Dental Materials Journal 22(3): 262-271, 2003Original paper

Control of Pore Size in L-lactide/ƒÃ-caprolactone Copolymer Foams

for Tissue Regeneration by the Freeze-drying Method

Hiroyuki NAKAO, Suong-Hyu HYON1, Sadami TSUTSUMI1,

Takuya MATSUMOTO and Junzo TAKAHASHI

Division of Biomaterials Science, Graduate School of Dentistry,

Osaka University, 1-8 Yamadaoka, Suita, Osaka 565-0871, Japan2Department of Medical Simulation Engineering

,

Institute for Frontier Medical Science, Kyoto University, 53 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan

Received March 3, 2003/Accepted Jane 23, 2003

In the regeneration and repair of missing tissues, synthetic polymer scaffolds need many pores

to involve cells and to supply cells with nutrients. The control of the pore size of biodegradable

L-lactide/ƒÃ-caprolactone copolymer foams was studied by changing the polymer concentration

and the cooling temperature in the freeze-drying method. The mixtures of polymer and 1, 4-

dioxane solution were poured into an 18-8 stainless steel Petri dish and frozen. The pore size

of a polymer foam tends to increase from the bottom towards the top of a Petri dish. The pore

size decreased to one-half with increasing polymer concentration (1 to 10wt%). The mean pore

size in foams of 8% polymer concentration decreased from 100ƒÊm to 20ƒÊm as cooling tempera-

ture was lowered. This suggests that the higher cooling rate due to lower cooling temperature

can produce smaller ice-crystals and result in smaller pores.

Key words: Pore size, Freeze-dry, PLCL

INTRODUCTION

Tissue engineering using several synthetic biodegradable polymers as scaffolds is im-

portant in the regeneration and repair of the missing tissues1-4). These scaffolds are

grafted in vivo with cells and growth factors, and become bases for tissue regenera-tion. Biodegradable polymer scaffolds require several characteristics such as cell ad-

herence, facilitation of cell proliferation and maintenance of the function of

differentiated cells. They require an adequate degradation rate and the mechanical

strength to maintain the desired shape. Furthermore, they need to be porous, be-

cause a proper pore size is required for cell seeding, cell growth and production of

extracellular matrix, and the paths for nutrients to enable cells to survive5,6).

Various processes for making biodegradable polymer foams have been reported

as follows; gas foaming/salt leaching method, solvent-casting/salt leaching technique,

etc7-10). In the gas foaming/salt leaching method, the semi-solidified polymer-salt

mixture is immersed in an aqueous solution of citric acid for the gas-forming process

at room temperature. Citric acid and ammonium bicarbonate have been used as ef-

fervescent salts due to their carbon dioxide and ammonium evolving property upon

contact in aqueous solution7). The solvent casting/salt leaching technique creates the

NAKAO et al. 263

desired pore size range by incorporating sieved sodium chloride, sodium tartrate and

sodium citrate particles of a desired size8,9). However, it is difficult to completely

eliminate these additives from foams in these methods using salt crystals. Moreover,

these methods require more preparation steps than the freeze-drying method.

The freeze-drying method was originally developed to preserve foods and drugs

such as antibiotics. Although there have been many studies on the physical property

and porosity of freeze-dried foams11), there have been few systematic studies on foam

pore size. This freeze-drying process has been difficult to reproduce and larger pore

sizes have been difficult to create12,13). The purpose of the present study was to iden-

tify a procedure for controlling the pore size of PLLA/PCL (PLCL: L-lactide/ƒÃ-

caprolactone) copolymer foams by the freeze-drying method.

MATERIALS AND METHODS

Preparation of copolymer (PLCL, 75: 25)

PLCL was synthesized by bulk ring-opening co-polymerization of L-lactide/ƒÃ-

caprolactone at 190•Ž for 5h in vacuum using stannous octoate as the catalyst14,15)

The mole ratio of L-lactide to ƒÃ-caprolactone was selected to be 75: 25, which has an

appropriate biodegradability16). All the polymerization products were purified by pre-

cipitation from a methylene chloride solution in methanol and then dried under re-

duced pressure. The weight average molecular weight (Mw) of the PLCL was

approximately 28•~104 as determined by gel permeation chromatography (HLC-8020,

Toyo Soda, Tokyo, Japan).

