Embed Size (px)
Citation preview
+"Indirect1Rapid1Prototyping1(iRP)1
• Molds1realised1with1RP1devices1(CAD/CAM)1
• CasCng1of1the1desired1(bioF)material1
• ExtracCon1of1the1final1object1
DW1Hutmacher1et1al.,1Trends1in1Biotechnology,122(7):3541–1362,120041
+"
• Wide1rage1of1(bioF)materials1
• Known1also1as1rapid1tooling1• Less1waste1• High1fidelity1• Microporosity1by:1– criCcal1point1drying1– freeFdrying11– leaching11
• Use1in1surgery1room1
iRP1–1General1concepts1
DW1Hutmacher1et1al.,1Trends1in1Biotechnology,122(7):3541–1362,120041
+"iRP1–1Stereolitography11
Casted"Materials" Extrac0on"method" Resolu0on"(μm)"
ThermoplasCc1Elastomer1
Mechanical1 ≈10001
HA1 Pyrolysis1 4001
PCL,1PLLA,1PLGA,1Chitosan,1alginate1
Basic1SoluCon1 200F4001
R1Sodian1et1al,.1ASAIO1Journal,148(1),120021TFMG1Chu1et1al.,1Journal1of1Materials1Science:1Materials1in1Medicine,112:471–478,1200111
YJ1Seol1Microelectronic1Engineering,186(4F6):1443–1446,1Apr1200911HW1Kang1and1DW1Cho.1Tissue1Eng1Part1C1Methods.120121Sep;18(9):719F291
!"#$ %%&'( )*+,) -*. /0+))1)./-"+2 -*0+3#* -*. /.2-.0/+4352 +6 7 89 "5:472-; -*. /.2-.0 /+4352 -+ -*. 4.6- 72<-*. .=-.0274 )3067/. -+ -*. 0"#*-$ >-07"#*- /*722.4) /72 '.)..2 -+ /+22./- -*. /.2-.0 /+4352 ,"-* -*. .=-.0274)3067/.$ !"#$ %%&/( )*+,) -*. .=-.0274 )3067/. +6 7 89"5:472-$ ?*. /*722.4) ,.0. 6+32< -+ .="- -*. "5:472- "2 7,.44170072#.< :7--.02$
?+ /*./@ -*. 32"6+05"-A +6 -*. )"2-.0"2# )*0"2@7#.; -*.<"5.2)"+2) +6 ?%; ?B; ?9 72< 8% ,.0. 5.7)30.< "2).C.074 4+/7-"+2)$ ?*. 0.)34-) 70. )*+,2 "2 ?7'4. DD$ ?*.4"2.70 )"2-.0"2# )*0"2@7#. +6 -*.). EF "5:472-) ,7)6+32< -+ 6744 "2 7 )5744 072#. '.-,..2 BG 72< BHI ,"-*72 7C.07#. +6 BJ!K!JI$ ?*. 0.)34- "2<"/7-.) 72 .C.2)"2-.0"2# )*0"2@7#. "2 -*. 5.7)30.< <"0./-"+2)$ L<"0./-"+2 "2 -*. -.-07#+274 "5:472-) ,7) 2+- 5.7)30.)<3. -+ )+5. 57-.0"74 4+)) <30"2# )75:4. :+4")*"2#$ ?+/+5:.2)7-. 6+0 -*. )*0"2@7#.; -*. 2.#7-"C. 5+4< <.)"#2)+6 8B 72< 89 ,.0. .=:72<.< ,"-* 7 )*0"2@7#. 67/-+0 +6%$9H '.6+0. 572367/-30"2# +2 -*. >M 7::707-3)$ ?*.5+4<) ,.0. 572367/-30.<; /7)- 72< /30.< 72< -*. #0..2'+<".) ,.0. )"2-.0.< 7) :0.C"+3)4A <.)/0"'.<$ ?7'4. DDD)*+,) 7/-374 )"2-.0.< <"75.-.0 6+0 8B 72< 89$ ?*.<"66.0.2/. '.-,..2 -*. <.)"#2 <"5.2)"+2 72< -*. )"2-.0.<<"5.2)"+2 0.<3/.< 60+5 BJI -+ BI 76-.0 -*. "5:4.15.2-7-"+2 +6 -*. )*0"2@7#. 67/-+0$
?7'4. DN )*+,) -*. 072#. +6 /*722.4 )"O. "2 744 )"=<.)"#2)$ ?*. 7C.07#. /*722.4 )"O.) "2 -*. -.-07#+274"5:472-) ,.0. '.-,..2 JP9 72< PJQ "5$ ?*. 7C.07#./*722.4 )"O.) "2 -*. /A4"2<0"/74 "5:472-) ,.0. '.-,..29JJ 72< RRR "5$ ?*. 47-.0 /*722.4 )"O.) 70. ,.44 "2 -*.072#. ,*.0. 5+)- '+2. "2#0+,-* ,7) 6+32< S9; P; BQ; BPT$?*. )-72<70< <.C"7-"+2) +6 -*. /*722.4 )"O. "2 744 <.)"#2)70. GK "5 +0 4.))$ U"-* <"66.0.2- /*722.4 /+22./-"+2:7--.02); "5:472- :+0+)"-A '.-,..2 BJI 72< GBI ,7)7/*".C.<$
