.0 ... .0.. ..0 0.0.0.0.. 0— .00.. E 0. 0 ......0. .0 . . _ . ... .- ...sfifi. ...... . .. .. 2: 0......y.« 0 I0. 0 . 0 06,00“. 0.‘ . .lrr...4u . . ...II 010‘ .l 00. . .. 0 ...0 00 . . . 5...u.. ....“ .. .. “:9... . ...: , 9.3 . 0.0.. 5...}...0. ... . . . . ‘- .. . . . . . 0 33..-... .... )1 . . .30. . . .0 ..0.LV..0. 0...... . . . . . . . 3.0.0.}.0000‘0300030 .0 .. . . .0. . .J 02.0.... . . .. . a 00... ... .. .0.....00. o ..0... ' 0000.,001000000; . .. . . .235... ......0... 3.2.4.... .0000...20. 000,00 I0 _ .0 ..1.....J...€0J agg i .3 . 23...... . ...... 3.01:1..1’0 ..900 . 00.000 . .. . . 00......3¢.300- ..0 000 ...... ..0 . 0. . 02000. 0.. . 00.00 09 .0 0 0 u 0. . '0 0. ...50000 . .00 .000 n 0 0 . .ul‘ 0‘0 . .... 0... .. . . .. . . .... , . . 00.2.... . .. . ..0 31.05.00? .0.- 0000. .. . , .. 0.05.0000! 0.0.0.. 03.0....) ... 0'. 0 0 0 0 0.. 400000000000.“ .. 0 . . 0 .0 . . .0 .0 I . 0 . . . 0.. . 0...0. . 2.00.0000. 0. .... . ... ... 0 9‘00 0. .. 030P§r . ..0 ..0 . . ......0. .. .. .. .... £0.........r.. 9.000,...00. 0..... . 3083;.rh0v: 0G0:- . .0“: 000. . . . . . .....0... ....003... 0 0 .....170. .... n . w. . ... . ......f.J.. . a .0 . an“ 100,. . .3007. 0.230.. .. .. 2.00200... . 0... 1‘. . .0 . u .0. 00. 10.0.0... n .... . ..0J.0..0.0£.0 0 .0 h . 4000. .. .. ...... .. .... . ...; 0.. . .0 . ..0 .. .00. 03.0.0.0... .00 £00.00. . .....vn. 0. L: . n. ...0 . .. 0... 0 0. ... 0. . . «0 .. 1.7.... . x . .... I... . .... 9 . . . . 0. ...0 .. ... 0 00. .0. .0. 000-00 00.003. . . . .L.nv. .01.. 0.0 .uo..0 .. 0.. _ . . v“: .an.0ovbl3’l.. 00.0.0.0 A ‘0... 00 0:. u .....00)‘. .. 0... .5. . "00“,... . “...... . .... .. ....0 . 58.0... 3““... . ‘ 00000 00 0000-1009. .I 0 ~ .. 0.0 .400. 0‘. ......0003 00 .0.0 ”.0 000 .’ .0. 000 .00 .00.. . .‘0. . 0:. 00.0 0 .... . 00. 0.. 00....030. ..0.....0. 70000.0 ... 0 v .00000. I .00 ...0 00. 0.... .0. .u. . 0.. . 0 L 0 n 0. . 0000 ..0 0000 00000000 0000.0 00.000...- 00.00 . 0.0.0 ......0 .0 . 000 0 .... 01. ..0»:.030..00.r. .. . L 0.... 3...... 0003.00.00: 0 00502.03... . ..0'. .0... 00 0... . . .000 .000. .00 0 . 000 u . 0 00.00 0 . 33.0.0 0.0. .0 .1. .0 0.0 ...... .00.. 0.3.? ... 0.0.... 0.0.05.3... . . ... . 300"».0003020: . . . . .. 0.0 \X. 0. 40100.00. .0. 008300.00. . 00.0.. . ...00 .00 0r 00 100500.? 0 ..0...0....000..‘........00 . 140.010.33.73. .810... . 0.. .. ..d...0...0.0...0000 .....0... . .03 ’200. .... ~ . . . ,. . 0 . 3 0. . .... . .. .. . .00.. . 0. 0...! . .... ‘23.. 00 $0300.30 000.0 .00....0w0‘300' .... 05.30. 00 . .. O... 0.70. 0 . ... J. 0 .000 0.!0 ....0. 0.. o. I...‘ 0... 00.. 0.00200. .00. 000.0 .9 0.00000 00. ...0‘ . .00001. 0 0...:00000’00 0.340 000L000 .. .. 201.00.. 0... 0‘. . ... 1.0.03.0. 0 $07201... .3 0 .. . . .0 0. 0.0.0....0...~000. .. . 0..... u. 0 . . . ...-00.0. ....0... . fun ..0 .00. 3.. 0. . . 0. .330 0.. 00... . .. 0. 0.. u: .... . 00 ..0....0...0.. .. ._ ....v . . .. ... . 00.. 00. ...... 0. .. .. 0 . .. . . 2.0. 0.00 0mm. 0.0 00.4.. . .. 00 0 0&0?! 0000 00,00 .0 0000-00.10. 1000.0.- 0 0 . v 0 . \ .... ..0 .0 . v . .00 .... . . . . . .. . 00 00.00 . . . 0 . ...: . 1.0 0. . . . . . . . . u. 0.1 00.00. .00 .00 0 ......3w 3’:- 0 0.. 0 ..0.. .4. .210; I 20.... .... 0.... t . 0 I . f . .21.... 0 . . .. . . ._ . ‘- 3.0.0.00 0. ...000. .00 0.0.0....70fl «0 30. H— 00 00.0 .0 . .0. 0 .. ... . .0 .0- 0.. 70.0.0 .. ...- . . . . . . .. .050. 0 . 0.00: 0.. «0.00.000 ”00... 00¢ 0 03 ,. .0 a I 000.. 0 .00 .. ,3..0.§..0 0.03.1000: .0. . . $000.00.. . 00000:... ..1...’0.000. .000... 0.!!00d' 00. 000.: 4.0.0 .0..O.q0000‘0’.-00 0~ '10. ’u I? 0 0 0 01.0 I... . 0.0 0 030 I00... .. .0... . 224.. .. . . .. . 00. .. 00.0 . .. .. 00.0.00- !IO... 3.000.000.0~.4 0‘ 0 . ‘00.!" 00 0. ..0 0:00:01: . . . .. . 0.0 ... 0 . 0 0.0.. .... .. .0 ...... .0 . _ ...: .. 0.00.0 .0009 .. \ 00 . 00.3.00‘.‘ 000‘ .0005. ’0000 L 0 I. 0 00.0 ..0. 0‘ 0 . 0 f... . .... a [0.0 0 9 0 . 0000.00. . .0 0 . . . .00 ... 0 0 0 ..00 .00. ‘0 . . . .. . . ..00 1.0 00.0.0000; 0 0.00“00 '0 .‘OC‘.00IO§D.- utuzvgldiira 9‘90- . -0... 00 . . . . ...-0.0 . . 0. .. ..0 ......0 ....1 .3. ....v .. ... .....0...00......-.. 0.0..0 .1 _ . . ........0.......0.......00...0 «..- 0...0I..30000.0.00...lu0 01.000 0 . ’00 0 4 00 . . 3.00.... .v . . .. . 1..-...0 ... .0. ... .0 . 00 .. . ... .... .0 ...: . ...... 0..... . . . .... ... 0. w . ... ...! ...... . 0.... ...-.2020... . 3.4.0.0... .. .0 0. .9”. "00. , _ . . 0000. .0... 0.0. ...-'0... 00 . . 10.0 I00 '9 .00..00 . . 0. .7000 0. .0 0... 0.. . 0 . 0.0000 .000 000.. .0 000.1. . 0.000..) ’000u3_00..000 003-0.... 0"}..0-00’50 ‘330 0 '00 000“. 0000. .000, .{:000{ . . . . 0 . . . . ...? 0 . ~ .0. .... .....0. ..0« . . . . . .u .00 u .. 0.1 000.... :0 0.0.1.0. .00 . % 03.0.0 3030.30.08 0 30.1-03.3 _ ... a .z....0. 0000. ... .. . .. 000.0. .. ... .Iz....0...... . . 020.320.... .SIEE 35.000303” _ .0 _ . P. ..J 0 .. . .. 0 .... ...: 0.0. 0.0 O 0 .000 «0.00.0 0.0...090.00. :01} oq 03.0-00.0 7 0‘..$.000 .... 10.. . .. . .. . 0. . . 0.. . . 0 ...: ... ...-...... a . .030. 3... . .- ...): .0203.0....h 9.0.00.0. .1000fill0‘.‘ $000.. “0000 . . . .... 0 r . .... ....a. . t... ...... 01.00. 0*... .. . ...... . ... . . . x ...«..........0....f.0.u0:x.... J.0......-l.:t.. .400. . .. ...0. 2f”. . 0.0 0. 0.. . .00. 3.. . 0.. u ' ‘0 . .0... 0. 0.01.0... .30.- 0...0.. 0 0.0000 0 000.. .01....0 0... . , . . . 03.0 0.0 4 0‘: 000.300.... 0000.0‘0.)0. 0030000022003. 0‘5“. . i 0 . . . .....I . .. ... .0... .0 .. .450... .. ...... . ... . ... .. . ..2...0.....0. 442.00.02.1«03. 0.4—...8. 0"......0.0ra06 91.158.30.11 .g 00 . . . . .. .33.. . ..0.0..... 3700.0 .. ..v ..0.0... 20.0.0 .0 ..000. . . . . ... .. 0 .0. 0 00 .. 300 .. . a .0 0 .. .... , ....q . . . 01...... 0 . o. - ... .. .. . .. . ...: .... . ... _. .. . . . . . . ..0......0...3........Jn.. 0.. . 60.00. h! 1 0..... . 0 00. f a. . . ......100 040... .. 0 011.. 00.0 . . . ..00 0... . .0.— 00 0. . ’0. . .....0 000.000 000000 . . . . . . . . . . , ... 1:1}...r .-..0.’ ... ... 0. . . .... . .. .. ). ...... .. ...? ...: C. .. . 0.13.1.3... . . . . . . _ 3.0. .... ..00 .0 . .. .. .... . ....I... . 0 . . .. . . 2 0.00 . ... 0.... 00.... ... ..20 00 . 3.00.. .000. 0:000. . . .. . ‘00. ...-00.0... . .0000. ‘00 .0 0. 0- ... . 0. on“...- 000... .. .00 00 .0 0 . . . y . . ... . ..0. .. . .. . ... u. ‘. . . . . . . Q0 0000“ ......5. 200.0... . . . . .. . . 0 . .. P000. .... .0..‘..r.0.0.... «3.00 v 00 .000...- 1100.}- . . .... .....9...§ aha-0m: . 00.. D 0 0 I... . c 0 0' 00”.?“00000. 0""..000rru0wnhlh00nnP0530C ..u‘... .~ .000 .0 . ‘00:...0 . 00 5..-... :00... 2 $0.03.. "1:22.000 00. 300).. . ..00...0.....0.3.P.000 Z “a .. ..0“...0 0. .. {000000.13 ...0H 0. 0.... . .07 .00 . . . . 0.0.0.0.. . 0.”..i-008.Iil\a~”rc=n:h. 000.000....50‘...0.000.0 . 00 .0120: 301.300.025.00) . $0.0 .. 0.0.5.0003...0.....00.1 0.. 10.. I .....00: ......0‘ .. 00. . .n. 0| 3.00:0...5 ...... ..0..0 00350000000 0"; .010- 000.. 0. . . 0.00. ...... 32.00.3000: 000.... 0 0 0‘30. .0 ....ll. 0.... .0394 . .. ... o . ... . . .. 0 .0 . 0....0 0 01.00... . 0 .00....00.0~000.r. 0000.03.02 . .0 00.0 3 0.. . 00 5.0.20: 0.. .0. 904001.00 .5000... .... 0. 31‘ ... ..00...0o .. I..400.0... 0.... ... . . 0.. . Y . :000 0 . . 0.... 00.0000. 0 -0010, ..00. . . {03.0.3.0 00.000000000010000 . 0 .0010... 00 . .0 0 .. 00. . 0 .000. 00 . . ..1. \ 0..00...00.0 .0 0.0.. 0.0 00-000 .0000... 0003 0. .03.." : ...... .00 090008063000..- 00.’00001 0.00 O. 05’Cr0‘0-000M0’g‘ .0 . . . . . - ... 2 39.11910... . . _ . . ...: e . 0. . .. . .. 2 . . d.-.0...000........0.0.00~ .0r0wli} V3... ... 0.00.01.10.00.) ._ . .1... . . .. . 0 . 9.. .. . . .0. . ... 1. .0.0I.00.00....00..000.: 510:...) 3.0000"? .0001: 0. >0... .7 1. ...}.J. 3 00 .. . .2. . 00“- 000 a .Ds..\.0000~ .. 0010.33003950 .0 ‘ 00 00.000; . . a .. 0 .. .... .. 0..... .. . .03 0000000. 0... ,. .0..s..0...00.fl..h... . 0..-00. 00200 . 0.00.3030. 000.00.000.00 . 0 J ... v 0.0 ...- .00... 0 00 . ........v..0t. .. .. ..0 i..0..0‘0..010:0.0.;0 3003000. 0.. . 4000.04 000.0... a. 00 400000.- ; . h. u .0 ...: . . .. .. . 0... ...... . 00010....s.....0‘0000.3.:..51.0.3831. ... 1.9.01: 2.. . ... .003 0“- ?!)I .0 . . 1 . ... .. ....00. ...30.1.fi~.0.0.090..0. 0.00.1090... 5.05.0.0}... ...-.000... I0. .0 L00. .\ . . . .... .... . ... .1; 00. . .00. 0....«0.ufi.0..u..000.3 ...-0... 300.0. .g..§003~.00120r1 0. 03-42% . . .0 , . :3. .. ... . . .... 0. 2:810 Eur-.3000. 10.. ... ......3 00.0... 00 «0.3302010. .00... {inf—00m. 00.10 K» t t. I. . o . K 0... . .0. 0.. . .0 ‘00 00. 000. 15.33.300.024. 0003G..0«u.!k0...‘0 0100904003,}..0001. .0. 00003400.. ’“0‘000 . .0, . u x .. 0.2.1.... .....1 0.. 00. 3.00.030... . .. ... .. 3.0.. 3.24 00.00 000.0, 00.... 0.00.0 .. 0... 0.... 00- “..000000050 .55.}..0300090 000:. .0000. .0000000 .00000000 0 . o '0. 0 u. . .... . ... a» 3'0 .11....0 ...00... 0.0100,...» 0...0....00.4.0..000000000.00ll 0. 1 0... 00.59.0005 .0 10.. 0.3.0.3333: 0.0.0.00 0.05.030. .13.. .. ..0.00..00000. 5.00...500’ .....00..0...’.w00v0 00 $0000 .000 00. .0 0.. . . l ..l... . 0.. ...0. . 00.00.. ..0000..)30.0...u0 00.00:: ..00...‘. 000..00..‘00.J 0 :0...“ 0 0 . a. . . ....00! . . ..00.0.0\0..0~00 uni.0’x-..03.S.C~. .0, 0." 0 00.. .0";.. d. . .. . . . .... 0.92:0. 3...... . . . . .... 0002000. 300.0..0011001Z-00 . 00000. ‘..570.°000 0 3.0.0. 0. ... .060 .900 .. 0... .. 0.. . n . 02.00 0.. . 000. . .. . .. in. o .50.}0‘..3.0‘l.r’0.’ 00300 0041 07.0000." g I... .331 ..ll. .. ... ...0. 50.42300: . 0.... 0... .500... . .... .000. ......0000050.0000.0000000 21030.0 9.0 .010: gur~30§ ..t... ... . a. . . 0 o . . .. .. . -20.... .... . . . 0 0.. 0.0.. .. .0H .0 0.0500. ....5'} .- . 3:00.000... ...-.0 3.1.0 . ... .x . . . . . 0.00. :2. 000.000 3.0... 3.00.3033... . 20...... 300720.010. 0.......B...0’0000.01.0.0 Cu! . ... . 0 I... 0 . . . v . . 0.... ... .. .. . ... . .. . .. . ... ...00..0.0..7.030...V}0...0.¢0. 10.000200. . .00 . ... . .. ... 0.... 030...... 0. 0......5 .00.. .... .. 0.0.. . .0..00’0..0..§60..00.0 o........0...‘...0...0. 0.3.000. . . .. .. . 1 I o -0 : . I .. . . .0 .. .. . . . . . . ... .. . .. . .. . . . . . 0......3 0‘0.300.0535500....00..030.0..0.' 0.5.01. -. . .0. ..0. . . .. . . . 00 0 ... . .0 0... .0 ... . 0.0. .. O... ......Q . . 00.00 0 ... 300.1030. . . . . , .‘0‘0‘,‘0"I 00.0.00"0!’50000’0n .000..0. 0.0“ ’00 w. . . .. . . A....... .. 0...... 0 ... .... ... . . . . . . . . 0. 0 0 . . . . . . . . 2 ....”0§.0000...0 003......0003’0 .“uo‘olu 00.. . '0 .. . . . , _. . ..... .. . . ....0 .- ... . . ... . . .00 .00. J... 000.20.100... ...}..000. ...0). ...-{0 1. 11.000.09.00 0.0 O 0.30.0000 n“. 0 o 00. . . .. .... .0. .. “0 .. . . 0 . . . . 0 . . u. . . 000.000.00.001,"}0‘0.§00 a .0 0.. «I 0 0...Da.7.flu..‘ . $0-0." . . .. . . . 1.00 .. .. 00 n 0 0. .0. 0.. . . .. 0. ..0 ...0 0 . 0. ... . . 00 0’0. (.000: 002000... . 0.: 0 j ’.“.d .00“ . . It .. ... ... .; . . . I . .0 ... ... .. ... ... 2 ... 0.....4?‘ ..00 00...S...0000n0.!k005.0.’u.0.1003 . . .00.!0I...’00’10005‘000£ . .. .n . .... .. .. . . .. 0. .. . 0 .. .. 0 . . . l . . : .. ......10... D 5.0) ...-£3003... ... :0 0551.0 ..‘00000. 00 .. . 0 o. . ... . . 200 .. 0. . .0. 0 . 0 . . 00.0. . 0.... 0. ..0 0.0... 000. ...8 0 0 . . . 0.8.0000 . .0“ O P'io‘I; 000 ..v. . ‘ .. . 4. .0. 0 . . .. 0 . .. . .. .. . '30 . .. . . 0‘ ..00... . “0500‘; 0:.‘0’.’ .0. .0 .... 0 0 .. 0 . . .. ... .0 .0 2...... . D .V 0 . .. . . ..00 .0... 0.0....0 00....- .. .0... 0. 0100 ..00...000-“.-.00.”.. ‘7 .... .0 0.....0 I... 0.... .- .... . a. . . . y . . ..0 _. . . . 00 0.00 000.0 . .. 0 I In... . 0.3. . .. . . 0 . .. 0.. . . ....0: . 0 02:0. . .. . 0.00 0.....4... 0.0 0 . 0 .000... .... . .0. ..0 ”v *3 . .... 0 r . . . . 1 0 ... . . . . .. ..0 0 . o 000.. .0.000.0..0..00"0I.. 0 ....0100.0.. . 01¢; ‘ 3. 3" u . .. . . .. . t ... . 0. .. .. «. ... .. . 0,. . 0. 0 0. . 0) 0 .0. 0 . .0.0 . . 0.. . . a... ....000 ..000 I001... .01 .0. . 05...: :11 ... _ 0.0. . . . -. .0. ..l 0 .. . . . . . . .000. . 0.. ..0. . .. .0. I n. 05000 00.0.3. . ‘4 i 000‘ .1300?! . 0‘00010. .0 r I. . I... ..0 .0. a . ... 0.. .. . . 0 .. 0 . . 000 .. 0 . .8 . 0! 0.. ....0..0.. . o. 000 .00. . .. 0300.10.53.50, 00000.0‘1‘00 h» .0 00. .0. - l0. . . . .0.. I . . I .0: .0 .. 00 0.. . .0 . .0. . 0 ..0 .. ......0001001? 01‘ 0.007.001}. . . .5. f'..."" i‘nti 000000 f . 0. 0 a . 0- . .. 0 ...... . .0... 00. .Q . .....t .0 .... . . . . ...0. 0000 .00 00’" 50 g u .. ... . . .. t . . .. - . . .. .. ...... ...“...H. . . .0000 0.30....k‘0. {.2033 .0. 0 Na 0 .0 . ..0. D 1.. . . .. . 0 . ... 0. .. . . 40L '0 .0 ‘ 00 on; k 0. 0 .. .. 0 . . u .. 0. 0 0 0 v. I .. . . 00 00. . .0: . . . n U. 0.00.0gq00“..\ru3v30 r. " ,0: 00.0 J .... .1.-. ... 0 . .. ..0 . .... .... . 0.0 . . o. . . . , . . . . . 0.0000..‘~h000l0u‘u‘..000000.01300'03 0 00.00 a . ..0. . .0 .. . . ... 0 ... 0.. . v .. . 0. . . . . .. . . . . . . .. ., . .0. if...‘ .003 .0 ”£0 1 . . .. . . .. . .0. . . ... . .. .. .0 . .. . . . . . . .. 019...? ... $0001...“ 00. 0.2.0..) 000000.155 ... . .. J . . l .0... 0 .0 . 0 .... . .00.. . 0 . 0.. ...00 .li‘....0v.. 0.3.0.0....00' 0. J 020. . . 3000001.“. 0000003. 00 ’10:.0000’ '5‘ . . 0 . 0 o ....0. . . . . .. .. . . .... . .. 0000...: 4 ...0....l 10.0 01...... . . . 5.71:0 00 , . . . .. . . 01.101001...- {00I‘Nln0 ‘00 " . . .... 5 a a . . . 0.7.... 0.. ..0 0:00. . ... .. . .. .... .0000... . .. .. ... . .07 ... .. . . .. 0.. .0 . . 1’00: .9000“- 00’ 300.70. ‘00-'00 ..-: . .. ... . .. . 0. . ... .. .. t . . . ... . .... 0 .. . . . «0nd; 3.100.034: :13)! 100 ... I . 0 . . X . . ... . u . . .. . v .. . . . . . _ . 00.1.2.0...02.0..00|00.0i.......0.)|..1|0..0‘3 r000. ... ...... . .. . . . 0 . . . . . . . . .... . 0 . I01. 00...!o..0.8\.-00.‘.0.000..... 10.0.4 .0" .00 _ . . . .0 . . ...-0:...0. 1;... . . ... . . 0 . 0.....0 0.. 0. 09.0.... 0..! 1...}:.J‘...s$)........0.0§u.0 “.00 . . . .. . .00 ... . 0. . . 3 . ... . -.. 0 .0 0 v . .. . ... 0. . .. .0 .00 .0. ‘.000n ) 0.0.51.3. 50‘ i 3 .‘0‘. .3 .0.. ... .. ..0 . 0.3. ..- .. .. ... . ..00 ..0 . u. .0 .n. .. . 1.. 0 0 . 0 0 .. . . .. .0 .. v. 0 .... . ... ... u. I . . . .0 o 00 A ... 0 0. . .... . . 0 .‘ .03 00.0000 003001.03 . . v 0 . :0 . . a 0 . .. . . . . 0. 0.00000 0 .0 . . .4 a .01 0 v. ’ 00. .. 3.00.. .J . 0 . .20.0’000...§0.0 . .0, . ... . 0 ... 0 .. o . I ... . 2...... £00.02... 4000 d.’0.u0...0l...:. . 000 0.‘.0‘m03 0000‘!‘\0.} . . . . . l v : 1 .. .. .... 0 A .. ......0 . 9 . 00.. .. 000.00%...” $0.00 100.030.200 €30.30. 0.03.0... 00.... ...-00.00.0000! 0. . . . ... . . .. . . i 0 .. . .. . . ... a . a . .. .. . : ... . . . .. 0.3.311.” .... {...-00.3.0.5}..39030. .020...) 00.... (0.3.0.. 0... . . . . . . . . . . . . y .7... . .........0..00.00.1..5.0..’.........0..... 0000.03.00.00.K..fi.‘ 00.0.3000}. .1. . 1 . 9 .0’0. . .. 0 . 0 . . .05 . . ..0 .. u .. . ... . . 0 . .. . 8.0.. . 0.300%.)030. oz?0\0...“00 .100.3;..(000.O‘0 .‘u.00.00.0.‘000.0‘00 0.... . . . . 0 0 .. . . . .. .0. . .. .0 .04 ... .0 . .00 ...! .003... . 3 . . .. 3.9.65.0; 00!.‘33-{3..00..‘..;.00’0.0- .‘p000.‘00000.0‘30.0.0‘ 08 . . . ..v 0.. . . . . .0. . ..v .. . . . .. 0.0.0 0.. 0.20.0 ..00 .... 0.. 11...... . 0. .0 ...: .. .0500. 0.000.. ...)...0...00000...0 . :0. 0.00.0000...0.....l0..i00..!0|. 0000.01.13.19 .13003005000032‘11‘ .0. . .0 0 . . 0 . . . .. ... 0 . .. ...0 0.....0 ... . 0 0.00! .. . 0.. . .00 ....0 .0. ... ..0 ... .... 0.. 0.00.0":.L00!|000. .0..0.000‘)\100 1“}:00‘I 0. 00 . . . 0....0 0 .00 .. ..00 . ...-0 ‘0 .... . . . .. . . ... 0.. t . 0.0 . 4 . ... .00... . . .. . .. .‘.0" 000000. g0..“"0’10‘ . . .. . . ... . .- . . . .. ...; o . .000. . ... 00 0 . . ..0. . 3.0. . CK..J.0.1 . . ...7 0... . .l’.000(.9|0..0{l00§|,0{0 ..0 _ .. ... . . a . .....l ... . 0.; ..00 0 . 0 .. . .. .. . ... .. .0 . ...... ..0 . ..0...~.. 0.000.01I..000.00..0_000000 . 0 0m 11 . . 0 . .. . 0 . . 0.. .l... .. . . . .0.. . . . .0 . . . .. .. .... $0,!!!i0iufln. .1 - . 0 . . . . . ......o . . . .. .. . . z . .... 0.. .00..)0...‘.2.00.0. 0. ...-l . ..0 . 0 . 7 0 . . . 0 .00 .. 0 . 0 0 00‘s..- ‘. ‘0- 01,02... . .. .0. o. 0 0 n . .00 0 n 0 ..0000. .00....0‘0033: .. ... . 0-0.4.0333. . .0 . . .. . . v 0 .I . .. .. 0 0 . 0-0.0. ... .1 000.0 0 Iv . .0 0, A. 00.. 0.0000‘00...‘,‘0’..0£.0000 .. .. . . . . ... a . 0 0. . ..l 0.. .00.... 0.0.4000. . 0.01.0 {.0503} 610.00.55.00. I . 0 o 0 O 0 r. . 0 0 .0 .0...'00.0. . {0.0) 0.0.. .0, .0 '03.},u000000000,000.0..d0l¢00 53.00000. . Q . . . 0 ... 0 0 00 .... .....\.0 07.0.3.0..0 ...Jl‘.0...... . ... 0 ... .. . . . 0 -... .0 .0 .o 0 . ‘0 . . 0:400... 0.03.3.3. 0... .0 3.0.000. 0000001.. 000000003 0 0 .. 1 . n. . . 0.00 000 . .0 00.2000.0.0.00.000.000.0000000.a... 30‘0‘li0- 0. 0 . 0 . . . . .0. a . 00 ‘00. . .. 00.0.0100! . ..0.000...!0000 0. .0) 10.0 0- 300-0000... >0...)..l..\.3 ’30::- 0.’.00\.“0.0A’ ...... . . . 0 . . . . ... .. ...... ’00.. . .0? 00... 0. . ... . . I‘llg5t¥‘.ni 00 0.0 0 _ 0 o . .. . 4 .. .0... . ..08 . .0. . ....30‘011100'fi03. . . ., .. . .0 . . . . . 00.. '0' . ... 0 0 .00 . 00.. . . . 0 . .00.. . . '- . . . 0 . .0. o . 0... . . 0 0 0.000.-.) 0...-.0‘0' . a. n 0.. . . .. 0 . . ... 10".. .0 . 00. . .0 ...-0 0 0 0 A . 0 . . . ... #000,010. .01 ... ... . 0. . _ . . . . .. ... . . .0. 5!... ‘ C . .0 .. 0 I ...-‘0. .. 00.00.00. 1.0.0 0.. 0.00..00:0.0 .- IVA .. .0. . . . . 0.000 ......k. [0:03.000 .. . . u .u I . .0 ..0 . . ..., I. . 0.0 .. .0... e 8 0 ... . .0 0100.0 . . .0 . 3... ... 30.23101... .02 . ‘00:... I.’0.00.V.0.\..0000’0H1..)000 W . u 0 . . . a . ... 0. . . 0 9.0 . o .0000. . . . .00 0 .100...- O.. 0010. 0000. 00‘300‘. I. 0.00.55.403‘0‘3005J 0.000‘10000'000‘0 . 0 0 . . 00 . . 0 ..0 0 n . . . ... .. 0. 0 . .0000‘ .000. . 00.00 00003000.. 0v0v‘00000’000.€4 0.00.0. 000 0 0. 00.9,..‘00300’.Ii . . 0.. . o0. . 0 0 0 . 13......0 0.. 0 . .00.....0. r0. ...0. . :fJ' .0.0100.0.0..0 0.0000. 0‘. 0.0.0.0’0000 0“ 000.0 .‘ 0 ... .. . . .. ... .... ...... ... . 1.00.0 .. s . 2.9.0.7.. 01.3.3.7... .0. . ... 0.0m\ .000.\. 0.0- . 0 . .. .. ... . . .0 .... .I? .. .. .. ... ...... ..0 ...; 3.092500000300013. 0000.:0f..~$0r§ 0.00. . .. . . . n ... . 0 . . . 0... 0 .. . . .. . _ . 0.0.‘(00045‘007100' £0.‘ .‘0,00. 00‘ ... 0 . ... 0 . 0 0 .0 . 0,0 ....00:90003. .0 30000.... .80 0. . ... . .. . . . . . . 5.... .....z .-....00... _. . ... .0 0 .. . 00. 0 . . 0 .b . . .. c . . 0 . . i .....0 20.0.0.1. .. .. I . .u . .y 0 . ....lu... .0...v.\0. ... .900 0‘0. 000 u _ 00.. . 0 . 0 v 00..0 . . ....0 . .00. . .. . 0 u . .. .. S . .. . .. 0 ...: ... _. I. 0 00. .0. 000 . 00.0 . 0 v . . .00.)..0. . .....l . ..0 .. ..00 00.0..0 . . 0v . . . 0 .0... 0 . . .. . . ... .0 . . ...0 u 0 0.0. u 00 .00 .... . 0. ... . . ..0 0. 0.. . 000 0. 00 0 .. ... .0 . .0. 0.. 0 .. . . .n _ . . . . ...... ....0. 0 . .. . 00 .... .. . 0....0 00......... 0.0 00.. 0.0.0.-.. ..000....0. . . .0. . .. _. v. . . . u . h p .. 0 . 0 .. . . .. 0 . . . a . . . . .. . . .. .0’.’0..03?. . . ..0. . .0 0 ... . . 0 v 00 .. . ... ..t 0 . 0 ......0. .1 0.0 0 00 . 0 ..0.0.0 0. . . ... ... 0000.0. 000.0233. .. . ...0 .0. 0‘0.)..0.1(J0 0. u. . 7.. 0 o . 0 . .‘ .0. v. . . .. . . .0; .. m 0 0 . ...v... ..0 0.. 0. I. 00 . .0 . . .... . u. 0 0 a 0 . 0 ..0 . .. L . . Q 0 . 0 — . ..0 . n .0 0 ... 0 .0 0 . 0 00 J 0 00 0 v . . _ ..0 0.0 0 .010 .. l . 0 .....00 I 00.0 ... 00...‘.. .0 ...... . w ... ... ..00 ... .. . .. .0- 0 .. I a . . 0.. .. . . v v . 0 . . 0 0. 0. a . 0 .. 00 00 . .. . .0. .. 0 . V0 00. 0 . 0. . . . .. . . 0 .. 0 . 0 . .... v . .. . 00 . ...: ..000. .0.0 . n. _ F 0 .0 ‘. .0... 0. 0.0.0.0.. . 0. .. 0.. A . 0 0 .0 .u 0 0. 0 0. 0 0 v ..0. 0. . ... .... . ....00 .3. 0 ..0 0...! . .0 .... .00. ..0.0..00I.0 .10 I i .. .....0 . 0 .0. .. .. 0 u . .. .0: . . . .- . .0 . . . . v. .. .0. . ......0. . .‘L c 0 . ...... . $0.0... . .........00 ...0.. ..0 . 0 .. . 0.. .0. ... .. I 00.. . . .. . 0 . 0.0.0. 0. 0 0 . . _ T v a 0 u . 0 . _ 1:00.. . .. .11 I t . .. . 0 00 . . .. ... .0A‘0A: “ 0 .0. O 00. _. . ..00 .0. .00.... 0 0.. . X . .0. ... . 0 . .. . .0 . . . 0... ._ .... .03 .0 .0. .00. . O. — ..0..0;0 . :0... . ..0..0. 0.....0. . .0. 0......0. . ..0... .00!‘000.0‘00 .' 0.. . .0. . . 0. . . .. 0. i . .0. r 0 .. . . . 0. 0 ..0... ...000000 0.. .. 0 ... )‘0 00 0..0 J00. 0 .0.. .. 10,... I... . .. ...... .... . ... 0. .0 ... . 0. ...00.0000 . .. . 0 . . .00. .0 . 0..... .. .. . .. . .u .. 0.0.: ..‘v 03: ..00..- it’0 _ . . ... . ..a . . . . . . . . . . . 0. .. . . 0. Z . . . .. 00 .- 0.0000'.£.00 00.00 ...00.0l&\.l.000.00.‘.01 . . . . .. . . 0 .00 . 0 0 0 . 0 . .. . . . . . . . . 0.. 1.00.00 Y..0\l 00000 l.). .l?’ .. ...... . o . .. ..0 . . . .. .. . .. .. . ....0... ......6. 0.1.5.0. ....00 00 u 0.. 00 . ....00). 0 I a. . . . . .5, . . . . . . . . . . . . . . . . .020 5‘0 ‘. t 0" .V...’ :0. .0)", V .. . ... 0.0.. 0 . ..0..0 0 . . . .0 ..0- .. .....0. .0... 0 0 .. 0 ..0..l ‘0 0 .000 5‘ 0". .|\ . ... ....0..0. .0. . .p . .0...0.. . 0.0....0 .. 0. .... 0.... 0 ... 0 0.0003 .. .. .. ..00\0000. 00-100 0. . t . . . . 0 ._ . 0. . . . 0 a ...... 0.. .. .0.0 00. -0004 {0.0.}. . 5?’1..0...,... 00$ . n 0 .. . ~.. .. .0 00. . .10 . ...... .. . . . . ..0h0...\0.0.0.0..0 ..0-0!‘I~.000.0v0.0.. ... 2.50.010: 9.. 5‘ I .00‘ IV\..41. . ..0 0.00. 00. .0. 0. . ... . .. A. . ...0... 35.4.. . ..00 0... .. n . .. _ .... .. .. . . 0 . 0 . I .0 . .000 .. . .. .. .000. . . x .. 0. 0. ..0 0 u .. 000.. . ......000 0‘. . . . . . . .0... 0 .0 0. .93.. n . . 00 .0 0. 0 .0... T .. . .... 0 . .. 0. .....0 .... ....0 . .00 ..V00 0. .. . \. . ... . 4. .v . 0 . .0 .. .. .0 .. ... .. 0 . . . ..0. .... .0... . . . ..v. .0... . ..0' s. .. 0 . . . . .. .13. .00.. 3.. . 10.0 00.. .... ,0 . . 5 . 0 .... ... . . ... o .. . . ... .00... 0 0 . 00 I 0 O Q 0 0 C O . u 0 l 0 0 . '00 I. .. . 0 0.0.. g . 0'40\ 0 . 0 00 0 . .0 . . .000.- 0. . . .0 .0. ... . .0 o . . . . . .0 0 . . .. . . . . . . .. ... 0. v . . 0 0 .. .. 0 .. . . . 0 . 0 .... .. 0. .0 . ..0 ... 0. _. .u. . . .. .. .. . . . . . . . . . .. . . . .. 0 0.0 . 0. .. .0 ...0 ..0‘0 . _ .. . o . .0 .. .0. 0 0 . .. 0 _ 0 0. . I .0 0 I 0 . . . . _ . . . . . . . . .. . . . . . .. . . 0 . . . 00.0 . . . . .. 0. ..00 . .. O 0 0s. 4 .l.. .. ...0 . . . . . . . . . . . . . .. . .. . . . . ... .. . ... . ..t . . . . . . . v0. .. ... .0 . . . . .. . . . . .. . . . . . 0. _. Y - V. .. 1. 0 .. .... . .. . . . . o. 0. . .. . .. . 2.0.0. . . .. . . . . . s . 0 .00.. . .. .0 ... .. .. . . .0 . .. .. . . ... . ..0 ..0 . .. . .. ... ..0 . 0. .. . . 0.. .0 .v .. 0... ...0 0. . ...-0 ..0 o 0. . ... . . O . I 0 r 0 .0 . 0 .. . . .. . . . . ..~ . .. . ..0 . . . . .. .. . ..... . .. _ .. . .00 . .‘0. 0...... 20 . .... 0.. .. 0. 0.... .. . . . .. ... .. 0 0 . . 4... .. 0 .. .. .. . . . . . 0. . s. . 0 . 0...‘ . ..n. . . o .. . 0. 0 . I A. 0 .0 . . .0 . l .v 0 00000. . 0 ..- . . . . . A1 . . 3.1. 0.. .0 .. .... .0 0'. a . . .. I... ......0.0.0.0.. ..o i . . .. .. .. . . ... n. . . ... .. 0 ..00 . ... T . . . .. . . .......0. . .. .. 0 . . .. 0 . . .....f. 9.0 .. . . . ..00 .0 0. .0 .0 0.! .. J . ..0.v .. .0 00. . . ... . . 0 . ..0. . 0... . 0:0 .0 . . u . . ll ......0 I . . 0 n . .. . . n . . 0. ..0 . . _ . . .. 000.. . . 0 . u 0. 0. 0 . 0 .. . . . ... 0 0 A u . : I0 . 0 u . . . 0. .. r. . I . 0... . ‘ . 0 0 . ..0 . . ... . .. ... ..0... 0. . . 0 ... . .. . v .0 ..0 . . 0 0. I 0 . 0. . ' I ... .00. .0 ...... ‘p. . .... .0.. .. . . . . . .00. .. . ... ... u .. .. . . . . 0 . . ... . on. . ... .. . A . I 0 0. . 0.. . . . . . f . . . . . . . . . . . 0 . . . q . . .. . .. . 0 0.. .303». . ... .000- . . . 0. ... 0. (”a 0 0..—.... .....0. 2 . . f... . . .v‘m...w0.0“0 ..0m.».. ..0 . 0 W.0....T..0.0 02.0.. . . 03.30... .0 .0 0 Q ....0 0 I I.“ . 0 J0 .0 0.0.0 0 I. 0.000 00 530‘ .... .00... . . .I... .0" .- :xanamu 1%"... .. mm. ....maaérnmfl ...... ...0: 100.61.... .00 .1 00.0 0.. 0.0 . . 0 a .0. p 0.0:: 00.....0 . ... . 0 0.10 . 0- 0 .... Z . . 0 . n .. o 0. 0 “3.1.00.1: 1.0. 0.3-714. 0810.0. .0 0&th 061.0 ...0 .... 0 .0 .0 ..0 . ....0.. ..0 .00... 00. .0: .. 0.... 0. 0 0.»... 10.00 J0 O0. . 0 .0 1V... .. a .. .0 0.....u....0 . . . .....33. .. . . .11.. ... .. 11.” .19... L9.- 0...0 .0. .00 A... . .1...0...0...0 ..0.010.0..0.0. \.. .00. . .. ... 0. 3.0.... 0.. ...030 .00 ..00..0 ......3 o 0 .0 ...010 .... . .0 .. .0 .l .0. \ Q0001. 0 0 000 0.0.0 0.0. 000 .0000 n. . \0.0-0 . .00.... 00.. 0 . 0 J (A -1) LIBRARY Michigan State University This is to certify that the dissertation entitled DEVELOPMENT AND CHARACTERIZATION OF FUNCTIONAL POLY(LACTIC ACID) COMPOSITES FOR PACKAGING APPLICATIONS presented by ISINAY EBRU YUZAY has been accepted towards fulfillment of the requirements for the Ph.D. degree in Packaging "-o---o-|'—.——‘--n- Major Professor’s Signature fly ’75} 4762/0 Date MSU is an Affirmative Action/Equal Opportunity Employer PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE A . 0 ‘ ragtime 5/08 KzlProleoc&PrelelRC/DateDue.indd ...—fin .___”_-_—_-- _ - - ...-....— DEVELOPMENT AND CHARACTERIZATION OF FUNCTIONAL POLY(LACTIC ACID) COMPOSITES FOR PACKAGING APPLICATIONS By Isinay Ebru Yuzay A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Packaging 2010 ABSTRACT DEVELOPMENT AND CHARACTERIZATION OF FUNCTIONAL POLY(LACTIC ACID) COMPOSITES FOR PACKAGING APPLICATIONS By Isinay Ebru Yuzay In this research several efforts have been directed to develop bio-based polymer materials with high performance and/or functionality and extend their applications to new packaging areas. The focus of the research is to investigate the interactions between mesoporous three- dimensional (3D) framework structures, zeolites, and aluminum oxide coating layers with bio-based poly(lactic acid)(PLA) polymer. It has been demonstrated that zeolites and thin aluminum oxide coating layers due to their unique features have great potential as functional fillers/materials for the PLA matrix that can be used not only in packaging but also in a plethora of applications. The initial part of this research provides an understanding about the fabrication method of PLA/zeolite composites and their effects on the existing properties of PLA. The second part focuses on aluminum oxide barrier layers grown with atomic layer deposition (ALD) and sol-gel techniques to form ftmctional PLA composites. It was found that PLA/synthetic zeolites, type 4A, and PLA/natural zeolites, chabazite, can be successfully fabricated using extrusion followed by injection molding. The morphological studies showed a homogenous dispersion of zeolite type 4A in the PLA matrix. As stress propagated through the composites, zeolite particles remained embedded into the matrix, indicating the existence of good interfacial adhesion between the zeolite particles and the PLA matrix. Incorporation of both type 4A zeolite and chabazite into PLA had significant effects on the cold crystallization and percent crystallinity of PLA. Both Type 4A and chabazite zeolite particles drastically accelerate the hydrolytic degradation of the PLA via surface hydrolysis. The apparent activation energies of thermal degradation of PLA and PLA/zeolite composites estimated using the Flynn-Wall-Ozawa and Kissinger models were found to increase in the order PLA/ type 4A < PLA/chabazite < PLA, facilitating the feedstock recyclability of PLA. In addition, ultrathin aluminum oxide (Ale) coatings were successfirlly deposited on PLA surfaces using an atomic layer deposition (ALD) technique at 42°C. The thickness of the coating was found to be 32 nm. The AFM results obtained from small scanning areas (10 x 10 umz) revealed that the surface appeared to be as smooth as neat PLA. The newly developed films exhibited remarkable improvement in gas barrier properties and good transparency. A stable aluminum oxide colloidal solution was also prepared through a sol-gel process. PLA pellets were coated with the aluminum oxide sol-gel solution using a dip-coating technique. The deposition of aluminum oxide on the PLA pellets was confirmed by FTIR and SEM-EDS. The atomic ratio of AVG obtained was found to be O.51:l:0.02. The aluminum oxide coated PLA pellets, particles obtained from the sol-gel solution, and calcinated aluminum oxide (or-A1203) were also used to form PLA/Ale composites through extrusion and injection molding processes. SEM studies of PLA/A10x composites showed that Ale particles were dispersed in the PLA matrix uniformly using melt compounding. It was found that Ale particles used at low loading level, i.e., 3 wt%, did not alter the melting temperature of PLA; however, the shape and intensity of the melting endotherrns were significantly changed, suggesting bimodal melting behavior. The results obtained from UV-visible transmission spectra also suggested that PLA/Ale composites may act as a better UV barrier than neat PLA. Com/right by ISINAY EBRU YUZAY 2010 To my parents, Sevgi and Fuat Yuzay and my brother Atalay Yuzay, whose hearts and prayers were always with me during my graduate studies ACKNOWLEDGMENTS During my graduate studies, I had lots of encouragements, words of wisdom, and gracious contributions from many people that made this research possible. But some of these people deserve special mentioning. I would like to thank my advisor, Dr. Susan Selke, for not only guiding me and believing me through the tough times, but also for giving me the opportunity to fulfill my dream. Her insights and helpful discussions made me a better researcher, writer, and critical-thinker. I am just hoping that I have learned a little bit of her endless knowledge and experience. I also would like to thank my dissertation committee members; Dr. Bruce Harte, Rafael Auras, and Dr. Thomas Pinnavaia for their deep insight and encouragement. Center for Food and Pharmaceutical Packaging Research (CFPPR) provided partially support funding for this research. This support is greatly acknowledged. Finally, I sincerely express my whole hearted gratitude towards my parents Sevgi and Fuat Yuzay along with my brother Atalay Yuzay for being extremely supportive throughout my life; without their love support, and reassurance, I would not be standing on my feet. vi TABLE OF CONTENTS LIST OF TABLES ................................................................................. xi LIST OF FIGURES ............................................................................... xiii CHAPTER 1 Research Motivation and Goals ...................................... , ............................... 1 CHAPTER 2 Background ......................................................................................... l4 Poly(lactic acids) ............................................... . ............................ 14 PLA Based Nanocomposites ............................................................ l6 Zeolites ....................................................................................... 25 Chabazites as a Model Natural Zeolite .................................................. 38 Literature Review on Pertinent Topics .................................................. 39 Thin Layer Deposition Techniques ................................................... 4O Atmospheric-Based Coatings /Liquid Coatings ....................................... 40 Aluminum Oxide ........................................................................... 47 Diamond-like Carbon (DLC) ............................................................ 52 REFERENCES ...................................................................................... 55 CHAPTER 3 Poly(lactic acid) and Synthetic Zeolite Composites Prepared by Melt Processing INTRODUCTION .................................................................................. 68 EXPERIMENTAL .................................................................................. 69 Materials .................................................................................... .69 Preparation of Composites ................................................................ 70 Characterization of Composites .......................................................... 73 Scanning Electron Microscopy (SEM) .................................................. 73 Transmission Electron Microscopy (TEM) ............................................. 73 Atomic Force Microscopy (AF M) ....................................................... 73 Tensile Properties ........................................................................... 74 Thermal Properties ......................................................................... 74 Differential Scanning Calorimetry (DSC) .............................................. 74 Thermogravimetric Analysis (TGA) .................................................... 75 Dynamic Mechanical Analysis (DMA) ................................................. 76 Optical Properties ........................................................................... 76 Melt Flow Index (MF I) Determination .................................................. 77 Molecular Weight Determination... . ............................................... 77 Barrier Properties ............................................................................. 78 Water Vapor Transmission Rate (WVTR) ..................................... 78 Oxygen Transmission rate (OTR) ...................................................... 78 vii Carbon Dioxide Transmission rate .............................................. 78 RESULTS AND DISCUSSION ................................................................... 78 Morphology .................................................................................. 78 Tensile Properties ........................................................................... 82 Thermal Analysis ........................................................................... 91 Dynamic Mechanical Thermal Analysis ................................................ 96 Optical Properties ......................................................................... 101 Melt Flow Index (MF I) Determination ................................................ 103 Molecular Weight Determination ...................................................... 104 BarrierProperties................. .......................................................................... 107 CONCLUSION ..................................................................................... 108 REFERENCES ...................................................................................... 109 CHAPTER 4 Effects of Synthetic and Natural Zeolites on Characteristic Properties of Poly(lactic acid) Composites INTRODUCTION ................................................................................. 1 13 EXPERIMENTAL ................................................................................. l 13 Materials ................................................................................................... l 13 Preparation of Composites ............................................................... 115 Characterization of Composites ......................................................... 114 Morphological Properties ................................................................ 1 14 Scanning Electron Microscopy (SEM) ................................................ 114 Fourier Transform Infrared Spectroscopy (FTIR) .................................... 114 Differential Scanning Calorimetry (DSC) ............................................. 114 Dynamic Mechanical Analysis (DMA) ............................................... 116 Optical Properties ......................................................................... 1 17 Molecular Weight Determination ................................................... 117 RESULTS AND DISCUSSION ................................................................. 118 Morphology ............................................................................... 118 FTIR Analysis ............................................................................. 121 DSC Analysis ............................................................................. 125 Dynamic Mechanical Thermal Analysis ............................................... 129 Optical Properties ........................................................................ 134 Molecular Weight Determination ................................................................... 135 CONCLUSION .................................................................................... 138 REFERENCES .................................................................................... 139 CHAPTER 5 Effects of Synthetic and Natural Zeolites on Thermal Degradation Behavior of Poly(lactic acid) Composites INTRODUCTION ................................................................................. 140 EXPERIMENTAL ................................................................................. 143 Materials ................................................................................... 143 Preparation of Composite Samples ..................................................... 143 CHARACTERIZATION ......................................................................... 144 viii Thermogravimetric Analysis (TGA) ................................................... 144 RESULTS AND DISCUSSION ................................................................. 145 CONCLUSION .................................................................................... 164 REFERENCES .................................................................................... 165 CHAPTER 6 Hydrolytic Degradation Behavior of Poly(lactic Acid)/Zeolite Composites INTRODUCTION ................................................................................. 168 EXPERIMENTAL ................................................................................. 169 Materials ................................................................................... 169 Preparation of Samples .................................................................. 169 CHARACTERIZATION ......................................................................... l 70 Weight Loss ............................................................................... 170 Morphological Evaluation ............................................................... 170 Scanning Electron Microscopy (SEM) ................................................ 170 Differential Scanning Calorimetry (DSC) Analysis ................................. 170 Molecular Weight Determination ...................................................... 171 RESULTS AND DISCUSSION ................................................................ 