Fabrication of PLCL foam

The copolymer was dissolved in 1, 4-dioxane solution at polymer concentration of 1,

2, 4, 8 and 10% (wt/wt) for 24h. The mixture (polymer solution) was poured into

an 18-8 stainless steel Petri dish with a cover (50mm in diameter•~15mm in depth)

and frozen at -196•Ž (liquid nitrogen was poured around the Petri dish in an ex-

panded polystyrene box), -85•Ž (Petri dish was placed on an aluminum shelf of a

deep-freezer (MFD-192, Sanyo, Tokyo, Japan)), -20•Ž (Petri dish was placed on a

plastic shelf in a freezing refrigerator (R-37V1, Hitachi, Tokyo, Japan), respectively.

Furthermore, to examine the effect of cooling rate, the Petri dishes with mixture at

polymer concentration of 8% were also placed on a plastic shelf in the freezing re-

frigerator at 4•Ž, and set in a corrugated carton (70•~70•~60mm) (-20•Ž -SC; SC:

slow-cooling) on a plastic shelf in the freezing refrigerator at -20•Ž, respectively.

The frozen contents were dried in a freeze-dryer (FDU-830, EYELA, Tokyo, Japan)

to yield porous foams at 24h4). Then, the obtained foams were warmed in a vacuum

drying oven (VO-320, ADVANTEC, Tokyo, Japan) for 24h and at 40•Ž to ensure

complete solvent removal. The size of PLCL foams was 40mm in diameter and

about 8mm in thickness as shown in Fig. 1.

264 CONTROL OF PORE SIZE IN FREEZE-DRIED FOAMS

Fig. 1 Appearance of the 8-mm-thick poly

(L-lactide/ƒÃ-caprolactone) foam.

This foam was fabricated by freeze-

drying a 1, 4-dioxane solution with 8

wt% copolymer (75:25).

Fig. 2 Three parts of the PLCL foam

observed by SEM.

The under part (Und) touched the

bottom of the stainless steel Petri

dish, the upper part (Upp) is the

liquid level facing to the cover of

the dish, and the middle part

(Mid) is the intermediate portion

between •gUpp•h and •gUnd•h. The

center portion of each part in the

PLCL foam (the size of 7•~7mm)

was observed.

Pore size characterization

The specimens were cut from three different parts of the PLCL foam, as shown in

Fig. 2. These parts of PLCL foams were the upper surface (•gUpp•h), the middle part

(•gMid•h), and the bottom surface (•gUnd•h) which touched the bottom of the stainless

steel Petri dish. The specimens were coated with platinum using a sputter coater,

and observed by scanning electron microscopy (SEM) (S-2380N, Hitachi, Tokyo,

Japan). The number of specimens was three in each condition. The area of an indi-

vidual pore in SEM photographs was determined using an image processing and

analysis program (NIH Image 1.62, National Institutes of Health, Bethesda, USA).

The shape of the pore was regarded as a circle. The pore size was expressed as the

diameter of a circle.

Examination of cooling rate and freezing time

The individual temperature-time curve was automatically measured using a thermo-

couple that was inserted into the middle of the solvent in a covered Petri dish (cor-

responds to •gMid•h in Fig. 2) three times at different cooling conditions. The cooling

rate just above the freezing temperature of the solvent (1, 4-dioxane) of approxi-

mately 11•Ž and the freezing time between two inflection points were calculated from

the temperature-time curve.

NAKAO et al. 265

Statistical analysis

The pore sizes are listed as mean values and S.D., and analyzed using nonparametric

tests because the data were not normally distributed. Significant differences between

groups were determined by Kruskal-Wallis one-way analysis of variance with the

Scheffe multiple post-hoc test to determine group difference. A value of P<0.05 was

considered statistically significant. Statistical analyses were performed with the

Excel Statistics 2002 program (SSRI Co., Tokyo Japan).

RESULTS

Fig. 3 a), b), and c) show the pore sizes in combinations of five copolymer concen-

trations and three different cooling temperatures, in three parts of •gUpp•h, •gMid•h

and •gUnd•h of PLCL freeze-dried forms, respectively. The result of the Scheffe's test

shows pairwise comparisons among three cooling temperatures of the same copoly-

mer concentration. In Fig. 3 a) and b), the pore size has significantly increased with

higher cooling temperature in almost all copolymer concentrations. However, in Fig.

3 c), the pore size at the cooling temperature of -85•Ž was smallest among the three

cooling temperatures. The pore size tends to increase from •gUnd•h toward •gUpp•h of

Fig. 3 The effect of cooling temperature on pore size of PLCL foams fabricated from solu-tions with various polymer contents.Figures a), b) and c) show the results of upper, middle and under parts, respec-tively. The pairwise comparisons were performed within each groups of the same

polymer contents (P<0.05).