!" #$%&'%%$()%?*. )"2-.0"2# +6 V"+-74 EF :+,<.0 *7) '..2 )-3<".< 'AW3A) !" #$% S9KT$ ?*. :+,<.0 ,7) 6+32< -+ .=*"'"- 7)"#5+"< )"2-.0"2# /30C. "2 -*. -.5:.07-30.1<.2)"-A <7-7;,"-* 7 :47-.73 "2 -*. %%KK !8 -+ %9GK !8 072#.$ ?*. X274)"2-.0.< <.2)"-A ,7) 6+32< -+ '. '.-,..2 Q%I 72< PPI$?*. )*7:. +6 +30 -.5:.07-30.1<.2)"-A /30C. ,7) )"5"470-+ -*") 0.).70/* '3- ,"-* 7 :47-.73 +//300.< 7- 7 *"#*.0-.5:.07-30.$ ?*. -.5:.07-30. 6+0 -*. :47-.73 ") @2+,2 -+
&'()*! + 80+))1)./-"+2 +6 /30.< EF )3):.2)"+2 "2)"<. -*. .:+=A 5+4<$
&'()*! ,- >"2-.0.< EF "5:472-)Y &7( ?%; ?B; ?9 <.)"#2Z &'( 8% <.)"#2Z&/( 8B <.)"#2Z &<( 89 <.)"#2$
!"#
uated by comparison with positive (copper disk) and negative(ceramic disk) controls. Fig. 6 shows the results.
3. Results
In this study, HA scaffolds with controllable architecture weresuccessfully fabricated using an indirect fabrication method thatcombined MSTL and molding technology. Microscopic images
showed that the fabricated scaffold (5.7 ! 5.7 mm) comprised afully interconnected porous structure with a pore size of 300 lmand line width of about 400 lm (Fig. 7). To solve the problem ofpartial filling of the mold with slurry (see Fig. 5), we will conductmore experiments in the future, in which the mold design willbe changed.
After sintering, shrinkage occurred due to removal of the resinand the free spaces between the HA particles. The shrinkage oc-curred at the same rate in repeated experiments, as shown in Table3. In addition, the shrinkage of all directions of structure occurredat same rate and the fabrication process was optimized, to fabri-cate a 3D structure with desired shape and mechanical properties.The optimal conditions were shown in Table 2. Thus, this fabrica-tion process could be used to fabricate 3D structures which had
Fig. 5. Fabricated HA scaffold.
Fig. 6. Agar-overlay test.
Fig. 7. Pores and lines of the HA scaffold.
Table 3Results for a HA scaffold.
Mold Scaffold Shrinkage (%)
Pore (lm) 700 Line (lm) 430 38Line (lm) 500 Pore (lm) 290 42Overall (mm) 9.4 Overall (mm) 5.7 39
Y.-J. Seol et al. /Microelectronic Engineering 86 (2009) 1443–1446 1445
For Peer Review
!"#$%&$%'()%*+,+-.%./011+23.%4567%*+,8%.598.%+1%:;;(<;;=>%10?,5/0683%-.5@A%BCDE%BDDFE%BD#FE%02A5@068E%
/756+.0@E%0@3%?+@8%/8>8@6%0.%?5+>068,502.$%%:&GHI<&>>%J&:%H%&:%)B"K%%
Page 29 of 40
Mary Ann Liebert, Inc.,140 Huguenot Street, New Rochelle, NY 10801
Tissue Engineering
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
+"iRP1–1Fused1DeposiCon1Modeling1
Casted"Materials" Extrac0on"method"
Resolu0on"(μm)"
Alumina,1TCP1 Pyrolysis11 300F5001
Agarose,1Alginate,1PEG,1Fibrin,1Matrigel1
Water1dissoluCon1 5001
S1Bose1et1al.1Materials1Science1and1Engineering:1C,123(4):4791–1486,120031JS1Miller1et1al,1Nature1Materials1Lefers:1Published1online111July120121
with any commercial hardware, as all the machines arecapable of making polymer prototypes. Moreover, this routeuses a traditional lost-mold ceramic processing approach inconjunction with slurry or gel casting.
