups are used in veterinary surgery, whereas other applications, such as sutures prepared from gelatin, are hindered by the poor mechanical properties of the material [9, 23]. The
use of cross-linking-under-deformation method allows to prepare gelatin films which display a significant improvement of the mechanical properties in the direction of deformation. In particular, cross-linked drawn gelatin films exhibit mechanical parameters which are comprised in the range of work of several connective tissues, including arterial walls and skin [24]. Thus, the use of this approach can be exploited in the design and preparation of biomaterials which could cover a wider field of applications than that characteristic of undrawn gelatin and offer a valuable and cheaper alternative to other materials, such as collagen itself.
Acknowledgements We thank Prof A. Ripamonti for useful advice and stimulating discussion. This research was supported by MURST, CNR (PF MSTA II) and the University of Bologna (Funds for Selected Research Topics). One of the authors (BB) carried out this research activity, thanks to a fellowship awarded by the Italgelatine S.p.A.
References [1] Ross PI. Gelatin. In: Mark HF, Bikales NM, Overberger CG, Menges G, editors. Encyclopedia of polymer science and engineering, vol. 7. New York: Wiley, 1987:488. [2] Veis A. The macromolecular chemistry of gelatin. New York and London: Academic Press, 1964. [3] Yao CH, Sun JS, Lin FH, Liao CJ, Huang CW. Biological effects and cytotoxicity of tricalcium phosphate and formaldehyde cross-linked gelatin composite. Mater Chem Phys 1996;45:6—14. [4] Fakirov S, Sarac Z, Anbar T, Boz B, Bahar I, Evstatiev M, Apostolov AA, Mark JE, Kloczkowski A. Mechanical properties and transition temperatures of crosslinked oriented gelatin. 1. Static and dynamic mechanical properties of cross-linked gelatin. Colloid Polym Sci 1996;274:334—41. [5] Pezron I, Tarawa T, Miyasaka K, Ishikawa K. X-ray diffraction of gelatin fibers in the dry and swollen states. Biopolymers 1974;13:1823—39. [6] Tanioka A, Miyasaka K, Ishikawa K. Reconstitution of collagen fold structure with stretching gelatin film. Biopolymers 1976;15:1505—11.
2340
A. Bigi et al. / Biomaterials 19 (1998) 2335—2340
[7] Rault I, Frei V, Herbage D, Abdul-Malak N, Huc A. Evaluation of different chemical methods for cross-linking collagen gel, films and sponges. J Mater Sci Mater Med 1996;7:215—21. [8] Khor E. Methods for the treatment of collagenous tissues for bioprostheses. Biomaterials 1997;18;95—103. [9] Nimni ME, Cheung DT, Strates B, Kodama M, Sheikh K. Bioprosthesis derived from cross-linked and chemically modified collagenous tissues. In: Nimni ME, editor. Collagen, vol. 3. Boca Raton: CRC Press, 1988:1—38. [10] Olde Damink LHH, Dijkstra PJ, Van Luyn MJA, Van Wachem PB, Nieuwenhuis P, Feijen J. Glutaraldehyde as a crosslinking agent for collagen-based biomaterials. J Mater Sci Mater Med 1995;6:460—72. [11] Ciferri A, Ward IM. Ultra high modulus polymers. London: Applied Science Publications, 1979. [12] Yang Y, Kloczkowski A, Mark JE, Erman B, Bahar I. A novel orientation technique for semirigid polymers. 1. Preparation of cross-linked cellulose—acetate and hydroxypropylcellulose films having permanent anisotropy in the swollen state. Colloid Polym Sci 1994;272:284—92. [13] Mark JE, Yang Y, Kloczkowski JE, Erman B, Bahar I. A novel orientation technique for semirigid polymers. 2. Mechanical properties of cellulose—acetate and hydroxypropylcellulose films. Colloid Polym Sci 1994;272:393—9. [14] Fakirov S, Sarac Z, Anbar T, Boz B, Bahar I, Evstatiev M, Apostolov AA, Mark JE, Kloczkowski A. Mechanical properties and transition temperatures of cross-linked oriented gelatin. II. Effect of orientation and water content on transition temperatures. Colloid Polym Sci 1997;275:307—14.
[15] Bigi A, Jeronimidis G, Purslow PP, Ripamonti A, Roveri N. Collagen orientation by X-ray pole figures and mechanical properties of media carotid wall. J Mater Sci 1981;16:2557—62. [16] Purslow PP, Bigi A, Ripamonti A, Roveri N. Collagen fibre reorientation around a crack in biaxially stretched aortic media. Int J Biol Macromol 1984;6:21—5. [17] Bigi A, Panzavolta S, Ripamonti A, Roveri N. Hydroxyapatite—gelatin films: a structural and mechanical characterization. Biomaterials 1998;19:739—44. [18] Bigi A, Cojazzi G, Roveri N, Koch MHJ. Differential scanning calorimetry and X-ray diffraction study of tendon collagen thermal denaturation. Int J Biol Macromol 1987;9:363—7. [19] Achet D, He XW. Brittleness of gelatin films. Ely Lilly Capsule Research and Development Internal Report. September 1993. [20] Achet D, He XW. Determination of the renaturation level in gelatin films. Polymer 1995;36:787—91. [21] Privalov PL, Tiktopulo EI. Thermal conformational transformation of tropocollagen. 1. Calorimetric study. Biopolymers 1970;9:127—39. [22] McPherson JM, Ledger PW, Sawamura S, Conti A, Wade S, Reihanian H, Wallace DG. The preparation and physicochemical characterization of an injectable form of reconstituted, glutaraldehyde cross-linked, bovine corium collagen. J Biomed Mater Res 1986;20:79—92. [23] Hastings GW, Ducheyne P. Macromolecular biomaterials. Boca Raton: CRC Press, 1984. [24] Vincent J. Structural biomaterials. Princeton: Princeton University Press, 1990.