266 CONTROL OF PORE SIZE IN FREEZE-DRIED FOAMS

Fig. 4 The effect of polymer content on pore size of PLCL foams fabricated by the freeze-drying method at different cooling temperatures.Figures a), b) and c) show the results of upper, middle and under parts, respectively. The pairwise comparisons were performed within each group of the same cooling temperature (P<:0.05).

a Petri dish at cooling temperatures of -196•Ž and -20•Ž.

Fig. 4 a), b), and c) show the results of Fig. 3 a), b), and c) under the same

cooling temperature but different copolymer concentrations. The results of the

Scheffe's test showed pairwise comparisons among five materials at the same cooling

temperature. The pore size in the group of lower copolymer concentration (1 and

2%) was approximately double that in the group of higher copolymer concentration

(4, 8, and 10%).

Fig. 5 shows SEM photographs at 5 different cooling temperatures and 3 differ-

ent parts of 8% PLCL foams. The pore size in Fig. 5 corresponds to that in Table

1. The pore size (approximately 1ƒÊm) in Fig. 5 f) is smallest among all conditions,

and the pore size (approximately 110ƒÊm) in Fig. 5 j) is largest among all conditions.

Cooling rates in the cooling conditions of -196, -85, -20, -20•Ž (SC), and 4

℃ were measured to be about 14.4, 6.8, 2.3, 0.9 and 0.3℃/min, respectively, as shown

in Table 1. Fig. 6 a), b), and c) express the effect of cooling rates on pore size in

three parts of •gUpp•h, •gMid•h and •gUnd•h of 8% PLCL freeze-dried forms, respec-

tively. The pore size significantly decreased with higher cooling rates in almost all

copolymer concentrations. Fig. 7 shows the relation between the square of the mean

pore diameter and the freezing time. The square of the mean pore diameter

NAKAO et al. 267

Fig. 5 SEM images of three parts from 8% PLCL foam freeze-dried at different cooling tem-

peratures: (a) -196•Ž-Upp, (b) -196•Ž-Mid, (c) -196•Ž-Und, (d) -85•Ž-Upp, (e) -85•Ž-

Mid, (f) -85•Ž-Und, (g) -20•Ž-Upp, (h) -20•Ž-Mid, (i) -20•Ž-Und, (j) -20•Ž SC-Upp,

(k) -20•Ž SC-Mid, (1) -20•Ž SC-Und, (m) 4•Ž-Upp, (n) 4•Ž-Mid, (o) 4•Ž-Und.

Each image was taken under different magnification.

268 CONTROL OF PORE SIZE IN FREEZE-DRIED FOAMS

Table 1 Cooling conditions and mean pore size of 8% PLCL foams fabricated

“-20 (SC)” indicates slow cooling in a corrugated box at-20℃.

Fig. 6 The effect of cooling rate on pore size of 8% PLCL freeze-dried foams.Figures a), b) and c) show the results of upper, middle and under parts, respectively. The pairwise comparisons were performed within each group of the same cooling temperature (P<0.05).

correlated closely with the freezing time.

DISCUSSION

A number of strategies in tissue engineering have focused on using synthetic biode-

gradable polymers as scaffolds for cell transplantation, cell growth and cell differen-tiation. PLLA, PLGA, PGA, and PLCL have been used as biodegradable poly-

mers5,17,18). These scaffolds need to be porous, with an adequate pore size to give cells

room, and allow paths for nutrients5). There have been various opinions on ideal

pore size but no consensus has been established6,12,19).

NAKAO et al. 269

Fig. 7 The relation between the freezing

time and the square of the mean

pore size of the middle part in 8% PLCL.

In the freeze-drying process, the pore structure after drying is a replica of the

ice crystal morphology after freezing20). Fig. 4 shows that the pore size decreased to

one-half with increasing copolymer concentration in 1, 4-dioxane solution, regardless

of the different cooling temperatures and different foam parts. It may be that the

number of ice-crystal nuclei depends on the polymer concentration. Fig. 4 also sug-

gests that foams of the same pore size but different strength can be fabricated, since

polymer concentration affects the mechanical properties of foams.