In the present work, fused deposition modeling (FDMk),one of the commercially available RP processes, was used tomake porous alumina and tricalcium phosphate (TCP)ceramic structures using the indirect route. These porousstructures have three-dimensionally interconnected porosity.The main advantage of this process is the simultaneousdesign of both the micro- and the macrostructure of the partusing the CAD program. During mold making, differentporosity parameters of the polymeric molds can be tailoredby optimizing the diameter of the extruded polymeric fila-ment (called ‘‘road-width’’), the distance between two poly-mer roads (called ‘‘road-gap’’), and the thickness of eachlayer (called ‘‘slice thickness’’) during the layered manufac-turing process. Moreover, by changing the deposition anglesof polymeric roads for each layer, the pores can be orientedlayer by layer. The present work on mechanical and bio-logical responses was carried out with porous ceramicstructures having constant pore sizes but different porevolumes, and constant pore volume with different pore sizes,using alumina and h-tricalcium phosphate (h-TCP)ceramics. Controlled porosity cylindrical structures were
tested under uniaxial compressive loading to understandthe influence of porosity parameters on the mechanicalproperties. In vitro assessment with the OPC1 human osteo-blast cell line was conducted to evaluate the influence ofporosity parameters on cell proliferation.
2. Processing of porous ceramics
Processing of porous ceramics has three steps including(a) mold microstructure design, (b) ceramic slurry develop-ment and (c) binder burnout and sintering. Fig. 1 shows theschematic of the process flow chart for porous ceramicstructures. The molds were made using a Stratasys FDM1650 machine with commercially available ICW06 thermo-plastic polymers. Polymeric molds were made with differentroad-gap and road-width to have different pore size andvolume fraction porosity in the final structures. Slice thick-ness was varied between 0.25 and 0.35 mm, road-gaps werevaried between 0.5 and 1 mm and road-widths were variedbetween 0.4 and 0.625 mm.Mold diameter was kept constantat 1.725 cm. Water-based ceramic slurries were made withhigh-purity 10D (surface area 10 m2/gm) alumina powderdopedwith 500 ppmMgO (Baikowski International, NC) andfood-grade TCP powders (Monsanto, CA) for alumina and
Fig. 1. The schematic of the process flow chart for porous ceramic structures. (a) Polymer filament. (b) Fused deposition modeling process. (c) Polymer mold.
(d) Slurry-infiltrated mold. (e) Porous ceramic structure with a porosity gradient from center to the outside.
S. Bose et al. / Materials Science and Engineering C 23 (2003) 479–486480
with any commercial hardware, as all the machines arecapable of making polymer prototypes. Moreover, this routeuses a traditional lost-mold ceramic processing approach inconjunction with slurry or gel casting.
In the present work, fused deposition modeling (FDMk),one of the commercially available RP processes, was used tomake porous alumina and tricalcium phosphate (TCP)ceramic structures using the indirect route. These porousstructures have three-dimensionally interconnected porosity.The main advantage of this process is the simultaneousdesign of both the micro- and the macrostructure of the partusing the CAD program. During mold making, differentporosity parameters of the polymeric molds can be tailoredby optimizing the diameter of the extruded polymeric fila-ment (called ‘‘road-width’’), the distance between two poly-mer roads (called ‘‘road-gap’’), and the thickness of eachlayer (called ‘‘slice thickness’’) during the layered manufac-turing process. Moreover, by changing the deposition anglesof polymeric roads for each layer, the pores can be orientedlayer by layer. The present work on mechanical and bio-logical responses was carried out with porous ceramicstructures having constant pore sizes but different porevolumes, and constant pore volume with different pore sizes,using alumina and h-tricalcium phosphate (h-TCP)ceramics. Controlled porosity cylindrical structures were
tested under uniaxial compressive loading to understandthe influence of porosity parameters on the mechanicalproperties. In vitro assessment with the OPC1 human osteo-blast cell line was conducted to evaluate the influence ofporosity parameters on cell proliferation.