172 CONCLUSION .................................................................................... 180 REFERENCES .................................................................................... 181 CHAPTER 7 Atomic Layer Deposition of Ultrathin ALOX Films on Poly(lactic Acid) INTRODUCTION ................................................................................. 182 EXPERIMENTAL ................................................................................. 183 Materials and Methods ................................................................... 183 RESULTS AND DISCUSSION ................................................................. 187 CONCLUSION .................................................................................... 188 REFERENCES .................................................................................... 189 CHAPTER 8 Poly(lactic acid)/ Aluminum Oxide Composites Fabricated by Sol-gel and Melt Compounding Processes EXPERIMENTAL ................................................................................. 190 Materials ................................................................................... 193 Methods .................................................................................... 193 Sol-gel Preparation ........................................................................ 194 PLA-Aluminum Oxide Composite Preparation ....................................... 195 CHARACTERIZATION ......................................................................... 197 Scanning Electron Microscopy (SEM) ............................................... 197 Fourier Transform Infrared (FTIR) Spectroscopy .............................. 197 Optical Properties ......................................................................... 198 Differential Scanning Calorimetry ...................................................... 198 X-ray Diffraction (XRD) ............................................................... 199 Thermogravimetric Analysis (TGA) .................................................. 199 Dynamic Mechanical Analysis (DMA) ............................................... 199 ix Molecular Weight Distribution .................................................. 200 RESULTS AND DISCUSSION ................................................................. 200 Morphology ................................................................................ 200 FTIR Analysis ............................................................................. 202 Optical Properties ......................................................................... 205 Thermal Analysis ......................................................................... 206 Thermogravimetric Analysis ............................................................ 212 Dynamic Mechanical Analysis ......................................................... 214 Molecular Weight Distribution ......................................................... 216 CONCLUSION .................................................................................... 218 REFERENCES .................................................................................... 220 CHAPTER 9 Conclusion and Suggestions for Future Work ............................................ 224 x Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 3.7 Table 4.1 Table 4.2 LIST OF TABLES Comparison of material properties between neat PLA and PLA with 4 wt % trimethyl octadecylammonium-modified montrnorillonite, adapted from Sinha Ray et al[30] Comparison of flexural modulus, strength, heat deflection temperature, and 02 gas permeability of neat PLA and nanocomposite films, adapted from Sinha et a1[35] Common zeolite types, polymer matrices, and their application areas ............................................................................... The kinetic diameter of some common molecules, adapted from Ullalandms], Semenova [139], and Cecopieri-Gomez et al.[140] Typical polymer zeolite composite membranes perrnselectivity values compared to the neat polymers, adapted from [85,130,137] .............. Comparison between uncoated PET bottles and Plasmax®coated PET bottles.[192]............................................. WVTR, OTR, COZTR values for A1203 ALD ............................. Micro extruder and injection molder processing parameters. . . . . . . . . The effect of zeolite content on tensile properties of PLA. N1, N2. . Thermal characteristics of PLA and PLA/zeolite composites.Nl, N2. The heat deflection temperature (HDT) of PLA/zeolite composites. N1, N2... ... .............................................................................................. Melt flow indices (MFI) of PLA and PLA/type 4A zeolite composites. GPC results of PLA and PLA/type 4A composites. ....................... Gas permeability values obtained from neat PLA, PLA/zeolite composite films, and commercially available zeolite containing film. Thermal characteristics of PLA and PLA/zeolite composites. . . . . . . GPC results of PLA and PLA/zeolite composites ........................... xi 18 2O 27 31 34 46 51 72 83 93 100 104 102 107 127 137 Table 4.2 Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 7.1 Table 8.1 Table 8.2 Table 8.3 GPC results of PLA and PLA/zeolite composites ........................... TGA analysis of PLA and PLA/zeolite composites. Heating rate: 20 C’C/min, in nitrogen flow ......................................................... TGA analysis of PLA and PLA/zeolite composites. Heating rate: 20 oC/min, in air flow Apparent activation energies of PLA and PLA/zeolite composites obtained by the F lynn-Wall-Ozawa and Kissinger methods (in nitrogen flow)................. .................................................................................. Apparent activation energies of PLA and PLA/zeolite composites obtained by the F lynn-Wall-Ozawa and Kissinger methods (in air flow)............... ..................................................................................... Oxygen and water vapor permeability of PLA and aluminum oxide coated PLA films ................................................................ Thermal characteristics of PLA and PLA/Ale composites. . . . . TGA analysis of PLA and PLA/Ale composites. ......................... GPC results of PLA and PLA/Ale composites. xii 137 147 163 163 149 188 206 213 217 Figure 1.1 Figure 1.2 Figure 1.3 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 LIST OF FIGURES Structures of optical isomers of lactic acids, lactide monomers, and poly(lactic acid) ............................................................... 2 Schematic representation of ALD cycle (layer by layer film growth ABAB..). Adapted from [50] .............................................. 7 Schematic of aluminum oxide sol-gel coated PLA production ....... 9 The basic tedrahedral building units of zeolite framework a=>b=>c Zeolite framework formation , adapted from Dyer[73] and van Bekkum.[67]......... ........................................................... 24 Schematic representation of typical zeolite pore sizes, adapted from [67,74] .................................................................. 25 Schematic diagrams of zeolite/polymer membrane morphology, adapted from [137,142] .................................................... 32 Framework structure of zeolite type 4A (d). Building unit of type 4A with 4-rings (a); Building unit with 6-rings; Building unit with 8-rings with the pore size of approximately 4 A.[159].................. 38 Framework schematic of chabazite (c). Building units (a, b) [159].. 39 SEM micrograph of type 4A zeolites. Magnification: x5000; Scale | bar: 10 um .................................................................... 70 DSM twin screw micro extruder (A) and injection molder (B). . . 71 Injection molded neat PLA and PLA/ zeolite composites. (A) PLA neat; (B) 1 wt% zeolite; (C) 3 wt%; (D) 5 wt% ................................. 72 TEM image of zeolite. Magnification: X 40,000; Scale barz500 nm. 80 SEM micrographs of neat PLA and PLA/zeolite composites. (A) PLA neat-200 (scale bar = 100 um) and 1000 (scale bar = 20 um) magnification, (D) PLA with 5 wt % zeolite- 200 (scale bar =100 pm), 1000 (scale bar = 20 pm), 3300 (scale bar = 5 pm), and 6000 (scale bar 5 pm) magnification ............................................ 81 xiii Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 3.11 Figure 3.12 Figure 4.1 Figure 4.2 Figure 4.3 Comparison of experimental tensile strength data with the Nicolais-Narkis model for PLA/zeolite composites. Experimental tensile strength data (0) and Nicolais-Narkis predicted data (I) with the adhesion K value of 1.21 ........................................ 87 The tensile stress of PLA/zeolite composites in the linear form of Eq.(3). The average interaction parameters (B) determined from the slopes of the straight lines. Note: The 95 percent confidence interval of the mean B values are 5.7, 0.7, and 1.1 at the zeolite volume fractions of 0.01 126(1 wt%), 0.03369(3 wt%), and 0.05601(5 wt%), respectively. 89 Experimental and predicted relative tensile yield stress (oc / om) of PLA composites as a function of zeolite volume fraction: - Experimental data; straight line according to the Pukanszky equation with B20; dotted and dashed lines according to the Pukanszky equation with B:3.0 and 6.0, respectively ................... 90 Representative DSC therrnograms of PLA pellet, PLA extruded, and PLA with 5 wt% zeolite composites (Heating Ramp: 10 OC/min) ....................................................................... 94 XRD patterns for PLA and PLA/zeolite composites. A: PLA; B: PLA/1 wt% type 4A; C: PLA/3 wt% type 4A; D: PLA/5 wt% type 4A ........................................................................... 96 Storage modulus (A), loss modulus (A), and tan delta (B) of PLA and PLA/zeolite composites as a function of temperature.(a: — PLA); (b:_ _ PLA with 1 wt% zeolite);(c: _ . .. PLA with 3 wt% zeolite); (d:— - - — PLA with 5 wt% zeolite).............. 98 UV/vis transmission spectra of PLA and PLA/type 4A composites. Sample thickness: 1: 0.15 mm, 11: 2 mm A: PLA; B: PLA/1 wt% type4A; C: PLA/3 wt% type4A; D: PLA/5 wt% type4A zeolites ......................................................... 102 Injection molded neat PLA, PLA/5 wt% type 4A, and PLA/5 wt% chabazite composite. ........................................................ 114 Photographs(l) and SEM micrographs(ll) of type 4A (a) and chabazite (b) zeolite powders. (In SEM micrographs, scale bar: 10 um ............................................................................. 119 SEM images of PLA and PLA/zeolite composites. xiv Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 4.10 Figure 5.1 Figure 5.2 Figure 5.3 Figure 5.4 Figure 5.5 A: PLA magnification: 1000X, scale bar 20 um; B: PLA/type 4A magnification: 1000X, scale bar:20 um; Cl: PLA/chabazite magnification: 200X scale bar: 100 um ;C2: PLA/chabazite magnification: 1000X scale bar: 20 um. Arrows indicate the zeolite particles. ..................................................................... F TIR spectra of (A) type 4A and (B) chabazite powders FTIR spectra of (A) PLA, (B) PLA/type 4A, and (C) PLA/chabazite composites ................................................. DSC thermograms of PLA, PLA/type 4A, and PLA/chabazite composites at a cooling rate of 0.5 C’C/min ............................... Storage and loss modulus of PLA neat (A), PLA/5 wt% type 4A (B), and PLA/5 wt% chabazite (C) composites as a function of temperature ................................................................. Tan delta versus temperature for PLA neat(A), PLA/ 5 wt% type 4A(B), and PLA/5 wt% chabazite(C) composites ..................... Typical HDT curve obtained from 3-point bending test on DMA for PLA (A), PLA/5 wt% type 4A (B), and PLA/5 wt% chabazite (C) composites ............................................................ UV/vis transmission spectra of PLA (a), PLA/type 4A (b), and PLA/chabazite (c) composites. Thickness: 2 mm. . . . . . . . . . .. TGA curves of type 4A, chabazite, and neat PLA. Heating rate: 20 o C O O C/mln, 1n mtrogen atmosphere. ........................................ Representative TGA and derivative thermogravimetric curves of PLA (A), PLA/type 4A (B), and PLA/chabazite (C) composites. . . -1 . . . Heating rate: 20 °C'm1n , 1n mtrogen and an atmosphere Representative TGA curves of PLA, PLA/type 4A, and PLA/chabazite composites at heating rates of 5, 8, 10, 15, and 20 °C/min, under nitrogen flow ................................................ log [3 vs 103 T" curves for (A) PLA, (B) PLA/type 4A, and (C) PLA/chabazite in nitrogen, using the F lynn-Wall-Ozawa method... Apparent activation energies (Ea) at different conversion values for thermal degradation of PLA, PLA/type 4A, and PLA/chabazite 120 122 124 128 131 132 133 134 146 148 151 156 Figure 5.6 Figure. 5.7 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Figure 7.1 Figure 8.1 composites as obtained fiom the Flynn-Wall-Ozawa method in nitrogen atmosphere ........................................................ Apparent activation energies (Ea at different conversion values for thermal degradation of PLA, PLA/type 4A, and PLA/chabazite composites as obtained from the Flynn-Wall-Ozawa method in air atmosphere. (*Experiments have not been performed on chabazite samples under air) .......................................................... Kissinger plots of PLA, PLA/ type 4A, and PLA/ chabazite composites. (with three replicates; the range of R2 values: PLA=0.9884-0.9999, PLA/type 4A= 0.9947-0.9976, PLA/chabazite=0.9829-0.9923 ) in nitrogen ............................ Weight loss data of PLA and PL/composites hydrolyzed at 23, 60, 90 °C in water and 0.01 N NaOH solution ........................... SEM micrographs of PLA and PLA/zeolite composites in water at 23 and 60°C, scale bar 20 um ............................................ SEM micrographs of PLA and PLA/zeolite composites in 0.01 N NaOH solution at 23 and 60°C, scale bar 20 um ...................... Weight average molecular weight data for PLA and PLA/zeolite composites at 23, 60, 90 0C, in water and 0.01 N NaOH solution... Number-average molecular weight data for PLA and PLA/zeolite composites at 23, 60, 90 C>C, in water and 0.01 N NaOH solution... Percent crystallinity of PLA and PLA/zeolite composites in 0.01 N NaOH solution at 23 and 60°C, scale bar 20 um ....................... Picture of the ALD system (a) and schematic diagram of the precursor injection system (b) ............................................. SEM images of PLA and PLA/Ale composites. A: PLA magnification: 200X, scale bar 100 um, B: PLA/SGl magnificationz200X scale bar:100 urn, C: PLA/SG2 magnificationz200X scale bar: 100 um, D: PLA/SG2 magnificationz600X scale bar:50 um, E: PLA/CALC magnificationz200X scale bar: 100 um (Arrows and circles indicate Ale particles) ........................................................... xvi 158 159 162 173 175 176 177 178 179 185 201 Figure 8.2 Figure 8.3 Figure 8.4 Figure 8.5 Figure 8.6 Figure 8.7 Figure 8.9 F TIR spectra of (a) Ale sol-gel solution (SGl), (b) dried Ale sol particles (SG2), and (c) calcinated aluminum oxide particles (CALC) ...................................................................... FTIR spectra of (a) PLA, (b) PLA composites with SGl, (c) PLA composite with SG2, and (d) PLA composite with calcinated aluminum oxide (CALC) ................................................... UV-visible transmission spectra of PLA and PLA/Ale composites. Thickness: 2 mm ............................................. DSC melting curves of PLA and PLA/Ale composites with a heating rate of 10 °C/min. a: PLA (lst heating cycle); b: PLA/SGl (lst heating cycle); c:PLA/SG2 (1 st heating cycle ); d: PLA/CALC (lst heating cycle); e: PLA (2nd heating cycle); f: PLA/8G1 (2nd heating cycle ); g: PLA/SG2 (2nd heating cycle); h: PLA/CALC (2nd heating cycle) ........................................................ Figure 8.7 X-ray diffraction patterns for the injection molded PLA and PLA/SGl ................................................................ TGA curves of PLA, PLA/8G1, PLA/S02, and PLA/CALC composites. Heating rate: 20 °C/min, in nitrogen atmosphere ........ Temperature dependence of storage and loss modulus (I) and tan delta (11) of PLA (a), PLA/8G1 (b), PLA/SG2 (c), and PLA/CALC ((1) composites ................................................................................. xvii 203 204 205 208 211 213 215 CHAPTER 1 Research Motivation and Goals An increased awareness of environmental issues is driving the desire to use bio- based polymers. Poly(lactic acid), PLA, is one of the most widely used bio-based polymers derived from renewable resources such as corn, sugar beets, wheat, and non- food biomass feedstocks (e.g., forest waste, agricultural waste which is an envisaged source in the near future). PLA is prepared by either condensation of lactic acid or ring opening polymerization of the cyclic lactide dimmer [1-3]. Having two optically active configurations for lactic acid (L- lactic acid and D- lactic acid) and three configurations for lactide (L-lactide, D-lactide, and meso-lactide) (Figurel.1) helps to tailor the molecular architecture in the final product, which directly affects the properties of the polymer; for example, depending upon the D/L ratio the resultant polymer can be semi- crystalline or completely amorphous. PLA has the advantage over traditional polymers that the production of PLA requires less energy and generates fewer greenhouse gas emissions than does that of traditional polymers [4,5]. In addition PLA can be compostable under specific temperature and relative humidity conditions. Currently, the largest PLA production plant in the world is operated by NatureWorks LLC in the US. with a capacity of 140,000 tonnes/year in 2009 [6], followed by Hisun Biomaterials Co., Ltd with 5,000 tonnes/year in China [7]. In November 2009, Futerro started up test production of PLA with an annual capacity of 1,500 tonnes in Belgium [8]. Pyramid Bioplastics Ltd. has also started up a 60,000 tonnes/year PLA plant in Guben, Germany, which is expected to be completed by 2012. Pyramid claims it will be the first and biggest industrial scale PLA production plant in Europe and its 60,000 tonnes/yr capacity is already close to sold out [9]. Several other smaller PLA producers are setting up facilities in Europe, Japan, and China. L-lactic acid O | I gfla C ;- I c l CH3 '3. \ C/ CH3 'C / H 'C 5"- g" \C \ / a l l _ I. c H o "E | l O L-lactide Mesa-lactide D-lactide /‘ \ H O 'r-ro -—c—c— o T—H \CHa _/ n Poly(lactic acid) Figure 1.1 Structures of optical isomers of lactic acids, lactide monomers, and poly(lactic acid). The increasing demand for PLA is mainly due to environmentally conscious company policies, governmental regulations to avoid fines or the fact that it is becoming price competitive with traditional polymers. For over 20 years, PLA has been mainly used in medical applications such as sutures, bone screws, and tissue implants; however, recent technological developments and the reasons mentioned above have made PLA widely available for commodity markets such as the textile, packaging, and automobile industries. Nowadays, several research efforts are directed to develop PLA based composites with high performance and/or functionality and to extend PLA’s current applications to new areas. The incorporation of various nano/micro inorganic fillers such as layered silicates (e.g., talc, montrnorillonite, mica, and other clays), calcium phosphate, and calcium carbonate, etc., have been shown to result in significant improvements in the properties of PLA [1,4, 10-12]. Recently, there has been increased interest in a different category of rnicroporous minerals, e.g., natural and synthetic zeolites, used as functional fillers in polymer matrices to enhance existing properties. Zeolites are crystalline porous nanostructures with pore sizes varying from about 3 to 15 Angstrom. In general, the structure of the zeolite consists of SiO4 and A104 tetrahedra, which build up a network of charmels and cavities. Extensive research has been done on natural and synthetic zeolites due to their intrinsic chemical reactivity, adsorptivity, and ion exchange capacity [13,14]. It is well documented that zeolite’s distinctive pore structure can accommodate different types of transition metal ions, gases, and liquids, such as C02, 02, N2, CH4, H28, NH3, VOCs, odors, etc [15-20] These features of the zeolites make their polymer composites functional. For instance, sodium ions present in zeolites can be substituted by silver ions (Ag+) and incorporation of silver ion-exchanged zeolites into the polymer matrix results in good antimicrobial properties [21]. Using various transition metal ions or dopants in zeolite pores also alters electrical conductivity properties of the polymer composites [22,23]. Selectively permeable zeolites are also being used in active packaging applications, particularly for fresh produce and vegetables which rapidly consume the oxygen in the package headspace, causing the deterioration of produce and vegetable quality [21]. Therefore, inclusion of zeolites into PLA holds promise for a wide range of polymer applications. To date, incorporation of zeolites into a PLA matrix for packaging applications has not been utilized. In this study, an attempt was made to develop an effective melt processing technique and evaluate the feasibility of using zeolites as fillers for the PLA matrix, which could bring potential new packaging applications. Zeolites have been used with various polymers, such as poly(dimethylsiloxane), PDMS[24], poly(ethersulfone), PES[25], poly(imide) [26], poly(aniline), PANI [27], low-density poly(ethylene), LDPE [21], high-density poly(ethylene), HDPE [28], poly(propylene), PP [29], poly(vinyl alcohol), PVOH [27], poly(vinyl acetate), PVAc [30], ethylene vinyl acetate, EVAc [31], poly(styrene), PS [32], cellulose acetate, CA [33], and poly(carbonate), PC [34] for gas and liquid separation purposes. In these studies, often a very large amount (up to 40-50 wt%) of zeolite was added to the polymer matrix. Most of the studies discussed the necessity of surface treatment of zeolites for better dispersion in the polymer matrix for improved properties [35]. Other studies argued that surface treatment may cause pore blockage, resulting in inferior gas and liquid separation [3 6]. Therefore, it was the purpose of this work to fabricate PLA composites containing a low amount of zeolites (0 to 5 wt%) in a twin-screw extruder, thus facilitating dispersion, and to evaluate the effect of zeolites on the physical, mechanical, thermal, and barrier properties of the PLA composites. In this research, the next approach to develop functional PLA composites was to explore the possibility of using an ultra-thin metal oxide coating film as a protective layer for PLA substrates in order to improve the gas barrier properties. Metal oxide coatings have been widely utilized as protective layer for a variety of applications including the fabrication of circuits, gas- and bio-sensors, fuel cells, coatings, solar absorbers, and insulators [37,38]. As the demand for the high barrier coating applications such as food, pharmaceutical, and medical packaging has increased, the technology has highly developed fi'om an aluminum foil or aluminum coated flexible substrates to glassy-like transparent metal oxide coatings, e.g., silicon oxide (SiO x) and aluminum oxide (Ale), which can meet high gas barrier requirements [39-41] In the literature, there are several studies concerning metal oxide coatings on polymers such as polycarbonates (PC), polyethylene terephthalate (PET), polymethylmethacrylate (PMMA), polydimethylsiloxane (PDMS), polyimide, polypropylene (PP), etc.[4,6,15-17] While in most of these studies SiOx was deposited on the polymers, Ale has received little attention. Yet, Ale has remained one of the most widely used metal oxides for non- polymeric surfaces due to its combination of wear resistance, high thermal and chemical stability, and electrical resistivity [42]. In addition, Ale coatings exhibit good biocompatibility and tribologic behavior in arthroplasty applications [42,43]. Therefore this part of the study aimed to evaluate the feasibility of coating biobased polymer, PLA, with an Ale layer. A variety of conventional deposition techniques including chemical vapor deposition (CVD), plasma enhanced chemical vapor deposition (PECVD), and RP. magnetron sputtering can be used for Ale coatings [44-49]. However, a common drawback of all these techniques is the brittle and less conformal nature of the resultant oxide coating layers [40,41]. In this study, the atomic layer deposition (ALD) technique was chosen as a means to conformally deposit Ale layers on the surface of the PLA substrate. ALD offers superior tmiformity of film thickness and flexibility on any substrate geometry. This unique feature of ALD makes it attractive for various industrial applications. ALD is based on sequential, self-limiting surface reactions [50]. The two precursors are introduced to a surface of the substrate in alternating pulses; therefore, their reactions take place on the surface of the substrate [50,51] Figure 1.2 shows a typical ALD cycle for a simple two precursor system. The details related to the ALD process can be found in chapters 2 and 7. 3m mafia FIT-FEW Figure 1.2 Schematic representation of ALD cycle (layer by layer film growth ABAB. . . . . ...). Adapted from [50]. For the last approach, we investigated the sol-gel coating process to develop fimctional PLA composites. In 1996, Kunitake’s [52-54] group introduced the surface sol-gel process for the preparation of ultrathin films of metal oxide-based gels, which enables the control of film thickness on hydroxylated surfaces. The film growth is achieved by repetition of the saturated adsorption of alkoxides and subsequent regeneration of a hydroxyl surfaces. They also claim that irrespective of the shape and size, if the substrate surface is modified with hydroxyl groups, this process can be used on various material surfaces to obtain controlled film growth of metal oxides. In the literature, numerous publications have reported the utilization of the sol-gel process on surfaces such as metallic surfaces [52-54], latex polymer surfaces [55,56], cellulose [57,58], and mesoporous silica materials [59,60]. The resemblance between the surface sol-gel process and ALD as well as it being inexpensive alternative to ALD led us to use the sol-gel process. Therefore this part of the study aimed to apply an Ale barrier layer directly on the surface of the PLA pellets using a modified sol-gel method,[61-64] which can be considered the most cost-efficient deposition technique. This study was also undertaken to achieve non-traditional core-shell polymer systems. A sol-gel coating technique was used to create a core-shell polymer system where the PLA is the core phase and Ale the shell phase, which is opposite of the traditional method. After coating PLA pellets with an Ale barrier layer, PLA pellets can be extruded using a twin screw extruder and then injection molded. During the extrusion process, the Ale shell on the PLA was expected to be broken and dispersed in the PLA matrix. A similar approach was used by Liang et al.[65,66] who coated micron-sized high density polylethylene (HDPE) particles with aluminum oxide by ALD in a fluidized bed reactor and then extruded thecoated HDPE particles into pellets or films, which resulted in good dispersion of aluminum oxide flakes in the HPDE matrix. It was expected that these flakes (broken shell particles) might create a torturous path for permeant gases and improve the barrier properties. However this concept needs to be proved. Figure 1.3 shows the schematics of the proposed concept. PLA matrix PLA pellets i- - . '.~ .2 i . » I ‘ .l‘ ' 4 \- L . L '1} Ls". , . ' ' 'i‘LL‘ . t , .. .-'.' r “- _ I." . . I ;t n .- . v‘ 1.3+ ‘ 'f ‘ u U 4' - , . _ . r :1 r. ‘ -' ‘3: . ,: ‘v'f 5 ,"I'! . - ' r‘ .: -. ' 21:. ~— . v.3: . I II". ~‘. 3 ’ :5:- F . '. {Ha I . :2? r \ - ' 2: . 11?.“ ll . ‘ ‘7 ' . _ . .‘ A‘s: . i ..z . . ’ , . f’ ‘ . 1"." ,. - . ' c 1 1 ..n'f“ , j . . . . . - "‘ " = -. ;'.' " r: -A-"ilfl' 0_ . ‘. . Uniform or non-uniform So called “ Broken egg shell” effect Figure 1.3 Schematic of aluminum oxide sol-gel coated PLA production. The research discussed here was primarily devoted to exploring the use of synthetic and natural zeolites, and aluminum oxide barrier layers grown with ALD and sol-gel techniques to form functional PLA composites. Since the combination of PLA and the materials mentioned above is a new research field, it is clear that there is a tremendous lack of fimdamental knowledge about the interactions among synthetic and natural zeolites, Ale, and the PLA matrix as well as lack of basic understanding of the prOperties of their composites. Therefore, this research overall aims to fill this gap. REFERENCES [1] [21 [3] [4] [5] [6] [7] [81 [9] [101 [11] [121 [13] [14] [15] [161 R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4, 835. B. Gupta, N. Revagade, J. Hilbom, Prog Polym Sci 2007, 32, 455. E. Rudnik, Ed. Compostable polymer materials, Elsevier Oxford 2008. S. Sinha Ray, M. Bousmina, Prog. Mater. Sci. 2005, 50, 962. M. Murariu, A. Da Silva Ferreira, P. Degee, M. Alexandre, P. Dubois, Polymer 2007, 48, 2613. NatureWorks,Available at http://www.natureworksllc.com/news-and- events/press-releases/2009/03-1 2-09-manufacturing—location2 .aspx.Access date; 12/10/2009. H. Biomearials, Available at htLp2//www.plaweb.com/english/CoOrganize.a_sp . Access date; 12/10/2009. Futerro, Available at http://www.futerro.cont/newsdetails.php?id=10. Access date; 12/10/2009. P. Bioplastics, Available at http://www.pyraplast.corrr/index.html. Access date; 12/10/2009. N. Ogata, G. Jimenez, H. Kawai, T. Ogihara, J Polym Sci Part B: Polym Phys 1997, 35, 389. S. Sinha Ray, P. Maiti, M. Okarnoto, K. Yarnada, K. Ueda, Macromolecules 2002, 35, 3104. S. Sinha Ray, S. Pouliot, M. Bousmina, L. Utracki, Polymer 2004, 45, 8403. R. M. Barrer, Zeolites and Clay Minerals as Sorbents and Molecular Sieves, Academic PresszLondon, 1978. R. T. Yang, Adsorbents: Fundamentals and Applications, Wiley: Hoboken, N.J., . 2003. N. De Kruijf, M. Van Beest, R. Rijk, A. Sipil, T. Inen-Malm, Food Add Contain 2002, 19, 144. T. Andersson, M. H. Holmgren, T. Nielsen, B. Wesslen, J Appl Polym Sci 2005, 95, 583. 10 [171 [18] [191 [20] [21] [221 [231 [241 [251 [26] [27] [28] [29] 1301 [31] [321 [33] [34] T. Andersson, T. Nielsen, B. Wesslen, J Appl Polym Sci 2005, 95, 847. S. C. Hodgson, R. J. Casey, S. W. Bigger, Polym-Plas Technol Eng 2002, 41, 795 R. M. Podhajny, "Efficacy of antirnicrobially treated substrates", presented at TAPPI PLACE Conference, Indianapolis, IN, 2004. D. H. Shin, H. S. Cheigh, D. S. Lee, J Food Eng 2002, 53, 229. A. Brody, E. Strupinsky, L. Kline, Active Packaging for Food Applications, Technomic Publishing Co, Lancaster, PA 2001. C. Chuapradit, L. R. Wannatong, D. Chotpattananont, P. Hiamtup, A. Sirivat, J. Schwank, Polymer 2005,46, 947. E. Vitoratos, S. Sakkopoulos, E. Dalas, P. Malkaj, C. Anestis, Curr Appl Phys 2007, 7, 578. G. Clarizia, c. Algieri, E. Drioli, Polymer 2004,45, 5671. Z. Huang, Y. Li, R. Wen, M. M. Teoh, S. Kulprathipanja, J Appl Polym Sci 2006, 101, 3800. T. W. Pechar, S. Kim, B. Vaughan, E. Marand, M. Tsapatsis, H. K. Jeong, J Membr Sci 2006, 277, 195. B. V. K. Naidu, S. D. Bhat, M. Sairam, A. C. Wali, D. P. Sawant, T. M. Aminabhavi, J Appl Polym Sci 2005, 96, 1968. H. Kim, J. Biswas, S. Choe, Polymer 2006, 47, 3981. F. Ozrnihci, Balkose, D., Ulku, S., J Appl Polym Sci 2001, 82, 2913. T. T. Moore, T. Vo, R. Mahajan, S. Kulkarni, D. Hasse, W. J. Koros, J Appl Polym Sci 2003, 90, 1574. Y. Yan, Y. Zhao, M. Liu, J Appl Polym Sci 2005, 96, 1460. M. J. A. Sales, S. C. L. Dias, J. A. Dias, T. A. Pimentel, Polym Degrad Stab 2005, 87, 153. J. M. Duval, A. J. B. Kemperman, B. Folkers, M. H. V. Mulder, G. Desgrandchamps, C. A. Smolders, J Appl Polym Sci 1994, 54, 409. D. Sen, H. Kalipcilar, L. Yilmaz, J Membr Sci 2007, 303, 194. ll [351 [36] 1371 [33] [39] [401 [411 [42] [431 [44] [45] [46] [47] [481 [49] [50] [51] [52] [53] R. Mahajan, W. J. Koros, Polym Eng Sci 2002, 42, 1432. T. T. Moore, W. J. Koros, J Mol Struct 2005, 739, 87. M. N. Rahaman, Ceramic Processing. Boca Raton: Taylor & Francis , 2007. G. Kickelbick, Prog Polym Sci 2003, 28, 83. H. Chatharn, Surf Coat Technol 1996, 78, l. M. Benmalek, H. M. Dunlop, Surf Coat Technol 1995, 76-77, 821. C. P. Barker, K. H. Kochem, K. M. Revell, R. S. A. Kelly, J. P. S. Badyal, Thin Solid Films 1995, 259, 46. E. Dorre, H. Hubner, Alumina: Processing, Properties, and Applications; Springer-Verlag:Berlin,1 984. X. Liang, S. M. George, A. W. Weimer, N.-H. Li, J. H. Blackson, J. D. Harris, P. Li, Chem Mater 2007, 19, 5388. S. J. Noh, S. K. Lee, E. H. Kim, Y. J. Kong, Curr Appl Phys 2006, 6, 171. M. Ritala, K. Kukli, A. Rahtu, P. I. Raisanen, M. Leskela, T. Sajavaara, J. Keinonen, Science 2000, 288, 319. J. W. Elam, Groner, M.D., George, S.M., Rev Sciec Instrum 2002, 73, 2981. M. D. Groner, F. Fabreguette, J. W. Elam, S. M. George, Chem. Mater. 2004 16, 639. L. Zhang, Jiang, H.c., Liu, C., Dong, J.V., Chow, P., Journal of Physics D: Applied Physics 2007, 40, 3707. A. Heyman, C. B. Musgrave, Journal of Physical Chemistry B 2004, 108, 5718. S. M. George, Chem . Rev 2009, 110(1), 111. B. S. Lim, A. Rahtu, R. G. Gordon, Nat Mater 2003, 2, 749. I. Ichinose, H. Senzu, T. Kunitake, Chem Mater 1997, 9, 1296. I. Ichinose, H. Senzu, T. Kunitake, Chem. Lett. 1996, 831. 12 [541 [55] [56] [57] [58] [59] [60] [61] [62] [63] [641 [65] [661 Y. Aoki, T. Kunitake, A. Nakao, Chem Mater 2005, 17, 450. Y. Li, T. Kunitake, S. Fujikawa, Coll Surf A 2006, 275, 209. S. Fujikawa, T. Kunitake, Langrnir 2003, 19, 6545. J. Huang, N. Matsunaga, K. Shimanoe, N. Yarnazoe, T. Kunitake, Chem Mater 2005,17, 3513. Y. Aoki, J. Huang, T. Kunitake, J Mater Chem 2006, 16, 292. W. Yan, B. Chen, S. M. Mahurin, E. W. Hagaman, S. Dai, S. H. Overbury, J Phys Chem B 2004, 2793. D. Sun, Z. Liu, J. He, B. Han, J. Zhang, Y. Huang, Micropor Mesopor Mater 2005, 80, 165. S. Y. Lee, J. D. Lee, S. M. Yang, J Mater Sci 1999, 34, 1233 K. Tadanaga, K. Kitamura, A. Matsuda, T. Minarni, J Sol-Gel Sc Technol 2003, 26, 705. Q. Hu, E. Marand, Polymer 1999, 40, 4833. J. Wen, Mark, J. 13., Journal of Applied Polymer Science 1995, 58, 1135. X. Liang, L. F. Hakim, D. Zhan, J. A. McCormick, S. M. George, S. M. Weimer, J. A. Spencer, K. J. Buechler, J. Blackson, C. J. Wood, J. R. Dorgan, J Am Cerarn Soc 2007, 90, 57. X. Liang, D. M. King, M. D. Groner, J. H. Blackson, J. D. Harris, S. M. George, A. W. Weimer, J Mern Sci 2008, 322, 105. 