Furthermore, the pore size decreased as cooling temperature decreased. The pore

size is determined from the size of the ice-crystal produced in the solidification proc-

ess during cooling and is hardly affected by the vaporization process of the ice-

crystal after freezing. Ice crystals become large after a long freezing time due to the

slow cooling rate. Table 1 and Fig. 6 show that the pore size increased with lower

cooling rate. The pore size (approximately 1ƒÊm) of the •gUnd•h part at a cooling

temperature of -85•Ž was smaller than at other cooling temperatures (-196 and

-20•Ž). The stainless steel Petri dish was placed directly on an aluminum shelf at

-85•Ž, and then the cooling rate of the •gUnd•h part at a cooling temperature of -85

℃ was considered higher than 6.8℃/min measured in the “Mid” part at a cooling

temperature of -85•Ž. Furthermore, when compared with the pore size of •gMid•h

part at a cooling temperature of -196•Ž, the cooling rate in this case was suggested

to be higher than 14.4•Ž/min. The difference in pore size due to the foam parts can

be explained from the finding that the cooling rate of each part decreases with dis-

tance from the bottom of the stainless steel Petri dish. This also supports the idea

that the pore size of freeze-dried foams can be regulated by the cooling rate.

In general, the freezing time depends on the cooling rate. A linear relation was

found to exist between the square of the mean pore diameter and the freezing time ,

as shown in Fig. 7. This means that as the freezing time increases, the ice crystals

270 CONTROL OF PORE SIZE IN FREEZE-DRIED FOAMS

become larger. On the other hand, the permeability of freeze-dried food samples was

reported to be in proportion to the freezing time21). Furthermore, the Carman-

Kozeny equation shows that the permeability is proportional to the square of the

pore diameter. k=ƒÃd2/16hCK, where k is the permeability, ƒÃ is the porosity, d is the

pore diameter, and hCK is a constant of Carman-Kozeny22). The present results as

shown in Fig. 7 also support the concept that the cooling rate regulates the pore size

of freeze-dried foams and that these pores are interconnecting.

The next aim was to identify the appropriate pore size for cell-incubation and

bone in-growth by making PLCL foams with larger pores using a slower cooling rate

and incubating the osteoblasts in the foams.

CONCLUSION

The 75: 25 L-lactide/ƒÃ-caprolactone copolymer was synthesized and dissolved in 1, 4-

dioxane solution at concentrations of 1, 2, 4, 8 and 10% (wt/wt). The mixtures of

polymer and 1, 4-dioxane solution were poured into an 18-8 stainless steel Petri dish

and freeze-dried at different cooling temperatures (4, -20, -85 and -196•Ž) and the

pore size of foams made by the freeze-drying method was evaluated by SEM images.

The pore size of a polymer foam tended to increase from the bottom towards the top

of the Petri dish. The pore size decreased to one-half with increasing polymer con-

centration, regardless of different cooling temperatures and different foam parts.

The mean pore size in foams of the 8% polymer concentration decreased from 100

μm to 20μm as the cooling temperature was lowered. This suggests that the high

cooling rate due to low cooling temperature can produce small ice-crystals and result

in small pores. It is suggested that the polymer concentration and cooling rate are

effective for controlling the pore size of polymer foams by the freeze-drying method.

ACKNOWLEDGEMENTS

This investigation was supported by JSPS Research for the Future Program, Biologi-

cal Tissue Engineering Project, No. JSPS-RFTF 98I00201 and Grant-in Aid for Scien-tific Research from MESSC Japan (14207084).

REFERENCES

1) Yaszemski, M., Payne, R., Hayes, W., Langer, R. and Mikos, A.: Evolution of bone

transplantation: Molecular, cellular and tissue strategies to engineer human bone,

Biomaterials 17 (2): 175-185, 1996.

2) Vacanti, C. and Vacanti, J.: Bone and cartilage reconstruction with tissue engineering

approaches, Otolaryngol Clin N Am 27 (1): 263-276, 1994.

3) Zhang, R. and Ma, P.X.: Poly (a-hydroxyl acids)/hydroxyapatite porous composites for

bone-tissue engineering, I. Preparation and morphology, J Biomed Mater Res 44: 446-455,

1999.

4) Honda, M., Yada, T., Ueda, M. and Kimata, K.: Cartilage formation by cultured

chondrocytes in a new scaffold made of poly (L-lactide-ƒÃ-caprolactone) sponge, J Oral

NAKAO et al. 271

Maxillofac Surg 58: 767-775, 2000.

5) Ma, P.X. and Langer, R.: Fabrication of biodegradable polymer foams for cell trans-

plantation and tissue engineering, In Morgan, J.R. and Yarmush, M.L.,: Tissue engi-

neering methods and protocols, Humana Press, New Jersey, 1999, pp. 47-56.