2. Processing of porous ceramics
Processing of porous ceramics has three steps including(a) mold microstructure design, (b) ceramic slurry develop-ment and (c) binder burnout and sintering. Fig. 1 shows theschematic of the process flow chart for porous ceramicstructures. The molds were made using a Stratasys FDM1650 machine with commercially available ICW06 thermo-plastic polymers. Polymeric molds were made with differentroad-gap and road-width to have different pore size andvolume fraction porosity in the final structures. Slice thick-ness was varied between 0.25 and 0.35 mm, road-gaps werevaried between 0.5 and 1 mm and road-widths were variedbetween 0.4 and 0.625 mm.Mold diameter was kept constantat 1.725 cm. Water-based ceramic slurries were made withhigh-purity 10D (surface area 10 m2/gm) alumina powderdopedwith 500 ppmMgO (Baikowski International, NC) andfood-grade TCP powders (Monsanto, CA) for alumina and
Fig. 1. The schematic of the process flow chart for porous ceramic structures. (a) Polymer filament. (b) Fused deposition modeling process. (c) Polymer mold.
(d) Slurry-infiltrated mold. (e) Porous ceramic structure with a porosity gradient from center to the outside.
S. Bose et al. / Materials Science and Engineering C 23 (2003) 479–486480
LETTERS NATUREMATERIALS DOI: 10.1038/NMAT3357
a
Top view Top view
Lumen Intervessel junction
c
b
d
Vascular unit cell
Interstitial zone
Vascular wall
Lumen
Inlet Blood
Junction
t = 0 s
t = 1.2 s
t = 0.4 s
t = 2.0 s
Interstitium
HUVEC(mCherry)
Vascular wall
10T1/2(EGFP)
HUVECinjection
Intervessel junction
e
Lumen
f
Lumen
Cross-section view Cross-section views
z-position = 10 µm z-position = 300 µm z-position = 950 µm
Figure 3 |Demonstrated control over the three key compartments of vascularized solid tissues. a, Schematic of these three compartments in a ’vascularunit cell’ consisting of the vascular lumen, endothelial cells lining the vascular wall, and the interstitial zone containing matrix and encapsulated cells.b, Patterned vascular channels support positive pressure and pulsatile flow of human blood with intervessel junctions supporting branched fluid flow (left).Spiral flow patterns (right, 0.4 s) are characteristic of non-laminar flow through cylindrical channels. See Supplementary Movies S2 and S3; Scale bars,1mm, left; 2mm, right. c, Control of the interstitial zone and the lining endothelium of vascularized tissue constructs is demonstrated by encapsulating10T1/2 cells (1.5⇥ 106 ml�1, constitutively expressing EGFP) in the interstitial space of a fibrin gel (10mgml�1) followed by seeding of HUVECs(constitutively expressing mCherry) throughout the vascular network via a single lumenal injection (see Methods). After one day in culture a confocalz-stack montage demonstrated HUVECs residing in the vascular space with 10T1/2 uniformly distributed throughout the bulk gel. Scale bar, 1mm. d, Apartial z-stack of two intersecting channels demonstrated endothelialization of channel walls and across the intervessel junction, while in the surroundingbulk gel 10T1/2 cells are seen beginning to spread out in three dimensions. See Supplementary Movie S4 for the complete 700 µm z-stack from d. e, Afternine days in culture, cross-section imaging of a representative channel (optical thickness and z-position= 10 µm) demonstrated that the endothelialmonolayer lining the vascular lumen became surrounded by 10T1/2 cells. Scale bar, 200 µm. f, Endothelial cells formed single and multicellular sprouts(arrowheads) from patterned vasculature, as seen in a z-stack (optical thickness= 200 µm) from deeper within the gel (z-position= 300 µm, left). Evendeeper imaging (z-position=950 µm, optical thickness= 100 µm, right) confirmed that the vascular lumen remained open throughout vessels andintervessel junctions and that endothelial cells also sprouted from larger vessels (arrowheads). See Supplementary Movie S5 for the complete 1mmz-stack from e,f.
tissues or non-perfusable control gels. The ability to access thisvascularization strategy without fabricating the networks in-housemay facilitate rapid adoption of the technology.
Engineered 3D constructs have gained increased attention asin vitro tools for the study of cell–cell and cell–matrix interactions,and are being explored for potential use as experimental models
772 NATUREMATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials
© 2012 Macmillan Publishers Limited. All rights reserved
+"iRP1–13D1prinCng1
Casted"Materials" Extrac0on"method" Resolu0on"(μm)"
PLGA1 Calcium1Reagent1 8001
and from left to right) using the XY coordinateinformation in the worksheet. The total cell numbercounted was normalized to analyzed area. Cell distribu-tion was evaluated in two different sections of each oftwo samples (n ¼ 4/culture time). Sections were takenfrom two different points (center and near edge ofscaffold) along the orthogonal direction of each sample.