13 CHAPTER 2 BACKGROUND Poly(lactic acids) Recently, interest in studies of biodegradable and compostable polymers has arisen. The most important group of biodegradable polymers is composed of aliphatic polyesters where the biodegradation mainly takes place on the hydrolysable ester bonds. PLA is one of the members of the family of aliphatic polyesters. It is produced from the cyclic dimer (lactide) of lactic acid, which is mainly derived from renewable resources such as corn, sugar beets, and rice [1]. Lactic acid (2-hydroxy propionic acid), which is known as the simplest hydroxy acid, has been used in polylactide production. Lactic acid has a chiral carbon atom, resulting in two enantiomers, the L(+) and D(-) lactic acid. The ratio of the enantiomers plays an important role in the lactic acid based polymers, determining the mechanical properties, biodegradability, and eventually the end use of the material [1-6]. In general, lactic acid is made by anaerobic fermentation of glucose, maltose, dextrose, lactose, or sucrose using microorganisms such as various types of lactobacilli strains: lactobacilli arnylophilis, lactobacilli bavaricus, lactobacilli casei, and lactobacilli maltaromicus [1,3,4] These microorganisms require specific fermentation conditions such as neutral pH, temperatures around 40°C, and low oxygen concentrations to yield high carbon conversion from feedstock [7]. After the fermentation is complete, any remaining cell biomass, protein and other insolubles need to be removed. The purification of the lactic cid is vital prior to polymerization in order to get ultimate efficiency from the lactic acid. 14 The purified lactic acid can be polymerized by two different methods. The first involves polycondensation of aqueous lactic acid [1,6,8-15] and the second involves ring-opening polymerization of cyclic lactide dimmers [1,6,8,14-l7]. In an early development stage of the direct condensation polymerization, the difficulties related to removal of residual water and catalyst resulted in an inability to obtain high molecular weight PLA polymers [8,13]. Mitsui Toatsu Chemicals has overcome this problem by developing an azeotropic distillation process which uses an organic solvent with a high boiling point for the removal of water and catalyst in the direct esterification process to obtain high molecular weight PLA. However, this method requires relatively large reactor volume and the need for evaporation and recovery of the solvent [lo-12,14]. Recently, researchers from Mitsui Toatsu Chemicals have proposed an improved direct condensation polymerization process which enables the production of high molecular weight PLA without using an organic solvent [9,14]. However, this method has not been extensively used for the manufacturing of PLA. The second method of converting lactic acid into polylactide is ring opening polymerization (ROP), which is a commercially viable process developed by Cargill Dow LLC [8,14,15,18,]9]. Before polymerization starts, lactic acid is converted into the lactide, the cyclic dimer of lactic acid, under controlled conditions, i.e. a temperature of 200-250 °C and a pressure of 01-003 mm Hg with a catalyst. The resultant mixture of lactide exists in four different stereoisomers as L-lactide (LLA), D-lactide (DLA), and meso—lactide (MLA), in addition to the racemic mixture D,L-lactide (DLLA). The molten 15 lactide is then purified by vacuum distillation. In the final stage, high molecular weight PLA (>100,000 Dalton) is produced by catalyzed ring opening polymerization in the melt. Each lactide rings opens up by means of initiators and adds to the growing chain to form PLA polymers. The polymerization mechanism involved in the ROP can be cationic, anionic, coordination-insertion mechanism, or enzymatic, depending on the catalyst/initiator systems used. Researchers have focused on developing new catalyst/initiators for ring-opening polymerization of lactides. Many catalysts/initiators have been cited in the literature for use in the ring Opening polymerization of lactides. These include lipase enzyme, SnClz, SnBrz, SnCl4, SnBr4, SnO, PbO, and metal alkoxides as well as metal complexes in the presence of alcohol [20]. Among these catalysts/initiators, metal alkoxides have the highest catalytic activity toward ROP of lactides [21,22]. The metal oxide catalyzed/initiated polymerization of lactide typically takes place at a temperature of around ISO-175°C to yield PLA with molecular weight up to 200,000 Dalton for a duration ranging from 2 to 6 hours [16,17]. PLA Based Nanocomposites Various inorganic and organic nano-particles have been used as possible fillers to enhance the PLA performance. Some examples of these particles are layered silicates, calcium phosphate, calcium sulphate, calcium carbonate, talc, potassium titanate, aluminum borate, polyhedral oligomeric silsesquioxane (POSS), and cellulosic nanofillers. Among all these fillers, layered silicates/layered clays and cellulosic fillers get most of the attention from industry. This is not only due to their wide availability and ow cost, but also due to their significant enhancements in mechanical and barrier properties [23-26]. Layered clays are hydrous aluminum silicates and their framework 16 layers are generated by a combination of two silica tetrahedral and one alumina octahedral sheets. A typical layer thickness can be around 10 angtrom while its lateral dimension may vary from 50 nm to several microns. Layered silicates including montrnorillonite (MMT), mica, saponite, and hectorite are the most commonly used mineral fillers [26]. Nowadays, researchers are faced with two challenges in developing polymer nanocomposites with layered clays: a) dispersion of the layered silicates in the polymers and b) the hydrophilic nature of the layered silicates, which causes incompatibility with the hydrophobic polymers. To overcome these problems, the surfaces of the silicates are organicallymodified and made compatible with the polymer [27]. Modification of the layered silicates can be done by a variety of methods. The most common method is ion exchange with cationic surfactants such as alkylarnmoniurn or alkylphosphonium cations. The cation exchange reaction between silicate cations and alkylammonium or alkylphosphonium salts lowers the surface energy and improves the wetting ability of the layered silicates, which results in the larger inter-layer spacing [24,26,28]. Ogata et a1. [29] were the first to prepare PLA/montrnorillonite composites by a solvent casting method. The mechanical properties of the PLA/montrnorillonite composites were not good enough due to lack of intercalation in the polymer matrix. Recent publications have reported the preparation of intercalated PLA/montrnorillonite nanocomposites with much improved mechanical and thermal properties. It is clear that surface modification and the type of the layered silicates play an important role in achieving improved mechanical and thermal properties. Extensive research work has 17 been performed by Sinha Ray et a1. [1,30-40] to develop PLA/organically modified layered silicate composites. As a starting point, montrnorillonite (MMT) modified with octadecylammonium cation (C18-MMT) was used and remarkable improvement was obtained in mechanical properties. In subsequent research [30], trimethyl octadecylammonium (C3C18-MMT) was used to get better mechanical properties. The storage modulus, flexural properties and heat deflection temperature of neat PLA and with 4 wt % of C3C18-MMT measured at 25°C are reported in 2.1. There is a significant increase of storage modulus, flexural properties, and heat deflection temperature for PLA with 4 wt% of trimethyl octadecylammonium-modified montrnorillonite compared to that of neat PLA. Table 2.1 Comparison of material properties between neat PLA and PLA with 4 wt % trimethyl octadecylammonium-modified montrnorillonite, adapted from Sinha Ray et a1. [30]. Material Properties PLA PLA ( with 4 wt % trimethyl octadecylammonium -modified montrnorillonite) Storage Modulus, GPa, 25 °C 163 2.32 Flexural Modulus, GPa, 25 °c 4-8 5.5 Flexural Strength, MPa, 25 0C 86 134 HDT °C 76.2 94 Chang et a1. [41] observed noticeable improvement in the 02 gas barrier of PLA clay nanocomposites made by a melt intercalation technique. With clay loadings ranging fi'om 0 to 10 wt %, the 02 permeability decreased from 777 to 327 cc/mZ/day for C16— MMT, to 330 cc/mz/day for DTA—MMT, and to 340 cc/mZ/day for Cloisite 25A. Even at 18 6 % clay loading, the permeability of the nanocomposite was reduced to about half of the PLA permeability value. This phenomenon is explained by the increase in the lengths of tortuous paths created in nanocomposites. Different types of layered silicates other than montrnorillonite have been studied by Sinha et a1. [35] Four different types of organically modified layered silicates (OMLS) used in this study were synthesized by replacing Na+ ions in different layered silicates with alkylammonium or alkylphosphonium cations by ion exchange. These clays were montrnorillonite modified with alkylarnmonium (ODA), montrnorillonite modified with alkylphosphonium cations (SBE), saponite (SAP), and synthetic fluorine mica (MEE). 2.2 shows the comparison of flexural modulus, strength, heat deflection temperature, and 02 gas permeability of neat PLA and various composites loaded with 4 wt % mineral. 19 Table 2.2 Comparison of flexural modulus, strength, heat deflection temperature, and 02 gas permeability of neat PLA and nanocomposite films, adapted fi'om Sinha et a1. [35]. Material Properties PLA PLA/ PLA/ PLA/ PLA/ ODA4 SBE4 SAP4 MEE4 Flexural Modulus, 4.84 5.66 5.57 4.5 6.11 GPa, 25 °c Flexural Strength, 86 132 134 93 94 MPa, 25 °C HDT, 76 94 93 98 93 with 0.98 MPa load, °C 02 gas permeability 200 172 172 120 71 coefficient, cc.mm.m‘2.day-l .MPam1 All these nanocomposites showed significant improvement of HDT in comparison to that of neat PLA. PLA/saponite composites had the highest HDT value. PLA filled with synthetic fluorine mica showed the lowest 02 permeability coefficient value. PLA/OMLS nanocomposites were also prepared with synthetic fluorine mica (OMSFM) with loadings of 0, 4, 8, 10 wt %. Adding organically modified fluorine mica in a PLA with D-lactide content of 1.1-1.7% promoted a higher l-IDT. The HDT increased sharply from 76 to 115 °c at the loading content of 10 wt%. [36]. Recently, cellulosic nanofillers have received attention from researchers for their remarkable reinforcing abilities. The use of cellulosic nanofillers has been proved to improve mechanical properties of polymers such as polyvinyl acetate (PVAc) [42,43], polyvinyl chloride (PVC) [43-45], polyoxyethylene (POE) [43,46], latex [47,48], cellulose acetate butyrate (CAB) [49,50], hydroxy propyl methyl cellulose (HPMC) [51], 20 and polypropylene (PP) [52,53]. In the literature, there exist different descriptors of cellulosic nanofillers such as cellulose whiskers, cellulose nanowhiskers (CNW), monocrystals, microcrystalline cellulose (MCC), and cellulose nanocrystallites. In this paper, they will be called cellulose whiskers. Plants, animals, and bacteria are generally used as a source of cellulose [43]. The cellulose whiskers are the crystalline portion obtained after acid hydrolysis of cellulose [43,47]. The main advantages of cellulose whiskers are their high aspect ratios (length/width: 30-150), large surface areas (>100 mz/g), and low densities (~l.56 g/cc) [48,54-56]. Polymer nanocomposites containing cellulose whiskers are usually made by a solution casting technique because of the difficulty of extracting whiskers. The addition of cellulose whiskers to the polymers mentioned above resulted in improvement in mechanical properties, except for plasticized starch[54] and sernicrystalline polyhydroxy alkanoate (PHA) [54,57]. The inclusion of cellulose whiskers did not significantly affect the glass transition temperature (Tg) regardless of the type of polymer. The thermal, mechanical, and barrier properties of cellulose whisker filled PLA nanocomposites were also studied by researchers [43,58-64]. Dorgan and Braun [61] used a different approach to prepare PLA nanocomposites. The polymerization reaction of lactide was initiated by the reactive groups (hydroxyl groups) on the surface of the cellulose whiskers [58-62]. The introduction of cellulose whiskers into the PLA matrix with this method did not affect the glass transition temperature, while the storage modulus of the composites containing 25 wt% cellulose whiskers prepared by compatibilization of 30 wt% preformed PLA and 70 wt% lactide (30/70 P/L) improved 21 by 53% [60,61]. In another study, N, N-Dimethyl acetarnide (DMAc) with 0.5 wt % lithium chloride (LiCl) was utilized as a swelling/separation medium for the cellulose whiskers [43 ,63,64]. The liquid suspension of the cellulose whiskers was pumped into a twin screw extruder after the PLA resins were introduced. The extruded materials were then compression molded. The mechanical results did not show significant change with addition of 5 wt % cellulose whiskers when compared to pure PLA. Furthermore the color of the samples was changed [63]. Petersson and Oksman [43] also compared the mechanical, thermal, and barrier properties of PLA filled with layered silicates and cellulose whisker composites. The nanocomposites were prepared using solution casting. The layered silicate system exhibited improvement in both tensile modulus and yield strength; however, the cellulose whisker system only improved the yield strength. The tan 8 peaks were shifted to higher temperature for both fillers. The oxygen permeability of PLA/layered silicates was reduced compared to pure PLA, whereas the oxygen permeability of PLA/cellulose whiskers increased [43]. Zeolites Zeolites are crystalline hydrated alurninosilicates with pore sizes varying from 3 to 15 angstrom. Zeolites can be obtained by chemical synthesis or naturally occur as minerals [65-68]. Currently there are 191 zeolite framework types which have been identified and approved by the Structure Commission of the International Zeolite 22 Association [69]. The general chemical formula of zeolites is represented by: M 2/n 0. A12 03 o xSiOz o szO ( l ) where M represents an exchangeable cation (e.g., Na+, K+, Ca2+, Mg2+), n is the cation charge, x is the number of Si02 tetrahedra, and y represents the number of water molecules in the tetrahedral framework [66,67,70,71]. The zeolite fiameworks can be classified on the basis of their structure, such as fibrous-chain structure, platelet-sheet structure, and cage-like structure. The primary building blocks of the zeolite fi'ameworks are the tetrahedra of S104 and A104 wherein a silicon or aluminum atom is at the center with four oxygen atoms at the corners (Figure la). In order to form a larger continuous framework, two or more tetrahedra are linked together by oxygen atoms (Figure 2.1b.). Eventually, the pores and channels are formed (Figure 1c) with no two aluminum atoms sharing the same oxygen [66 ,72]. 23 Oxygen atoms Figure 2.1 The basic tedrahedral building units of zeolite framework. a=>b=>c Zeolite fi'amework formation , adapted from Dyer [73] and van Bekkum [67]. As can be seen from the general chemical formula, the void spaces within the zeolite fi'amework are mainly occupied by cations and water molecules. An aluminum atom-containing primary building unit bears a negative charge which is balanced by a mobile cation, typically an alkali or alkaline earth metal cation, within the framework [67,73]. These cations (e.g., Na+, K+, Ca2+, Mg2+) are weakly bonded to the tetrahedral framework and can be exchangeable easily by washing with a solution of another ion. Possessing the ability to exchange ions can be considered the basis of most of the applications such as water softerners, municipal waste water treatment, removing harmful ions from radioactive waste, or removing ammonium ions from agricultural waste. The channels and interconnected voids in the zeolite frameworks can also accommodate water molecules which are loosely tied by hydrogen bonding to anionic framework atoms. Water molecules in zeolites can be removed by heating without collapsing the framework 24 structure, or can be substituted by other molecules. The pore structure and size vary from one zeolite to another. Typical zeolite pore sizes using oxygen-packing units are illustrated in Figure 2.2. The size of the pore is determined by the number of oxygens in the ring. For example, small pore zeolites (e.g., zeolite A) are formed by 8 member rings of oxygen atoms with fiee diameters of about 3.5-4.5 angstrom. Intermediate pore zeolites (e.g., ZSM-S) are formed by 10-membered rings with free diameters of approximately 4.5-6 angstrom. Large pore zeolites (e.g., zeolite X and Y) with 12-membered oxygen rings are about 6-8 angstrom in free diameter [67,70,73,74]. 12-Ring 1 O-Ring 603. .0 3 8 Diameter:3.5-4.5 angstrom Diameter: 4.5-6 angstrom Zeolite A ZSM—S 34.: {7:- Diameter: 6 - 8 angstrom Zeolite X or Y Figure 2.2 Schematic representation of typical zeolite pore sizes, adapted from [67,74]. The actual pore diameter also depends on the type of cation present in the void spaces. Smaller cations can be placed in pores which make them wider. For instance, the 25 + pore diameter of zeolite A can vary from about 4.3 angtrom for zeolite 5A (Mg2 , Ca2+ form) to 3.5-4 angstrom for zeolite 4A (Na+ form)[68]. It is this pore dimension that determines the size of the molecules that can be adsorbed or excluded from the interior of zeolites. This feature makes them valuable as molecular sieves, desiccants, odor absorbers, and catalysts. The overall properties of zeolites can be tailored based on their end use. By changing the location and size of the cations within the framework and the ratio of Si/Al, the ion exchange capacity, conductivity, interaction between the zeolite and adsorbed molecules, and the hydrophilic or hydrophobic properties can all be changed. Zeolite Based Composites Zeolite based composites have been prepared with a vide variety of polymer matrices, including polydimethylsiloxane (PDMS), sulfonated polyether ether ketone (SPEEK), polyethersulfone (PES), polyimide, epoxy, polyaniline (PANI), polythiophene (PTP), low density polyethylene (LDPE), high density polyethylene (HDPE), polypropylene (PP), polyvinyl alcohol (PVOH), polyvinyl acetate (PVAc), ethylene vinyl acetate (EVAc), polystyrene (PS), cellulose acetate, and polycarbonate (PC). The list shown in Table 3 gives a general idea of the types of zeolites involved along with their field of application. The majority of studies on zeolite based polymer composites have been focused on the separation process of gas mixtures and liquid mixtures. 26 In general, the composite polymer membranes are prepared by dispersion of the zeolite particles into a polymer solution followed by a casting evaporation process which produces a membrane with a thickness of approximately 1-6 mil. The zeolite loadings are varied in the range of 5-50 wt%. Table 2.3 Common zeolite types, polymer matrices, and their application areas. Polymeric Matrix Type of Zeolite Application Areas / References Comments Silicon Rubber ZSM-S, Silicalite- Ethanol/water separation [75,76] Polydimethylsiloxane- 1 used in biomass (PDMS) fermentation Isopropanol/water [7 7,7 8] separation Extraction of VOCs from [79-82] aqueous waste stream Na-Y, 13X, 4A Removal of aroma [83,84] compound Industrial gas separation [8 5] Poéyethylenimine MCM-41 Separation of €02 from [86,87] (P I) gas—mixtures Sulfonated polyether Na-Y High proton conductivity [88] ether ketone (SPEEK) in hydrogen firel cells Polyethersulfoue 4A, 5A, 13X H e /N 2’ H 2 /N 2, H e/CO 2, [89-94] (YES) H2/C02 Gas separation Separation of natural gas [95] Polyimide 4A, 5A, 13X,Na-Y Gas separation [96-100] Silicalite-l Isopropanol/water [ l 01 ] separation Epoxy Clinoptilolite Gas separation [ 102,103] ZSM-5 Electrical conductive [104] polymers, light weight batteries Polyaniline (PANI) ZSM-S, Na-Y High electrically [104] conductive polymers 1 3X Water—acetonitrile (ACN) [l 05] separation Polythiophene (PTP) 13X High electrically [106] conductive polymers 27 Table 2.3 (cont’d). separation 28 Low density Zeolite containing Antimicrobial [7 1 ,107, 1 08] polyethylene Ag+ ion (LDPE), High density polyethylene ( HDPE) 4A, 13X Removal of odor [1 08] C2H4/02 selectivity [109] Modified atmosphere packaging Aldehydes absorber, Odor [71] reduction Breathable microporous [1 10] films Odor causing compounds [1 1 1-1 13] removal in water pipes 002/ 02 selectivity [1 14] Modified atmosphere packaging Anti-blocking agents [1 15] Clinoptilolite Modified atmosphere [1 16] packaging Ethylene Scavengers [1 17,1 18] Polypropylene (PP) Clinoptilolite Nucleating agents [1 19] High water vapor [120] permeability Flame retardant [121,122] Zeolite containing Antimicrobial [107,108,123,124] Ag+ ion Polyvinyl alcohol ZSM-5- and Na-Y Water—acetonitrile (ACN) [105] (PVOH) separation Polyvinyl acetate 4A, SSZ-13, Natural gas separation [125] (PVAc) Ethylene vinyl Zeolite containing Antimicrobial [126] acetate (EVAc) Ag+ ion £51310“ acetate Zeollte-X C 0 2 /CH4 separation [100] Chitosan Zeolite containing Antimicrobial [127] Ag+ ion Na-Y Isopropanol —water [128] mixture separation HY Ethanol-water mixture [129] Table 2.3 (cont’d). Polycarbonate (PC) 4A Hz/Nz, 02/N2, COz/Nz, [130,131] H2/CH4, C02/ CH4 improved gas separation Polystyrene (PS) ZSM-5,Na-Y, 13X Good selectivity for [132,133] aromatic species Recycled Clinoptilolite Improved flow behavior [134] PP/LDPE/HDPE Nucleating agent blend HDPE] PS blend Clinoptilolite Improved impact behavior [134] Researchers suggested that incorporating the excellent separation properties of zeolites into polymers would improve the gas separation and liquid separation of neat polymer membranes [90,135]. A wide range of industrially important multi-component gas and liquid mixture separations have been investigated involving gas pairs: 02/N2, Hz/Nz, He/Nz, COz/Nz, COz/CH4, Hz/CO, COz/CH4, H2/C02, H2/CH4, H20/CH4, and solvent-water mixtures. The separation process depends on several parameters such as sorbent-sorbate interaction, molecular dimensions, partition firnctions, type and size of zeolites, etc. The ability of a membrane to separate two molecules (molecule A and molecule B) is called permselectivity and it can be considered one of the useful parameters to compare the performance of the membrane materials. The permselectivity is defined as the ratio of the permeabilities [91,136,137]. 29 aA/B= PA/PB (2) Since the permeation of molecules through the membrane is controlled by the diffusivity coefficient (D) and solubility coefficient (S), Eq. 2 may also be written as (1 ME = PA/P B = DA*SA/DB*SB (3) It is also worth mentioning that the selection of zeolite type is crucial for the separation process. Zeolite itself exhibits a high affinity to certain gases and vapors. Basically, its effective pore openings may determine what size of molecules will be entrapped or excluded inside the pores and cages. For example, type 4A zeolites having a pore opening of around 4 angstrom will only adsorb molecules with kinetic diameters of less than 4 angstrom, whereas any molecules with larger kinetic diameter will be excluded. Table 2.4 provides kinetic diameters of several common molecules used in separation processes. 30 Table 2.4 The kinetic diameter of some common molecules, adapted from Ullaland [138], Semenova [139], and Cecopieri-Gomez et al. [140]. Kinetic Kinetic Kinetic Molecule Diameter Molecule Diameter Molecule Diameter (angstrom) (angstrom ) (angstrom ) H2 2.89 HCl 3.2 H20 2.65 ()2 3.46 HBr 3.5 CC14 5.90 N2 3.64 CO 3.76 H28 3.60 C12 3.20 C02 3.30 802 3.60 Br2 3.50 CH4 3.80 C6H6 6.60 He 2.60 Csz 3.30 N20 3.30 Ne 2.75 C2H4 3.90 NO 3.17 Ar 3.40 C3H6 4.50 C4H10 4.30 Kr 3.60 C3H3 4.30 NH3 2.60 Xe 3.96 C4H6 4.40 C4H3 5.60 Many researchers have studied the effect of zeolites in polymeric membranes using permeation experiments where the selectivity and permeability of the membranes are determined for particular separation processes. For practical gas separation, the composite membranes should combine high permselectivity with high diffusion flux [137,141] The morphology of the interface between zeolites and the polymer membrane is also one of the major determinants of the overall separation performance. Chung et al. [137] and Moore et al. [142] discussed four different scenarios, providing a better 31 understanding of this complicated system. Figure 2.3 schematically illustrates what kinds of morphologies are expected for zeolite/polymer membranes. Case 1 shows an ideal morphology where the compatibility between zeolite crystals and polymer chains is perfect. Case 2 represents poor compatibility between zeolites and the polymer matrix, thus causing void formation at the interface. Case 3 indicates the possibility of reducing polymer chain mobility near the zeolite interface in comparison with the chain mobility of the bulk polymer. It is believed that the rigidified polymer layer around the zeolite causes a reduction in overall membrane permeability [136,142-144]. Case 4 exhibits a situation in which the pores of the zeolites have been blocked by solvents, contaminants, polymer chains or modifiers. The partial pore blockage may lead to a decrease in selectivity and permeability of the membranes. Sometimes, at least two or three cases mentioned previously take place simultaneously, which makes it difficult to predict zeolite filled membranes’ behavior with existing conventional methods. CASE 1 CASE 2 Interface Voids Partial Pore Blockage Figure 2.3 Schematic diagrams of zeolite/polymer membrane morphology, adapted from [137,142]. 32 Mahajan et al. [143] and Moore et al. [136] studied the effect of type 4A zeolites incorporated into polyvinylacetate (PVAc). They reported that addition of 40 wt % type 4A zeolite into polyvinylacetate (PVAc) substantially increased the oxygen (02) selectivity over nitrogen (N2) [permselectivity (a 02/N2)l from 5.5 to 11. Even at 15 wt % zeolite loading, the permselectivity was found to be 6.7. A similar improvement in permselectivity was observed by Suer et al. for zeolite 13X and 4A filled polyethersulfane (PES) relative to the pure PBS [90,97]. Duval et a1. [85,145] also studied the permeation rates of C02, 02, N2, and CH4 of polydimethylsiloxane (PDMS) membranes filled with zeolite 13X and silicalite-l. They found that both C02/CH4 and 02/N2 separation selectivities increased with silicalite-l loadings. Adding zeolite 13X into polydimethylsiloxane (PDMS) membranes only affected the C02/CH4 selectivity. 33 Table 2.5 Typical polymer zeolite composite membranes permselectivity values compared to the neat polymers, adapted from [85,130,137]. Permselectivity, a Polymer Zeolite Gas Pair Neat Polymer Composite (loading, wt %) Celulose acetate Silicalite-1 (25) Oz/Nz 3.00 4.30 Polyvinyl acetate 4A (40) 02/N2 5.50 11.00 Polysulfone 4A (25) OZ/NZ 5.90 7.70 Polysulfone 3A (41) H2/C02 1.60 13.00 Polyethersulfane 5A (50) Oz/Nz 5.80 7.40 Polyethersulfane 4A silver ion C02/CH4 5.30 40.40 exchange(50) Polydimethylsiloxane Silicalite-1 (70) OZ/NZ 2.14 2.92 Polydimethylsiloxane Silicalite-1 (70) C02/CH4 3.42 8.86 Polydimethylsiloxane Silicalite -l (30) C02/CH4 3.42 4.10 Polycarbonate 4A (30) H2/N2 56.70 73.20 Polycarbonate 4A (30) C02/N2 32.60 39.10 Polycarbonate 4A (30) C02/CH4 23.50 37.60 As can be seen from Table 2.5, the incorporation of zeolites into various polymer families resulted in better permselectivities compared to the neat polymers. This is also evidence of good compatibility between zeolites and polymer matrices and uniform zeolite particle distribution (Case 1) [83,85,91,98,137,146]. However, some of the zeolite filled membranes did not show successful separation performance for specific gas pairs. The primary reasons of this diminished performance were the poor contact between zeolite particles (Case 2) and the polymer, and rigidification of the polymer matrix around the zeolite particules (Case 3) [74,92,96-98,13l,137,147,148]. Poor interfacial contact results in non-selective voids at the interface. These non-selective voids permit 34 the transport of gases indiscriminately, which reduces the overall selectivity of the membranes [90,130,143,144,146]. Uneven shrinkage and stresses that arise at the interface during the preparation of composite membranes are believed to be the major culprits in void formation and polymer chain rigidification [92,136,137,l42]. Significant efforts have been devoted to overcoming some of these issues and promoting interfacial contact. Using appropriate compatibilizers - low molecular weight additives - or coupling agents, such as 2,4,6- triarninopyrimidine [96], p—nitroaniline [130], and amino functional silanes [92,100,145], using high membrane formation temperatures[100], and fabricating the membranes using a melt extrusion process [90,96] have resulted in relatively improved adhesion between zeolites and the polymer matrix. However, the possibility of blocking the pores by coupling agents, additives or rigidified polymer chains increased [97,137,148]. Pore blockage significantly deteriorates the selectivity of zeolites. In some cases, this negative effect of pore blockage of zeolites may be desirable in order to increase the selectivity, such as when the original pore size of the zeolite is significantly larger than the kinetic diameter of the tested molecules. Clearly, it requires intensive exploratory and comparative studies to evaluate these systems. In the literature, there also exist a few studies concentrated solely on thermoplastics and related to packaging applications. Zeolites have been incorporated in polyethylene and polypropylene for a variety of interesting packaging applications such as antiblock agents [115,149], antimicrobial agents [108,124,150], fillers [1 12,1 13,1 19,120,151], and odor and ethylene scavengers 35 [108,109,111,113,114,152,153]. Van Essche [115] reported that zeolites can be used as an antiblocking agent, creating a micro rough surface that reduces the adhesion between the film layers and thus lowers the blocking force. The use of zeolites as antimicrobial agents in polyolefins has been extensively studied [124,154-156], In its structure, zeolite contains sodium ions that can be replaced with silver ions to form Ag-zeolite which exhibits antimicrobial activity [108,124]. In modified atmosphere packages, C02 can also be considered one of the important components in the gas mixtures due to its antimicrobial activity. Depending upon the nature of the food product, the package permeability ratio of C02 to 02 can be varied. The right combination can prolong the shelf-life and preserve the quality of the food product. For instance; 1-3% 02 and 0-3% C02 content extend the storage life of apples by inhibiting ethylene production which causes ripening [157]. A number of patents describe the use of zeolites in polyolefins to adsorb products formed during the ripening and degradation of packaged food products [71,111]. Villberg et al. [111] studied the potential use of zeolites to selectively adsorb undesirable VOCs originating from high density polyethylene itself during processing. It was found that off-odor and off-taste compounds in HDPE packaging materials can be eliminated with the aid of 9, zeolite. Numerous patents and articles describe the effectiveness of “Abscents zeolites in reducing odor in the polymer substrates. 36 Although over a hundred zeolites have been synthesized or naturally occur, in this study we chose to use one synthetic and one natural zeolite as functional fillers to incorporate into a PLA matrix. The next section will describe the basic characteristics of these zeolites. Type 4A as a Model Synthetic Zeolite Zeolite type 4A was used in this study. Each unit cell of zeolite 4A contains 24 tetrahedral units to form a cubic structure with a Si/Al ratio of 1, a BET surface area of 560 mz/g and a total pore volume of 0.2 cm3/g. Figure 2.4 shows a fiamework of the type 4A zeolite. In order to visualize the pore openings, the building units of type 4A are also shown in Figure 2.4. The unit cell of type 4A (Figure 2.4 (1) consists of one large cage with 8 rings (Figure 2.4 c), which are located in the center of the unit cell, and small cages with 6 rings (Figure 2.4 b) located at the comers of the unit cell The typical chemical composition of type 4A can be written as: Na12A1128112048-27H20. Since the Si/Al ratio of the framework is 1, type 4A contains the maximum amount of exchangeable cations, balancing the negative charge of aluminum. Type 4A is also considered hydrophilic in nature due to its low Si/Al ratio [158]. 37 Figure 2.4 Framework structure of zeolite type 4A (D). Building unit of type 4A with 4- rings (A); Building unit with 6-rings(B); Building unit with 8-rings (C) with the pore size of approximately 4 angstrom [159]. Chabazites as a Model Natural Zeolite Chabazite is a natural zeolite which is often found in l) alteration of siliceous volcanic pyroclastics on the ocean floor, saline, alkaline lakes, and on the surface by ground water; 2) deposition from hydrothermal solutions, including hot springs [160]. Chabazite consists of a three-dimensional framework of interconnected double-6-ring units linked with a tilted 4 ring unit which form a large ellipsoid open cage with a 6.7 angstrom pore window and which can be described as a rhombohedral [161]. Its general chemical formula can be written as (K2, Nag, Ca, Mg) [AIZSi4012].6H20. Depending on the source, the ratio of cations may vary. Figure 2.5 shows the framework structure of 38 chabazite. It is stable at a pH of 2.5 which makes it suitable for removing water from hydrogen chloride gas streams, as well as trace gas removal, such as nitrogen removal from argon. Figure 2.5 Framework schematic of chabazite (c). Building units (a, b) [159] Literature Review on Pertinent Topics The present research seeks to establish a coating technique and material that can be utilized with poly(lactic acid) for packaging applications; therefore, it is important to evaluate the existing technologies and studies that are related to this topic. The first section will review previous work on thin layer deposition techniques. In the second section, the deposition materials, silicon oxide, aluminum oxide, and diamond like carbons will be introduced and related literature on their uses in packaging applications will be reviewed. 39 Thin Layer Deposition Techniques Currently, there are two common methods used to deposit thin films of a variety of materials: atmospheric and vacuum-based deposition. Atmospheric-based techniques consist of applying a chemical solution followed by thermal or UV curing. This coating uses a liquid phase as the mass transfer media. Vacuum-based coating techniques can be divided into three categories: physical vapor deposition (PVD), chemical vapor deposition (CVD), and atomic layer deposition (ALD). All of them use a gas phase to transport molecules to the surface of the substrate where a thermal reaction/deposition occurs. Silicon oxide, aluminum oxide, and diamond—like carbon (DLC) have been deposited on PET substrates for the purpose of improving barrier properties [162-164]. Atmospheric-Based Coatings [Liquid Coatings In the mid-19908, PPG Industries Inc. developed an epoxy-amine organic coating material called Bairocade® which can be sprayed onto the external surface of PET bottles. It is reported that the coating thickness is about 6-8 microns and it offers excellent C02 and 02 barrier [165]. A typical Bairocade® coating operation starts with feeding the bottles to the coating unit. The bottles are held by the neck in individual chucks. The chucks give an electrostatic charge to the plastic bottle. Then the bottles proceed through the coating unit where they are rotated and sprayed with the Bairocade®. The coated bottles pass into the curing oven where they are heated at 65 0C (149 0F) for 3 to 8 minutes to evaporate the organic solvents and crosslink the polymer film. The 40 crosslinked 6-8 micron thick coating provides a high barrier. Depending on the coating thickness and specific product requirements, these coatings can reduce the oxygen transmission to a ratio of 1/2 to 1/20, and carbon dioxide transmission to a ratio of 1/2.5 when compared with standard uncoated PET bottles. Bairocade® has the potential to give a more than 13 month shelf-life for acidic products such as juices and a 9 month shelf-life for beers. The Bairocade® technology has been commercialized by Graham Packaging in the US. for 12 to 20 oz juice bottles. PepsiCo uses Bairocade® for single serve carbonated soft drink (CSD) bottles in Saudi Arabia. SIPA, the Italian blow molding machine manufacturer, has developed its own coating technology, Smart Coa ®, based on a Bairocade® coating unit integrated into a PET bottle blowing line. The capacity of the line is 6000 to 36000 bottles per hour, while occupying only 170 m2 [166-169]. Dow has also developed an amorphous thermoplastic epoxy resin (BLOX®) which can be used either as a barrier layer in multilayer PET structures or as a coating material for the bottles. It is claimed that BLOX®’s oxygen and C02 barrier properties are about 10 times higher than these of PEN. It is also said to be more cost competitive than the alternative barrier polymers. Dow and Tetrapak worked together to produce PET preforms with a layer of BLOX®, called Sealica. Special equipment is used to injection mold the PET preform and then over-mold it with BLOX® resin [165,170,171]. 