6) Schugens, C., Maquet, V., Grandfils, C., Jerome, R. and Teyssie, P.: Biodegradable and

macropore polylactide implants for cell transplantation, 1. Preparation of macroporous

polylactide supports by solid-liquid phase separation, Polymer 37: 1027-1038, 1996.

7) Yoon, J.J. and Park, T.G.: Degradation behaviors of biodegradable macroporous scaf-

folds prepared by gas foaming of effervescent salts, J Biomed Mater Res 55: 401-408,

2001.

8) Mikos, A.G., Thorsen, A.J., Czerwonka, L.A., Bao, Y., Langer, R., Winslow, D.N.,

Vacanti, J.P.: Preparation and characterization of poly (L-lactic acid) foams, Polymer

35: 1068-1077, 1994.

9) Lu, L., Peter, S.J, Lyman, M.D, Lai, H.L., Leite, S.M., Tamada, J.A., Vacanti, J.P.,

Langer, R. and Mikos, A.G.: In vitro degradation of porous poly (L-lactic acid) foams,

Biomaterials 21: 1595-1605, 2000.

10) Martin, I., Shastri, V.P., Padera, R.F., Yang, J., Mackay, A.J., Langer, R., Vunjak-

Novakovic, G. and Freed, L.E.: Selective differentiation of mammalian bone marrow

stromal cells cultured on three-dimensional polymer foams, J Biomed Mater Res 55:

229-235, 2001.

11) Goldblith, S.A., Rey, L. and Rothmayr, W.W.: Freeze drying and advanced food tech-

nology, Academic Press, London, New York, San Francisco, 1975.

12) Hu, Y., Grainger, D.W., Winn, S.R. and Hollinger, J.O.: Fabrication of poly (ƒ¿-

hydroxy acid) foam scaffolds using multiple solvent systems, J Biomed Mater Res 59:

563-572, 2002.

13) Schoof, H., Apel, J., Heschel, I. and Rau, G.: Control of pore structure and size in freeze

dried collagen sponges, J Biomed Mater Res (Appl Biomater) 58: 352-357, 2001.

14) Matsui, T., Nakamura, T., Tsuda, T., Le, Y.H., Shimizu, Y., Cha, W.I., Hyon, S.H.,

Ikada, Y. and Nishiya, K.: Bioabsorption of poly-ƒÃ-caprolactone and L-lactide, J Jpn

Soc Biomater 6: 330-337, 1993. (in Japanese)

15) Nakamura, T., Shimizu, Y., Takimoto, Y., Tsuda, T., Li, Y.H., Kiyotani, T.,

Teramachi, M., Hyon S.H., Ikada, Y. and Nishiya, K.: Biodegradation and

tumorigenicity of implanted plates made from a copolymer of ƒÃ-caprolactone and L-

lactide in rat, J Biomed Mater Res 42: 475-484, 1998.

16) Hyon, S.H., Nakamura, T., Cha, W.I. and Ikada, Y.: Synthesis and properties of

elastomeric lactide caprolactone copolymers, J Jpn Soc Biomater 14: 216-223, 1996. (in

Japanese)

17) Ma, P.X., Schloo, B., Mooney, D. and Langer, R.: Development of biomechanical prop-

erties and morphogenesis of in vitro tissue-engineered cartilage, J Biomed Mater Res

29 (12): 1587-1595, 1995.

18) Ishaug, S.L., Crane, G.M., Miller, M.J., Yasko, A.W., Yaszemski, M.J. and Mikos,

A.G.: Bone formation by three-dimensional stromal osteoblast culture in biodegradable

polymer scaffolds, J Biomed Mater Res 36 (1): 17-28, 1997.

19) Itala, A.I., Ylanen, H.O., Ekholm, C., Karlsson, K.H. and Aro, H.T.: Pore diameter

of more than 100ƒÊm is not requisite for bone ingrowth in rabbits, J Biomed Mater Res

(Appl Biomater) 58: 679-683, 2001.

20) Schoof, H., Bruns, L., Fischer, A., Heschel, I. and Rau, G.: Dendric ice morphology in

unidirectionally solidified collagen suspensions, J Crystal Growth 209: 122-129, 2000.

21) Araki, T., Sagara, Y., Abdullah, K. and Tambunan A.H.: Transport properties of cellu-

lar food materials undergoing freeze-drying, Drying Techno 19: 297-312, 2001.

22) Mauran, S., Rigaud, L. and Coudevylle, O.: Application of the Carman-Kozeny Correla-

tion to a High-Porosity and Anisotropic Consolidated Medium: The Compressed Ex-

panded Natural Graphite, Transport in Porous Media 43: 255-376, 2001.