3. Results and discussion
The CAD model for intestinal villi and the resultantplaster mold made by 3DP is shown in Fig. 2. The widthof each printed strut was approximately 300 mm, about150 mm larger than that of the original CAD model.
Because of the additional strut width, the inter-strutspacing also decreased correspondingly from 850 mm inthe CAD model to 700 mm in the 3DP mold. Since theaverage particles in the powder bed are approximately20 mm, the additional 150 mm comprises of multiple extraparticles stuck to each side of the strut. This differencemay be attributed to lateral binder migration, and thesubsequent incorporation of loose powder on each sideof the strut. Liquid migration into porous media is awell-documented phenomenon in inkjet printing [15].Briefly, when a liquid droplet impacts the loose powdersurface, it decelerates and deforms from a sphere to alarger diameter disk, imbibes downwards and migrateslaterally into the inter-particulate void space due tocapillarity as the advancing contact angle changes toreceding. Depending on inertial forces, the droplet maysplash during initial impact, and/or create a crater in theloose powder bed. At typical inkjet droplet’s dimensions(o 10"4m), velocity (#1m/s), density (#1 g/ml), andsurface tension (#10"2N/m), low Weber number regimedominates and hence splashing is an unlikely cause for
ARTICLE IN PRESS
Fig. 2. (a) CAD model of mold for villi-shaped scaffold and (b)printed plaster mold.
Fig. 3. Villi-shaped PLGA scaffold generated from plaster mold (a)and SEM image of surface of scaffold (b).
M. Lee et al. / Biomaterials 26 (2005) 4281–42894284
and from left to right) using the XY coordinateinformation in the worksheet. The total cell numbercounted was normalized to analyzed area. Cell distribu-tion was evaluated in two different sections of each oftwo samples (n ¼ 4/culture time). Sections were takenfrom two different points (center and near edge ofscaffold) along the orthogonal direction of each sample.
3. Results and discussion
The CAD model for intestinal villi and the resultantplaster mold made by 3DP is shown in Fig. 2. The widthof each printed strut was approximately 300 mm, about150 mm larger than that of the original CAD model.
Because of the additional strut width, the inter-strutspacing also decreased correspondingly from 850 mm inthe CAD model to 700 mm in the 3DP mold. Since theaverage particles in the powder bed are approximately20 mm, the additional 150 mm comprises of multiple extraparticles stuck to each side of the strut. This differencemay be attributed to lateral binder migration, and thesubsequent incorporation of loose powder on each sideof the strut. Liquid migration into porous media is awell-documented phenomenon in inkjet printing [15].Briefly, when a liquid droplet impacts the loose powdersurface, it decelerates and deforms from a sphere to alarger diameter disk, imbibes downwards and migrateslaterally into the inter-particulate void space due tocapillarity as the advancing contact angle changes toreceding. Depending on inertial forces, the droplet maysplash during initial impact, and/or create a crater in theloose powder bed. At typical inkjet droplet’s dimensions(o 10"4m), velocity (#1m/s), density (#1 g/ml), andsurface tension (#10"2N/m), low Weber number regimedominates and hence splashing is an unlikely cause for
ARTICLE IN PRESS
Fig. 2. (a) CAD model of mold for villi-shaped scaffold and (b)printed plaster mold.
Fig. 3. Villi-shaped PLGA scaffold generated from plaster mold (a)and SEM image of surface of scaffold (b).
M. Lee et al. / Biomaterials 26 (2005) 4281–42894284
M1Lee1et1al.,1Biomaterials,126:4281–4289,1Jan120051
+"iRP1–1BallisCc11
Casted"Materials" Extrac0on"method" Resolu0on"
Alumina1 Pyrolysis11
TCP1
Casted"Materials" Extrac0on"method" Resolu0on"(μm)"
HA,1TCP1 Pyrolysis11 300F4001
Collagen,1Silk,1PLLA1 Organic1Solvent1 200F4001
E1Sachlos1et1al.,1Biomaterials,124(8):1487–197,1Apr120031MJJ1Liu1et1al.,Med1Eng1Phys,1Nov1201111
M1Schumacher1et1al.,1J1Mater1Sci:1Mater1Med,121(12):3119–3127,1Dec120101JM1Taboas1et1al,1Biomaterials,124(1):1811–1194,120031
morphology, as in Fig. 8, but also allowing designcontrol of the macroscopic shape of the scaffold, asillustrated in Fig. 10 for the case of a blood vesselscaffold. Moreover, the process can easily be interfacedwith CT or MRI scans to define patient-tailoredreconstruction shapes. Controlling the internal mor-phology of the scaffold may be the key to successfullygrowing thick cross-sections of tissue because nutrientand oxygen flow through the internal channels may beable to support cellular migration, proliferation anddifferentiation deep inside the scaffold.