41 Another relatively new exterior coating comes fi'om InMat, which developed an aqueous suspension of nano-dispersed silicates such as vermiculite and montrnorillonite in a polyester matrix (N anolok PT®) [7-10% solids content]. It can be applied by a spray coating or gravure coating process to the PET film. When it dries, a 0.25-2 micron thick coating can be formed on the PET. The coating contains nanodispersed silicate platelets which give a tortuous path for molecules such as oxygen, carbon dioxide and aromatics. Thus, the barrier properties of the substrate can dramatically increase. InMat suggests that the Nanolak® coatings have the highest gas barrier level amongst the other polymeric coatings available on the market. It can be up to 100 times better than the standard uncoated PET substrate [172]. Meanwhile, Microcoating Technologies, now known as nGimat, has introduced a new development which utilizes a combustion chemical vapor deposition technique (CCVD) to deposit nanopowders on polymers. When compared to traditional chemical vapor deposition techniques, the CCVD process offers uniqueness because it can operate in an open atmosphere (non vacuum). There is no data available regarding the nature of the nanopowders used for food packaging applications, yet it is reported to provide a barrier to C02 and 02. The shelf-life of 20 02. bottles is claimed to be increased from 10 weeks to 30 weeks [173,174]. Vacuum-Based Coatings As mentioned earlier, vacuum-based coating techniques basically involve depositing SiOx, A1203, and diamond-like carbon (DLC) from the gas phase. In these 42 systems, the precursors are heated to form a gas and then deposited as a solid on the surface of the substrates, usually under vacuum. There may be deposition by condensation (physical vapor deposition-PVD) or by chemical reaction to form a new product which differs from the precursor that was volatilized (chemical vapor deposition- CVD) [162-164]. Oxides are recognized for their chemical inertness, good high temperature properties, and resistance to oxidation. Most oxides have a significant degree of ionic bonding, since oxygen is the most electronegative divalent element. Therefore, they generally have the characteristics of ionic crystals, for example, optical transparency when pure, high electrical resistivity, low thermal conductivity, and chemical stability [163,164]. CVD and PVD can be considered major processes for oxide deposition. In packaging, silicon oxide and aluminum oxide coatings have been used for gas barrier improvement of plastic films and bottles. Thus, only these oxide coatings are evaluated in this chapter. Silicon Oxide Silicon oxide is a major industrial coating material in many applications, particularly in optics and microelectronics. In high barrier packaging applications, silicon oxide -SiOx (l10 (BIF) for 02 Barrier improvement factor - >4 (BIF) for C02 Shelf life, weeks 8-9 >25 46 NatureWorks LLC presented a new development at Nova-Pack Americas 2006, the 20th International Conference on PET Containers for Food and Beverages. They have used the Plasmax 12D® coating system for PLA bottles. The banier improvement factor was found to be 1.25 for C02 transmission, and 92 for OTR [193]. Aluminum Oxide Aluminum oxide (A1203), also known as alumina, is a highly stable compound with many industrial applications. Aluminum oxide is resistant to oxidation and has extremely low permeability to oxygen. It has a melting point of 20150C, a coefficient of thermal expansion of 7-8.3 x 10 -6/ 0C, a thermal conductivity of 25-29 W/moC, a flexural strength of 421 MPa, a modulus of 378 GPa, and a vicker’s indentation hardness of 18.73 GPa [163] Aluminum oxide can be formed by using precursors such as aluminum trichloride (AlCl3), H2 and C02 at a temperature of 1000°C at low pressure (approximately 2—3 kPa) by CVD. This system is mainly used for tool coating and electronic applications [163]. There exists another CVD technique which operates at lower temperatures for A1203 deposition. In this deposition technique, aluminum isopropoxide (Al(0C3H7)3) is used as a precursor. The formation of A1203 begins at approximately 300 oC. Sometime, it is necessary to deposit A1203 at low temperatures which are compatible with thermally susceptible plastic substrates. In this case, plasma 47 enhanced chemical vapor deposition (PECVD) is preferred because it does not use high temperature to stimulate the chemical reaction which takes place on the substrate surface. In 1997, Montgomery, in US. Patent No. 5691007 [194], disclosed a method for applying a plasma enhanced chemical vapor deposited barrier film coating onto three- dimensional items, including low melting temperature polymer items. Several researchers also studied the possibility of A1203 deposition on polymers by atomic layer deposition (ALD). ALD is akin to plasma enhanced chemical vapor deposition (PECVD). Both techniques use the gas phase to deliver molecules to the substrate surface where chemical reaction/transformation occurs, leaving the desired solid-state composition as a thin film. The distinction between these techniques is that ALD has intrinsic deposition uniformity on the substrates, especially complex 3-D surfaces, whereas PECVD exhibits less conformal deposition. The reason behind the uniformity of ALD is the sequential cycle of precursors. The ALD process uses the sequential precursor gas cycles to form a coating one monolayer at a time. Two precursors such as trimethylalurninum (TMA) [A1(CH3)3] and H20 are alternately introduced into the chamber. The first precursor (TMA) reacts with the surface, which possesses hydroxyl groups due to contact with air, yet it does not react with itself. Thus, it forms a monolayer on the surface. After obtaining a saturated monolayer, all of the excess TMA and methane, which is the by-product of the reaction, are removed. Then, the second precursor (H20) is introduced and reacts with the TMA to produce a saturated monolayer of oxygen. Again, after obtaining a saturated monolayer, the excess amounts of H20 and methane are evacuated. The cycles are repeated until the desired thickness is reached. As a result, the coating thickness is 48 controlled by the number of deposition cycles, which provides uniformity on the surface [195-198]. In 2004, Groner et al. [195] and Elam et al. [199] used the ALD technique for the first time to deposit aluminum oxide on 510 mL PET bottles at 58°C and investigated the barrier properties of this coating. During the deposition, the bottles were pressurized with N2 in order to prevent bottle deformation. The coating growth rate was characterized by the quartz crystal micro balance (QCM) which was integrated into the system. The use of QCM in the ALD flow reactor has given quite a lot of challenge to researchers, yet the modification of the system provided accurate, calibrated mass measurements in the ALD flow reactor. A more detailed explanation can be found in Elam et al. [199] The QCM measurements for A1203 ALD at 58 °C showed linear growth of the aluminum oxide film over many cycles. The average mass gain per cycle was found to be 30 ng/cmz. Almninmn oxide coating produced with 300 cycles and having a thickness of 360 angstrom reduced C02 permeability, providing a 1.6 fold improvement in barrier. Although A1203 ALD has been studied on several polymers including polyethylene naphthalate (PEN), polyamide (PA), and polyethersulfane (PES), polypropylene (PP), polystyrene (PS), poly(vinyl chloride)(PVC), low-density polyethylene (LDPE), the studies related to their gas barrier properties are limited. Table 2.7 summarizes the available literature values of WVTR, OT R, and C02TR for the A1203 ALD on several polymers along with the deposition thicknesses and temperatures. It should be noted that 49 the Table 2.7 gives just a very general idea. Comparing the results from these reported values was impossible due to lack of important parameters such as the experimental conditions and the thickness of the polymer substrates. Cleaning and handling the polymer surfaces before the deposition also plays a vital role in getting diverse permeation values. Nevertheless, it can be concluded that A1203 ALD provides improved barrier properties. A different approach was also taken by Hegeman and Summer [183] in order to reduce the rate of permeation of gases through PET bottles. They developed a gas impermeable layer system consisting of acrylate and aluminum oxynitride (AleNy). The PET bottle is coated with a 0.2-1.5 pm thick acrylate layer, on which is applied a 1-100 nm thick aluminum oxynitride layer. Above this layer another acrylate layer having a thickness of 0.2—1 .5 pm is applied by sputtering techniques. It is also claimed that coated bottles can withstand a pasteurization process. This coating provides elasticity with no crack formation. Henry et al. [200] studied a wide range of oxides deposited on PET substrates. They found that AleNy coatings exhibited the highest resistance to water vapor transmission. 50 Table 2.7 W V'I'R, OTR, C02TR values for A1203 ALD. Substrate A1203 deposition Gas Barrier Reference # thickness (11m) PEN 0 WVTR Carcia et al.[201] 0.72 g/ mz/day 10 WVTR Carcia et al.[201] <5X10D5 g/ mz/day 20 WVTR Langereis et (at 23 °C) 5x10'3 g/tnz/day “1207-1 25 WVTR Carcia et al.[203] (at120°C) 1.7x10'5 g/ mz/day 38°C, 90%RH 25 WVTR Carcia et al.[203] (at 120°C) 6.5x10'5 g/mz/day 60°C, 90%RH 125m 26 WVTR Darneron et (at 175°C) 1X10’3 g/mZ/day a1-[204] 23°C 10 OTR Carcia et al.[201] (at 120°C) <1x10'2 cm3/m2/day <25 OTR Mclean et al [205] <5x10'3 cm3/m2/day 23°C, 50%RH 25 OTR Carcia et al.[201] (at 120°C) <5x10'3 cm3/m2/day 51 Table 2.7 (cont’d). Substrate A1203 deposition Gas Barrier Reference # thickness (um) PBS 30 nm WVTR Park et al.[206] (Both side) 0.3 g m2/day 38°C, 90%RH PA 26 WVTR Darneron et 50 pm (at 175°C) 1X10-3 g/mZ/day al.[204] 60°C, 90%RH PET Bottle 36 C02TR Groner et al.[l95] (at 58°C) BIF:1.6 fold PET 10 OTR Carcia et al.[201] (100°C) <5X10-3 cm3/m2/day 25 OTR Carcia et al.[201] (100°C) <5X10-3 cm3/m2/day Diamond-like Carbon (DLC) Diamond-like carbon (DLC) films are amorphous and characterized by their high hardness, chemical inertness, and thermal stability [207,208]. They have a mixture of diamond-like (sp3) bonds and graphite (spz) bonds. Unlike diamond, DLC can be deposited at room temperature. Recently, studies have indicated that DLC films have not only good mechanical properties but also high gas barrier properties. DLC coating layers have been applied to food packaging such as plastic bottles [184,209-212]. A number of techniques have been used for coating substrates with DLC films, including PECVD, ion implantation, and filtered cathodic vacuum arc (FCVA) techniques. PECVD is the currently favored method in the packaging industry. The coating process basically consists of the deposition of a very thin layer of diamond-like carbon on the interior of 52 the PET container. The bottle is placed into a vacuum chamber. The precursor gas such as acetylene, methane or ethylene is then injected into the bottle. Microwave or radio fiequency energy is applied to excite the gas into an ionized state, creating a low temperature plasma. The carbon ions coalesce onto the inner surface of the bottle in an amorphous structure. The final thickness of the coating varies between 0.02 and 0.1 micron [184,210,212]. Boutroy et al. [209] studied the different precursors mentioned above and found that the highest deposition rate was obtained by combining acetylene gas precursor and a microwave plasma unit. The DLC fihns provided a significant reduction in oxygen transmission rates. The C02 loss of the bottles was also measured, and the shelf life of the soft drinks and beer was calculated. It was found that an uncoated bottle has 10 weeks shelf-life and a coated bottle has 44-45 weeks for soft drinks. For beer, an uncoated bottle has 4 weeks and a coated bottle has 38 weeks shelf-life. Research has shown that coating with DLC films dramatically improves the gas barrier properties. There are currently two commercially available DLC coating processes: Actis® (Sidel)[213] and D.L.C. ( Kirin and Mitsubishi Shoji) [214]. DLC coating technologies developed by Sidel and Kirin/Mitsubishi Shoji used plasma vapor deposition to provide an internal coating of a thin layer of highly hydrogenated amorphous carbon. These technologies are reported to improve C02 barrier 7 fold and 02 banier around 20-30 fold [165,171,209-211,213,214]. Although the DLC coatings are clear, they exhibit a yellow tint [184]. Sidel’s Actis coating system wa used for PLA bottles by NatureWorks. This method has been claimed to improve 02 barrier properties of PLA bottles by a factor of 10, and water vapor barrier 53 properties by a factor of 4 [193]. The information gathered from studies mentioned above led us to focus on aluminum oxide ALD on PLA substrates. Chapter 8 will describe them in detail. 54 REFERENCES [11 [2] [3] [41 [5] [6] [7] [8] [9] [10] [111 [12] [131 [141 [151 [161 [17] [18] R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4, 835. R. H. Datta, Michael, J Chem T echnol Biotech 2006, 81, 1119. R. E. Drumright, P. R. Gruber, D. E. Henton, Adv. Mater. 2000, 12, 1841. D. Garlotta, J Polym Environ 2001, 9, 63. 0. Wolf, Available at http.'//www. chem. uu.nl/nws/www/publica/ Publicatie32005/E2005-28I . p Access date 11.07.2007. A. Sodergard, Stolt, M., Prog Polym Sci 2002, 27, 1123. A. K. Mohanty, Misra, M., Hinrichsen, G., Macromol Mater Eng 2000, 276- 277, 1. 0. Coulembier, Degee, P., Hedrick, J .L., Dubois, P. , Prog Polym Sci 2006, 31, 723. B. Gupta, N. Revagade, J. Hilbom, Prog Polym Sci 2007, 32, 455. H. S. Kim, Hwang, M.O., Kim, M.N., Yoon, J.S., J Appl Polym Scie 2006, 100, 466. K. W. Kim, W00, 8., I. , Macromol Chem Phys 2002, 203, 2245. H. Suizu, Takagi, M., Ajioka, M., Yarnaguchi, A., US. Patent No: 5679767 1995. J. Tuominen, Kylma, J ., Seppala, J ., Selin, J .F., (1.5. Patent No. 7256250 2007. E. T. H. Vink, Rabago, K. R., Glassner, D.A., Gruber, P.R., Polym Degrad Stab 2003, 80, 403. L. S. Nair, Laurencin, C.T., Progress in Polymer Science 2007, 32, 762. H. R. Kricheldorf, Berl, M., Scharnagl, N., Macromolecules 1988, 21, 286 H. R. Kricheldorf, Chemosphere 2001, 43, 49. K. Marcincinova, Boustta, M., Coudane, J. Vert, M., in Polymers fi'om Renewable Resources .' Biopolyesters and Biocatalysis, (Ed: C. Scholz, Gross, R.A), Washington, DC : American Chemical Society : Distributed by Oxford University Press, 02000. 55 [191 [20] 121] [22] [231 [24] [25] [261 [271 [231 [291 [30] [311 [32] [331 [341 [351 [361 P. R. Gruber, Kolstad, J. J ., Iwen, M., Benson, R.L, US Patent Number 5142023 1992. P. R. Gruber, Kolstad, J. J ., Ryan, C.M., Hall, E., US. Patent No. 5484881 1996. J. Wu, Yu, T.,Chen, C.,Lin, 0, Coordination Chemistry Reviews 2006, 250, 602. M. H. Chisholm, Gallucci, J. C., Krempner, C., Polyhedron 2007, 26, 4436. A. Sorrentino, Gorrasi, G., Vittoria,V., Trends in Food Science & Technology 2007, 18, 84. F. Hussain, Hojjati, M., 0kamoto, M., Gorga, R., J Compe Mater 2006, 40, 151 1. J. J. Jordan, Tannenbaurn, R., Sharaf, M., Jasiuk, 1., Mater Sci Eng A 2005, 393,1. S. C. Tjong, Materials Science and Engineering: R: Reports 2006, 53, 73. J. Weiss, Takhistov, K., Mcclements, J ., Journal of Food Science 2006, 71, R107. S. Sinha Ray, S. Pouliot, M. Bousmina, L. Utracki, Polymer 2004, 45, 8403. N. Ogata, G. Jimenez, H. Kawai, T. Ogihara, J Polym Sci Part B: Polym Phys 1997, 35, 389. S. Sinha Ray, Yarnada,K., 0kamoto,M., Ueda, K., Nano Letters 2002, 2, 1093. S. Sinha Ray, M. Bousmina, Prog. Mater. Sci. 2005, 50, 962. S. Sinha Ray, 0kamoto, M., Macromolecular Materials and Engineering 2003, 288, 936. S. Sinha Ray, Yarnada, K.,0kamoto, M.,Fujimoto, Y., Ogami, A.,Ueda, K., Polymer 2003, 44, 6633. S. Sinha Ray, P. Maiti, M. 0kamoto, K. Yarnada, K. Ueda, Macromolecules 2002, 35, 3104. S. Sinha Ray, 0kamoto, M., Macromolecular Rapid Communications 2003, 24, 815. S. Sinha Ray, K. Yarnada, M. Okamoto, A. Ogami, K. Ueda, Chem Mater 2003, 15, 1456. 56 [37] [38] [39] [40] [41] [421 [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] S. Sinha Ray, M. Okamoto, Progress in Polymer Science (Oxford) 2003, 28, 1539. S. Sinha Ray, Yamada, K., Okamoto, M., Ueda, K., Macromolecular Materials and Engineering 2003, 288, 203. S. Sinha Ray, Yarnada, K., Okamoto, M., Ueda, K., Polymer 2003, 44, 857. S. Sinha Ray, M. Okamoto, Macromolecular Rapid Communications 2003, 24, 815. J. Chang, Y. U. An, G. S. Sur, J Polym Sci Part B: Polym Phys 2003, 41 , 94. N. L. G. Rodriguez, Thielemans, W., Dufi'esne, A. , Cellulose 2006, I3, 261. L. Petersson, Oksman, K., Composites Science and Technology 2006, 66, 2187. L. Chazeau, Cavaille, J. Y., Terech, P., Polymer 1999, 40, 5333. L. Chazeau, Cavaillé, L.Y., Perez, J ., Journal of Polymer Science Part B: Polymer Physics 2000, 38, 383. M. A. S. Samir, Alloin, F.,Sanchez, J -Y., Dufresne, A., Polymer 2004, 45, 4149. V. Favier, Canova, G. R., Cavaille, J. Y., Chanzy, H., Dufiesne, A., Gauthier, C., Polymers for Advanced Technologies 1995, 6, 351. P. Hajji, Cavaille, J. Y., Favier, V., Gauthier, C., Vigier, G., Polymer Composites 1996, I 7, 612. K. Oksman, Abstracts of Papers, 233rd ACS National Meeting, Chicago, IL, United States, March 25-29, 200 7, CELL 108. M. Grunert, Winter, W.T. , Journal of Polymers and the Environment 2002, 10, 27. N. Dogan, McHugh, T. H., Journal of Food Science 2007, 72, E016. N. Ljungberg, Bonini, C., Bortolussi, F., Boisson, C., Heux, L., Cavaille, J. Y., Biomacromolecules 2005, 6, 2732. N. Ljungberg, Cavaille, J. Y., Heux, L., Polymer 2006, 47, 6285. M. N. Angles, Dufresne, A., Macromolecules 2001, 34, 2921. V. Favier, Chanzy, H., Cavaille, J. Y., Macromolecules 1995, 28, 6365. 57 [56] [57] [53] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] V. Favier, Canova, G. R., Shrivastava, S. C., Cavaille, J. Y., Polymer Engineering and Science 1997, 37, 1732. A. Dufresne, Kellerhals, M. B., Witholt, B., Macromolecules 1999, 32, 7396. B. Braun, Dorgan, J .R., Available at: http://www.nt.ntnu.no/users/skoge /prost/proceedings /aiche-2004/pdfliles/papers/004apdf Access date: 03/26/2007. B. Braun, Dorgan, J .R., Available at: http://aiche. confex. com/aiche/2006/techprogram/P735 65.H T M. Access date: 03/26/200 7. B. Braun, J. R. Dorgan, D. M. Knauss, J. Polym. Environ. 2006, 14, 49. J. R. Dorgan, Braun, B., US. Patent Number: 0235113, 2006. J. R. Dorgan, "Ecobionanocomposites: A New Class of Green Materials", Cincinnati, OH, United States, 2005. K. Oksman, Mathew, A. P., Bondeson, D., Kvien, 1., Composites Science and Technology 2006, 66, 2776. L. Petersson, Oksman, K., in ACS. symposium series, Vol. 938, American Chemical Society, Washington, DC. 2006, 132. Y. Kumzerov, Vakhrushev, S., In Encylopedia of Nanoscience and Nanotechnology Nalwa, H.S. (ed), Stevenson Ranch: Califomia; 2007, 811. S. Polarz, Smarsly, B., J Nanosci Nanotechno 2002, 2, 581. H. Van Bekkum, Flanigen, E.M., Jansen, J .C., Introduction to Zeolite Science and Practice, Elsevier: New York: 1991. R. T. Yang, Adsorbents: Fundamentals and Applications, Wiley: Hoboken, N.J., 2003. The International Zeolite Association. Available at http://www.iza-online.org/, Access date: 12/23/2007 G. V. Tsitsishvili, Andronikashvili, T.G., Kirov, G.N., Filizova, L.D., Natural zeolites, Ellis Horwood, New York 1992. S. C. Hodgson, R. J. Casey, S. W. Bigger, Polym-Plas T echnol Eng 2002, 41 , 795 Y. Yan, Wang, H., In Encylopedia of Nanoscience and Nanotechnology Nalwa, 58 [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [38] HS. (ed), Stevenson Ranch:California; 200 7, 762. A. Dyer, In Mineral Surfaces, Vaughan, D. .I. and Patrick R.A.D. (eds), Chapman&.I-Iall: London, 1995, 333. x. Xu, Wang, J., Long, Y., Sensors 2006, 6, 1751. M. Jia, Peinemann, K.V., Behling, R.D., Journal of Membrane Science 1991, 57, 289. M. D. Jia, Peinemann, K.V., Behling, R.D., Journal of Membrane Science 1992, 73, 119. A. A. Kittur, S. K. Choudhari, M. Y. Kariduraganavar, Comp Interf 2006, 13, 507. W. Kujawski, Ostrowska-Gumkowska, B., Separation Science and Technology 2003, 38, 3669. H. Yang, Nguyen, Q.T., Ping, 2., Long, Y., Hirata, Y., Materials Research Innovations 2001, 5, 81. M. V. Chandak, Lin, Y. S., Ji, W., Higgins, R. J ., Journal of Membrane Science 1997, I33, 231. B. Moermans, Beuckelaer, W.D., Vankelecom, 1.F.J., Ravishankar, R., Martens, J. A., Jacobs, P.A., Chemistry Communication 2000, 2467. L. E. M. Gevers, Vankelecom, 1.F.J., Jacobs, P.A., Chemistry Communication 2005, 2500. S. Tantekin-Ersolmaz, Atalay-Oral, C., Tatlier, M., Brdem-Senatalar, A., Schoeman, B., Sterte, J ., Journal of Membrane Science 2000, I 75, 285. I. F. J. Vankelecom, De Beukelaer, S., Uytterhoeven, J .B., Journal of Physical ChemistryB 1997, 101, 5186. G. Clarizia, C. Algieri, E. Drioli, Polymer 2004, 45, 5671 . X. Xu, Song, C., Andresen, J .M., Miller, B., Scaroni, A.W., International Journal of Environmental Technology and Management 2004, 4, 32. X. Xu, Song, C., Andresen, J.M., Miller, B., Scaroni, A.W., Microporous and Mesoporous Materials 2003, 62, 29. M. I. Ahmad, Zaidi, S. M. J ., Rahman, S. U., Desalination 2006, 193, 387. 59 [39] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] Z. Huang, Y. Li, R. Wen, M. M. Teoh, S. Kulprathipanja, J Appl Polym Sci 2006, 101, 3800. M. G. Suer, N. Bac, L. Yilrnaz, Journal of Membrane Science 1994, 91, 77. C. M. Zimmerman, A. Singh, W. J. Koros, Journal of Membrane Science 1997, 13 7, 145. Y. Li, H. Guan, T. Chung, S. Kulprathipanja, J Membr Sci 2006, 275, 17. Y. Li, Chung, T., Kulprathipanja, S., AIChE Annual Meeting, Conference Proceedings, American Institute of Chemical Engineers, New York, United States 2005, 2062. Y. Li, Chung, T., Cao, C., Kulprathipanja, S., Journal of Membrane Science 2005, 260, 45. L. Yi, Tai-Shung, C., Santi, K., American Institute of Chemical Engineers. AIChE Journal 2007, 53, 610. H. H. Yong, Park, H. C., Kang, Y. S., Won, J ., Kim, W. N., Journal of Membrane Science 2001, 188, 151. T. W. Pechar, S. Kim, B. Vaughan, E. Marand, M. Tsapatsis, H. K. Jeong, J Membr Sci 2006, 277, 195. T. W. Pechar, M. Tsapatsis, E. Marand, R. Davis, Desalination 2002, 146, 3. I. Vankelecom, VandenBroeck, S., Merckx, E., Geerts, H., Grobet, P., Uytterhoeven, J .B., Journal of Physical Chemistry B 1996, 100, 3753. J. M. Duval, A. J. B. Kemperman, B. Folkers, M. H. V. Mulder, G. Desgrandchamps, C. A. Smolders, J Appl Polym Sci 1994, 54, 409. X. Qiao, Chung, T.S., Rajagopalan, R., Chemical Engineering Science 2006, 61, 6816. J. Y. Lee, Shim, M.J.,Kim, S.W., Polymer Engineering and Science 1999, 39, 1993. J. Y. Lee, Shim, M.J., Kim, S. W. , Journal of Materials Science 2001, 36, 4405. B. Sreedhar, S. Palaniappan, S. Narayanan, Polym Adv Techn012002, 13, 459. B. V. K. Naidu, Bhat, S. D., Sairam. M., Wali, A.C., Sawant, D.P., Halligudi, S. 60 [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] B., Mallikarjuna, N.N., Aminabhavi, T.M., Journal of Applied Polymer Science 2005, 96, 1968. N. Ballav, Biswas, M., Materials Science and Engineering B: Solid-State Materials for Advanced Technology 2006, 129, 270. M. A. Nobile, M. Cannarsi, C. Altieri, M. Sinigaglia, P. Favia, G. Lacoviello, R. D'Agostino, J Food Sci 2004, 69, 379. A. Brody, E. Strupinsky, L. Kline, Active Packaging for Food Applications, Technomic Publishing Co, Lancaster, PA 2001. Y. Wang, Easteal, A. J ., Chen, X. D., Packaging Technology and Science 1998, II, 169. H. Kim, J. Biswas, S. Choe, Polymer 2006, 47, 3981. K. Villberg, A. Veijanen, I. Gustaffson, Polym Eng Sci 1998, 38, 922. T. Andersson, T. Nielsen, B. Wesslen, J Appl Polym Sci 2005, 95, 847. T. Andersson, M. H. Holmgren, T. Nielsen, B. Wesslen, J Appl Polym Sci 2005, 95, 583. A. P. Jacomino, Sarantopoulos, d. L., Grigol, C.I., Sigrist, J.M.M., Kluge, R.A., Minami, K., Journal of Plastic Film and Sheeting 2001, 1 7, 6. G. Van Essche, Kromminga, T., Schmidt, A., Journal of Plastic Film and Sheeting 2000, I6, 155. S. N. Dirim, Ozden, H.O., Bayindirli, A., Esin, A., Journal of Food Engineering 2004, 63, 9. M. Ozdemir, Floros, J ., Critical Reviews in Food Science and Nutrition 2004, 44, 185. B. Erdogan, M. Sakizci, E. Yorukogullari, Applied Surface Science, In Press, Corrected Proof. F. Ozrnihci, Balkose, 1)., Ulku, s., JApplPolym Sci 2001, 82,2913. H. Pehlivan, F. Ozrnihci, F. Tihminlioglu, D. Balkose, S. Ulku, J Appl Polym Sci 2003, 90, 3069. H. Demir, Balkose, D., Ulku, S., Polymer Degradation and Stability 2006, 91, 1079. 61 [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] H. Demir, Arlds, E., Balkose, D., Ulku, S., Polymer Degradation and Stability 2005, 89, 478. H. Pehlivan, D. Balkose, S. Ulku, F. Tihminlioglu, J Appl Polym Sci 2006, 101, 143. H. Pehlivan, D. Balkose, S. Ulku, F. Tihminlioglu, Comp Sci T echnol 2005, 65, 2049. T. T. Moore, T. Vo, R. Mahajan, S. Kulkarni, D. Hasse, W. J. Koros, J Appl Polym Sci 2003, 90, 1574. Y. Yan, Y. Zhao, M. Liu, J Appl Polym Sci 2005, 96, 1460. J. W. Rhim, Hong, S. 1., Park, H. M., Ng, P. K. W., Journal of Agricultural Food Chemistry 2006, 54, 5814. A. L. Ahmad, Nawawi, M.M.G., So, L. K., Journal of Applied Polymer Science 2006, 99, 1740. X. Chen, Yang, H., Gu, Z., Shao, 2., Journal of Applied Polymer Science 2001, 79, 1144. D. Sen, H. Kalipcilar, L. Yihnaz, J Membr Sci 2007, 303, 194. D. Sen, H. Kalipcilar, L. Yilmaz, Desalination 2006, 200, 222. M. J. A. Sales, S. C. L. Dias, J. A. Dias, T. A. Pimentel, Polym Degrad Stab 2005, 8 7, 153. H. L. Frisch, Y. Xue, J Polym Sci Part A: Polym Chem 1995, 33, 1979. S. Djoumaliisky, Zipper, P., Macromolecular Symposia 2004, 21 7, 391 . B. V. K. Naidu, S. D. Bhat, M. Sairam, A. C. Wali, D. P. Sawant, T. M. Aminabhavi, J Appl Polym Sci 2005, 96, 1968. T. T. Moore, R. Mahajan, D. V. Vu, W. J. Koros, AIChE Journal 2004, 50, 311. T. S. Chung, Jiang, L.Y., Li, Y., Kulprathipanja, 8., Progress in Polymer Science 2007, 32, 483. O. Ullaland, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 2005, 553, 107. 62 [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [15m [152] usn “sq nsn nsq nsn nsa S. I. Semenova, Journal of Membrane Science 2004, 231, 189. M. Cecopieri-Gomez, Palacios-Alquisira, J ., Dominguez, J. M., Journal of Membrane Science 2007, 293, 53. R. H. Perry, Green, D., McGraw-Hill Inc. New York, 1984. T. T. Moore, W. J. Koros, J Mol Struct 2005, 739, 87. P. Mahajan, R. Burns, M. Schaefia, W. J. Koros, J Appl Polym Sci 2002, 86, 881. R. Mahajan, W. J. Koros, Polym Eng Sci 2002, 42, 1432. J. M. Duval, Folkers, B., Mulder, M. H. V., Desgrandchamps, G., Smolders, C. A., Journal of Membrane Science 1993, 80, 189. R. Mahajan, W. J. Koros, Ind. Eng. Chem. Res. 2000, 39, 2692. P. Pandey, R. S. Chauhan, Progress in Polymer Science 2001, 26, 853. S. Kim, Marand, 13., Ida, J ., Guliants, V. V., Chemical Materials 2006, 18,1149. P. P. Kundu, Biswas, J., Kim, H., Shim, S.E., Choe, S., Lee, D.S., Advances in Polymer Technology 2004, 23, 230. A. Top, Ulku, S., Applied Clay Science 2004, 27, 13. J. Biswas, Kim, H., Choe, S., Journal of Applied Polymer Science 2006, 99, 2627. D. S. Lee, Shin, D. H., Lee, D. U., Kim, J. C., Cheigh, H. S., Journal ofFood Engineering 2001, 48, 183. S. N. Dirim, A. Esin, A. Bayindirli, Turkish Journal of Engineering and Environmental Sciences 2003, 27, 1. Y. Kimura, US. Patent No: 7202293 2007. N. De Kruijf, M. Van Beest, R. Rijk, A. Sipil, T. Inen-Malm, Food Add Contam 2002, I9, 144. S. Quintavalla, L. Vicini, Meat Science 2002, 62, 373. R. M. Beaudry, Postharvest Biology and Technology 1999, I5, 293. D. A. Faux, J Phys Chem B 1999, 7803. 63 [159] [160] [161] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [175] [176] IZAStructureCommission, Available at: http://izasc.ethz.ch/frni/xsl/IZA- SC/ftc fl.xsl?-db=Atlas_main&-lay=fw&-max=25 &ST C =LTA&-find. Access date: 11/10/2007. J. Zhanga, R. Singha, P. A. Webley, Micropor Mesopor Mater 2008, I I I, 478. D. E. Akporiaye, I. M. Dahl, R. Wendelbo, H. B. Mostad, J Phys Chem 1996, 100, 4153. H. Pierson, Handbook of Chemical Vapor Deposition, Noyes Publication, New York 1999. K. Schuegraf, Handbook of Thin-Film Deposition Processes and Techniques, Noyes Publications, Carson, Ca 1988. W. D. Brooks, Giles, G., Ed. PET Packaging T echnolgy, Sheffield,UK 2002. N. Huige, Master Brewers Association of the Americas 2002, 39, 218. W. Lange, Wyser, Y., Packaging Technology and Science 2003, 16, 149. R. Leaversuch, Available at httv://www._ptonline.com/articles/article__Lrint1.qfin. Access date: 09/20/2006 PPG, Available at: www.ppg.com, Access date: 09/20/2006. Anonymous, Available at http://www.ptonline. com/article/kuw/I 325 6. Access date: 01/2 7/2007 G. Giles, Bain, D., Ed. Materials and Development of Plastics Packaging for the Consumer Market, , Sheffield Academic Press, Sheffield, UK 2000. Inmat-, Available at http:/MWw.inmat.com/tech-packaging.shtml. Access date:01/29/2007 Microcoatinglnc, Available at http://www.mdatechnology. net /techsearch, asp?articleid =5 63. Access date:01/29/200 7. G. D. Dimitroulas, Badeka, A.B., Kontominas, M.G., Polymer Journal 2004, 36, 198. T. Hachisuka, Yoshida, S., Okawara, C., "New Development of Heat Sterilization Durable SiOx Vacuum Coated Film", Denver, CO, United States, 2005. 64 [177] [178] [179] [1 80] [181] [182] [183] [184] [185] [186] [187] [188] [189] n90] H91] [192] D. Howells, Henry, B. M., Assender, H. E., Medico, L., Leterrier, Y.,Manson, J. A. E., "The Influence of the Polyester Substrate on the Structure and Performance of Vacuum-deposited Coatings", Denver, CO, United States, 2005. R. W. Knoll, Theil, J. A., "Effects of Process Parameters on PECVD Silicon oxide and Aluminum Oxide Barrier Films", Chicago, IL, USA, 1995. M. Minelli, Grazia De Angelis, M., Baschetti, M., Doghieri, F., Sarti, G., "Gas Barrier Properties of Polymeric Films with Hybrid Organic/inorganic Coatings for Food Packaging Applications", Cincinnati, OH, United States, 2005. A. P. Roberts, Henry, B. M., Sutton, A. P., Grovenor, C. R. M. ,Briggs, G. A. D., Miyamoto, T., Kano, M., Tsukahara, Y.,Yanaka, M., Journal of Membrane Science 2002, 208, 75. M. Walker, Meermann, F., Schneider, J ., Bazzoun, K., Feichtinger, J ., Schulz, A.,Kruger, J ., Schurnacher, U., Surface and Coatings Technology 2005, 200, 947. S. Dhoot, Freeman, B.D., Stewart, M.E., Available at: http://www. mrw. interscience. wiley. com/emrw/9 7804 71 4 4 0260/epst/article/pst025 /curren t/pdf Access date: 12/10/2006. T. Hegeman, Summer, E., US. Patent Number 0027450 2006. A. Shirakura, Nakaya, M., Koga, Y., Kodarna, H., Hasebe, T., Suzuki, T. , Thin Solid Films 2006, 494, 84. US. Patent Number 01 775 75 . Japan Patent Number 2003328131 . US Patent Number 2006/01 77575 A I . D. G. Shaw, Dawson, E., Cline, D., Langlois, M., US. Patent No. 6231939 1996. D. G. Shaw, Dawson, E., Cline, D., Langlois, M., US. Patent No. 6420003 2000. PRnewswire, Available at http://www. highbeam. com/doc/I G1-80561 014. htrnl. Access date; 12/10/1 00 7. SIG, Available at http:/wnw.sigzllasmax.com .Access date: 12/10/2006 SIGCorpoplast, Available at 65 [193] [194] [195] [196] [197] [198] [199] [200] [201] [202] [203] [204] [205] [206] [207] http://www.sigcor_poplast. com/download/prospekte/si g. biz%2 002_05te.pdf Access date: 12/10/2006. Anonymous, Available at: http:/ www.schotland.com/_pr_challenge.html. Access date: 11/15/2006. A. Montgomery, US. Patent No. 5691007 1997. M. D. Groner, F. Fabreguette, J. W. Elam, S. M. George, Chem. Mater. 2004 16, 639. A. Niskanen, Arstila, K., Ritala, M., Leskela, M. , Journal of Electrochemistry Society 2005, I 52, F90. S. M. George, Chem . Rev 2009, 110(1), 1 1 1. D. Monsma, J. Becker, Mater Matt 2006, I , 5. J. W. Elam, Groner, M.D., George, S.M., Review of Scientific Instruments 2002, 73, 2981 . B. M. Henry, Erlat, A. G., Grovenor, C. R. M., Briggs, G. A. D., Miyamoto, T, Tsukahara, Y., "Permeation Studies of Transparent Barrier Coatings", San Francisco, CA, United States, 2003. P. F. Carcia, R. S. McLean, M. D. Groner, A. A. Darneron, S. M. George, J Appl Phys 2009, 106, 023533. E. Langereis, M. Creatore, S. B. S. Heil, M. C. M. Van de Sanden, W. M. M. Kessels, Appl Phys Lett 2006, 89, 081915. P. F. Carcia, R. S. McLean, M. H. Reilly, M. D. Groner, S. M. George, Appl Phys Lett 2006, 89, 031915. A. A. Darneron, S. D. Davidson, B. B. Burton, P. F. Carcia, R. S. McLean, S. M. George, J Phys Chem C 2008, 112, 4573. R. S. McLean, P. F. Carcia, S. G. Park, Y. Senzaki, Proceedings of the ALD 2003 Conference, American Vacuum Society Topical Conference, San Jose, August 2003. S.-H. K. Park, S.-H. Oh, C.-S. Hwang, J.-I. Lee, Y. S. Yang, H. Y. Chu, Electrochem. Solid-State Lett 2005, 8, H21. G. Abbas, Roy, S., Papakonstantinou, P., McLaughlin, Carbon 2005, 43, 303. 66 [208] [209] [210] [211] [212] [213] [214] A. Kimura, Kodama, H., Suzuki, T., Journal of Vacuum Science and Technology 2003, 21, 515. N. Boutroy, Pemel, Y., Rius, J. M., Auger, F., Bardeleben, H. J., Cantin, J. L., Abel, F ., Zeinert, A., Casiraghi, C., Ferrari, A. C., Robertson, J., Diamond and Related Materials 2006, 15, 921. H. Kodama, Nakaya, M., Shirakura, A., Hotta, A., Hasebe, T., Suzuki, T., New Diamond and Frontier Carbon Technology 2006, I 6, 107. H. Kodama, Shirakura, A., Hotta, A., Suzuki, T., Surface and Coatings Technology 2006, 201, 913. S. Yamamoto, Kodama, H., Hasebe, T., Harakura, A., Suzuki, T., Diamond and Related Materials 2005, 14, 1112. Sidel, Available at: http://www.sidel.com/en/Equipment_sheet_Actis.aspx. Access date: 11/15/2006. A. Ueda, Nakachi, M., Goto, S., Yamakoshi. M., Shirakura, A., Mitsubishi Heavy Industries, Ltd. Technical Review 2005, 42, 1. 67 CHAPTER 3 Poly(lactic acid) and Synthetic Zeolite Composites Prepared by Melt Processing *Extended version of the article Yuzay, I.E., R. Auras, and S. Selke, Journal of Applied Polymer Science, 2009. 115(4): p. 2262-2270. INTRODUCTION Recently, several research efforts have been directed to develop PLA based composites with high performance and/or functionality and extend PLA’s current applications to new areas. The incorporation of various nano/micro inorganic and organic fillers has been shown to result in significant improvements in the properties of PLA [1,2]. As mentioned in Chapter 2, zeolites can be also considered functional fillers. They are based on tetrahedral networks with channels and cavities which can accommodate molecules with various size and shapes. Extensive research has been done on natural and synthetic zeolites due to their intrinsic chemical reactivity, adsorptivity, and ion exchange capacity [3,4]. The unique features of zeolites make them ideal for multi-functional polymer composites. In this study, an attempt has been made to develop an effective melt processing technique and evaluate the feasibility of using synthetic zeolites as functional fillers for the PLA matrix. The objectives of the work presented in this chapter are to: (l) fabricate PLA composites containing a low amount of synthetic 68 zeolites (0 to 5 wt%) in a twin-screw extruder; (2) investigate the morphology of the PLA/zeolite composites in order to assess the dispersion of the zeolites and interfacial adhesion between zeolite particles and the PLA matrix, which provides information on mechanical properties; and (3) evaluate the effect of synthetic zeolites on the physical, mechanical, thermomechanical, thermal, and banier properties of the PLA composites; (4) compare the experimental data on tensile strength to theoretical predictions applying the Nicolais-Narkis [5] and Pukanszky [6,7] models. EXPERIMENTAL Materials Poly(lactic acid) resin produced from 94% L-lactide was obtained from NatureWorks LLC (Blair, NE, USA). Type 4A zeolites were chosen to be used as a model synthetic zeolite for this research. The type 4A synthetic zeolites contain a high amount of aluminum which basically implies its hydrophilic nature and a large number of cations (Na+). It is well known that the localized electrostatic poles between the positively charged cations and the negatively charged zeolite framework strongly attract highly polar molecules, resulting in a hydrophilic structure. Synthetic zeolite (type 4A) (Si/Alzl) with a pore size of 3.8- 4 angstrom was supplied by UOP LLC, (Des Plaines, IL, USA) in the form of powder. An SEM micrograph of type 4A zeolite particles with an average particle size of 1- 2 pm is shown in Figure 3.1. Prior to processing, PLA resin pellets and zeolite powders were dried in a vacuum oven at 60°C for 4 h and at 100°C for 24 h, respectively. 69 Figure 3.1 SEM micrograph of type 4A zeolites. Magnification: x5000; Scale bar: 10 um. Preparation of Composites PLA composites containing 0 to 5 wt% zeolite type 4A were prepared using a micro extruder (DSM Research, The Netherlands) equipped with co-rotating twin-screws having lengths of 150 mm, L/D ratio of 18, and capacity of 15 cc (Figure 3.2). The extrusion was carried out at 185°C and residence time of 5 min at a screw rotation speed of 100 rpm for both PLA and PLA/zeolite composites. After the set extrusion time, the extrudates were collected from the die and transferred into a mini-injection molder (DSM Research, The Netherlands) by a pre-heated transfer cylinder in order to prepare test specimens for physical and mechanical property evaluation. The optimized injection pressure was 896 KPa and the mold temperature was kept at 30°C. The operational 70 parameters for the micro-extruder and injection molder are listed in Table 3.1. Test specimens were wrapped in aluminum foil and stored at 23°C and 50% relative humidity for not less than 40 h prior to testing in accordance with ASTM D 618-03 [8] (Standard Practice for Conditioning Plastics for Testing). The injection molded test specimens are shown in Figure 3.3. ”’I’I'. ‘ ‘ \ 31 K \ \ \ 1 ‘. . , «Qt .4 .2 v > u v r _ m: ,, ~.' . .nz- s16 , ~ .g. ’751‘.-,‘ [1:17; ‘1‘ f .411 : .fi‘fmazfihafi Figure 3.2 DSM twin screw micro extruder (A) and injection molder (B) 71 Table 3.1 Micro extruder and injection molder processing parameters. Micro Extruder Injection Molder Top : 190 Processing temperature profile, Middle: 190 0C Bottom: 190 Melt: 1 83-1 85 Screw speed (rpm) 100 Cycle time (min) 5 Tranfer cylinder temperature, °C 195 Injection pressure, KPa 896 Mold temperature, °C 30 Residence time in the mold, sec 15 Figure 3.3 Injection molded neat PLA and PLA/ zeolite composites. (A) PLA neat; (B) 1 wt% zeolite; (C) 3 wt%; (D) 5 wt%. 72 Characterization of Composites Morphological Properties Scanning Electron Microscopy (SEM): Morphological analyses were done using a SEM JSM 6400 (JOEL, Tokyo, Japan) equipped with a LaB6 emitter on the fracture surfaces of PLA and PLA/zeolite composites. The Izod impact fractured surfaces facing upwards were mounted onto aluminum stubs with carbon adhesive tape and then sputter coated with an approximately 15 nm layer of gold using an Emscope SC500 sputter coater (Emscope Laboratories Ltd, Ashford, UK) operated at 20 mA for 3 min. Finally, SEM micrographs were collected at an accelerating voltage of 15 kV and a working distance of 15 mm. Transmission Electron Microscopy (TEM): TEM bright field images were taken using a JOEL 100CX (JOEL, Tokyo, Japan) operated at an accelerating voltage of 150 kV. PLA/zeolite composite samples (the middle part of tensile dumbbell specimens) were ultra-microtomed with a diamond knife to give approximately 70 nm thick sections, and then the sections floating on the distilled water were transferred to carbon coated copper grids of 300 mesh. Atomic Force Microscopy (AFM): The surface topography of neat PLA and PL/zeolite composite samples was examined using atomic force microscopy (AF M). The AFM measurements were done with a Nanoscope IIIA (Digital Instruments, Santa Barbara, CA) operating in contact mode, at ambient conditions. Images of 30x30 umz 2 . . and 5x5 mm were scanned on two drfferent locatlons of each sample. The root-mean- 73 squared (RMS) roughness calculations were performed by NanoScope software (Digital Instruments, Santa Barbara, CA). Composite film samples with 0.15 mm thickness were used for the AF M studies. Injection molded disks were used to produce thin composite films using compression molding. One injection molded disk was placed between 2 metal plates covered with Teflon sheets and then inserted into a hydraulic press (Hydraulic Unit model #3925, Carver Laboratory Equipment, Wabash, IN, US). The press jaws were set to 190 c’C. First, the metal plates were left in the press without any pressure for 5 min to allow the PLA disk to melt and then a compression pressure of 3 MPa was applied for 3 min in order to obtain thin film samples. Finally, samples were cooled for 15 min. AFM samples were taken from the compression molded film samples. Tensile Properties Dumbbell shaped specimens obtained by injection molding were used for the tensile tests. Tensile properties such as tensile strength, modulus of elasticity, and elongation at break PLA/zeolite composites containing 0, 1, 3, and 5 wt% zeolite were measured in accordance with ASTM D 638-03 [9] using an Instron Model 5565 tensile tester (Instron Corp., Canton, MA, US.) with a load cell of 5 KN. The initial grip distance was 50 mm and the crosshead speed was 50 mm/min. Five specimens were tested for each composite and the mean values and standard deviation are reported. Thermal Properties Differential Scanning Calorimetry (DSC). The thermal analyses of PLA/zeolite composites were performed on a differential scanning calorimeter (DSC) Q100 (TA Instruments, NewCastle, DE, US.) in accordance with ASTM D3418-03 [10]. 