Removal of the mould is essential to reveal theinternal features and ethanol is used to dissolve themould material away. Ethanol treatment also exposesthe porosity in the volume of collagen present around
Contamination of collagen from dye in ProtoBuildTM
602 612 622 632 642 652 662 672wavelength (nm)
Abso
rban
ce (a
rbitr
ary
units
) collagen + PB/CPD
collagen + dPB/CPD
collagen
Fig. 7. UV-vis spectra of unprocessed collagen (collagen), collagen films immersed in 0.2%w/v ProtoBuildt and CPD (collagen+PB/CPD) andcollagen films immersed in 1%w/v dyeless ProtoBuildt and CPD (collagen+dPB/CPD). The broad peak at approximately 640 nm present in‘collagen+PB/CPD’, is not found in ‘collagen’ and ‘collagen+dPB/CPD’.
Fig. 8. Artificial vasculature created by (a) designing the CAD model to possess a series of branched shafts with decreasing dimensions, (dimensionsare in micrometers), (b) SEM image of the printed mould; the shafts designed to be 200mm are approximately 220mm, (c) SEM image showingcollagen scaffold with predefined internal channels. The smallest channels in the collagen scaffold are approximately 135 mm.
Fig. 9. SEM image of CPD collagen scaffold. Note the open-cellporosity.
E. Sachlos et al. / Biomaterials 24 (2003) 1487–14971494
Please cite this article in press as: Liu MJJ, et al. The development of silk fibroin scaffolds using an indirect rapid prototypingapproach: Morphological analysis and cell growth monitoring by spectral-domain optical coherence tomography. Med Eng Phys (2011),doi:10.1016/j.medengphy.2011.09.029
ARTICLE IN PRESSG Model
JJBE-1996; No. of Pages 10
M.J.J. Liu et al. / Medical Engineering & Physics xxx (2011) xxx– xxx 7
Fig. 8. (a) Printed thermoplastic mould (after removal of support material) and (b) SF scaffold obtained after using the RP-fabricated mould.
opposed to that of Wang et al. [27] who inferred that a wider peakreflects increased ability to retain water.
In a similar study, it was proposed that the high freezing ratemight lower the opportunity to crystallize as the porosity and heat
kinetics energy are low [28]. To circumvent the issues on low crys-tallinity of SF foams, the current study produced highly cross-linkedspecimens by subjecting the SF foams to boiling ethanol for anhour. Based on the observations from this study, it is suggested
Fig. 9. OCT image of the (a) RP-fabricated SF tissue scaffold, (b) after 21 days and (c) 35 days of cell culture. The arrows illustrate the macro-channels which correspond tothe struts of the thermoplastic moulds.
revealed a slight distortion after sintering resulting in non-
parallel surfaces in the x/y-plane (Fig. 6, 2nd and 3rdsample from right). This may be a result of a density
gradient established during casting or a temperature gra-
dient that occurred during sintering. However, such
distortion could have strong influence on the results of the
subsequent compression testing.All pores had the predetermined shape (Fig. 7). Pore
diameters (Table 1) as well as absolute dimensions of
the sintered specimens differed between the different
Fig. 5 SEM micrographs ofR90 wax moulds, representing ahighly textured surfaceperpendicular to the buildingplatform (a, c). Sintered ceramicspecimens (b, d) show the samesurface texture, indicating theaccuracy of the casting process
Fig. 6 Photograph of R90samples manufactured fromHA, HA80, HA60, HA40,HA20 and TCP (left to right)