74 Approximately 5-8 mg of injection molded samples were heated from room temperature to 190°C at a rate of 10 °C/min under a constant nitrogen flow (70 mL/min). For each sample, the glass transition temperature (Tg), cold crystallization temperature (ch), melting temperature (Tm), and enthalpies of cold crystallization (AHCC) and melting (AHm) were evaluated from the DSC thermograms. The degree of crystallinity (Xc) of PLA and PLA/zeolite composites was calculated based on the following equation: Xc (%) = [( AHm -AHcc)/(AHf(1- X) )] * 100 (1) where AHf is the enthalpy of fusion of an 100 % crystalline PLA which is 93.7 J/g [11,12], and x is the weight fraction of zeolite in the composite. At least five specimens were tested for each composite combination. The degrees of crystallinity of samples were also evaluated using X-ray diffraction (XRD). The X-ray diffraction measurements were performed for PLA and PLA/zeolite composites using an X-ray diffractometer Rigaku 200B (Rigaku Corp, Tokyo, Japan) operated at a voltage of 45 kV and a current of 100 mA, equipped with Cu K a radiation source (1 =1.541 nm). The diffraction data were collected from 20: 2 to 350 with a step width of 0.020 and a step time of 0.4 s. Thermogravimetric Analysis (TGA). TGA was performed using a TGA 2950 (TA Instruments, New Castle, DE, USA) with a heating rate of 20 oC/min from room temperature to 600 C)C under nitrogen flow of 70 mL/min. Three specimens were tested 75 for each composite. The percent weight loss and derivative weight loss were plotted against temperature in order to evaluate the onset and final degradation temperatures at 5% weight loss and the maximum decomposition temperatures of the neat PLA and its composites Dynamic Mechanical Analysis (DMA) The temperature dependence of the storage modulus (E'), loss modulus (E"), and damping factor (tan delta) of PLA/zeolite composites was evaluated using a DMA Q800 (TA Instruments, NewCastle, DE, USA). The test was carried out by heating the samples at a rate of 2 °C/min from room temperature to 90°C. The samples were tested in a dual cantilever mode at an oscillating amplitude of 15 um and a frequency of 1 Hz. A minimum of three injection molded specimens with dimensions of 58 mm x 12 mm x 2 mm (length x width x thickness) were tested for each composite combination. The heat deflection temperature (HDT) of PLA/zeolite composites was also measured by a DMA Q800 (TA Instruments, NewCastle, DE, US.) operating in three- point bending mode according to ASTM D648-03 [13]. The specimens were heated at a rate of 2 C’C/min from room temperature to 80°C under a constant load of 0.455 MPa, and the HDT was determined as the temperature at which the specimens reached 0.2% strain. Three test specimens with dimensions of 58 mm x 12 mm x 2 mm were tested for each composite combination. Optical Properties The transmission of ultraviolet (UV) and visible (Vis) light was measured on a Lambda 25 UV-Visible spectrometer from Perkin-Elmer Instruments (W ellesley, MA, 76 USA). The samples were scanned at a rate of 480 nm/min, over the spectral range from 200 to 800 nm. Melt Flow Index (MFI) Determination The MFI measurements were done to evaluate the effects of zeolites on the extrusion process. A Ray-Ran Melt F low Indexer MK (TMI Testing Machines Inc., USA) was employed to measure the melt flow index of extruded PLA and PLA/zeolite composites in accordance with ASTM D123 8-04 [14], under the conditions of 190°C and 2.16 kg load. Molecular Weight Determination The number-average (Mn) and weight-average (MW) molecular weights and the polydispersity index (PDI=Mw/M,,) of the PLA and PLA/zeolite composites were determined using a gel permeation chromatograph (GPC) equipped with 2414 reflective index (RI) detector (Waters, Milford, MA, USA) and using a series of three columns (HR4, HR3, and HR2). The analysis was conducted at room temperature using tetrahydrofuran (THF) as the mobile phase at a flow rate of 1.0 mL/min. Approximately 35 mg of PLA and PLA/zeolite composite samples were dissolved in 15 mL THF. In the case of PLA/zeolite composites, the zeolite particles were extracted fi'om the solution by centrifugation and then filtration through a 0.45 pm PTFE filter before injection into the GPC. The GPC system was calibrated using polystyrene standards. 77 Barrier properties Water Vapor Transmission Rate (WVTR). The W V'I'R was measured using a Permatran W3/33 from Mocon Inc. (Minneapolis, MN) in accordance with ASTM F 1249-06 “Standard Test Method for Water Vapor Transmission Rate Through Plastic Film and Sheeting Using a Modulated Infrared Sensor”[15] Temperature and relative humidity of the test were 378°C and 90 % RH, respectively. Oxygen Transmission rate (OTR). The OTR was tested using an Illinois 8001 (Illinois Instruments, IL) in accordance with ASTM D 3985-02 “Standard Test Method for Oxygen Gas Transmission Rate Through Plastic Film and Sheeting Using a Coulometric Sensor” [16] Test conditions were 23°C and 0 % RH. Carbon Dioxide Transmission rate (C02 TR). The C02 TR was measured using a Permatran C 4/41 from Mocon Inc. (Minneapolis, MN) according to ASTM F 2476-05 “Test Method for the Determination of Carbon Dioxide Gas Transmission Rate (C02 TR) Through Barrier Materials Using An Infrared Detector”[17] Temperature and relative humidity of the test were 23°C and 0 % RH, respectively. RESULTS AND DISCUSSION Morphology The fracture surfaces of PLA and PLA/zeolite composites were examined using SEM and TEM in order to evaluate the size, shape, and dispersion of zeolites in the PLA matrix. Zeolite type 4A particles exhibited a cubical shape with an average particle size ranging from 0.7-2 [1m (Figures 3.4 and 3.5). Good particle dispersion in the PLA matrix 78 was visually observed. As the fracture stress propagated in the composites, zeolite particles remained embedded into the PLA matrix. At lower magnifications, neither void formation around the zeolite particles nor cavities in the PLA matrix were observed, indicating the existence of good interfacial adhesion between zeolite type 4A particles and the PLA matrix (Figures 3.4 and 3.5). Mahajan and Koros[18] also observed similar improved adhesion between poly(vinyl acetate), PVAc, and zeolite type 4A mixed matrix membranes. The good contact between PVAc and zeolite type 4A was attributed to the affinity of PVAc for alumina, the flexibility of the polymer during membrane formation, and molecular adsorption of the polymer onto the zeolite surface. However, other attempts [19,20] at developing mixed matrix membranes, which combined glassy polymers such as polyimide and poly(ethersulfone),PES, with zeolite type 4A, resulted in extremely poor interfacial adhesion between zeolite and polymer matrices. This was mainly attributed to weak interaction between the zeolite and the polymer matrices, the very rigid nature of the polymers (Tg values varied from 200 to 305°C), and uneven shrinkages and stresses generated during the solvent removal after casting the membranes [21]. Using PLA as a polymer matrix, whose Tg value is around 60°C, and melt compounding the PLA /zeolite composites helped us to obtain good interfacial adhesion. This good interfacial adhesion may also be due to dipolar interaction or hydrogen bonding between PLA and zeolite type 4A. 79 Figure 3.4 TEM image of zeolite. Magnification: X 40,000; Scale bar: 500 nm 80 Figure 3.5 SEM micrographs of neat PLA and PLA/zeolite composites. (A) PLA neat- 200 (A1) (scale bar = 100 um) and 1000 (A2) (scale bar = 20 um) magnification, (D) PLA with 5 wt % zeolite- 200 (D1) (scale bar =100 pm), 1000 (D2) (scale bar = 20 pm), 3300 (D3) (scale bar = 5 pm), and 6000 (D4) (scale bar 5 pm) magnification. Arrows indicate the zeolite particles. 81 The surface topographies of compression molded neat PLA and PLA/zeolite film samples were examined using AF M. The surface roughness (RMS) values, root—mean- square averages of height deviations taken from the x-y plane, were calculated for both neat PLA and PLA/zeolite composites. The RMS values give us an idea whether or not topological changes occurred on the PLA surfaces with the zeolite incorporation. The average RMS values obtained from the scan sizes of 30x30 um2 and 5x5 [m2 were about 164i20 and 85:1:8 nm for the neat PLA, respectively. In the case of PLA with 5 wt% type 4A zeolite, the RMS values were 263i17 nm and 67:1:7 nm for the scan sizes of 30x30 um2 and 5x5 umz, respectively. It is interesting to note that the RMS values measured at a large scan size were higher than those of low scan size for both neat PLA and PLA with 5 wt% zeolite samples which shows the scan size dependence of the measured RMS values. One should note that the presence of surface contaminants such as dust as well as the compression molding process itself could be the main reasons for such high RMS roughness values. If we consider the RMS values taken from the scan size of 5x5 umz, the neat PLA and PLA with 5 wt% zeolite films exhibited similar roughness values. The AFM results suggest that the inclusion of 5 wt% type 4A zeolites may not affect the printability and sealability of PLA. Tensile Properties Table 3.2 presents a comparison of tensile properties of PLA and PLA/zeolite composites with different zeolite loadings. It can be seen that the addition of zeolite type 4A particles increased the tensile strength and modulus of elasticity, while decreasing the 82 elongation at break. The tensile strength of PLA/zeolite composites slightly increased with the addition of 3 wt% zeolite; however, the inclusion of 3 wt% zeolite increased the modulus of elasticity by 20%. With further addition of zeolite (i.e., 5 wt%), the value of the modulus of elasticity leveled off. This amount might be the threshold for the formation of agglomerates. Adding 5 wt% zeolite decreased the elongation at break for the PLA from 10.96 :1: 2.3 to 6.60 d: 0.8%. In general, the addition of rigid particulate fillers increases the tensile modulus, but drastically reduces the elongation at break. In many cases, filler particles also decrease the tensile strength of the polymer matrix, but there are some exceptions. In these cases, the tensile strength may increase due to good adhesion between the filler and matrix [22,23]. It is clear that the interfacial adhesion between the zeolites and the PLA matrix and the dispersion of zeolite particles in the PLA matrix were good enough to maintain the tensile strength. Metin et al. [24] and Sun et a1. [25] reported a decrease in tensile strength when combining poly(propylene), PP, with natural zeolite, and chitosan with H-ZSM-S, synthetic zeolite Si/Al ratio of 25, respectively. The investigators tried to overcome the reduction in the tensile strength of PP and chitosan matrices by using several silane coupling agents, improving the filler- polymer adhesion at the interface. 83 Table 3.2 The effect of zeolite content on tensile properties of PLA. N1’ N2 Zeolite Loadings Tensile Strength Elongation at Modulus of Elasticity wt % (MPa) Break (%) (MPa) 0 62.2 d: 5.4 a 10.9 4 2.3 5 1,221 4 64.5 a 1 62.5 i 3.2 a 6.9 4 2.0 a 1,295 2‘: 127.1 a 3 67.9 3: 1.4 b 6.9 3: 0.2 a 1,465 4 147.5 b 5 64.2 2: 3.4 a b 6.6 :1: 0.8 a 1,465 4 56.1 b N1: Mean :1: standard deviation N2: Means with different superscript letters in the same columns are significantly different (P < 0.05, Tukey) A number of theories and empirical models have been developed to evaluate the mechanical and interfacial properties of polymers filled with particulate fillers. Such models usually assume either that there is no interaction between the filler and the matrix or that there is good adhesion between the filler and the matrix. Since the nature of interactions between zeolite and PLA matrix is not yet clear, we considered all assumptions as plausible scenarios for PLA/zeolite composites when using individual models to predict the tensile strength. Although cubical zeolite particles were introduced into the PLA matrix, the models developed for spherical particles were used to predict the mechanical properties. It was assumed that the models developed for the spherical fillers would be applicable to zeolite particles since cubical particles have the same dimension in all directions. The Nicolais-Narkis model [5] and its modifications are the most widely used models to predict the effects of interfacial interaction on tensile yield strength for the case of poor adhesion between the spherical filler particle and the polymer matrix. In 84 the Nicolais-Narkis model, the tensile strength values of the composite can be predicted by the following equation: _ _ 2/3 (2 ) ac — am (1 K V f ) where (Sc and cm are the tensile strength of the composite and matrix, respectively, and Vf is the volume fraction of filler. K is a parameter expressing the filler-matrix interaction. According to this model, the value of K becomes 1.21 in the absence of adhesion between the filler and the matrix, which means the load is sustained only by the polymer matrix and addition of fillers leads to reduction in the tensile strength of the composite [26,27]. If the value of K is less than 1.21, the tensile strength of the composite increases, which is evidence of improved adhesion between the filler and the matrix. In the case of strong filler-matrix interfacial adhesion, a simple model developed by Pukanszky et a1. [6,7] to predict the tensile strength of the polymer composites can be expressed by the following equation: a 1—V, C a 1+ 251/, exp(BVf) (3) m where (Sc and Om are the tensile yield stress of the composite and the matrix, respectively, V f is the volume fraction of the filler, and B is an interaction parameter, which depends on the surface area of the fillers, the thickness of the interface, filler density, and interfacial bonding energy. The pre-exponential term in Eq. (3) is related to 85 the decrease of the effective load-bearing cross-section of the matrix while the exponential term, exp (BVf ), describes the interfacial interactions in the composite [26,28,29]. Eq. (3) can be rearranged in a linear form. Ln [(O'c(1+2.5 Vf)/( Gm(l-Vf)] is plotted as a function of Vfand then the interaction parameter, B, can be calculated from the slope of the straight line. In general, if the value of B is zero, the filler acts as a void. The higher B value indicates stronger interfacial interactions between the filler and the matrix. It was shown in the literature [24,30,31] that if the value of B is more than 3, the filler-matrix interface is considered strong and a reinforcing effect of filler can be obtained. In this study, both the Nicolais-Narkis and Pukanszky equations were applied to the tensile stress data of PLA/zeolite composites in order to evaluate the interfacial interaction between the zeolites and the PLA matrix. It can be seen from Figure 3.6 that the Nicolais-Narkis model based on the hypothesis of zero adhesion between the zeolite and the PLA matrix predicted a decrease in tensile strength that was not observed, which is evidence of the interfacial adhesion between zeolite particles and the PLA matrix. 86 Tensile strength, MPa 8 8 s 8 0 2 4 6 Zeolite content, vol % Figure 3.6 Comparison of experimental tensile strength data with the Nicolais-Narkis model for PLA/zeolite composites. Experimental tensile strength data (0) and Nicolais- Narkis predicted data (I) with the adhesion K value of 1.21. The Pukanszky model was also applied to experimental tensile data under the assumption that good bonding exists between the zeolites and the PLA matrix. The plot of Ln [(O’c(1+2.5 ij/( Gm(1-Vf)] as function of Vfis shown in Figure 3.8. The average B values calculated from the slopes of the straight lines for each zeolite volume fraction are also indicated on Figure 3.7. All of the average B values were higher than 3, which confirms the presence of interfacial interactions between the zeolite particles and the PLA matrix. The composites containing 3 wt% zeolite exhibited the highest B value, 6.0, implying the strongest interfacial interaction compared with PLA/zeolite 1 wt% and PLA/zeolite 5 wt% composites. This prediction agrees with the slightly higher tensile 87 strength measure of PLA/zeolite 3 wt% shown in Table 3.1. Increasing the filler content from 3 to 5 wt% led to a reduction in the B value, yet still the B value was higher than 3, which might prevent the occurrence of debonding at the zeolite-PLA interface. The comparison of the experimental data with the Pukanszky model is also shown in Figure 3.8. The experimental data points remain between the straight lines obtained from the Pukansky equation using the B values of 3.0 and 6.0. In the study of Metin et a1. [24], PP composites containing untreated and treated natural zeolites were evaluated in terms of interfacial adhesion. PP with untreated natural zeolite composites exhibited a B value of minus 9. Treatment of the natural zeolites with an amino functional silane coupling agent increased the B value to 2.15 which was considered an indication of strong interaction. Similar results were also reported by Demjen et al. [32] for PP/CaCO3 composites. Thus, the B values obtained from the present study without using any coupling agents indicated good interfacial interactions between the zeolite particles and the PLA matrix. 88 0.30 s? E 0.20— is e; 0.10- 3 E 0.00 " T l 0.00 0.02 0.04 0.06 Volume fraction of zeolite, V, Figure 3.7 The tensile stress of PLA/zeolite composites in the linear form of Eq.(3). The average interaction parameters (B) determined from the slopes of the straight lines. Note: The 95 percent confidence interval of the mean B values are 5.7, 0.7, and 1.1 at the zeolite volume fractions of 0.01126(1 wt%), 0.03369(3 wt%), and 0.05601(5 wt%), respectively. 89 1.25 0 Experimental - - - -B: 6.0 ------- 8:3.0 ——8: 0 1.15 7 1.05 — ocl O'm 0.95 .. 0.85 . 0.75 1 1 0.00 0.02 0.04 0.06 Volume fraction of zeolite, V, Figure 3.8 Experimental and predicted relative tensile yield stress (O'c / O’m) of PLA composites as a function of zeolite volume fraction: 0 Experimental data; straight line according to the Pukanszky equation with B:0; dotted and dashed lines according to the Pukanszky equation with 323.0 and 6.0, respectively. 90 Thermal Analysis The thermal characteristics of PLA and PLA/zeolite composites were evaluated using DSC. The transition temperatures along with the calculated percentages of crystallinity of PLA and PLA/zeolite composites are summarized in Table 3.3. The zeolite type 4A content did not affect the glass transition and melting temperature of the PLA matrix; however, the cold crystallization and percent crystallinity were affected. PLA and PLA/zeolite composites exhibited not only well-defined Tg values at around 58°C but also endothermic peaks immediately after the glass transition temperatures. These endothermic peaks represent excess enthalpy relaxation resulting fi'om the thermal and mechanical history of PLA and PLA/zeolite composites prepared by extrusion followed by injection molding. The representative DSC heat flow curves for the unprocessed PLA (PLA pellets), PLA, and PLA with 5 wt% zeolite composites are shown in Figure 3.10. As expected, there was no endothermic peak following the Tg for the unprocessed PLA due to the lack of thermal and mechanical history. In addition, a weak and broad glass transition appeared for the PLA pellets. The deflection point at the base line was difficult to observe. This could be attributed to a high degree of crystallinity of the pellets. The mobility of the molecules in the amorphous region could be confined by the presence of the crystalline regions, which makes the Tg difficult to detect [33]. The endothermic peaks at around 58°C for PLA and PLA with 5 wt% zeolite were clearly evident in the thermograms. Upon heating from room temperature, the PLA molecular chains gained mobility after the glass transition and then began formation of crystallites in the region of the cold crystallization exotherm. As can be seen from Table 91 3.3, all of the injection molded samples, excluding PLA pellets, underwent cold crystallization at the heating rate of 10 °C/min. In general, a low cold crystallization temperature is considered as an indirect indication of fast crystallization. With the addition of 5 wt% zeolite, the cold crystallization temperature of PLA shifted from 123.6 :1: 4.0 to 114.3 d: 16°C. This reduction in the cold crystallization temperature suggests a nucleating effect of the zeolite particles. A similar nucleating effect of zeolites in PP crystallization has been reported by Pehlivan et al. [34]. In a study of mesoporous zeolites (MGM-41), Run et al. [3 5] prepared PET/mesoporous zeolite composites using in-situ polymerization and observed that the mesoporous zeolites acted as nucleating agents in the PET matrix. Another interesting phenomenon is the bimodal (double peak) melting endotherrns of PLA and PLA/zeolite composites. Although the unprocessed PLA exhibited only one melting endotherm at 158.6 i 10°C, the injection molded PLA samples showed the main melting peakat 148.0 at 37°C and an additional peak (a small shoulder) at 153.8 i 07°C. Several researchers have observed similar bimodal melting behavior for PLA and its composites [36-39]. These peaks were attributed to the existence of different crystalline structures (i.e., variation in thiclmess of the larnellas and size of spherulites) being formed during heating and cooling in the DSC scans or processing [39-41]. The lower temperature peak, Tm], usually corresponds to [3 crystals and the upper temperature peak, Tmz, corresponds to 0. crystals of PLA as established in 92 the literature [39,42,43]. In the case of PLA—filler composites, the inclusion of fillers might also accelerate the growth of additional crystalline forms by creating defects in the structure [33]. Table 3.3 Thermal characteristics of PLA and PLA/zeolite compositesm’ N2 332?. Tm<°C> wt% Tg (0C) ch(°C) Tml Tm2 Xe (%) 0* 56.4 4 0.9** - ' 158.6 4 1.0 a 37.3 4 1.6 a 0 58.54 0.7 a 123.6440a 148.04 3.7a 153.8407 b 3.2407b 1 57941.7al 119.0412b 149.9408a 153.9411b 5241.1cd 3 57.5 41.5 8 116.7434" b 149.14 1.6 a 154.7 4 2.2 b 6.2 4 2.2 d e 5 58741.7a 114.3416c 148.0414a 155.0412b 7.6412 c * PLA pellet (unprocessed PLA); ** Week and broad heat increment N1: Mean i standard deviation; N2: Means with different superscript letters in the same columns are significantly different (P < 0.05, Tukey) As can be seen from Table 3.3 and Figure 3.9, there were no significant changes in these melting temperatures with the addition of zeolites into the PLA matrix; however, the double melting peaks were more pronounced upon increasing the zeolite content. The percent crystallinity of PLA pellets and injection molded PLA samples were found to be 37.3 i 1.6 and 3.2 i: 0.7%, respectively. DCS results suggest that injection molded PLA test specimens were nearly amorphous and the percent crystallinity of injection molded PLA significantly increased with increasing zeolite. 93 0.5 PLA pellet 0.01 \, ........ 1 ~~~~~~~~~~~~~~~~~ ”i ‘*~~‘ / \\ I \\ l ‘ \ 'PLA 5 wt% w . E B 2 Fit .... -1.0- a o m .15. ' lj PLA \ extruded 20 IIIIIII I rrrrrrr I f 111111 r r j l 30 70 110 150 190 EXO UP Temperature, °C Figure 3.9 Representative DSC therrnograrns of PLA pellet, PLA extruded, and PLA with 5 wt% zeolite composites (Heating Ramp: 10 °C/min). 94 DSC results were also confirmed by the x-ray diffraction measurements performed on injection molded PLA and PLA/zeolite composites. Figure 3.10 shows the XRD patterns for the neat PLA and PLA/zeolite composites. As can be seen, the broad peak observed at approximately 26=15~16° with no distinct crystal peak suggests that the neat PLA has a predominantly amorphous structure. In general, the crystalline peak of the PLA appears at 26=16.6°, corresponding to diffraction of the (200) and/or (110) plane of typical orthorhombic crystals.[44,45] On the other hand, after inclusion of 3 and 5 wt% type 4A zeolites in the PLA matrix, four narrow sharp diffraction peaks at 7.84, 8.78, 22.96, and 23.820 were detected, which can be ascribed to the zeolites’ characteristic crystalline peaks [46]. 95 Intensity (A.U) 2 1 1 20 29 2 Theta (degrees) Figure 3.10 XRD patterns for PLA and PLA/zeolite composites. A: PLA; B: PLA/1 wt% type 4A; C: PLA/3 wt% type 4A; D: PLA/5 wt% type 4A Dynamic Mechanical Thermal Analysis Figure 3.11 (A) shows the storage and loss modulus of PLA and PLA zeolite composites at various zeolite loadings. The storage modulus denotes the maximum energy stored in the material during straining [the oscillation cycle or cycle of sinusoidal deformation]. When the applied mechanical energy is not stored, it is converted into heat via molecular fiiction. Therefore, this energy dissipated as heat is called the loss modulus. As can be seen in Figure 3.11 (A), the storage modulus of PLA/zeolite 96 composites was higher than that of the neat PLA matrix. Addition of 5 wt% zeolite type 4A increased the storage modulus of PLA fiom 3100 to 3854 MPa at 30°C; however, the storage modulus was not affected by low zeolite content (i.e., 1 wt%) at room temperature. When the PLA was heated through the glass transition region, zeolites had more pronounced effects on the storage modulus. In the vicinity of 60°C, the storage modulus of PLA with 5 and 3 wt% zeolites dropped abruptly while the storage modulus of neat PLA exhibited a sudden decrease at about 49°C. As can be seen, the stiffening effect of the zeolites was significant at higher temperatures. The increase in storage modulus with increasing zeolite content means that the stress was transferred from the matrix to the zeolite particles that introduced stiffness into the matrix. Similar behavior was reported by other researchers using fibers, talc, and clays [47 ,48]. The loss modulus values also increased with zeolite loading. The loss modulus of PLA and PLA with 5 wt% zeolite composites reached a maximum value of 668 MPa at 590°C and 901 MPa at 606°C, respectively. These results show that PLA/zeolite composites have better ability to dissipate the vibrational energy as heat than does the neat PLA. 97 . 1000 g: f9 r S C 800 v? . 3 . fl - ha I: ~ 600 E O ' a 2 ~ g o ‘ -- g0 - 400 I; s - g 03 . m m - 200 O , u—l :13 o G E a [.1 20 Temperature, °C Figure 3.11 Storage modulus (A), loss modulus (A), and tan delta (B) of PLA and PLA/zeolite composites as a function of temperature.(a— PLA); (b:— _ PLA with 1 wt% zeolite);(c:— - — PLA with 3 wt% zeolite); (d'_. . . — PLA with 5 wt% zeolite) 98 Figure 3.11(B) exhibits the tan delta temperature curves for PLA and PLA/zeolite composites. The tan delta denotes material damping characteristics. It is mainly a measurement of the ratio of loss modulus to storage modulus. The tan delta peak of the PLA with 5 wt% zeolite composite shifted to a slightly higher temperature. Additionally, the intensity of the tan delta peak was increased in comparison with the neat PLA. This high magnitude of tan delta may suggest that the PLA with 5 wt% zeolite composites has better damping properties than the neat PLA. Nielsen and Landel suggest [22] that the effects of rigid fillers on the damping of polymers can be estimated by the following equations: (‘15,- = tandeltamp = (tandeltamm * omw) + (tan delta fine, * (”fizz”) camp (4) E : tandeltawmp 2 tandeltamatrix *(I_ (”fill”) (5) camp where E" and E. are the loss and storage modulus, respectively, and (0mm and (0 filler are the volume fractions of matrix and filler, respectively. Since the damping of most rigid fillers is very low compared to the damping of the polymers, the second term in Eq. (4) can be neglected. Therefore Eq. (4) can be rewritten as Eq. (5). This equation suggests that the incorporation of fillers decreases the peaks of energy dissipation curves due to the decrease in volume fraction of the matrix. In the literature, there are several studies showing a decrease in tan delta values after inclusion of fillers [47,49]. In our study, the effect of the zeolites on the dissipation energy was an exception; the zeolite 99 increased the damping properties. Nielsen and Landel [22] suggest that the increased damping may result from induced thermal stresses or changes in polymer conformation or morphology. Table 3.4 The heat deflection temperature (HDT) of PLA/zeolite composites. N1’ N2 Zeolite Content (wt%) HDT ( °C) 0 5514018b 1 55.5405b 3 54.24058 5 54.4403a N1: Mean i standard deviation N2: Means with different superscript letters in the same columns are significantly different (P < 0.05) The heat deflection temperature (HDT) of neat PLA was found to be 55.1 4 01°C. The 3-point bending test showed that the HDT value of PLA was not much affected by the presence of zeolite (Table 3.4). This implied that either the amount of zeolite was not sufficient to improve the HDT or that zeolites are incapable of raising the HDT. The HDT might be improved by raising the degree of crystallinity, glass transition and melting temperatures, or a combination. While the percent crystallinity of PLA increased fiom 3.2:t0.7 to 7.6i1.2 with 5 wt% zeolite, this small increase did not result in any change in the HDT value. 100 Optical Properties UV-Vis spectra of neat PLA and PLA/type 4A composites at a film thickness of 0.15 and 2 mm are shown in Figure 3.12. At the film thickness of 0.15 mm, the transmission spectrum of PLA in the visible light region (400-800 nm) is not affected by the presence of the type 4A zeolite loadings. The PLA and all PLA/type 4A composites allowed about 93-95 % transmission in the visible light region, indicating their high transparency. On the other hand, the addition of 5 wt% type 4A zeolites reduced the UV light transmission of PLA from 82i1 to 75:1:0.7% at 250 nm. In the case of using thicker (2 mm) composites samples, there was a large reduction in the amount of visible and UV light transmission. PLA, PLA/1 wt% type4A, PLA/3 wt% type4A, and PLA/5 wt% type4A composites had a transmission of 80 :t 0.9, 75 :1: 0.3, 65 i 0.1, and 48 d: 0.4% at 400 nm, respectively. The optical clarity of PLA decreased as a function zeolite loading, reflecting that there is strong scattering of zeolite particles resulting in lower transparency. At 300 nm, the neat PLA transmitted 33 d: 0.5% of the light while the PLA/5 wt% type 4A composites transmitted only 10 :t 0.4% of the light. This strong absorption value in the region of UV light suggests that the PLA with 5 wt% zeolite composites provide better UV barrier than the neat PLA. The effects of UV light on nutritional values, flavor, and colors of packaged food and beverage products are well known. The improvement that we observed in PLA/ type 4A zeolite samples helps to maintain quality of the packaged product through an extended shelf life. 101 90 80' 60 Transmission, % if fi 300 400 500 600 700 800 900 Wavelength, nm UV REGION 100 "' [if/4" / 60 * II Transmission, % 200 300 400 500 600 700 800 Wavelength, nm Figure 3.12 UV/vis transnnssion spectra of PLA and PLA/type 4A composites. Sample thickness: 1: 0.15 mm, 11: 2 mm A: PLA; B: PLA/1 wt% type4A; C: PLA/3 wt% type4A; D: PLA/5 wt% type4A zeolites 102 Melt Flow Index (MFI) Determination The melt flow index (MFI) is generally defined as the weight of the polymer extruded in 10 min through an extrusion plastometer whose dimensions are specifically defined by ASTM D 1238-04. The MFI is an inverse measure of the melt viscosity which is a basic quality control parameter and widely used in plastics industry to evaluate polymer processability for application such as blown film extrusion, cast fihn extrusion, extrusion blow molding, injection molding, injection blow molding, fiber spinning, etc. In general, the lower MF I polymers are preferred for blow molding or extrusion processing systems. On the other hand, higher MFI polymers are used with injection molding. The effect of type 4A zeolite on the extrusion process was determined using a melt flow indexer (extrusion plastometer). Table 3.5 shows the melt flow indices in g/10 nrin obtained at 190 0C using a 2.16 kg load for neat PLA pellets (as received), neat PLA extruded and PLA/type 4A zeolite composites at 5 wt% loading level. The MFI of the extruded neat PLA was found to be somewhat higher than that of the neat PLA pellets, indicating the effect of the melt processing. With the addition of type 4A zeolites, the MFI of the neat PLA increased from 9.23 d: 0.75 to 56.17 i 4.55 g/ 10min; in other words, inclusion of type 4A zeolites significantly reduced the viscosity of the PLA matrix. As mentioned earlier, having higher MF I values can be very useful for high-speed and multi- cavity injection molding systems where a shorter cycle time and higher output is needed. High MFI values also allow running at lower melt temperatures which reduces the cooling time, and therefore significant energy savings can be achieved. On the other 103 hand, an increase in the MFI may create challenges for extrusion and extrusion blow molding where the high melt strength is necessary. It should be also noted that the reduction in melt viscosity did not detrimentally affect the mechanical and thermomechanical properties of the PLA. PLA/zeolite composites maintained their tensile strength and even improved the storage modulus (i.e., stiffness). Table 3.5 Melt flow indices (MFI) of PLA and PLA/type 4A zeolite composites. Zeolite Content, MFI, g/10 min wt. % 0* 7.57408a 0” 9.23 4 0.7 b 5 C 56.17 4 4.55 *PLA neat pellets (as received), MPLA neat extruded Molecular Weight Determination GPC was used to determine the number-average molecular weight (Mn), weight- average molecular weight (MW), and polydispersity index (PDI) of PLA and PLA/type 4A composites. Table 3.6 summarizes these molecular weight parameters. For comparison purposes, PLA pellets (unprocessed) were also analyzed. The PLA pellets had Mn=106,574 4 11,966 g/mol, Mw=162,598 4 4,437 g/mol, and PDI=1.53 4 0.13 as determined by GPC. Approximately 16% and 9% reductions in M11 and MW values were observed during the extrusion/injection molding process due to the thermomechanical 104 degradation of PLA. A similar reduction in molecular weight due to the melt processing of PLA was reported by several researchers [11,50,51]. PLA/ 1 wt%, PLA/ 3 wt%, PLA/5 wt% zeolites composites exhibited a decrease in Mn and MW when compared with the processed PLA. The reduction in MD and MW values were 6.4, 33, 43% and 10, 33, 42%, respectively. Although the changes in Mn and MW were significant for extruded PLA and PLA/zeolite composites, polydispersity indices remained constant. It is interesting to note that with low molecular weight values, the PLA with 5 wt % type 4A sustained the mechanical properties of neat PLA. Sinha Ray et al. [11], Pluta [50], and Hiroi et al. [52] have reported that the incorporation of organically modified layered silicates (i.e., synthetic fluorine mica and montrnorillonite, Cloisite 30B) and titanates have resulted in similar reductions in Mn and MW of the PLA matrix. The combined effects of the shearing forces during melt processing and the interaction of PLA with hydroxy groups of fillers were found to be responsible for the decrease in molecular weight. 105 $33. .36 v 4: 28.6446 32:85ch 0.8 4.532 Brahma Eofitmw firs .8828 06% 05 5 $822 6622433 @8286 5:» 6.888 23:65 mofie> H 802 @6me005 .44.: 4 2.4 4 43154 6.54 24.4 4 4432132 .3644 84.: 4 332152 8:8 <44 4 3o 4 m: N... 4 $6.... 4 Show a. E? .4 4 4.3.8 5.2:. $4: 343 4 mod 4 43 mm 6 $3 4 444.66 mm 6 £44 4 £34 «.425 $2., 93.— 4 36 4 S; 2 a 54.4 4 646.42 4.6 a :44 4 62.44 38.5 $4.: :3.— 4 3.6 4 8.2 - 4 so 4 and: - 4 $3 4 24.44 483.8 .34 £52.13— :38: o . o 956 12 432V 5.— .25» .32 e m .25» :2 _ m .x. a\.. 484858 <4 252: 4:4 <54 .6 £38 040 6.4 634.4 106 Barrier Properties The results of water vapor, oxygen, and carbon dioxide permeability measurements and the ratio of C02/02 permeability (i.e., permselectivity) of PLA, PLA/zeolite and commercially available zeolite filled film (Evert-fresh Green Produce Bags from Evert-Fresh Corp, Houston, TX) are shown in Table 3.7. Incorporation of 5 wt% type 4A into the PLA matrix resulted in higher water vapor, oxygen, and carbon dioxide permeability values than those of neat PLA. The permselectivity (COz/OZ) for the PLA/zeolite composite was found to be 0.16, implying their potential suitability for modified atmosphere packaging systems. In these systems, the 02 consumption rate and C02 production rates should be in balance with the 02 and C02 permeability of the packaging films. In general, polymer films used for food packaging have a ratio of C02/02 permeability approximately 4-8:1 [53]. In many cases, this ratio is not very favorable because it allows C02 permeation at higher rates than 02. Therefore, zeolite can be used for altering the COz/Oz permselectivity of the modified atmosphere packaging systems. Table 3.7 Gas permeability values obtained from neat PLA, PLA/zeolite composite films, and commercially available zeolite containing film. Permeability Permselectivity Water Oxygen" Carbon vapor* [cc.mil/ dioxide*** a [g.mil/ mzday.atm] [cc.mil/ C02/02 mz'day.atm] m2°day.atm] PLA 553 d: 122 1,444 i 335 3,590 i 230 ~2.48 PLA/ 5 wt 656 :1: 118 67,728 :1: 6750 11,366 i 920 ~0.16 type 4A Evert-fresh 23.5 i 1.7 24,433 i 8427 28,385 4 3220 ~1.16 *38.7 °C, 100%RH, ** 23 °C, 0% RH, ***23 °C, 0%RH, 107 CONCLUSION PLA/type 4A synthetic zeolite composites were successfully fabricated using extrusion followed by injection molding processes. The morphological studies showed a homogenous dispersion of zeolite type 4A in the PLA matrix. As the stress propagated through the composites, zeolite particles remained embedded into the matrix, indicating the existence of good interfacial adhesion between zeolite particles and the PLA matrix. The percent crystallinity of the PLA increased with the proportion of zeolites while no significant changes occurred in the glass transition, melting and heat deflection temperature. With addition of 5 wt % zeolite, the storage modulus, loss modulus and damping were enhanced. The modulus of elasticity was also positively correlated with zeolite content while the elongation at break was reduced with increasing zeolite loading. The results obtained from UV-visible transmission spectra also suggested that PLA/ type 4A composites may act as a better UV barrier than neat PLA. Thus, combining PLA and type 4A synthetic zeolite particles could bring several potential applications. 108 REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] 111] [12] [13] [14] R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4, 835. S. Sinha Ray, M. Bousmina, Prog. Mater. Sci. 2005, 50, 962. R. M. Barrer, Zeolites and Clay Minerals as Sorbents and Molecular Sieves, Academic PresszLondon, 1978. R. T. Yang, Adsorbents: Fundamentals and Applications, Wiley: Hoboken, N.J., 2003. L. Nicolais, M. Narkis, Polym Eng Sci 1971, I 1, 194. B. Pukanszky, B. Turcsanyi, F. Tudos, Efl'ect of interfacial interaction on the tensile yield stress of polymer composites. In Interfaces in polymer, ceramic and metal matrix composites; Ishida H., Ed. ; Elsevier: Amsterdam, 1988. p. 46 7—77. B. Turcsanyi, B. Pukanszky, F. Tudos, J Mater Sci Lett 1988, 7, 160. American Society for Testing and Materials. Standard Practice for Conditioning Plastics for Testing. D 618-03. In Annual Book of ASTM Standards. AST M: Philadelpia, PA, 2004: 8:01. American Society for Testing and Materials. Standard Test Method for Tensile Properties of Plastics. D 638-03. In Annual Book of ASTM Standards. ASTM: Philadelpia, PA, 2004: 8:01 . . American Society for Testing and Materials.Standard Test Method for Transition Temperatures and Enthalpies of Fusion and Crystallization of Polymers by Differential Scanning Calorimetry. D 3418-03. In Annual Book of ASTM Standards. AST M: Philadelpia, PA, 2004: 8:02 S. Sinha Ray, K. Yarnada, M. Okamoto, A. Ogami, K. Ueda, Chem Mater 2003, 15, 1456. R. Auras, B. Harte, S. Selke, J Appl Polym Sci 2004, 92, 1790. American Society for Testing and Materials. Standard T est Method for Deflection Temperature of Plastics Under F lexural Load in the Edgewise Position. D 648- 03. In Annual Book of AST M Standards. AST M: Philadelpia, PA, 2004: 8:01 . American Society for Testing and Materials. Standard Test Method for Melt Flow Rates of Thermoplastics by Extrusion Plastometer. D 1238-04. In Annual Book 109 [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] of ASTM Standards. ASTM: Philadelpia, PA, 2004: 8:01. American Society for Testing and Materials. Standard Test Method for Water Vapor Transmission Rate Through Plastic Film and Sheeting Using a Modulated Infrared Sensor. F 1249-06. In Annual Book of ASTM Standards. ASTM: Philadelpia, PA, 2006: 15.02. American Society for Testing and Materials. Standard Test Method for Oxygen Gas Transmission Rate Through Plastic Film and Sheeting Using a Coulometric Sensor. 03985-02. In Annual Book of AST M Standards. AST M: Philadelpia, PA, 2004: 15.10. American Society for Testing and Materials. T est Method for the Determination of Carbon Dioxide Gas Transmission Rate (C0 2T R) Through Barrier Materials Using An Infiared Detector. F 24 76-05. In Annual Book of AST M Standards. ASTM: Philadelpia, PA, 2006: 15.10. R. Mahajan, W. J. Koros, Ind. Eng. Chem. Res. 2000, 39, 2692. T. W. Pechar, S. Kim, B. Vaughan, E. Marand, M. Tsapatsis, H. K. Jeong, J Membr Sci 2006, 277, 195. T. T. Moore, W. J. Koros, J Mol Struct 2005, 739, 87. Y. Li, H. Guan, T. Chung, S. Kulprathipanja, J Membr Sci 2006, 275, 17. L. E. Nielsen, R. F. Landel, Mechanical Properties of Polymers and Composites, Marcel Dekker, New York 1994. D. Sekutowski, in Plastics Additives and Modifiers Handbook, (Ed: J. Edenbaum), Van Norstrand Reinhold, New York 1992, 493. D. Metin, F. Tihminlioglu, D. Balkose, S. Ulku, Comp Part A: Appl Sci Man 2004, 35, 23. H. Sun, L. Lu, X. Chen, Z. Jiang, Appl Surf Sci 2008, 254, 5367. S.-Y. Fu, X.-Q. Feng, B. Lauke, Y.-W. Mai, Comp Part B: Eng 2008, 39, 933. J. L. Willett, F. C. F elker, Polymer 2005, 46, 3035. L. Danyadi, K. Renner, J. Moczo, B. Pukanszky, Polym Eng Sci 2007, 4 7, 1246. A. Kiss, E. Fekete, B. Pukanszky, Comp Sci T echnol 2007, 67, 1574. T. Kaully, A. Siegrnann, D. Shacham, Polym Comp 2008, 29, 396. 110 [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] G. C. Papanicolaou, A. G. Xepapadaki, A. Kotrotsos, D. E. Mouzakis, J Appl Polym Sci 2008, 109, 1150. Z. Demjen, B. Pukanszky, J. Nagy, Comp Part A: Appl Sci Man 1998, 29, 323. T. Hatakeyama, F. X. Quinn, Thermal Analysis: Fundamentals and Applications to Polymer Science, Wiley, Chichester, England 1999. H. Pehlivan, D. Balkose, S. Ulku, F. Tihminlioglu, Comp Sci T echnol 2005, 65, 2049. M. Run, S. Wu, D. Zhang, G. Wu, Polymer 2005, 46, 5308. Y. T. Shieh, G. L. Liu, J Polym Sci Part B: Polym Phys 2007, 45, 1870. M. D. Sanchez-Garcia, E. Girnenez, J. M. Lagaron, Carbohydr Polym 2008, 71 , 235. G.-X. Chen, H. Shimizu, Polymer 2008, 49, 943. S. Jacobsen, H. G. Fritz, P. Degee, P. Dubois, R. Jéréme, Polymer 2000, 41 , 3395. M. Yasuniwa, S. Tsubakihara, Y. Sugimoto, C. Nakafuku, J Polym Sci Part B: Polym Phys 2004, 42, 25. Y. Wang, J. F. Mano, Eur Polym J 2005, 41 , 2335. P. Pan, B. Zhou, W. Kai, T. Dong, Y. Inoue, J Appl Polym Sci 2008, 107, 54. M. L. DiLorenzo, Macromol Sym 2006, 234, 176. Y. Furuhashi, Kimura, Y., Yarnane, H., J Polym Scie Part BPhysics 2007, 45, 218. J. Y. Nam, Okamoto, M., Okamoto, H., Nakano, M., Usuki, A., Matsuda, M., Polymer 2006, 4 7, 1340. M. M. J. Treacy, J. B. Higgins, Collection of simulated XRD powder patterns for zeolites, Elsevier, Amsterdam 2001. M. S. Huda, L. T. Drzal, A. K. Mohanty, M. Misra, Comp. Part B: Eng. 2007, 38, 367. P. V. Joseph, G. Mathew, K. Joseph, G. Groeninckx, S. Thomas, Comp Part A: Appl Sci Man 2003, 34, 275. 111 [49] [50] [51] [52] [53] J. Lu, T. Wang, L. T. Drzal, Comp Part A: Appl Sci Man 2008, 39, 738. M. Pluta, J Polym Sci Part B: Polym Phys 2006, 44, 3392. V. Taubner, R. Shishoo, J Appl Polym Sci 2001, 79, 2128. R. Hiroi, S. S. Ray, M. Okamoto, T. Shiroi, Macromol Rapid Commun 2004, 25, 1359. T. Al-Ati, Hotchkiss, J. H., Agricul Food Chem 2003, 51, 4133. 112 CHAPTER 4 Effects of Synthetic and Natural Zeolites on Characteristic Properties of Poly(lactic acid) Composites INTRODUCTION Chapter 3 explored the possibility of melt compounding the synthetic zeolite (type 4A) and PLA matrix and evaluated the morphological, mechanical, thermal, and banier properties of the composites. The current chapter is devoted to naturally occurring zeolites and PLA composites. Chabazites were used as model natural zeolites. The specific goals of this study were to fabricate PLA composites containing 5 wt% chabazite using a twin-screw extruder and injection molder and compare and contrast the effects of type 4A and chabazite zeolites on some of the characteristic properties of the PLA matrix. EXPERIMENTAL Materials Poly(lactic acid) resin produced from 94% L-lactide was obtained from NatureWorks LLC (Blair, NE, USA). Synthetic zeolite (type 4A) (Si/Alzl) with a pore size of 3.8- 4 angstrom was supplied by UOP LLC, (Des Plaines, IL, USA) in the form of powder. Natural zeolite chabazite (CHA) (Si/A1:2.4-3.4) powders were obtained from a mine located at a latitude of 24.70 north and a longitude of 109.50 west in Divisaderos, Sonora, Mexico. Prior to sample preparation, PLA resin pellets and zeolite powders were dried in a vacuum oven at 60°C for 4 h and at 100°C for 24 h, respectively. 113 Preparation of Composites As stated in the previous chapter, the composite samples with 5 wt % type 4A and chabazites were prepared using a micro extruder (DSM Research, Geleen, The Netherlands) equipped with co-rotating twin-screws having lengths of 150 mm, UD ratio of 18, and capacity of 15 cc. The extrusion was carried out at 185°C for 5 min at a screw rotation speed of 100 rpm for both PLA and PLA/zeolite composites. After the set extrusion time, the extrudates were collected fiom the die and transferred into a rnini- injection molder (DSM Research, Geleen, The Netherlands) by a pre-heated transfer cylinder in order to prepare test specimens for the property evaluation. The injection pressure and mold temperature were 896 KPa and 30°C, respectively. Test specimens were wrapped in aluminum foil and stored at 23°C and 50% relative humidity for not less than 40 h prior to testing in accordance with ASTM D 618-03 [1] (Standard Practice for Conditioning Plastics for Testing). The injection molded neat PLA, PLA/5 wt type 4A, and PLA/5 wt chabazite specimens are shown in Figure 4.1. Figure 4.1 Injection molded neat PLA, PLA/5 wt% type 4A, and PLA/5 wt% chabazite composite. 114 Characterization of Composites Morphological Properties Scanning Electron Microscopy (SEM): Morphological analyses were done using a SEM JSM 6400 (JOEL, Tokyo, Japan) equipped with a LaB5 emitter on the fracture surfaces of PLA and PLA/zeolite composites. The Type 4A and chabazite powders and the Izod impact fiactured surfaces of composites were mounted onto aluminum stubs with carbon adhesive tape and then sputter coated with an approximately 15 nm layer of gold using an Emscope SC500 sputter coater (Emscope Laboratories Ltd, Ashford, UK) operated at 20 mA for 3 min. Finally, SEM micrographs were collected at an accelerating voltage of 15 kV and a working distance of 15 mm. An elemental analysis by energy dispersive spectroscopy (EDS) was also carried on the SEM equipped with an EDX detector. The analysis was conducted at an accelerating voltage of 20 kV and working distance of 15 mm. Fourier Transform Infrared Spectroscopy (FTIR) The surface structure of PLA and PLA/zeolite composites was examined by a Shimadzu IRPrestige-21 spectrometer (Columbia, MD, USA) used in the attenuated total . . -l reflection mode. Each spectrum was collected at a resolution of 4 cm and a scan rate of 40 over the range of 4000-550 cm-l. Differential Scanning Calorimetry (DSC) Analysis DSC studies of PLA/zeolite composites were performed on a differential scanning calorimeter (DSC) Q100 (TA Instruments, NewCastle, DE, USA) in accordance with 115 ASTM D3418-03 [2]. Approximately 5-8 mg of injection molded samples were heated from room temperature to 190°C at a rate of 10°C/min under a constant nitrogen flow (70 mL/min). For each sample, the glass transition temperature (Tg), cold crystallization temperature (ch), melting temperature (Tm), and enthalpies of cold crystallization ’(AHC'C) and melting (AHm) were evaluated from the DSC thermograms. The degree of crystallinity (Xc) of PLA and PLA/zeolite composites was calculated based on the following equation: Xe (%) = [( AHm -AHcc)/( AHf(1- X) )1 * 100 (1) where AHf is the enthalpy of fusion of an 100 % crystalline PLA which is 93.7 J/g [3,4], and x is the weight fiaction of zeolite in the composite. At least five specimens were tested for each composite combination. Dynamic Mechanical Analysis (DMA) The temperature dependence of the storage modulus (E'), loss modulus (E"), and damping factor (tan delta) of PLA/zeolite composites was evaluated using a DMA Q800 (TA Instruments, NewCastle, DE, USA). The test was carried out by heating the samples at a rate of 2 oC/min from room temperature to 90°C. The samples were tested in a dual cantilever mode at an oscillating amplitude of 15 um and a frequency of 1 Hz. A minimum of three injection molded specimens with dimensions of 58 mm x 12 mm x 2 mm (length x width x thickness) were tested for each composite combination. 116 The heat deflection temperature (HDT) of PLA/zeolite composites was also measured by a DMA Q800 (TA Instruments, NewCastle, DE, USA) operating in three- point bending mode according to ASTM D648-03 [5]. The specimens were heated at a rate of 2 °C/min from room temperature to 80°C under a constant load of 0.455 MPa, and the HDT was determined as the temperature at which the specimens reached 0.2% strain. Three test specimens with dimensions of 58 mm x 12 mm x 2 mm were tested for each composite combination. Optical Properties The transmission of ultraviolet (UV) and visible (V is) light was measured on a Lambda 25 UV-Visible spectrometer from Perkin-Elmer Instruments (W ellesley, MA, USA). The samples were scanned at a rate of 480 nm/min over the spectral range fi'om 200 to 800 nm. Molecular Weight Determination The number-average (Mn) and weight-average (MW) molecular weights and the polydispersity index (PDI=Mw/Mn) of the PLA and PLA/zeolite composites were determined using a gel permeation chromato graph (GPC) equipped with 2414 reflective index (RI) detector (Waters, Milford, MA, USA) and using a series of three columns (HR4, HR3, and HR2). The analysis was conducted at room temperature using tetrahydrofuran (THF) as the mobile phase at a flow rate of 1.0 mL/min. Approximately 117 35 mg of PLA and PLA/zeolite composite samples were dissolved in 15 mL THF. In the case of PLA/zeolite composites, the zeolite particles were extracted from the solution by centrifirgation and then filtration through a 0.45 pm PTFE filter before injection into the GPC. The GPC system was calibrated using polystyrene standards. RESULTS AND DISCUSSION Morphology The SEM analysis indicated that type 4A and chabazite zeolite particles were a cubical shape with an average particle size ranging fiom 0.7-2 um and an irregular rhombohedral shape with an average particle size range from 2-10 pm, respectively (Figure 4.2). SEM images of the cross-sectional surface of neat PLA and PLA/5 wt% zeolite composites are also shown in Figure 4.3. The neat PLA and PLA with zeolite type 4A specimens exhibited typical brittle fracture surfaces; however, the incorporation of chabazite changed the appearance of the fracture surface. The formation of a fibrillar structure was observed at the edge of the fracture lines (Figure 4.3) due to the stretching of the PLA matrix. This suggests that the cross-sectional surface of impact specimens with chabazite displays more ductility compared to the PLA and PLA/type 4A composites. This may be due to the structural make-up, a higher Si/Al ratio, a lower acidic nature of the chabazite, plasticizer effects of the zeolitic water or higher molecular weight of the PLA/5 wt% chabazite composites; however, further investigation is needed in order to identify the cause. It was also visually observed that both type 4A and chabazite particles were well dispersed within the PLA matrix. 118 Figure 4.2 Photographs (I) and SEM micrographs (II) of type 4A (A) and chabazite (B) zeolite powders. (In SEM micrographs, scale bar: 10 11m) 119 Figure 4.3 SEM images of PLA and PLA/zeolite composites. A: PLA magnification: 1000X, scale bar 20 um; B: PLA/type 4A magnification21000X, scale bar:20 um; Cl: PLA/chabazite magnificationz200X scale bar: 100 um ;C2: PLA/chabazite magnification:1000X scale bar: 20 um. Arrows indicate the zeolite particles. 120 SEM-EDS elemental analyses were done on both types of zeolite particles. Type 4A and chabazite powders had Si/Al ratios of 1.1 d: 0.2 and 2.9 :t 0.5, respectively. EDS measurements also showed that the type 4A zeolites contained alkali metal, e.g., Na+ ions . . . +2 +2 . . whereas chabazrtes contarned alkalrne earth metal, e. g., Ca , Mg rons and alkalr + + , +2 , metal, e.g., Na , K rons, as well as a trace amount of Fe 1011. EUR Analysis The FTIR spectra were collected to examine the molecular interactions between PLA and zeolites. Figure 4.4 shows the FTIR spectra of type 4A synthetic and chabazite natural zeolites dried under vacuum at 100 °C for 24 h. In general, the absorption bands appearing at around 3300-3600 cm-1 and 1660 cm'1 can be assigned to the stretching and deformation vibration of hydroxyl (-OH) groups of zeolites, respectively [6]. The FTIR spectrum of chabazite exhibited strong absorption peaks centered at 3396 and 1647 cm-1 whereas type 4A zeolites showed only a weak peak at 1655 cm-1, suggesting that the chabazites retain more water molecules than type 4A zeolites even after they were dried. The intensive absorption band at around 1000 cm.1 can be attributed to the asymmetric stretching of the Si-O and Al-O bonds belonging to the SiO4 and A104 tetrahedra [7]. The bands found at 750-880 and 500-665 cm'1 are associated with the symmetric stretching of the Si-O bond and oscillations of chains of aluminosilicate oxygen tetrahedrals, respectively [6-8]. 121 120 100 (D O O) O Transmission, °/o .b O N O 0 i i t i i i t 4500 4000 3500 3000 2500 2000 1 500 1 000 500 1 Wavenumber, cm' Figure 4.4 FTIR spectra of (A) type 4A and (B) chabazite powders Figure 4.5 compares the spectra of neat PLA and PLA with 5 wt % type 4A and chabazite zeolites. FTIR spectra of PLA exhibited a broad absorption band in the region of 3600-3100 cm.1 which can be assigned to the hydroxyl groups (-OH) of PLA. The intensity of this band slightly decreased upon incorporation of zeolites. This may be caused by hydrogen-bonding or ionic interactions between PLA and zeolites. The PLA spectrum in the region of 3000-1200 cm.1 can be characterized by absorption bands at 2995 and 2947 cm-1 arising from the C-H stretching vibrations of CH3 groups in the side chains as well as a band ascribed to the stretching vibration of C=O groups at 1751 cm-1, bending vibration of CH3 groups at 1452 cm-1, and absorption bands at 1382 and 1365 122 arising from the C-H deformation and asyrnmetric/symmetric bending, respectively [9]. In this particular region, the intensities of the PLA characteristic peaks notably increased with the addition of 5 wt% zeolite. In the region of 1200-600 cm-l, the characteristic peaks of zeolites could not be discriminated since they were superimposed by the PLA spectrum. However, higher peak intensities were also observed in that region for PLA/zeolite composites. A new absorption peak at 565 cm.1 appeared in spectra of PLA/zeolite composites, which is attributed to the oscillations of chains of almninosilicate oxygen tetrahedrals. 123 com Door 88226266 66 6:4 .3. 6632: a: .536 2:6 8884 a: 3. 8:45 F oomr ooom .50 9383853 66mm 6666 6666 6664 6664 6 . 66 I. . mu 64 MW MI 8 - _m. 66 m“ an 66 664 111..:11111.11111;11:1 11:11111111. 664 124 DSC Analysis Table 4.1 summarizes the thermal characteristics (Tg, ch, Tm) along with the percentages of crystallinity of injection molded PLA and PLA/zeolite composites obtained fiom the first heating cycle. The Tg and Tm values were not significantly altered by the addition of either type 4A or chabazite zeolites; however, the cold crystallization and percent crystallinity were significantly affected. All of the injection molded samples exhibited cold crystallization peaks. The ch of PLA/type 4A and of PLA/chabazite composites was found to be ll4.3:1:1.6 and 110.8 :t1.2°C, respectively, which is about 10°C lower than that of neat PLA. In general, a low cold-crystallization temperature (obtained during the heating cycle) is considered as an indirect indication of fast crystallization. Therefore, the observed reduction in the cold-crystallization-temperature suggests that type 4A and chabazite may enhance the nucleation of PLA crystallites. PLA is notorious for its slow crystallization which can be an obstacle in some processing systems such as extrusion and injection molding. In order to confirm the effects of the zeolites on the crystallization rate of PLA, cooling DSC thermograms were also investigated. After the injection molded samples were heated from room temperature to 190°C and kept at this temperature for 2 min to remove the thermal history, the samples were cooled to 10°C at a cooling rate of 10 oC/min. The crystallization exotherms were not detected at a cooling rate of 10 oC/rnin for the neat PLA samples. It was obvious that cooling rate was not enough to generate crystallization exotherrns for the PLA. However, a cooling rate of 0.5 C’C/min was able to create crystallization exotherrns for the neat PLA samples. Figure 4.6 shows the DSC therrnograrns of the cooling cycle for PLA and PLA/zeolite composites at a cooling rate of 0.5°C/min. A lower crystallization 125 temperature (during the cooling cycle) indicates slower crystallization, which is the opposite of the case in the heating cycle. As can be seen from Figure 4.6 the crystallization peaks of PLA/zeolite composites became narrower, suggesting smaller homogenous spherulite formation. The crystallization temperatures of the PLA/type 4A and chabazite composites were all higher than that of the neat PLA, confirming a nucleating effect of the synthetic type 4A and naturally occurring chabazite zeolite particles. 126 Ammmrnoxsp. HQovmv 658.46% buqeomiwmm 86 6:828 0566 65 5 6.832 63.866236 6:86th 8:3 6:82 8:633. 26956 4 :82 o E 4 E a 4.6 4 $2 a 6.6 4 4.6.1 o S 4 4.6: a _.6 4 4.64 6:42.26 $63 624.. 63 46.4 a: 46.42 a: 46.4: 46.4 4 2.: 45 424 S. 25 so? 62.: a s6 4 6.6 a 4.6 4 4.42 a 2 4 6.4: a 6.... 4 6.4.2 4 4.6 4 36 . 36 exp ox 6663.4 60:39 Goran. 6666. 6.6846 Oovah 6866388 828N544: Ea <4.“ .60 6266566865 .6585. :4 634% 127 41.1 98.37°C Cooling Cycle PLA 101.70%: . \ ,x’J’Iiv ___________________________ 0.2- ------------- / “\ A /' in PLA/type E 03‘ 107.34°c 1:. H 8 ,4. I / \.‘__ .............. 04- /' . PLA/chabazrte 4,5...,...,...,...,.......,..4,-4.... 20 40 60 80 100 120 140 160 180 200 Exo Up Temperature, (°C) Figure 4.6 DSC therrnograrns of PLA, PLA/type 4A, and PLA/chabazite composites at a cooling rate of 0.5 oC/min. As can be seen in Table 4.1, there were no significant changes in melting terIilperatures of PLA, PLA/type 4A, and PLA/chabazite composites; however, all of these composites exhibited two melting peaks located at about 148 and 155°C. Several reSearchers have observed similar bimodal melting behavior for PLA and its composites [1 0- l 2]. This phenomenon was linked to various reasons including (a) the effect of low cryStallization temperature which caused the formation of the disordered alpha phase of 128 PLA; (b) the existence of more than one crystal structure; (c) the effect of melting the original crystals, recrystallization, and remelting the recrystallized crystals during the heating scan; ((1) the presence of different larnellae morphologies formed prior to the heating scan; and (e) the processing conditions or molecular weight distribution [1 1,13,14]. The percent crystallinity of the neat PLA was found to be 3.2 4 0.7% which suggests that injection molded specimens were nearly amorphous. With the addition of 5 wt% type 4A and chabazite the percent crystallinity of injection molded PLA increased to 7.6 :t 1.2 and 7.1 :1: 1.4%, respectively. Dynamic Mechanical Thermal Analysis The viscoelastic nature of PLA and PLA/zeolite composites were studied by DMA. As oscillating force was applied to a specimen, the resultant strain was recorded as a function of temperature. As shown in Figure 4.7, in the glassy state region (<50-600C), the storage modulus of naet PLA was lower than PLA/zeolite composites. The storage mOdulus of PLA/5 wt% chabazite and PLA/5wt% type 4A composites increased by about 9% and 21%, respectively. This trend suggests that the incorporation of zeolites in PLA forfiled good dispersed composites with improved modulus. At 50°C, the neat PLA and PL-‘\/5 wt% chabazite showed a sharper drop in storage modulus; however, PLA/5 wt% type 4A exhibited a sudden decline at 60°C, which indicates that PLA/type 4A col“posites maintain their stiffness longer than PLA/chabazite composites, which reflects the Presence of stronger interactions between the type 4A and PLA matrix. PLA chains may adhere easily to the surface of type 4A causing an increase in modulus. The 129 511‘ the [e] magnitude of loss modulus values also increased with the addition of type 4A and chabazite zeolites. The loss modulus of PLA, PLA with 5 wt% type 4A, and PLA with 5wt% chabazite zeolite composites reached a maximlnn value of 668i23 MPa at 5941.0°c, 9014 9MPa at 6140.5°C, and 760415 MPa at 6140.4, respectively. These results suggest that PLA/type 4A and chabazite composites have better ability to dissipate the vibrational energy as heat than does the neat PLA. Figure 4.8 exhibits the tan delta versus temperature curves for PLA and PLA/zeolite composites. As 5 wt% type 4A and chabazite particles were introduced into the PLA matrix, the tan delta peak temperatures slightly shifted to higher temperatures indicating that zeolite particles restrict the mobility of PLA segmental chains. Additionally, the intensity of the tan delta peak increased in comparison with the neat PLA. Possible reasons for the increased intensity were discussed in chapter 2. 130 4000 \. . 100 -\\\\ '\\ ]] i800 3000 . ‘-4.\\ \i ' ~ * \ 1 1 r a? <%:U \‘\4[ii 600 23 “7' \\ [‘11 \\‘l g 2 XIV“ ‘1‘] '5 .3 2000 . 1],],1‘] 3 4’ I111. 111 2 ll \, \ 1.] a 3, / 1 1 H] to E Ill/ [‘1 ‘1]! 3 m l/ ‘ 1\ 10004 ’I’ " 1' - I/ l 1““ \- ~200 20 40 60 80 100 Temperature, °C Figure 4.7 Storage and loss modulus of PLA neat (A), PLA/5 wt% type 4A (B), and PLA/5 wt% chabazite (C) composites as a function of temperature. 131 2.5 1.5 . Tan Delta 0.5 . 20 Temperature, °C Figure 4.8 Tan delta versus temperature for PLA neat(A), PLA/5 wt% type 4A(B), and PLA/5 wt% chabazite(C) composites. Figure 4.9 shows the typical HDT behavior of PLA and PLA/zeolite composites obtained from the 3-point bending test. The HDT of neat PLA, PLA/type 4A, and PLA/chabazite composites were found to be 55.ld:O.1, 54.4d:0.3, and 53-2iO-20C, reSpectivel y. At a 5 wt% loading level, no improvement was observed in the HDT values for either PLA/type 4A or PLA/chabazite composites. This may be due to the low degree 0f Clyewlllinity that PLA possesses (initially) and addition of just 5 wt % zeolite was not SufliCiem to improve the HDT. 132 1.5 1.0- C 0-5.- 53.39 °c 0.20% .\° .5“ g 00- ~— m -o.5‘ -1.0 ‘ 20 30 4o 50 60 70 80 Temperature, 00 Figure 4.9 Typical HDT curve obtained from 3-point bending test on DMA for PLA (A), PLA/5 wt% type 4A (B), and PLA/5 wt% chabazite (C) composites. 133 Optical Properties UV-Vis spectra of neat PLA and PLA/type 4A and PLA/chabazite composites at a film thickness of 2 mm are shown in Figure 4.10. At 400 nm, PLA, PLA/5 wt% type 4A, PLA/5 wt% chabazite composites had a transmission of 80i0.9, 4810.4 and 21i0.2%, respectively. In the region of UV light (at 300 nm), the neat PLA transmitted 33i0.5% of the light while the PLA/5 wt% type 4A and PLA/5 wt % chabazite composites transmitted only 10:}:O.4 and 4.6i0.2% of the light, respectively. This result implies that PLA with 5 wt% chabazite provides better UV barrier than the PLA/type 4A composites. C Transmission, % 200 400 600 800 1 000 Wavenumber, nm Figure 4.10 UV/vis transmission spectra of PLA (a), PLA/type 4A (b), and PLA/chabazite (c) composites. Thickness: 2 mm. 134 Molecular Weight Determination Table 4.2 summarizes the molecular weight parameters for the PLA and PLA/zeolite composites. In chapter 2, number-average and weight-average molecular weight reductions due to the extrusion processing were discussed. In order to evaluate the effects of cycle time on the molecular weight parameters of PLA/type 4A composites, the extrusion cycle time was reduced from 5 min to 1 min. For 1 min and 5 min cycle times, the number-average molecular weights reduced by 36 and 43 %, respectively, when compared with extruded neat PLA. The weight-average molecular weight values also exhibited the same trend (by 34 and 42%, respectively). Although using 1 min cycle time slightly improved the reduction in molecular weight, poor dispersion of the zeolite particles was observed; therefore this condition is not recommended. Surprisingly, inclusion of 5 wt% chabazites caused only 0.6 and 10% reductions in number-average and weight-average molecular weights, respectively. Several studies reported the drastic effects of melt processing and inorganic fillers on the molecular weight of PLA. However chabazites tend to limit the mechanical degradation of PLA. The study conducted by Mathieu et al. [15] and Ara et al. [16] reported that the presence of slightly basic calcium phosphates and weakly basic calcium carbonates stabilizes the carbonyl end groups of degradation products and hence limits the polymer chain length . . . . . +2 +2 . . . reduction. Chabazrte mlnerals contain alkaline earth metal, e. g., Ca , Mg ions 1nthe1r frameworks which might be the reason for this stabilization. This phenomenon deserves more investigation, and the next chapter deals with the effects of type 4A and chabazites 135 on the thermal degradation behavior of PLA. 136 .mcoumSQe pawgm 5:5 2:88 8865 $2“? 802 .Bfigxo <1Hn~ no @0me DdOfi 0.53 mflOfi—mfifiommo SOUOQBH “GOOHDAH ”OHOZ So a :3 2 £3 « 3&2 so $3 a swag as: 2&0 ss 3 8552—9 $3 3:: ed a m: as $3. a gem a. a; a $4.8 as: 2&0 ss 3 <9 2:. $3 3:: sod an s: 3 :3 a £33 on 23 a Show as: 2&0 as a 3. 25 $3 3:: sod « of - as a :3: - $3 a 3qu 32:23 3.. 25 a m2 3? a 332 8a.: a :32 3:2. 3.. «fleas: s .3 a :—m e\.. flaw—0:60“ 35% E Eaton-flou— —OE\M 2 .mozmomEoo ofioonzqm use «3.30 8:52 0&0 NV 03$ 137 CONCLUSION PLA can be effectively melt compounded with both type 4A and chabazite zeolites. The SEM results showed that the fractured surface of the PLA/ 5wt% type 4A zeolite samples were more susceptible to crazing and brittle surface deformation than PLA/chabazite composites. Incorporation of both type 4A zeolite and chabazite into PLA had no significant effect on the Tg and Tm of PLA; however, the cold crystallization and percent crystallinity were significantly changed. Both type 4A and chabazite promoted the crystallization of PLA. DMA reveals that PLA/type4A composites have higher storage and loss modulus as compared to PLA/chabazite composites with the same loading level. Since favorable interactions exist between type 4A and PLA, their composites exhibited acceptable thermal and thermomechanical properties; however type 4A reduced the number-average and weight-average molecular weight of PLA. On the other hand, chabazites seemed to stabilize the reduction in molecular weight of PLA. 138 \i REFERENCES [1] [3] [4] [5] [6] [7] [8] [9] [10] r1 1] r12] [13] [14] [15] [16] American Society for Testing and Materials. Standard Practice for Conditioning Plastics for Testing. D 618-03. In Annual Book of ASTM Standards. ASTM: Philadelpia, PA, 2004: 8:01 . American Society for Testing and Materials.Standard T est Method for Transition Temperatures and Enthalpies of Fusion and Crystallization of Polymers by Differential Scanning Calorimetry. D 3418-03. In Annual Book of ASTM Standards. AST M: Philadelpia, PA, 2004: 8:02 S. Sinha Ray, K. Yarnada, M. Okamoto, A. Ogami, K. Ueda, Chem Mater 2003, 15, 1456. R. Auras, B. Harte, S. Selke, J Appl Polym Sci 2004, 92, 1790. American Society for Testing and Materials. Standard Test Method for Deflection Temperature of Plastics Under F lexural Load in the Edgewise Position. D 648- 03. In Annual Book of AST M Standards. AST M: Philadelpia, PA, 2004: 8:01. G. V. Tsitsishvili, Andronikashvili, T.G., Kirov, G.N., Filizova, L.D., Natural zeolites, Ellis Horwood, New York 1992. A. Aronne, S. Esposito, C. Ferone, M. Pansini, P. Pernice, J Mater Chem 2002, 12, 3039. K. B. Payne, T. M. Abdel-Fattah, J Environ Sci Health, Part A: 2005, 40, 723. R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4, 835. I. E. Yuzay, R. Auras, S. Selke, J Appl Polym Sci 2009, 115, 2262. M. L. DiLorenzo, Macromol Sym 2006, 234, 176. X. Ling, J. E. Spruiell, J Polym Sci Part B: Polym Phys 2006, 44, 3200. M. Yasuniwa, K. Lura, Y. Dan, Polymer 2007, 48, 5398. J. Zhang, K. Tashiro, H. Tsuji, A. J. Domb, Macromolecules 2008, 41, 1352. L. M. Mathieu, P. E. Bourban, J. A. E. Manson, Comp Sci T echnol 2006, 66, 1606. M. Ara, Watanabe, M., Imai, Y., Biomaterials 2002, 23, 2479. 139 CHAPTER 5 Effects of Synthetic and Natural Zeolites on Thermal Degradation Behavior of Poly(lactic acid) Composites *Extended version of the article: Yuzay, LE, Auras, R., Selke, S., Polymer Degradation and Stability, 2010 (Submitted) INTRODUCTION In recent years, there has been growing emphasis on using bio-based polymers. Poly(lactic acid) (PLA) can be considered one of the most important bio-based polymer substitutes for petroleum-based polymers due to its biocompatibility, biodegradability, compostability, and relatively good mechanical strength [1,2]. PLA is generally produced either by direct polycondensation of L-lactic acid or by the catalytic ring-opening polymerization of the lactide monomers (the cyclic dimer of lactic acid) [3,4] The structure and properties of PLA can be modified by controlling the composition of the L- and D-isomers. It can also be a promising material for feedstock recycling into L-lactic acid or L-lactic based compounds by hydrolysis [5,6] and hydrothermal depolymerization [7], and into cyclic dimers by pyrolysis [8,9]. Several researchers studied the capability of feedstock recyclability of PLA via thermal degradation. It is well established that the thermal degradation of PLA is due to several reasons: (i) hydrolysis by trace amounts of Water [10,11], (ii) unzipping polymerization (backbiting reaction) starting from the hydroxyl ends of the chains, catalyzed by residual polymerization catalyst [8,12-14], (iii) 140 intra- and intermolecular transesterification reactions leading to the formation of cyclic oligomers of lactic acid and lactides [11,14,15], and (iv) random chain scission (cis- elimination) of the ester groups, which results in a small amount of acrylic acid and acrylic oligomers [13,14,16]. It was also agreed by several investigators [17-21] that these mechanisms may occur concurrently, which makes the degradation behavior of PLA very complex. There are numerous publications that have been devoted to investigating the effects of metal (e.g., Al, Ti, Sn, Zn, Fe, Zr) [12,13,18,21], metal oxides, e.g., calcium oxide (CaO) [18], magnesium oxide (MgO) [18,20], aluminum hydroxides Al(OH)3 [9], and montrnorillonite [22] on the thermal degradation behavior of PLA, using thermal analysis techniques including pyrolysis-mass spectrometry (Py-GCMS), differential scanning calorimetry (DSC), and thermogravimetric analysis (TGA). However, there is no result reported regarding the effects of synthetic and natural zeolites on the thermal degradation behavior of PLA although it has been shown that PLA can form composites with synthetic zeolites without significant deterioration of mechanical properties. Hence it is important to understand their effect on the thermal degradation of PLA. Zeolites are hydrated crystalline aluminoslicates with pore sizes ranging from 3- 15 angstrom. The structure of zeolites consists of 3-dimensional fiameworks made up by A104 and SiO4 tetrahedra. The cages and channels within the zeolite framework have the Capability of adsorbing or rejecting molecules depending on their size, shape, and polarity [23,24]. There are a variety of naturally occurring as well as synthetic zeolites with 141 differing pore structure, crystal size, chemical composition and ion exchange properties. Zeolites have found widespread industrial applications ranging from adsorbents to separation of gases and purification. Furthermore, zeolites are used to catalyze a variety of hydrocarbon reactions, including cracking, hydrocracking, alkylation, and isomerization, based on the combination of high thermal stability and high acidity with the shape selectivity of zeolites [25]. Incorporation of zeolites into various petroleum- based polymer matrices has been well explored to produce mixed membrane matrices for gas and liquid separation purposes. Zeolites have been also used as functional fillers to improve the properties of polymers. Therefore, inclusion of zeolites into bio-based PLA may hold promise not only for a wide range of polymer applications, but also for improving the feedstock recyclability of PLA. In chapters 3 and 4, the preparation of PLA/synthetic zeolite (Type 4A) and PLA/natural zeolites (chabazite) composites using a melt compounding process and the effects of Type 4A synthetic zeolites on the morphology, structural, physical-mechanical properties of PLA were examined. This chapter aims to investigate the thermal degradation behavior of PLA and PLA/zeolite composites under non-isothermal conditions using TGA and to estimate the apparent activation energies of the thermal degradation process using both the Flynn-Wall-Ozawa [26-28] and Kissinger [29] methods. 142 EXPERINIENTAL Materials Poly(lactic acid) resin produced fiom 94% L-lactide was obtained fi'om NatureWorks LLC (Blair, NE, USA). Synthetic zeolite (type 4A) (Si/Alzl) was supplied by UOP LLC (Des Plaines, IL, USA), in the form of powder. Natural zeolite chabazite (CHA) (Si/Al:2.4-3.4) powders were obtained from a mine located at a latitude of 24.70 north and a longitude of 109.50 west in Divisaderos, Sonora, Mexico. Prior to sample preparation, PLA resin pellets and zeolite powders were dried in a vacuum oven at 60°C for 4 h and at 100°C for 24 h, respectively. Preparation of Composite Samples PLA composites containing 5 wt% zeolite type 4A and chabazite were prepared using a micro extruder (DSM Research, Geleen, The Netherlands) equipped with co- rotating twin-screws having lengths of 150 mm, W ratio of 18, and capacity of 15 cc. The extrusion was carried out at 190°C at a screw rotation speed of 100 rpm and a cycle time of 5 min. After the set cycle time, the extrudate was collected from the die and transferred into a mini-injection molder (DSM Research, Geleen, The Netherlands) by a pre-heated transfer cylinder in order to prepare test specimens. The optimized injection pressure was 0.896 MPa and the mold temperature was kept at 30°C with a residence time of 15 sec. In order to prevent moisture uptake, test specimens were wrapped in aluminum foil and stored at 23°C and 50% relative humidity for not less than 40 h prior 143 to testing in accordance with ASTM D 618-03 [30]. CHARACTERIZATION Thermogravimetric Analysis (TGA) Two sets of TGA experiments were performed, the first of which was to evaluate the influence of both synthetic and natural zeolites on the thermal decomposition temperature of PLA at a constant heating rate. TGA was performed using a TGA 2950 (TA Instruments, New Castle, DE, USA) with a heating rate of 20 °C/rnin from room temperature to 600°C under nitrogen and air flow of 70 mL/min. Three specimens were tested for each composite. The percent weight loss and derivative weight loss were plotted against temperature in order to evaluate the onset and final degradation temperatures at 5% weight loss and the maximum decomposition temperatures of the neat PLA and its composites. The second set of experiments was to establish the activation energies of thermal degradation of PLA and PLA/zeolite composites. The experiment was carried out on a TGA 2950 with heating rates of 5, 8, 10, 15, and 20 °C/min from 27 to 600°C under nitrogen and air flow of 70 mL/min. The apparent activation energy (E) values of decomposition of PLA and PLA/zeolite composites were estimated using both the Flynn-Wall-Ozawa [26-28] and Kissinger [29] methods. 