Fig. 7 Micrographs of sinteredspecimens show the round andsquare pore geometry of R90(left), S90 (middle) and S60(right, lateral view) samples
J Mater Sci: Mater Med (2010) 21:3119–3127 3123
123
+"iRP1–1main1problem1
• Difficulty1to1extract11the1final1object1– Mechanical1
– Pyrolysis1– (Organic)1solvent1
• IDEA:11CASTING"INTO"LOW"MELTING"POINT"MOLDS"
+"
• Manufacturing1steps1for1investment1casCng1that1uses1rapidFprototyped1wax1parts1as1blanks.11This1method1uses1a1flask1for1the1investment,1but1a1shell1method1also1can1be1used.11Source:11Courtesy1of13D1Systems,1Inc.1
Manufacturing,1Engineering1&1Technology,1Fikh1EdiCon,1by1Serope1Kalpakjian1and1Steven1R.1Schmid.1ISBN10F13F148965F8.1©120061Pearson1EducaCon,1Inc.,1Upper1Saddle1River,1NJ.11All1rights1reserved.1
Investment1CasCng1Using11RapidFPrototyped1Wax1Parts1
+"PAM21
• Modular1CAD/CAM1system1
• A13Faxes1roboCc1stages:1– posiCon1±501mm;1
– velocity10F151mm/s;1
– resoluCon111μm;1
– different1extrusion1modules;11
– layerFbyFlayer1processing.1
3D1roboCc1stage1
Pressure1 Force1 Temperature1 Light1 Tirella1A,1De1Maria1C,1CriscenC1G,1Vozzi1G,1Ahluwalia1A.1The1PAM21system:1a1mulClevel1approach1for1fabricaCon1of1complex1threeFdimensional1microstructures.11Rapid1Prototyping1J12012;18(4):5F51
+"Temperature1Controlled1Syringe1
3D1roboCc1stage1
Pressure1 Force1 Temperature"
Dispenser" DeposiCon1Plane11
Light1
• NTC1temperature1sensor11• The1control1algorithm1based1on1a1
step1strategy1• From1room1temperature1up1to190°1• CalibraCon1with1different1material1
inside1the1syringe11
TCS1extrusion1module1Temperature1Controlled1Syringe1
+"Temperature1Controlled1Syringe1
3D1roboCc1stage1
Pressure1 Force1 Temperature"
Dispenser" DeposiCon1Plane11
Light1TCD1module1Temperature1Controlled11DeposiCon1plane11
• PelCer1Cell1• HFBridge1drivers1• Control1algorithm1based1
on1step1strategy1• ±140°C1respect1to1room1
temperature11
� �
+"
PloFng""low"mel0ng"point"waxes"
+"Drop1formaCon1
d(MV )dT
= γ (πD0 )cos(θc )± 2πµLV
ln DD0
!
"#
$
%&
−Mg
Chang1B1et1al.,1Commun1Nonlinear1Sci1Numer1Simulat1171(2012)12045–20511
+"Drop1formaCon1
1.61
1.71
1.81
1.91
21
2.11
2.21
01 11 21 31 41 51 61 71 81
Drop
"diameter"(m
m)"
Pressure"(kPa)"
T=70°C1 T=80°C1 T=90°C1
• Just1before1drop1detachment:1
γπD0 cos(θ ) = ρgπD3
6!
"#
$
%&
+"System1calibraCon11preliminary1tests1
• Volumetric1shrinking1– CasCng1into1an1aluminium1mold1
• Contact1angle1– Several1material1tested11
0.0731
0.476481
0.3583331
01
0.11
0.21
0.31
0.41
0.51
0.61
0.71
40°C1 25°C1 10°C1
volumetric
"shnk
inkg"(%
)"
01
201
401
601
801
1001
1201
1401
1601
glass1 aluminium1 stainless1steel1
PDMS1 LDPE1 PMMA1 PVC1 derlin1
α (d
eg)"
T=125°C1
Bhola1R,1Chandra1S,1J.1MATER1SCI1341(1999)14883–148941
+"Processing1parameter1in1PAM21
01
1001
2001
3001
4001
5001
6001
01 0.51 11 1.51 21 2.51 31
linew
idth"(m
m)"
pressure(kPa)"
71mm/s1 91mm/s1 111mm/s1
Needle1!15001um1
+"Low1melCng1point1mold1–12D1• Hexagonal1path:1– Side121mm,1linewidth14001μm11
– Side111mm,1linewidth13001μm1
• 60%1W/V1HA1in1gelaCn1gel1(5%1W/V),11crosslinked1with1genipin10.51W/V1
• Volumetric1change1<11%1
+"Low1melCng1point1mold1–12D1• Hexagonal1path:1– Side121mm,1linewidth14001μm11
– Side111mm,1linewidth13001μm1
• Pentagonal1path:1– Side121mm,1linewidth13001μm1
• 5%1gelFcollagen11:11+10.2%1GP1• Volumetric1change1<161%1
GelificaCon1
+"Low1melCng1point1mold1for1microfluidic1devices1• SerpenCne1path1– Length1201mm,1height121
• CasCng1with15%1w/v1gelaCn11gel1crosslinked1with1genipin1
Huang1et1al.,1BiofabricaCon,120111vol.131(1)1pp.10120011