144 RESULTS AND DISCUSSION Thermal Degradation Analysis Figure 5.1 shows the TGA weight loss thermograms obtained in a nitrogen atmosphere for the dried type 4A, chabazite and neat PLA samples. The TGA thermograms of type 4A and chabazite powders exhibited 1.9 and 16.4% weight loss at 600°C. Figure 5.2 illustrates the representative temperature-dependent weight loss and the derivative of the weight loss curves for PLA and PLA/zeolite composites in nitrogen and air at a heating rate of 20 oC/min. Temperatures of onset, 5% weight loss (T5% weight loss): and maximum decomposition (Tdmax) of all three composites are also summarized in Tables 5.1 and 5.2. As can be seen from Figure 5.2, the first derivative of the TGA curves exhibited only one peak for the PLA and its zeolite composites both under nitrogen and air, which is evidence that the complete thermal degradation occurred in a single step for all these samples. PLA and PLA/zeolite composites were relatively stable up to temperatures around 290-3000C. Only a negligible amount of weight loss (up to 0.6%) was observed at temperatures around 190-2000C, which was the operational temperature for the twin-screw extruder and injection molder. At temperatures above 300°C, PLA/type 4A composites were thermally decomposed more easily than the PLA and PLA/chabazite composites in both nitrogen and air. It was found that the onset degradation temperatures at 5 % weight loss of PLA/type 4A composites were about 58 and 38°C lower than that of neat PLA in nitrogen and air environments, respectively. However, the onset temperatures of PLA/chabazite composites were almost the same as that of neat PLA samples. This result suggests that type 4A zeolites may be used as a depolymerization catalyst for PLA. 145 120 ‘ Residue: ‘ 98.08% Type 4A ] soj o\° \ Residue: ...— - 83.56% chabazite f» \ (7.116mg) .5 .4 .\ 3 x 40 s '\ l. . l. l. l 0 i . . , r Y r ' \;——r-_I._r.v__f_ 200 400 600 Temperature,°C Figure 5.1 TGA curves of type 4A, chabazite, and neat PLA. Heating rate: 20 °C/min, in nitrogen atmosphere. 146 Table 5.1 TGA analysis of PLA and PLA/zeolite composites. Heating rate: 20 °C/min, in nitrogen flow. T onsets T 5% weight loss, T max weight loss 9 PLA 376.0 21: 2.18 391.0 :t 0.13 433.0 :1: 0.3a PLA/5M% type 4A 318.03: 3.5b 348.0116b 415.021: 0.9b PLA/ 5 wt% 378.0 2t 0.4a 396.0 :k 0.8c 440.0 :1: 10° chabazite Mean :1: standard deviation Means with different superscript letters in the same columns are significantly different (P<0.05; Tukey-HSD) Table 5.2 TGA analysis of PLA and PLA/zeolite composites. Heating rate: 20 °C/min, in air flow. T onsets T 5% weight loss, T max weight loss 9 °C °C °C PLA 341.0 3; 1.921 358.6 21: 0.1a 405.0 3: 0.4a PLA! 5 wt% type 4A 303.4 i 2.5b 328.5 i 1.3b 397.0 3:: 1.2b PLA/ 5 wt% 338.0 i 0.4a 360.0 i 0.8a 407.3 :1: 1.0ac chabazite Mean i standard deviation Means with different superscript letters in the same columns are significantly different (P<0.05; Tukey-HSD) 147 120 100 4 Nitrogen __ _ 80 .. o\° so - 15' .9 o _ 3 40 20 .. o ..t 0 100 200 300 400 500 600 700 Temperature, °C 3.0 2.5 ‘ Nitrogen 2.0 a 9 \ 1.5 r 32 E. '6 1.0 " 3 g 0.5 - D 0.0 - 3 -0.5 T l I I I I I I 100 200 300 400 500 600 700 Temperature, °C Figure 5.2 Representative TGA and derivative thermogravimetric curves of PLA (A), PLA/type 4A (B), and PLA/chabazite (C) composites. Heating rate: 20 °C-min", in nitrogen and air atmosphere. [Solid line: A; Dash linezB; Dot line:C] 148 [\O ..f—Cmfli\ 20 °C-min1 g 60 3a 15 °C-min’1 06 '0 4° 8 OC-min'1 - g 10 °c~min1 20 5 OC-min‘ o ' 106 I ' '2'00 ' 3'06 460 V 560 ' 600 0 Temperature. C Figure 5.3 Representative TGA curves of PLA, PLA/type 4A, and PLA/chabazite composites at heating rates of 5, 8, 10, 15, and 20 °C/min, under nitrogen flow. 151 The kinetics of thermal degradation of polymers is generally expressed by the following typical kinetic equation [34-36]: da ‘dt—=kxf(a) (2) where (l is the conversion degree or the fraction decomposed [Cl = (W0.Wt)/(WO-Wf), W0, Wt, and Wf are the initial, time t, and final weights of the polymer], f(a) is the kinetic model function, da/dt is the rate of conversion, and k is the temperature dependent degradation rate constant which can be expressed as a function of temperature using the Arrhenius equation: E k = Aexp —-1 V RT (3) where A is the pre-exponential factor, Ea is the apparent activation energy of the degradation reaction, R is the universal gas constant, and T is the absolute temperature. The substitution of Eq. 3 into Eq. 2 gives da E 71?: Aexp(—-I§]f(a) (4) Eq. 4 can be considered as a general expression for isothermal conditions. In the case of non-isothennal conditions where the sample is heated with a constant heating rate, B= dT/dt, Eq. 4 can be transformed into an equation describing the degradation reaction rate as a function of temperature and then can be rewritten as follows: 152 da A E, — = —ex - — f(a) dT [3 RT (5) Although the polymer degradation is complicated, in many studies the degradation reaction is assumed to be a simple n'h order reaction and the kinetic model function, f( a), is expressed as (1- 0t)n where n is the order of reaction [35,37]. Numerous methods using different approaches (model-fitting and model-free isoconversional) have been proposed to determine the kinetic parameters (A, Ba, and n) from isothermal and non-isothermal TGA data using equations 4 and 5. These methods attempt to solve the equations by (a) integration, (b) differentiation, or (c) approximation [36-3 8]. The model-fitting approach involves fitting various models to conversion versus temperature curves and simultaneously determining the activation energy and fi'equency factor. Several researchers heavily criticized model-fitting methods for producing unreliable kinetic parameters and using only one TGA curve recorded at a certain heating rate [39,40]. On the other hand, the model-free approach has gained popularity in the scientific community due to the ability to estimate Ea values as a function of conversion without assumption of a kinetic model (f(a)). The main advantage of this approach is that one can easily assess if the kinetic parameters remain constant; furthermore the type of reaction model can be evaluated. For instance, if the activation energy varies with the conversion, the process can be considered complex (multi-step reaction); similarly, if the activation energy does not depend on conversion, the process is simple (single step reaction) [40,41 ]. 153 The Flynn-Wall-Ozawa [26-28] and Kissinger [29] methods are well-known representatives of model-free approaches. In this study, these two methods were used to evaluate the thermal degradation behavior of PLA and PLA/zeolite composites. The Flynn-Wall-Ozawa method is an integral method that utilizes Doyle’s linear approximation [42] and the following equation can be obtained: 2315 0.4513,, Rg(a) ' RT where B, A, Ba, and R are the heating rate, the pre-exponential factor, the apparent 108%) = log (6) activation energy, and the gas constant, respectively, and a. is the conversion (fractional weight loss). The apparent Ea can be obtained fi'om the slope of the plot of log [3 against VT for any level of conversion. One should not expect that the kinetic parameters obtained from TGA data directly represent the energy barriers and collision theories generally associated with Arrhenius parameters as in gas and liquid kinetics. In solid- state kinetic studies, the “apparent” Ea is defined as the average excess energy obtained fiom the vibration of an atom or molecule at a certain temperature. This energy can also be related to the rupture of chemical bonds [36]. The activation energy is denoted as apparent activation energy throughout this study since the activation energy derived from TGA data is the sum of activation energies of chemical reactions and physical processes [43]. The effects of zeolites on the thermal degradation behavior of PLA were analyzed from TGA curves obtained at heating rates of 5, 8, 15, 10, and 20 C)C°min-1 under nitrogen and air (Only the PLA mass fraction was taken into account since the total mass 154 [l of the zeolites did not change during the entire experiment). Figure 5.4 illustrates the typical plots of log [3 against III" at 13 conversion levels in the range of 2.5-95% which were used to estimate apparent activation energies under nitrogen. The dependence of the apparent Ea on conversion for PLA and PLA/zeolites composites is presented in Figures 5.5 and 5.6. As can be seen from the plot of apparent Ea values as a function of % conversion, the neat PLA exhibited the highest apparent Ea values during the entire thermal degradation process under a nitrogen environment. The initial apparent 15a of thermal degradation of PLA was found to be 110 kJ/mol at the 2.5 % conversion level (i.e., 97.5 % residual weight) and gradually increased up to about 120 kJ/mol at 10% conversion, then reached a plateau in the region of 10-95% conversion. The apparent Ea values obtained were relatively similarly to previously reported values for PLA [11,13,15,21,44]. 155 I09(5) I000” 1.5 l l 1.1 l 0.9 0.7 -2.5% A5% X10% 115% 020% 030% +40% -50% -60% 070% I80 % 590% XQ5% 1.5 1.3 1.1 0.9 0.7 1.35 I I I 1.45 1.55 1.65 1000rr, K'1 1 .75 l l I'2.5% A556 x10% ‘15% .2096 .30% +40% '50% “60% 070% '80% #9096 X9596 I I 1 .55 1 .75 1000rr, K'1 2 R values over the conversion range of 2.5 - 95% A: 0.9883 - 0.9970 ; BI 0.9021 - 0.9989; CZ 0.97l3 - 0.9961 156 1.95 Figure 5.4. log [3 vs 103 T-1 curves for (A) PLA, (B) PLA/type 4A, and (C) PLA/chabazite in nitrogen, using the Flyrm-Wall-Ozawa method.(Cont’d). '09”) 1.5 1.3 - 1.1 _ 0.9 — 0.7 - '2.5% ‘5% x10% I915% ‘20% 030% +40% '50% '60% .7096 '80% 590% x9596 0.5 , 1.35 1.45 'Figure 5.4. (Cont’d). 1 .55 1ooorr, K'1 157 1 .75 14o . 120 r ‘ ‘ 3 9 9 ¢ ¢ 9 ¢ ‘4 100 ~ '5 80 - E g 60 d + PLA neat “3' +PLA+5wt%type4A 4° “ .... PLA + 5wt% chabazite 20 - o 1 l L - 1 o 20 40 so 80 100 Conversion, % Figure 5.5 Apparent activation energies (Ea) at different conversion values for thermal degradation of PLA, PLA/type 4A, and PLA/chabazite composites as obtained from the Flynn-Wall-Ozawa method in nitrogen atmosphere. 158 140 0 0 0 0 1 0 1} l 120 - on O O O l l Ea (kJ Imo_l) O) O + PLA neat +PLA+5wt%type4A 20 ~ o I I j , o 20 4o 60 so 100 Conversion, % Figure 5.6 Apparent activation energies (Ea) at different conversion values for thermal degradation of PLA, PLA/type 4A, and PLA/chabazite composites as obtained fiom the Flynn-Wall-Ozawa method in air atmosphere. (*Experiments have not been performed on chabazite samples under air) 159 In the case of addition of 5 wt% type 4A and chabazite zeolite, the initial apparent Ea values of the composites were found to be about 40 and 10 kJ/mol lower than that of the neat PLA, respectively. As % conversion increased from 10-95, PLA/zeolite composites had nearly constant apparent activation energy, showing the same trend as that found in PLA samples. PLA/type 4A and PLA/chabazite composites leveled off at around 100 and 110 kJ/mol, respectively. This suggests the presence of only one dominant degradation mechanism over the conversion range of 10-95%. For the comparison purposes some of the experiments were also carried out in air; due to time limitation the experiment could not performed on chabazite samples under air. Table 5.4 and Figure 5.6 summarize the results obtained under air for the neat PLA and PLA/type 4A composites. On the other hand, the Kissinger [29] method is related to the variation of the peak temperature (Tmax) at the maximum rate of conversion (decomposition) with heating rate. The apparent activation energy can be estimated using the following equation [29,45]: AR (7) é = Inf-Hm n(l-amax)n_1] - Rf,“ a where [3, Tmax, and A are the heating rate, the temperature at maximum rate of weight loss, and pre-exponential factor, respectively. R is the gas constant, Ea is the apparent 160 activation energy, amax is the extent of conversion at Tmax, n is the reaction order. According to Eq. 7, a straight line should be obtained when In (B / T2 ) is plotted max against the reciprocal of the Tmax, The apparent Ea for decomposition can be obtained from the slope of the straight line of the plots. Figure 5.7 shows plots of In (B / T Sax ) versus 1/Tm for PLA and PLA/zeolite composites. The values of the apparent Ea obtained from the slope of these plots are also listed in Table 5.3. The apparent Ea of PLA, PLA/type 4A, and PLA/chabazite composites were found to be 105i 3.5, 862t4.0, and 94i2.0 kJ/mol, respectively. The apparent Ea values of PLA decreased by 10 and 18% with the incorporation of 5 wt % chabazite and type 4A zeolites. As can be seen in Table 5.3, the Kissinger method produced somewhat lower Ea values for all composites than that obtained from the F lynn- Wall-Ozawa method, yet the order of the apparent Ea of the composites (PLA/type 4A< PLA/chabazite < PLA) was the same. 161 PLAIchabazite PLA 2 In ( BI Tmax ) 30 O) -10 - -1o.4 . ° 40.8 : f ‘ r 1.35 1.4 1.45 1.5 1.55 1.6 100mm,, K'1 Figure. 5.7 Kissinger plots of PLA, PLA/ type 4A, and PLA/ chabazite composites. (with three replicates; the range of R2 values: PLA=O.9884-O.9999, PLA/type 4A= 0.9947- 0.9976, PLA/chabazite=0.9829-09923 ) in nitrogen. 162 Table 5.3 Apparent activation energies of PLA and PLA/zeolite composites obtained by the Flynn-Wall-Ozawa and Kissinger methods (in nitrogen flow) Apparent activation energy (kJ°mol-l) Sample Flynn-Wall-Ozawa’s Method Kissinger's Method (meaniSTD) PLA 110-120 (119.0i2.9) 105.0zt3.5 PLA/ 5 wt% type4A 70-101 (92.0 i 9.5) 86.0 :t 4.0 PLA/ 5 wt% chabazite 98-110 (112.0 i 4.8) 94.0 i 2 .0 Table 5.4 Apparent activation energies of PLA and PLA/zeolite composites obtained by the Flynn-Wall-Ozawa and Kissinger methods (in air flow) Apparent activation energy (kJ-mol-l) Sample Flynn-Wall-Ozawa’s Method Kissinger's Method (meaniSTD) PLA 101-131(125.6i:9.4) 117.03: 1.35 PLA/ 5 wt% type4A 58-114 (93.2 :t 16.8) 92.0 :t 1.39 (*Experiments on chabazite samples have not been performed under air) Consequently, the effects of type 4A synthetic zeolites were more pronounced in the thermal degradation process. In other words, using type 4A zeolites in the PLA matrix increased the thermal degradation rate of the PLA matrix. On the other hand, this result suggests that chabazites had better thermal resistance than zeolite type 4A in the PLA matrix, which resulted in an increase in the amount of energy required for the degradation 163 of the composites. This might be attributed to factors such as molecular weight, particle size, the number of available Bronsted acid sites, or heat and mass transfer phenomena during the thermal degradation process. CONCLUSION Thermal degradation properties of PLA and PLA/zeolite composites were studied by non-isothermal thermogravimetric analysis (TGA). TGA results showed that at temperatures above 300 °C, PLA/type 4A synthetic zeolite composites were thermally decomposed more easily than the PLA and PLA/chabazite natural zeolite composites. Inclusion of type 4A markedly reduced the thermal degradation temperature of PLA. The apparent activation energies of thermal degradation of the PLA and PLA/zeolite composites estimated using the F lynn-Wall-Ozawa and Kissinger methods increased in the order of PLA/ type 4A < PLA/chabazite < PLA. As a result, the effects of both type 4A and chabazite zeolites on the thermal degradation behavior of PLA provide a foundation for additional detailed investigation of the PLA/zeolite systems which can facilitate the chemical recycling of PLA. 164 REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4, 835. D. Garlotta, J Polym Environ 2001, 9, 63. R. E. Drumright, P. R. Gruber, D. E. Henton, Adv. Mater. 2000, 12, 1841. A. Sodergard, Stolt, M., Prog Polym Sci 2002, 27, 1123. A.-F. Mohd-Adnan, H. Nishida, Y. Shirai, Polym Degrad Stab 2008, 93, 1053. H. Tsuji, H. Daimon, K. Fujie, Biomacromolecules 2003, 4, 835. T. Saeki, T. Tsukegi, H. Tsuji, H. Daimon, K. Fujie, Kobunshi Ronbunshu 2004, 61 561. Y. Fan, H. Nishida, Y. Shirai, Y. Tokiwa, T. Endo, Polym Degrad Stab 2004, 86, 197. H. Nishida, Y. Fan, T. Mori, N. Oyagi, Y. Shirai, T. Endo, Ind. Eng. Chem. Res. 2005, 44, 1433. E. Rudnik, Ed. Compostable polymer materials, Elsevier Oxford 2008. O. Wachsen, K. Platkowski, K. H. Reichert, Polym Degrad Stab 1997, 5 7, 87. D. Cam, M. Marucci, Polymer 1997, 38, 1879. F. D. Kopinke, M. Remmler, K. Mackenzie, M. Moder, O. Wachsen, Polym Degrad Stab 1996, 53, 329. X. Liu, Y. Zou, W. Li, G. Cao, W. Chen, Polym Degrad Stab 2006, 91 , 3259. A. Babanalbandi, D. J . Hill, Hunter S, Kettle L, Polym Int 1999, 48, 980. F. D. Kopinke, K. Mackenzie, J Anal Appl Pyrolysis 1997, 40-41, 43. Y. Aoyagi, K. Yamashita, Y. Doi, Polym Degrad Stab 2002, 76, 53. Y. Fan, H. Nishida, T. Mori, Y. Shirai, T. Endo, Polymer 2004, 45, 1197. T. Mori, H. Nishida, Y. Shirai, T. Endo, Polym Degrad Stab 2004, 84, 243. 165 [20] [21] [22] [23] [24] [25] [26] [27] [23] [29] [31] [32] [33] [351 [36] [37] [38] [39] [40] [41] T. Motoyarna, T. Tsukegi, Y. Shirai, H. Nishida, T. Endo, Polym Degrad Stab 2007, 92, 1350. H. Nishida, T. Mori, S. Hoshihara, Y. Fan, Y. Shirai, T. Endo, Polym Degrad Stab 2003, 81 , 515. K. Okamoto, K. Toshima, S. Matsumura, Macromol Biosci 2005, 5, 813. D. W. Breck, W. G. Eversole, R. M. Milton, T. B. Reed, T. L. Thomas, J Ameri Chem Soci 1956, 78, 5963. G. Kickelbick, Prog Polym Sci 2003, 28, 83. S. Polarz, Smarsly, B., J Nanosci Nanotechno 2002, 2, 581. T. Ozawa, Bull Chem Soci Japan 1965, 38 1881. J. Flynn, H. , J Therm Anal Cal 1983. J. Flynn, L. Wall, J Res Natl Bur Stand 1966, 40, 487. H. Kissinger, J Res Natl Bur Stand 1956, 5 7, 217. N. Ignjatovic, E. Suljovrujic, J. V. Simendic, I. Krakovsky, D. Uskokovic, J Biomed Mater Res B 2004, 713, 284. A. Mohamed, S. Gordon, G. Biresaw, J Appl Polym Sci 2007, 106, 1689. Y. Fan, H. Nishida, Y. Shirai, Y. Tokiwa, T. Endo, Polym Degrad Stab 2004, 86, 197. A. Khawam, D. R. Flanagan, J Phys Chem B 2006, 110, 17315. J. P. Redfem, in Diflkrential thermal analysis, Vol.1: Fundamental aspects, Vol. 1 (Ed: Mackenzie RC), Academic Press, London 1970, 134. N. C. Billingharn, in Encyclopedia of polymer science and technology, Vol. 6:Degradation. New York: Wiley-Interscience, 2002. p.15. J. H. Flynn, Thermochimica Acta 1997, 300, 83. P. Budrugeac, Polym Degrad Stab 2000, 71, 185. A. Khawam, D. R. Flanagan, Thermochim Acta 2005, 436, 101. P. Budrugeac, Polym Degrad Stab 2005, 89, 265. 166 [42] [43] [44] [45] C. D. Doyle, J Appl Polym Sci 1962, 6, 639. B. Andricic, T. Kovacic, Polym Degrad Stab 1999, 65 , 59. I. C. McNeil], H. A. Leiper, Polym Degrad Stab 1985, I I, 309. A. Ortega, IntJ Chem Kinetics 2002, 34, 193. 167 CHAPTER 6 Hydrolytic Degradation Behavior of Poly(lactic Acid)/Zeolite Composites INTRODUCTION Ideally degradable plastics are expected to be stable and preserve their integrity during their use, yet quickly disintegrate, hydrolyze after their use life. Developing a plastic material that serves both these purposes can be challenging. Therefore, it is vital to understand the hydrolysis processes of PLA/zeolite composites and explore their degradation in an aqueous environment. The hydrolytic degradation mechanism of PLA in various environments and conditions has been widely studied. The main findings from these studies can be summarized as [1-4]: 1) During hydrolytic degradation two reactions take place; random chain scission and chain-end hydrolysis. 2) The amorphous regions of the PLA are more susceptible to hydrolysis than the crystalline regions 3) The rate of hydrolysis is dependent on a number of physical characteristics such as the degree of crystallinity and the thickness of the sample. The presence of crystallinity in PLA tends to slow down the degradation rate. Thick samples degrade faster than thin films due to having a higher concentration of catalytic acid end-groups that tend to stay longer inside the bulk. 4) Hydrolysis is catalyzed by the increasing number of carboxylic acid end- 168 groups. 5) The rate of hydrolysis is much greater above the glass transition temperature than below. In the literature, mainly three experimental design conditions for hydrolysis were used: melt hydrolysis, solution hydrolysis of PLA dissolved in solvents, and solid state hydrolysis in which the solid samples were suspended in aqueous media and/or solid samples were exposed to humidity. This chapter aimed to study solid-state hydrolyis of PLA and PLA/zeolite composites. EXPERIMENTAL Materials The materials that mentioned in previous chapters were used. Preparation of Samples The samples of PLA neat and PLA with 5 wt% type 4A and chabazite were prepared as stated in previous chapters. The samples ( ~1 cm x~ 0.5 cm x ~0.25 cm) were cut from injection molded test specimens, and then were placed into vials containing 15 mL deionized water (pH: 6.8) and 0.01 N NaOH solution (pH:11.8). The vials were stored at 23, 60, and 90 0C for 96 hrs. Three replicates were used for each composite. The vials were agitated every 5 hrs. At the end of the hydrolysis experiment, samples were removed from the solutions, rinsed with distilled water and surface dried using filter paper. The samples were then dried under vacuum at 30 0C for 3 days to constant weight. 169 CHARACTERIZATION Weight Loss The weight loss of PLA and PLA/zeolite composites after degradation was calculated according to the equation: % weight loss= [(mi-md)/mi] x 100 (l) where mi and md are initial and dry weight, respectively. Morphological Evaluation Scanning Electron Microscopy (SEM): Morphological analyses were done using a SEM JSM 6400 (JOEL, Tokyo, Japan) equipped with a LaB6 emitter on the surfaces of PLA and PLA/zeolite composites. Samples were sputter coated with an approximately 15 nm layer of gold using an Emscope SC500 sputter coater (Emscope Laboratories Ltd, Ashford, UK) operated at 20 mA for 3 min. SEM micrographs were collected at an accelerating voltage of 15 kV and a working distance of 15 mm. Differential Scanning Calorimetry (DSC) Analysis DSC studies of PLA/zeolite composites were performed on a differential scanning calorimeter (DSC) Q100 (TA Instruments, NewCastle, DE, USA) in accordance with ASTM D3418-O3 [9]. Approximately 5-8 mg of injection molded samples were heated from room temperature to 190°C at a rate of 10 oC/min under a constant nitrogen flow 170 (70 mL/min). The degree of crystallinity (Xe) of PLA and PLA/zeolite composites was calculated based on the following equation: Xe (%) = [( AHm -AHcc)/(AHf(1- X) )l * 100 (2) where AHf is the enthalpy of fusion of 100 % crystalline PLA which is 93.7 J/g [10,11], and x is the weight fraction of zeolite in the composite. At least 3 specimens were tested for each composite. Molecular Weight Determination The number-average (Mn) and weight-average (MW) molecular weights and the polydispersity index (PDI=Mw/Mn) of the PLA and PLA/zeolite composites were determined using a gel permeation chromatograph (GPC) equipped with 2414 reflective index (RI) detector (Waters, Milford, MA, USA) and using a series of three columns (HR4, HR3, and HR2). The analysis was conducted at room temperature using tetrahydrofuran (THF) as the mobile phase at a flow rate of 1.0 mL/min. Approximately 35 mg of PLA and PLA/zeolite composite samples were dissolved in 15 mL THF. In the case of PLA/zeolite composites, the zeolite particles were extracted fi'om the solution by centrifugation and then filtration through a 0.45 um PTFE filter before injection into the GPC. The GPC system was calibrated using polystyrene standards 171 RESULTS AND DISCUSSION Figure 6.1 shows the weight loss for PLA and PLA/zeolite composites after degradation. The weight loss of all samples was less than 2.5 % except for the samples stored at 90°C in water and in 0.01 N NaOH solution. Among all the samples, PLA/type 4 A samples exhibited the highest weight loss values at 90°C in both water and NaOH solution. Tsuji et al. [1] reported that the weight loss can be considered as an index for the content of water-soluble oligomers formed by hydrolysis and released from the PLA itself into surrounding media. However, they also reported that no weight change was observed at 37°C in phosphate-buffered solution for 18 months. On the other hand, the study conducted by Cam et a1 [12] showed that 10 % weight loss occurred after 25 days when PLA samples were incubated in an alkaline medium at 37°C. 172 lPLA . 40 (1) lSynZe05M% DNahralZeoSwt‘h c\° 30 :11 an G 1-1 E N 20 '5 3 See Figure 6.1 (ii) A 0 T M A B C D E F G H I 2.5 2 « ' (ii) 1.5 a __ Weight Loss, % A l A ‘ B C D E F G ’ Figure 6.1 Weight loss data of PLA and PLA/composites hydrolyzed at 23,—60, 90 dCin water and 0.01 N NaOH solution. A: 23°C; B: 23°C water; C: 23°C 0.01 N NaOH sol;D: 60°C; E: 60°C water; F: 60°C 0.01 N NaOH sol; G: 90°C; H: 90°C water; I: 90°C 0.01 N NaOH sol. The magnified portion of the graph (i) is shown in (ii). 173 The morphology of PLA and of PLA/zeolite composites was observed by SEM. Figures 6.2 and 6.3 show micrographs of samples that were stored in water and in NaOH solution at 23 and 60°C for 96 hrs. The SEM pictures showed the smooth nature of the surface of the neat PLA samples after degradation in water. There were no signs of erosion or pore formation. However, the samples with zeolites, especially type 4A, exhibited an irregular coarse grained structure at 230C, and pore openings at 60°C. The effect of the 0.01 N NaOH was much more drastic than water. The neat PLA at 23°C started to have small grains. Samples with type 4A had a completely sponge-like surface and it was clear that degradation started at the PLA and particle interface. Samples with chabazite at 60°C also showed a few large pores around the particles. The SEM micrographs are not shown for the samples kept at 90°C since they had totally degraded into a powder-like form. As can be seen from the DSC data the percent crystallinity of these samples was found to be 75-90%. 174 PLA PLA/5 Wl% type 4A PLA/5 wt% chabazite Figure 6.2 SEM micrographs of PLA and PLA/zeolite composites in water at 230C (A) and 60°C (B), scale bar 20 pm. 175 PLA/5 wt% type 4A PLA/5 wt% chabazite Figure 6.3 SEM micrographs of PLA and PLA/zeolite composites in 0.01 N NaOH solution at 23°C (C) and 60°C ( D), scale bar 20 pm. 176 It is interesting to note that the weight average and number average molecular weights of samples in water and NaOH decreased at 60°C with less than 2.5 % weight loss (Figure 6.1 and 6.4). Kanasewa and Doi [13] also found a similar result where the hydrolysis of P(3HB-o-3HV) fibres at 37 and 60°C caused a reduction in molecular weight with no measurable weight loss. The reason behind this phenomenon is not yet clear. [ I PLA u PLA+5 wt% Zeolite SYN o PLA +5 mes Zeolite NAT 1 160' 140 l 1204 + . 100, 80. 60‘ Mw x 1000, g/mol 40, 20‘ __-:L_.-:L A B CID E FGH I Figure 6.4 Weight average molecular weight data for PLA and PLA/zeolite composites at 23, 60, 90 °C, in water and 0.01 N NaOH solution. A: 23°C; B: 23°C water; C: 23°C 0.01 N NaOH sol; D: 60°C; E: 60°C water; F: 60°C 0.01 N NaOH sol; G: 90°C; H: 90°C water; I: 90°C 0.01 N NaOH sol. 177 __,| IPLA IPLA+5 wt% Zeolite SYN UPLA+5M%Zeolite NAL 100 _- - -.__. 80+ Mn x 1000, g/mol 20 Figure 6.5 Number-average molecular weight data for PLA and PLA/zeolite composites at 23, 60, 90 °C, in water and 0.01 N NaOH solution. A: 23°C; B: 23°C water; C: 23°C 0.01 N NaOH sol; D: 60°C; E: 60°C water; F: 60°C 0.01 N NaOH sol; G: 90°C; H: 90°C water; 1: 90°C 0.01 N NaOH sol. 178 8 so IPLA umuuxzmesvn 9‘ uPLA+m2eaiew 370‘ e\° :50: Q far .5 i T: I“ a m I E“ '30‘ U I ms 10: o- A B C D E Figure 6.6 Percent crystallinity of PLA and PLA/zeolite composites in 0.01 N NaOH solution and water at 23, 60, and 90°C. A: 23°C; B: 23°C water; C: 23°C 0.01 N NaOH sol; D: 60°C; E: 60°C water; F: 60°C 0.01 N NaOH sol; G: 90°C; H: 90°C water; I: 90°C 0.01 N NaOH sol. lzPLA, 2: PLA+5 wt% Zeolite SYN; 3: PLA+5 wt% Zeolite NAT 179 CONCLUSION From the above mentioned results the following overall conclusion can be drawn: Bulk hydrolysis occurs for the neat PLA sample whereas surface hydrolysis predominates in PLA/zeolite samples. Type 4A zeolite accelerates the hydrolytic degradation of the PLA matrix. 180 REFERENCES [1] [2] [3] [4] l5] [6] [7] [8] [9] [10] [11] [12] [13] H. Tsuji, K. Nakahara, K. Ikarashi, Maromol Mater Eng 2001, 286, 398. G. L. Siparsky, K. J. Voorhees, F. Miao, J Polym Environ 1998, 6. D. Henton, P. R. Gruber, J. Lunt, J. Randall, In Natural Fibers, Biopolymers and Biocomposites;Mohanty, A. K. ;Misra, M ;Drzal, L. T. Eds. CRC: Boca Raton, FL, 2005; Chapter I. S. Huang, in Handbook of biodegradable polymers, Bastioli C. ed. Rapra:Shropshire UK 2005. Chapter 9. T. Ozawa, Bull Chem Soci Japan 1965, 38 1881. J. Flynn, H. , J Therm Anal Cal 1983. J. Flynn, L. Wall, J Res Natl Bur Stand 1966, 40, 487. H. Kissinger, J Res Natl Bur Stand 1956, 5 7, 217. American Society for Testing and Materials. Standard Test Method for Transition Temperatures and Enthalpies of Fusion and Crystallization of Polymers by Diflcrential Scanning Calorimetry. D 3418-03. In Annual Book of AST M Standards. AST M: Philadelpia, PA, 2004: 8:02 S. Sinha Ray, K. Yarnada, M. Okamoto, A. Ogami, K. Ueda, Chem Mater 2003, I5, 1456. R. Auras, B. Harte, S. Selke, J Appl Polym Sci 2004, 92, 1790. D. Cam, S.-h. Hyon, Y. Ikada, Biomaterials 1995, 16, 833. Y. Kanesawa, Y. Doi, Makromol Chem Rapid Commun 1990, I I, 679. 181 CHAPTER 7 Atomic Layer Deposition of Ultrathin ALOX Films on Poly(lactic Acid) INTRODUCTION There has been a growing interest in ceramic-like coatings on polymers using various techniques in order to improve the barrier properties of polymers. Aluminum oxide is one of the most widely used ceramic materials as a protective layer for electronic, optical, biomedical, and packaging applications [1-7]. Currently, aluminum oxide coatings are deposited using various techniques such as chemical vapor deposition (CVD), plasma enhanced chemical vapor deposition (PECVD), and RF. magnetron sputtering [1,3-7]. In this chapter, we explore the possibility of using the atomic layer deposition technique (ALD). ALD is akin to plasma enhanced chemical vapor deposition. Both techniques use the gas phase to deliver molecules to the substrate surface where chemical reaction/transformation occurs. However, the distinction between these techniques is that ALD is capable of producing pinhole free and highly conformal thin film coatings on the substrates, whereas PECVD or CVD exhibit less conformal deposition. The reason behind the uniformity of ALD is the surface limited sequential cycle of precursors. Each precursor gas is introduced onto the substrate surface and reacts with the active sites until the surface is completely saturated. This saturation is very important for the film growth. The sequential precursor gas cycles are repeated until the desired thickness is reached. On the other hand in deposition techniques such as CVD and PECVD, the thickness is dependent on the deposition time [1,4,7]. The ALD 182 technique is used to deposit aluminum oxide films on various substrates including glass, metals, ceramics, nanotubes, granular materials, and powders [5,8,9]. However, there are relatively few studies that focus on aluminum oxide ALD on polymers. As mentioned in chapter 2, polymers such as low-density polyethylene (LDPE), polyethylene terephthalate (PET), polyethylene naphthalate (PEN) , polyamide (PA), polyethersulfane (PES), polystyrene (PS), poly(vinyl chloride)(PVC), and polypropylene (PP) have been employed as a substrate for the aluminum oxide ALD [9,10] In this study, aluminum oxide ALD was used on a biobased polymer, poly(lactic acid) (PLA). Although PLA has reasonable good mechanical, physical, and banier properties for a wide range of applications, in some cases the barrier properties of the PLA need to be increased in order to compete with petroleum based polymers. Nowadays, extensive work has been concentrated on the improvement of PLA barrier properties by blending it with other polymers and/or incorporating various inorganic organic particles. The present study explores the possibility of the deposition of thin layer aluminum oxide on PLA surfaces in order to improve the barrier properties of PLA. EXPERIMENTAL Materials and Methods Poly(lactic acid) (94% L-lactide) samples with a thickness of 0.304 mm were provided by Wilkinson Manufacturing Company (Fort Calhoun, NE, USA). Trimethyl aluminum (TMA) and water were selected as precursors. Previous studies showed that employing TMA and water results in higher growth rates and higher refractive index at low temperatures compared to other precursors such as aluminum 183 trichloride (AlCl3) [7,11]. Moreover the TMA and water can be considered more compatible with food packaging applications than chloride containing precursors. Alternating exposures of TMA and water were utilized to deposit aluminum oxide according to the following reactions [5,7,9] : A]-OH* + A1(CH3)3 —> Al-O-A1(CH3)2 * + CH4 [1] Al-CH3* + H20 —-> Al-OH* + CH4 [2] where the asterisks denote the surface species. Each reaction is self-limiting and produces a linear, dense, and pinhole-free growth on surfaces. Figure 7.1 illustrates the ALD system and the schematics of the precursor injection mechanism. Transparent aluminum oxide coating film was grown on a PLA surface using a prototype ALD system developed by SVT Associates, Eden Prairie, MN, USA. The largest sample size that this prototype ALD chamber can handle was 10.16 cm in diameter. A typical ALD process, one growth cycle process consisted of four main steps: 1) introduction of TMA onto the PLA surface and formation of one monolayer on the PLA surface, then surface saturation; 2) nitrogen purge to remove the excess TMA and by—products; 3) introduction of water to produce a saturated monolayer and 4) nitrogen purge to evacuate the excess water and by-products. For each gas injection line, two fast acting ALD specific valves from Swagelok were used to obtain millisecond actuation which helped to reduce the interrnixing of precursors. 184 Particle filter / Needle valve Shut-off valve Precursor bottles: r A and B To injector Fast actin \‘ valve g F 3 Mechanical pm" (b) Figure 7.1 Picture of the ALD system (a) and schematic diagram of the precursor injection system (b) 185 The surface topographies of aluminum oxide deposited PLA and neat PLA surfaces were examined using atomic force microscopy (AFM). The AFM measurements were done with a Nanoscope IIIA (Digital Instruments, Santa Barbara, CA) operating in contact mode, at ambient conditions. Images of 30x30 um2 and 10x10 umz were scanned on two different locations of each sample. The root-mean-square (RMS) roughness calculations were performed by NanoScope software (Digital Instruments, Santa Barbara, CA). Spectroscopic ellipsometry was also used to measure the refiactive index and thickness of the aluminum oxide deposition. Elemental analysis by energy dispersive spectroscopy (EDS) was carried out on a scanning electron microscope (SEM) (model J SM 6400 (JOEL, Tokyo, Japan) equipped with an EDS detector in order to confirm the aluminum oxide deposition. The analysis was conducted at an acceleration voltage of 20kV and working distance of 15 mm. The PLA neat substrate and aluminum oxide coated PLA were carbon coated before the analysis. The WV'IR was measured using a Permatran W3/33 from Mocon Inc. (Minneapolis, MN, USA) in accordance with ASTM F 1249-06 “Standard Test Method for Water Vapor Transmission Rate Through Plastic Film and Sheeting Using a Modulated Infrared Sensor” [12] Temperature and relative humidity of the test were 378°C and 90% RH, respectively. The OTR was tested using an Illinois 8001 (Illinois Instruments, IL, USA) in accordance with ASTM D 3985-02 “Standard Test Method for Oxygen Gas Transmission Rate Through Plastic Film and Sheeting Using a Coulometric Sensor” [13] Test conditions were 23°C and O % RH. 186 RESULTS AND DISCUSSION Transparent aluminum oxide coating film was grown on a PLA surface at a temperature of 42°C using an ALD system developed by SVT Associates. Aluminum oxide film was grown on a PLA surface at a temperature of 42°C using an SVTA ALD reactor with trimethyl aluminum and water. The deposition growth rate was approximately 1.28 angstrom per coating cycle. After 250 cycles, the thickness of the aluminum oxide film was found to be approximately 32 nm. The ellipsometry results showed that the refractive index of the aluminum oxide coating was close to that of bulk alumina A1203 (1.796). EDS elemental analysis confirmed the presence of aluminum oxide depositions on the PLA surface. However, the stoichiometric ratio of Al/Si could not be determined. Since the penetration of EDS beam was expected to be around 1 pm (with 20kV of energy used for the measurements), the amount of oxygen determined may also come from the base substrate which is PLA. Therefore, the oxygen peak spectrum can be considered the combination of the coating layer and the base PLA film. For comparison purposes the EDS spectrum of neat PLA was also measured. Nevertheless, the presence of Al Ka peak centered at 1.5 KeV and oxygen peak centered at 0.5 KeV confirmed some aluminum oxide formation. Therefore, hereafter, aluminum oxide ALD is referred as Ale in this text. AFM was used to study the surface structure of the aluminum oxide layer on the PLA. The coated surfaces were smooth and defect-free. The root-mean-squared (RMS) 187 surface roughness values obtained from the scan size of 30x30 um2 and 10x10 um2 were about 5.43:0.8 and 3:1:05 nm, respectively, for the neat PLA. In the case of coated PLA, the RMS values were found to be 11.7:t0.9 nm and 3.7 3:06 nm, respectively. The barrier properties of the coated PLA films were also investigated (Table 7.1). The deposition of aluminum oxide substantially reduced the oxygen permeability of the PLA samples from 1180 i 9 to 276 i 3 cc-mil/ m2.day.atm. A similar reduction was also observed for the water vapor permeability. The water vapor permeation decreased fiom 451 i 5 to 91 i 10 g-nril/m2.day.atm. Table 7.1 Oxygen and water vapor permeability of PLA and aluminum oxide coated PLA films. Oxygen Permeability * Water Vapor Pemeability (cc . mil)/(m2 . day . atm) (g . mil)/(m2 . day . atm) PLA 1180i9 451i5 PLA 276 :1: 3 91 :1: 10 A10, coated CONCLUSION Ultrathin aluminum oxide coatings were successfully deposited on PLA surfaces using an atomic layer deposition technique at 42°C. The thickness of the coating was found to be 32 nm. The AF M results obtained from small scanning areas (10 x 10 umz) revealed that the surface appeared to be as smooth as neat PLA. The newly developed films exhibited high gas barrier properties and good transparency. 