+"
• Hexagonal1path1– Side121mm,1linewidth14001um1
• CasCng1with1PDMS11
Low1melCng1point1mold1for1microfluidic1devices1
SHAPE"DEPOSITION"MANUFACTURING"(SDM)"
+"
• Rapid1producCon1method1with1repeCCve1addiCon1and1selecCve1removal1of1materials.11
• It1uses1convenConal1machining1faciliCes,1hence1also1achieves1the1same1order1of1machining1tolerances.11
• MulCFmaterial1parts1can1be1created1to1compose1funcConal1mechanisms,1also1with1embedded1funcConal1parts1such1as1sensors1and1actuators.11
• CrossFboundary1embedding1is1the1key1for1realizing1highly1integrated1structures.11
hfp://wwwFcdr.stanford.edu/biomimeCcs1Motohide1Hatanaka:[email protected]
Shape1DeposiCon1Manufacturing1
+"Shape1DeposiCon1Manufacturing1
• Developed1at1Stanford1&1Carnegie1Mellon1• Is1it1a1pure1SFM1process?1
1. DeposiCon1F1material1is1added1by1plasma1or1laser1based1welding1techniques1
2. Filler1material1is1deposited1around1part1and1Material1is1shaped1using11convenConal1CNC1
3. Solid1is1stress1relieved14. Components1can1be1embedded15. Filler1is1removed1to1leave1only1finished1
part1
+" Why1not1add1material1in1bulk1and1then1selecCvely1remove?1
+" Why1not1add1material1in1bulk1and1then1selecCvely1remove?1
+"
RP1– Limited1material1variaCon1
– Limited1fabricaCon1tolerance1
– Requires1special1equipment1
SDM1– Wide1variety1of1materials1– FabricaCon1tolerance1comparable1to1convenConal1machining1
– ConvenConal1machining1tools1used.1
– Can1embed1parts1(sensors,1actuators,1reinforcement)1
Comparison1between1common1RP11methods1and1SDM1
+"SDM1capabiliCes1
MulCFmaterial1molding1
Component1embedding1
+"FabricFreinforced1flexural1hinges1
• Lek:1KinemaCc1prototype1of1stroke1extension1linkage1with1311parts1
• Center:1Single1component1SDM1linkage1with1thick1flexures1
• Right:1SDM1linkage1with1thin1fabricFreinforced1flexures1(2001)1
+" Biology1is1a1target1for1complex1integrated1structures1manufacturing11
+" SDM1is1suited1for1complex1integraCon1
• Material1properCes1can1be1locally1altered1by1mulCFmaterial1fabricaCon.11
• Components1can1be1assembled1without1fasteners,1hence1easier1and1more1room1for1complex1integraCon.1
• SemiFautomated1process1allows1detailed1fabricaCon.1
+"CrossFboundary1embedding1
• SelecCvely1adding,1removing1or1otherwise1processing1material1around1the1flexible1strands1without1damaging1them1or1being1hindered1by1them.1
Material1A1
Material1B1
Insert1
+"SelecCve1deposiCon1
ParCally1embedded.1
CrossFboundary1embedded.1
Embedded1component1
Sacrificial1material1
SelecCve1deposiCon1
SelecCve1deposiCon1of1part1material1
SelecCve1deposiCon1of1sacrificial1material1
+" Capillary1effect1for1selecCve1deposiCon1
Example:11Small1stringFsuspended1gimbals1with1two1rotaConal1degrees1of1freedom.1Developed1for1aztude1control1of1solar1panels1on1a1small1satellite1(100mmFside1cube).1
+"SelecCve1deposiCon1
+"SelecCve1removal1
Excessive1deposiCon1
SelecCve1removal1
SelecCve1removal1of1part1material1
SelecCve1removal1of1sacrificial1material1
+"Manual1SelecCve1Removal1
Example:11SpringFloaded1hinge1with1parCally1embedded1coilFspring1and1fiberFreinforced1flexure.1Developed1for1deploying1solarFpanels1for1a1small1satellite.1
+"SelecCve1removal1
+" Material1property1alteraCon1by1postFprocessing1
Material1property1alteraCon.1E.g.1by1heat1treatment,1light1exposure,1or1material1addiCon.1
+" SelecCve1removal1by11photoFlithography1
Example:11A1small1flexural1hinge1with1embedded1electrical1wires.11
+" SelecCve1removal1by11photoFlithography1