188 REFERENCES [1] [2] [3] [4] [5] [6] l7] [8] [9] [101 [11] [121 [13] s. J. Noh, s. K. Lee, E. H. Kim, Y. J. Kong, Curr App Phys 2006, 6, 171. A. Niskanen, K Arstila, M. Ritala, M. Leskela , J ElectrochemistrySoc 2005, I52, F90. M. Ritala, K. Kukli, A. Rahtu, P. I. Raisanen, M. Leskela, T. Sajavaara, J. Keinonen, Science 2000, 288, 319. J. W. Elam, M.D. Groner, S.M George, Rev Sci Instrum 2002, 73, 2981. M. D. Groner, F. Fabreguette, J. W. Elam, S. M. George, Chem. Mater. 2004 I6, 639. L. Zhang, H.C. Jiang, C. Liu., J.V. Dong, P Chow., J Phys D: Appl Phys 2007, 40, 3707. A. Heyman, C. B. Musgrave, J Phys Chem B 2004, 108, 5718. X. Liang, S. M. George, A. W. Weimer, N.-H. Li, J. H. Blackson, J. D. Harris, P. Li, Chem Mater 2007, I9, 5388. S. M. George, Chem . Rev 2009, 110(1), 111. C. A. Wilson, R. K. Grubbs, S. M. George, Chem. Mater. 2005, 17, 5625. M. Ritala, M. Leskela, Handbook of thin film materials, Academic Press, 2002. American Society for Testing and Materials. Standard Test Method for Water Vapor Transmission Rate Through Plastic Film and Sheeting Using a Modulated Infrared Sensor. F 1249-06. In Annual Book of AST M Standards. AST M: Philadelpia, PA, 2006: I 5. 02. American Society for Testing and Materials. Standard Test Method for Oxygen Gas Transmission Rate Through Plastic Film and Sheeting Using a Coulometric Sensor. D 3 985 -02. In Annual Book of AST M Standards. AS T M: Philadelpia, PA, 2004: 15.10. 189 CHAPTER 8 Poly(lactic acid)/ Aluminum Oxide Composites Fabricated by Sol-gel and Melt Compounding Processes *Extended version of manuscript: Yuzay, I.E., Auras, R, Selke, S. Macromolecular Materials and Engineering, 2010 DOI :10.1002/mame.200900223 INTRODUCTION There has been an increasing interest in metal oxides due to their properties such as optical, magnetic, electrical, thermal, mechanical, and catalytic. Metal oxides play a leading role as nomnetallic materials in a variety of applications including the fabrication of ceramics, circuits, gas- and bio-sensors, fuel cells, coatings, solar absorbers, and insulators [1,2]. Among these applications, metal oxide coating on polymers is a rapidly expanding technology that can be used as a protective layer for medical devices and a gas banier layer film for food and pharmaceutical packaging applications [3-6]. There are two common methods used to deposit thin films of a variety of metal oxides: vacuum and atmospheric-based coating techniques. Vacuum coating techniques can be divided into three categories: physical vapor deposition (PVD), chemical vapor deposition (CVD), and atomic layer deposition (ALD). All of these deposition techniques use a gas phase to transport molecules to the surface of the substrate where a thermal reaction/deposition occurs [2,7,8]. They have a number of advantages; however, the slow deposition rate and the requirement of high pressure and temperature makes them 190 commercially unattractive. Plasma-assisted CVD and ALD permit the production of metal oxide coatings on polymers at relatively low pressures and temperatures, but they are still expensive and cumbersome, especially for coating large areas [4,9,10]. On the other hand, atmospheric-based coating techniques consist of applying a chemical solution followed by thermal or UV curing. The sol-gel method is one atmospheric-based deposition technique for preparation of metal oxide coatings. It has very attractive features such as easy preparation and scale-up, low processing temperature, capacity for entrapment of various components, and good chemical homogeneity [2,11,12]. Metal oxide coatings derived fi'om the sol-gel process are formed through the hydrolysis and condensation of precursors such as aqueous metal salts and, especially, alkoxides [1,11,13]. Metal alkoxides are members of the family of metal-organic compounds having the general formula M(OR)n, where M, R, and n represent a metal ion, an alkyl group, and the valence of the M ion, respectively. Hydrolysis of metal alkoxides followed by condensation of hydrolyzed metal oxides leads to formation of metal-oxygen-metal bonds and eventually forms particulate or polymeric gel networks [1,2,11]. The sol-gel solution can be applied directly on the substrates by spin, dip, or spray coating [14] Further heat treatment enables the densification of the sol-gel solution which results in formation of a protective layer on the polymer substrate. In the literature, there are many investigations of metal oxide coatings on polymers such as polycarbonates (PC), polyethylene terephthalate (PET), polyrnethyl methacrylate (PMMA), polydimethylsiloxane (PDMS), polyimide, polypropylene (PP), etc.[4,6,15- 17]. While in most of these studies, silicon oxide (SiOx) was deposited on the polymers, 191 aluminum oxide (Ale) has received little attention. Yet, Ale has remained one of the most widely used metal oxides for non-polymeric surfaces due to its combination of wear resistance, high thermal and chemical stability, and electrical resistivity [18]. In addition, Ale coatings exhibit good biocompatibility and tribologic behavior in arthroplasty applications [1 8,19]. Therefore, this chapter aimed to evaluate the feasibility of fabrication of bio-based polymer/Ale composites using two different sol-gel based processing methods and one standard mixing method. As a bio-based polymer, poly(lactic acid) (PLA) was chosen since it is one of the most promising biodegradable polymer alternatives to traditional petroleum based polymers. PLA is an aliphatic polyester and can be derived from renewable resources such as corn and sugar beets [20-22]. Several research efforts have been directed towards developing PLA based composites with high performance and/or firnctionality. Various nano/micro organic and inorganic fillers have been incorporated into the PLA matrix. A detailed discussion of these investigations can be found elsewhere [23-27]. In this study, the PLA/Ale composites fabricated by sol-gel and melt compounding methods were characterized using various analytical techniques in order to investigate the effect of Ale particles on the morphological, physical, mechanical, and thermal properties of the PLA matrix. 192 EXPERIIVIENTAL Materials Poly(lactic acid) resin produced from 94% L-lactide was obtained fiom NatureWorks LLC (Blair, NE, USA). Aluminum isopropoxide [AIP, Al[OCH(C2H6)]3, Sigma-Aldrich, 298% purity], acetic acid [AcAc, CH3C02H, Sigma-Aldrich, Z 99% purity ], and distilled water were used for aluminum oxide (Ale) sol-gel preparation. Calcinated aluminum oxide (a-A1203) was purchased from Sigma-Aldrich and used as received. Methods Three approaches were used to incorporate Ale into the PLA matrix. The first approach was based on a non-traditional core-shell polymer system. A sol-gel coating technique was used to create a core-shell polymer system where the PLA was the core phase and aluminum oxide was the shell phase. After coating PLA pellets with an aluminum oxide barrier layer, PLA pellets were extruded using a mini-twin screw extruder and then injection molded. During the extrusion process, the Ale shell on the PLA was broken and dispersed in the PLA matrix. A similar approach was used by Liang et al. [10,28] who coated micron-sized high density polylethylene (HDPE) particles with aluminum oxide by ALD in a fluidized bed reactor and then extruded the coated HDPE particles into pellets or films, which resulted in good dispersion of aluminum oxide 193 flakes in the HPDE matrix. It was expected that these flakes (broken shell particles) might create a torturous path for permeant gases and improve the barrier properties. The second approach was direct mixing of the PLA with Ale particles precipitated from a sol-gel solution, and then extrusion using the mini-twin screw extruder mentioned above and finally injection molding. The last approach was introducing calcinated aluminum oxide particles (a-AIZO3), which is the most thermodynamically stable form of aluminum oxide, into the PLA matrix during extrusion. The following section describes in details the three process steps. Sol-gel Preparation The method used to prepare aluminum oxide sol-gel was adopted from procedures described by Yoldas [29]. AIP was chosen as the precursor since it reacts readily with water and the by-product, isopropyl alcohol, can be removed from the system during drying. Acetic acid was also used as a peptizer to obtain a sol having stable colloidal particles [30]. AIP was dissolved in distilled water at 80°C. While constantly stirring, acetic acid was added to the solution, and then refluxed at 80°C for 12 h. The molar ratio of water to alkoxide precursor to acetic acid was 100:1:O.36. The resulting sol exhibited high stability; neither gelation nor precipitation was observed after 5 months of storage at room temperature. 194 PLA-Aluminum Oxide Composite Preparation After the aluminum oxide sol-gel solution was prepared, PLA pellets were submerged into the sol- gel solution at room temperature for 3 h. Special care was taken to completely cover the PLA pellet surfaces with the sol-gel solution. The PLA pellets were removed from the solution and kept in a hood for 1 h and then dried in an oven at 60°C overnight. The sol-gel coated PLA pellets were then extruded using a micro extruder (DSM Research, Geleen, The Netherlands) equipped with co-rotating twin screws having lengths of 150 mm, W ratio of 18, and capacity of 15 cc. The extrusion was carried out at 190°C at a screw speed of 100 rpm and a cycle time of 5 min. After the set cycle time, the extrudate was collected from the die and transferred into a mini-injection molder (DSM Research, Geleen, The Netherlands) by a pre-heated transfer cylinder in order to prepare test specimens for physical and mechanical property evaluation. The optimized injection pressure was 0.896 MPa and the mold temperature was kept at 30°C with a residence time of 15 see. In order to prevent moisture uptake, test specimens were wrapped in aluminum foil and stored at 23°C and 50% relative humidity for not less than 40 h prior to testing in accordance with ASTM D 618-03 [31] (Standard Practice for Conditioning Plastics for Testing). The preliminary thermogravimetric analysis (TGA) revealed that the weight content of Ale in the specimens obtained from sol-gel coated PLA pellets was approximately 3 wt%. For comparison purposes, the particles precipitated from the sol-gel solution and calcinated aluminum oxide were also introduced into the PLA matrix. The Ale sol-gel 195 solution used to coat PLA pellets was dried at 60°C for 3 h and the remaining flaky precipitates were also dried at 100°C overnight in a vacuum oven. No heat treatment was applied to the calcinated aluminum oxide particles (or-A1203), which were used as purchased. The sol-gel precipitates and the a-A1203 particles were mixed with PLA pellets at a loading of 3 wt%, and then the PLA/Ale composites were fabricated by the DSM microextruder and injection molding system using the same processing conditions as mentioned above. It is well known that the particles derived from a sol-gel solution are in a hydrous state and undergo several transition phases such as boehmite, 'y-, 11-, 5-, p-, X“: 0-, K-, and a-A1203 when they are subjected to heat treatment. The a-A1203 is the most stable structure, and occurs around 1200 oC [18,30,32] In this research, the crystallographic phase structure of the particles derived from the sol-gel solution was not characterized. Hereafter, for the sake of simplicity, the PLA composites made fi'om the sol-gel coating method will be referred as PLA/SGl, the composites prepared by mixing the precipitated sol-gel particles as PLA/S62, and those made by adding calcinated aluminum oxide particles as PLA/CALC. PLA/SGI, PLA/SG2, and PLA/CALC composites altogether as a group are referred to as PLA/Ale composites. The loading level of 3 wt% was selected for the entire study because it was the approximate amount 196 of deposition determined from the preliminary thermogravimetric analysis (TGA) measurements for the PLA/8G1 composites. CHARACTERIZATION Scanning Electron Microscopy (SEM). Morphological analyses were done using a SEM JSM 6400 (JOEL, Tokyo, Japan) equipped with a LaB6 emitter on the fracture surfaces of PLA and PLA/Ale composites. The Izod impact fractured surfaces facing upwards were mounted onto aluminum stubs with carbon adhesive tape and then sputter coated with an approximately 15 nm layer of gold using an Emscope SC500 sputter coater (Emscope Laboratories Ltd, Ashford, UK) operated at 20 mA for 3 min. Finally, SEM micrographs were collected at an accelerating voltage of 15 kV and a working distance of 15 mm. An elemental analysis by energy dispersive spectroscopy (EDS) was also carried on the SEM equipped with an EDX detector. The analysis was conducted at an accelerating voltage of 20 kV and working distance of 15 mm. Dried sol-gel particles were carbon coated before the analysis. Fourier Transform Infrared (FT IR) Spectroscopy. The surface structure of PLA and aluminum oxide composites was examined by a Shimadzu IRPrestige-Zl spectrometer (Columbia, MD, USA) used in the attenuated total reflection mode. Each spectrum was collected at a resolution of 4 cm.1 and a scan rate of 40 over the range of 4000550 em'l. 197 Optical Properties. The transmission of visible and UV light was measured on a Lambda 25 UV-visible spectrometer from Perkin-Elmer Instruments (W ellesley, MA, USA). The spectra was recorded at a scanning rate of 480 nm-min-l , from 200 to 800 nm. Differential Scanning Calorimetry. Thermal analyses of PLA/Ale composites were performed on a differential scanning calorimeter (DSC) Q100 (TA Instruments, NewCastle, DE, USA). Approximately 5-8 mg of injection molded samples were heated from room temperature to 180°C at a scanning rate of 10 OC/rnin, kept at this temperature for 2 min and then quenched to -30 0C. Then a second heating scan was performed from 10 to 180°C at a rate of 10 0C/min. All measurements were conducted under a constant nitrogen flow (70 mL/ min). For each sample, the glass transition temperature (Tg), cold crystallization temperature (ch), melting temperature (T m), and enthalpies of cold crystallization (AHCC) and melting (M1,) were evaluated from the DSC thermograms. The degree of crystallinity (Xc) of PLA and PLA/Ale composites was calculated based on the following equation: Xc(°/o)=[(AHm-AHccl/(AHf(1-X))l"‘ 100 (1) where AHf is the enthalpy of fusion of 100% crystalline PLA which is 93.7 J ~g-1[33,34], 198 and x is the weight fraction of Ale in the composite. At least 3 specimens were tested for each composite combination. X-ray Diffraction (XRD). The x-ray diffraction measurements were performed for PLA and PLA/A10, composites using an x-ray diffractometer Rigaku 2OOB (Rigaku Corp, Tokyo, Japan) operated at a voltage of 45 kV and a current of 100 mA, equipped with Cu K a radiation source (ll =1.541 nm). The diffraction data were collected fiom 26: 2 to 35° with a step width of 0.020 and a step time of 0.4 s. Thermogravimetric Analysis (TGA). TGA was performed using a TGA 2950 (TA Instruments, New Castle, DE, USA) with a heating rate of 20 °C/min from room temperature to 600°C. All TGA measurements were performed under nitrogen flow (inert atmosphere) of 70 mL/min. Three specimens were tested for each composite. Dynamic Mechanical Analysis (DMA). The temperature dependence of the storage modulus and loss modulus of PLA and PLA/Ale composites were evaluated using a DMA Q800 (TA Instruments, NewCastle, DE, USA). The samples were heated at a rate of 2 oC/min from room temperature to 90°C and tested in the dual cantilever mode at an oscillating amplitude of 15 um and frequency of 1 Hz. Three specimens were tested for each composite. 199 Molecular weight distribution. The number-average (Mn) and weight-average (MW) molecular weights and the polydispersity index (PDI=Mw/Mn) of the PLA and PLA/Ale composites were determined using a gel permeation chromatograph (GPC) equipped with 2414 reflective index (RI) detector (Waters, Milford, MA, USA) and using a series of three columns (HR4, HR3, and HR2). The analysis was conducted at room temperature using tetrahydrofuran (THF) as the mobile phase at a flow rate of 1.0 mL/min. Approximately 35 mg of PLA and PLA/Ale composite samples were dissolved in 15 mL THF. In the case of PLA/Ale composites, the Ale particles were extracted from the solution by centrifugation and then filtration through a 0.45 pm PTFE filter before injection into the GPC. The GPC system was calibrated using polystyrene standards. RESULTS AND DISCUSSION Morphology The fracture surfaces of PLA and PLA/Ale composites were examined using SEM in order to evaluate the size, shape, and distribution of Ale particles in the PLA matrix (Figure 8.1). It appeared that particles obtained fiom SGl and SG2 were flake- like, irregular in shape, and ranged from 1-10 pm in size, whereas calcinated aluminum oxide particles were spherical in shape with an average particle size of 10-15 um. As can be seen from Figure 8.1, all of the PLA and PLA/Ale specimens exhibited a typical brittle fracture with very smooth fracture surfaces. The Ale particles were fairly unifome dispersed in the PLA matrix without any agglomerates. This indicates that the Ale shell coating on the PLA pellets can be broken into small pieces and distributed homo genously during the melt compounding process. 200 Figure 8.1 SEM images of PLA and PLA/Ale composites. A: PLA magnification: 200X, scale bar 100 um, B: PLA/SGl magnification:200X scale bar:100 mm, C: PLA/SG2 magnificationz2OOX scale bar:100 um, D: PLA/SG2 magnificationz6OOX scale bar:50 um, E: PLA/CALC magnification2200X scale bar: 100 um (Arrows and circles indicate Ale particles) 201 SEM-EDS measurements confirmed the presence of O and A1 in the dried sol-gel solution (SG2) with the atomic ratio Al/O of 0.51i0.02 which is somewhat lower than the theoretical stoichiometric value for aluminum oxide, A1203 (0.67). It is obvious that using the sol-gel method without the calcination process makes it diffith to reach the theoretical value. FTIR Analysis The infi'ared spectra of Ale sol-gel solution (SGl), dried Ale sol particles (SG2), and calcinated aluminum oxide particles are shown in Figure 8.2. The spectrum of SGl shows two broad absorbance bands. The first is located in the region of 1000-500 cm"1 and includes the O-Al-O bending mode in the range of 650-700 cm.1 and the Al-O stretching mode at 750-850 cm.1 [35,36]. The bands observed in this region can also have contributions from C-O stretching modes, C-H rocking motions, and C-C stretches of isopropyl groups [37,38] The overlapping of these modes results in a broad peak in this particular region. Another broad band is found in the 3800-2600 cm-1 region, which can be assigned to the OH stretching mode [38]. The peak at 1645 cm-1 is also attributed to the O-H bending mode [3 7]. The spectrum of calcinated aluminum oxide is similar to the spectrum of SG2. It also does not show any absorption peak at about 3800- 2600 cm.1 and at 1645 cm'l. In addition, all of the alkyl stretching frequencies have disappeared. The only broad peak left is near 500- 900 cm"1 which defines the characteristic aluminum oxide structure [39]. 202 (C) (b).. . . . . ‘21 (a) I \A Transmittance 1111111111111111111111111111111111 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber, cm'1 Figure 8.2 FTIR spectra of (a) Ale sol-gel solution (SGl), (b) dried Ale sol particles (SG2), and (c) calcinated aluminum oxide particles (CALC). Figure 8.3 depicts the FTIR spectra obtained for PLA and its composites with aluminum oxides (SGl, SG2, and CALC). A slight increase in the peak intensity is observed in the region of 3800-2600 cm-1 alter inclusion of SGl and SG2, indicating the formation of hydrogen bonds between the PLA and aluminum oxide sol-gel and particles. The intensity of the broad peak at about 900-500 cm.1 is also increased with the addition of SGl , SG2, and CALC, due to the aluminum oxide vibrations. 203 (d) . <0) 0 C 5 5” 33 i Cb.) E Z ‘ “- m L g : (a) In ' ' _ l-t ' W’ TL: 3550 2550 1550 0 Wavenumber, cm'l 1m WW4 “1 100 90- 95$ 90 94-§ t °\° 80* °\° 92*: 0. . . I u 3 90:; 5 70 t H a a : 3:1 .13 88‘: E ; n 60. § we 3 a : ;- 5 841 l I 50 , 82: I : l L 80“ W n .‘ 40L r .44 3800 3000 3200 3000 2000 2600 85° 75° 55° 55° 1 Wavenumber, cm'l Wavenumber, cm' Figure 8.3 FTIR spectra of (a) PLA, (b) PLA composites with SGl, (c) PLA composite with SG2, and (d) PLA composite with calcinated aluminum oxide (CALC). 204 Transmission, % Optical Properties Figure 8.4 shows the UV-visible transmission spectra of PLA and PLA composites with 3 wt-% Ale at a film thickness of 2 mm. The PLA, PLA/8G1, PLA/SG2 and PLA/CALC have a transmission at 650 nm of 90 i 1.1, 83 i 0.8, 83 :h 1.0, and 49 d: 0.6%, respectively. At 300 nm, the neat PLA transmitted 33 i 0.5% of the light while the PLA/CALC transmitted only 11 i: 0.3% of the light. The % transmission for both PLA/SGl and PLA/SG2 also at 300 nm was about 13 i 0.2%. These strong absorption values suggest that the PLA/A10x composites provide better UV barrier than the neat PLA. PLA 1°05 4‘ @861 802 PLA/SG2 . so: . PLA/CALC 4O : 1 200 300 £00 550 600 foo 600 900. Wavelength, (nm) Figure 8.4 UV-visible transmission 0)) ectra of PLA and PLA/Ale composites. Thickness: 2 mm 205 Thermal Analysis The thermal characteristics of injection molded PLA and PLA/Ale composite samples evaluated using a DSC are shown in Table 8.1, which summarizes the transition temperatures along with the calculated percentages of crystallinity of PLA and PLA/Ale composites obtained fi'om the first and second heating cycles. The T g was not .. significantly altered by the addition of Ale; however, the melting temperature and I! . percent crystallinity were affected in both the first and second heating cycles. l Table 8.1. Thermal characteristics of PLA and PLA/Ale composites. 18‘ Heating Cycle Sample Tg(°C) ch(°C) Tmi(°C) T...2(°C) X1st(%) PLA 58.5:t0.7a 123.6i4.0a 148.0i3.7ab 153.8i0.7a 3.2i0.7a PLA/8G1 60.2:b2.0a 113.5:1:O.7b 149.0:b0.1a 153.5:h0.7a 18.5:l:2.0b PLA/SG2 59.2i0.53 1 12.9:t0.6b 150.7:t0.6b 155.2i0.0b 15.4i4.9bc PLA/CALC 57.2:l:1.0a 122.7:1:0. 1 a 151 .8i0. la - 10.9i0.3c 206 Table 8.1.(Cont’d) 2'"I Heating Cycle Sample Tg(°C) ch(°C) Tmi(°C) Tm2(°C) X1st(%) PLA 60.0i0.5a 123.4:t2.2a 151.9:|:O.8a - 4.25:0.8a PLA/SGI 59.7i0.5a - 152.9:t0.1a - 4.43:0.1a PLA/SG2 59.5:t0.7a - 153 .2d:0.4a - 3.9i0.7a PLA/CALC 59.2i0.4a 130.9i0.6b 153.1:t0.1a - 48$] .03 Note: Values indicate means with standard deviations. Means in the same columns with different Superscript letters are significantly different (P < 0.05) ~Tm1 and Tmz values indicate the first and second melting peaks, respectively. Figure 8.5 -abcd shows the melting endotherrns obtained from the first heating cycle for the PLA and PLA/Ale composites. The PLA, PLA/8G1, and PLA/SG2 composites exhibited bimodal melting endotherms. The occurrence of bimodal melting behavior of PLA and its composites was reported by several researchers [40-44] This phenomenon was linked to various reasons [45-52] : 1) the effect of melting the original crystals, recrystallization, and remelting the recrystallized crystals during the heating scan; 2) the presence of different larnellae morphologies formed prior to the heating scan; 3) the existence of more than one crystal structure (polymorphism); 4) processing conditions or molecular weight distribution; and 5) the effect of low crystallization temperature which caused the formation of the disordered alpha phase of PLA . 207 a PLAtstheating b ..... \ g, ------- PLNSG1tstheating 0‘ W4" 0 _\ /___ PLNSGZ1stheating \ i l d \55/ . “'-~-~\§ §/--——- PLNCALC1sthoating e -\ s~—-- PLAanheating -2- f -—-———-—-.;.___§,—— ----- PLNsmzndheating 0 — ----- 3.,-— PLAlsczmdheaung " PLNCALCanheating l .4 . . . . I . . . r r r . ff. . . . . 110 135 160 185 210 EXO UP Temperature (°C) Figure 8.5 DSC melting curves of PLA and PLA/Ale composites with a heating rate of 10 °C/min. a: PLA (lst heating cycle); b: PLA/8G1 (lst heating cycle); c:PLA/SG2 (lst heating cycle ); d: PLA/CALC (lst heating cycle); e: PLA (2nd heating cycle); f: PLA/SGl (2nd heating cycle ); g: PLA/SG2 (2nd heating cycle); h: PLA/CALC (2nd heating cycle) 208 As shown in Figure 8.5, the injection molded PLA and PLA/8G1 composites exhibited two melting peaks located at about 149 and 154°C. The prominent endothermic peaks were seen at the lower melting temperature of 149°C and additional small peaks (shoulder) were seen at around 154°C. Yasuniwa et al. [45,46] reported that in bimodal melting systems, the first endotherm is usually attributed to the melting of some original crystals, and the later one is ascribed to the melting of crystals formed during the melting process. The imperfect or unstable crystals formed during the heating scan start turning into more stable crystals; consequently, the recrystallization takes place druing the melting process. Then, the recrystallized crystals again start melting which forms a melting peak at higher temperature. In many cases, the rate of melting and rate of melt- recrystallization compete with each other during a heating scan which results in changing the shape and intensity of endothermic peaks [42,45,47,51]. These changes in endothermic peaks were also attributed to the existence of different crystalline structures (i.e., variation in thickness of the larnellas ) being formed during processing or in the DSC scans [41-44]. For the PLA/SG2 composites, the peak shape and the intensity of the prominent endothermic peak corresponding to the lower-temperature melting peak remained almost unchanged while the shape and the intensity of the small endothermic peak corresponding to the high melting peak changed. On the other hand, the PLA samples with calcinated aluminum oxide particles showed a single melting endotherm centered at 151°C. This result suggests that calcinated aluminum oxide particles lead to a more perfect crystal formation and homogenous crystalline phase [47,53]. 209 Figure 8.5-efgh shows the melting endotherrns obtained from the second heating cycle for the PLA and PLA/Ale composites after quenching from the melt. The double melting peak features for the PLA, PLA/SGl and PLA/SG2 composites totally disappeared. Since the rapid quenching did not allow the PLA and PLA/Ale composites to crystallize from the melt, the presence of single melting endotherrns for PLA and PLA/Ale composites suggests that the crystals that formed during the 2nd heating cycle were intrinsically stable. In the second heating cycle, the percent crystallinity of PLA and PLA/Ale composites were found to be approximately 4%, suggesting that the injection molded PLA and PLA/Ale composites were nearly amorphous. This result was also confirmed by the x-ray diffraction measurements performed on PLA and PLA/8G1 samples. PLA/SGl composites were selected for the XRD studies since they exhibited about the same percent crystallinity in the second heating cycle as well as the highest percent crystallinity values in the first heating cycle. Therefore they were expected to show the most difference from the PLA. However, the broad peak observed at approximately 28:15-16o suggests that the PLA and PLA/SGl samples have a predominantly amorphous structure (Figure 8.6). In the composites, the amount of SGl, SG2 and CALC may not be sufficient to sufficiently accelerate the crystal growth. Similarly in the study by Streller et al. [54], where boelunite fillers with five different crystallite sizes (10, 20, 40, 53, and 60 nm) were used in isotactic polypropylene (iPP) with a loading level of 10 210 wt%, the degree of crystallinity of iPP was not affected by the fillers. However, it was reported that the boehmites were effective nucleating agents for iPP, increasing the melt crystallization temperature by about 14°C. Bhimaraj et a1. [55] showed that the addition of aluminum oxide nanoparticles (a—A1203) into PET did not increase the percent crystallinity; furthermore, it did not cause PET to heterogeneously nucleate. 2000 1500 5 <1 £91000 5 E 500 0 l l l l l l T 0 5 10 15 20 25 30 3540 2 theta (degrees) Figure 8.6 X-ray diffraction patterns for the injection molded PLA and PLA/SGl. 211 It is also interesting to note that after the PLA and PLA/Ale composites were quenched quickly from the melt to prevent any crystallization during the cooling cycle, the cold-crystallization exotherms disappeared for PLA/SGl and PLA/SG2 in the second heating scan. We hypothesize that the heating scan of 10 oC/min was not slow enough to create cold-crystallization exotherms for the PLA/SGl and PLA/SG2 composites, yet it was slow enough for the neat PLA and PLA/CALC composites. Thermogravimetric Analysis Thermal stabilities of neat PLA and PLA/Ale composites were investigated using thermogravimetric analysis. Figure 8.7 illustrates temperature-dependant weight loss curves of PLA, PLA/SGI, PLA/SG2, and PLA/CALC composites. The complete thermal degradation of PLA and PLA/Ale composites occurs in a single step. Temperatures 0f 5% weight 1053 (T 5% weight loss), 50% weight 1053 (T 50% weight loss), and maximum degradation (T dmax) of all samples are summarized in Table 8.2. PLA and PLA/Ale composites were relatively stable up to temperatures of about 350°C. At temperatures above 350°C, PLA/CALC composites were thermally decomposed more easily than the PLA, PLA/8G1, and PLA/SG2 composites. For neat PLA, the temperatures at 5% and 50% of weight loss were around 391.010.] and 426.0:l:0.l°C, respectively. The maximum degradation temperature (T dmax) of PLA, also called the temperature of maximum weight loss rate, and ascribed to the maximum peak position of the derivative of the TGA curve, was found to be 433.0 i 03°C. 212 120 ; 4 100 1 ‘1" °0 » Pwsct ; . ,1 _ 40m 3.3 PLA/30w 11111312 80 t s, -—PWCALC 8 j“, . ,_1 . X b E :11 m“ 501 1302* i’ 111 8 I '5 if? 1—1 ’ a 11 1 E 40 i >' /’1 .g ; '5 1 WC ,1"; PM a 1 Q will 20 t l . . 1' I t ‘ I 1 . , _ , , 0 . 111,.i., ... .... 0 , 0 100 200 300 400 500 600 200 300 Temperature, °C 400 500 600 Temperature, °C Figure 8.7 TGA curves of PLA, PLA/SGl, PLA/SG2, and PLA/CALC composites. Heating rate: 20 oC/nrin, in nitrogen atmosphere. Table 8.2 TGA analysis of PLA and PLA/Ale composites. Sample T 5 % weight loss T 50 % weight loss Tdmax PLA 391.0 d: 0.1 a 426.0 i 0.1 a 433.0 i 0.3 a PLA/8G1 384.0 3: 1.0 b 417.0 3: 0.5 b 424.0 d: 0.4 b PLA/SG2 384.0 d: 1.4 b 419.0 :1: 0.1 c 424.0 :t 0.1 b PLA/CALC 373.0 d: 0.1 c 410.0 3: 0.2 (1 417.0 a: 0.00 c Note : Values indicate means with standard deviations. Means in the same columns with different superscript letters are significantly different (P < 0.05) 213 Incc temperature about 433.1 obvious th reduction i [57], Chan a filler in investigati Dynamic Fi; mOdUluS, aluminum 3460 and the lowe decrease storage transit“ Incorporation of 8G] and SG2 into the PLA resulted in slightly lower degradation temperatures. However, the maximum degradation temperature of PLA decreased from about 433.0i0.3 to 417.0:t0.l 0C with the presence of calcinated aluminum oxide. It is obvious that the calcinated aluminum oxide catalyzed the PLA pyrolysis. A similar reduction in thermal degradation temperature was also reported by Nakayama [56], Wu [57], Chang [58], and Ogata [59] when titanium oxide or clay nanoparticles were used as a filler in the PLA matrix. The cause of this reduction is unclear and requires further investigation. Dynamic Mechanical Analysis Figure 8.8 illustrates the temperature dependence of the storage modulus, loss modulus, and tan delta for PLA and PLA/Ale composites. Addition of calcinated aluminum oxide and SG2 increased the storage modulus of neat PLA from about 3173 to 3460 and 3310 MPa at 30 °C, respectively; however, the PLA/3G1 composites exhibited the lowest storage modulus value (3096 MPa) at 30°C. The storage modulus values decreased with increased temperature in the glassy region, yet the reduction seen in the storage modulus was constant below T g. In the vicinity of Tg and in the glassy-rubbery transition region, the differences among PLA/SGl, PLA/SG2, and PLA/CALC composites completely disappeared. 214 4000 800 3000 ‘ ’ . _ 600 2000 ‘ ~ ~ 400 F 1000 * * . .200 0 . , . - 0 20 80 100 2.5 2.0; g 1.5", o . D d t: S 1.0, 0.5: ' H 0.0' g f 1 . 0.0 40 Temperature (°C) Figure 8.9 Temperature dependence of storage and loss modulus (I) and tan delta (II) of PLA (a), PLA/SGI (b), PLA/SG2 (c), and PLA/CALC (d) compoosites. 215 As can be seen in Figure 8.9 (I), the maximum value of the loss modulus for neat PLA, PLA/SGl, PLA/SG2, and PLA/CALC composites was 686 MPa at 59°C, 677 MPa at 61°C , 709 MPa at 60°C, and 733 MPa at 60°C, respectively. These results suggest that PLA/CALC composites have better ability to dissipate the vibrational energy as heat than does the neat PLA. As shown in Figure 8.9 (11), there was no significant change in tan delta peaks for PLA/8G1, PLA/SG2, and PLA/CALC composites; however, a shift of tan delta peaks to higher temperature was observed in comparison with neat PLA, which is evidence of interaction between PLA and Ale fillers. Molecular Weight Distribution The number-average molecular weight (Mn), weight-average molecular weight (MW), and polydispersity index (PDI) of PLA and PLA/Ale composites are presented in Table 8.3. One may expect that the shear mixing and high temperature in the extruder result in a reduction in molecular weight of the PLA matrix. Hence, for comparison purposes, the PLA pellets (unprocessed) were also analyzed by GPC. The Mn, MW, and PDI values of unprocessed PLA were found to be 106,574 3: 11,967, 162,598 :1: 4,437, and 1.53 i 0.13 g/ mol, respectively. Approximately 16% and 9% reductions in Mn and MW values were observed after the extrusion/injection process. A similar reduction in molecular weight due to the melt processing of PLA was reported by several researchers [60,61]. 216 Table 8.3 GPC results of PLA and PLA/Ale composites. Sample Mn, g-niol'l MW, g-nnol'l PDI (Mw/ Mn) PLA * 88,833 :1 2,543 a 148,551 i 691 a 1.66 :1 0.04 a PLA/SCI 83,156 1 1,941 b 137,354 :1: 1,219 b 1.65 9. 0.02 a PLA/SG2 76,170 1 1,344 c 127,475 a: 724 c 1.67 1. 0.02 a PLA/CALC 77,623 1 2,771 c, b 131,101 :1: 1,054 d 1.69 1 0.05 a * PLA processed Note : Values indicate means with standard deviations. Means in the same columns with different superscript letters are significantly different (P < 0.05) PLA/SGl, PLA/SG2, and PLA/CALC composites also exhibited a slight decrease in Mn and MW when compared with the processed PLA. The reduction in Mn and Mw values were 6, 14, 12% and 8, 14, 12%, respectively. Although the changes in Mn and Mw were significant for extruded PLA and PLA/Ale composites, polydispersity indices remained constant. A lower Mn and MW reduction is observed with the PLA/SGl than PLA/SG2, suggesting that the sol-gel coating method may reduce the effect of Ale on increasing PLA degradation as compared with the standard direct mixing method. Sinha Ray et a1. [33], Pluta [60], and Hiroi et a1. [62] have reported that the incorporation of organically modified layered silicates (i.e., synthetic fluorine mica and montrnorillonite, Cloisite 30B) and titanates have resulted in similar reductions in Mn 217 and MW of the PLA matrix. The combined effects of the shearing forces during melt processing and the interaction of PLA with hydroxy groups of fillers were responsible for the decrease in molecular weight. CONCLUSION A stable aluminum oxide colloidal solution was prepared through a sol-gel process using aluminum isopropoxide as a precursor and acetic acid as a peptizer. PLA pellets were coated with the aluminum oxide sol-gel solution using a dip-coating technique. The deposition of aluminum oxide on the PLA pellets was confirmed by FTIR and SEM-EDS. The atomic ratio of AVG obtained was found to be 0.51d:0.02. The aluminum oxide coated PLA pellets, particles obtained fi'om the sol-gel solution, and calcinated aluminum oxide (a-A1203) were used to form PLA/Ale composites through extrusion and injection molding processes. SEM studies of PLA/Ale composites showed that Ale particles were dispersed in the PLA matrix uniformly using melt compounding. it was found from TGA analysis that PLA and PLA/Ale composites were relatively stable up to temperatures of about 350°C. At temperatures above 350°C, the PLA composites with a-A1203 were thermally decomposed more easily than those with SCI and SGZ. In addition, the combined effects of the mixing during melt compounding and the interaction of PLA with aluminum oxides resulted in a decrease in molecular weight of the composites. The incorporation of 3 wt% Ale particles into the 218 PLA, whether a-A1203 or derived from sol-gel, did not alter the T m of PLA. However, the shape and intensity of the melting endotherrns was significantly changed, suggesting bimodal melting behavior. In addition, the storage modulus, loss modulus, and tan delta values of PLA/Ale composites increased with the addition of Ale particles. The results obtained from U‘V-visible transmission spectra also suggested that PLA/Ale composites may act as a better UV barrier than neat PLA. Thus, combining PLA and Ale particles could bring several potential applications. 219 REFERENCES [1] [2] [3] [4] [6] [7] [8] [9] [10] 1111 1121 [13] [14] [15] [16] G. Kickelbick, Prog. Polym. Sci. 2003, 28, 83. M. N. Rahaman, Ceramic Processing, 1St edition, Taylor &Francis, Boca Raton 2007, p. 2. A. G. Erlat, B. M. Henry, C. R. M. Grovenor, A. G. D. Briggs, R. J. Chater, Y. Tsukahara, J. Phys. Chem. B 2004, 108, 883. J. D. Mackenzie, E. Bescher, J. Sol-Gel Sci. T echnol. 2003, 2 7, 7. [5] C. Marano, F. Briatico-Vangosa, M. Marini, F. Pilati, M. Toselli, Euro. Polym. J. 2009, 45, 870. M. Toselli, M. Marini, P. Fabbri, M. Messori, F. Francesco Pilati, J. Sol-Gel Sci. T echnol. 2007, 43, 73. H. Pierson, Handbook of Chemical Vapor Deposition, 1st edition, Noyes Publication, New Jersey 1999. K. Schuegraf, Handbook of Thin-Film Deposition Processes and Techniques, 1St edition, Noyes Publication, New Jersey 1988. M. D. Groner, F. Fabreguette, J. W. Elam, S. M. George, Chem. Mater. 2004, I6, 639. X. Liang, L. F. Hakim, D. Zhan, J. A. McCormick, S. M. George, S. M. Weimer, J. A. Spencer, K. J. Buechler, J. Blackson, C. J. Wood, J. R. Dorgan, J. Am. Ceram. Soc. 2007, 90, 57. C. J. Brinker, G. W. Scherer, Sol-gel Science: The Physics and Chemistry of S01- gel Processing, lst edition, Academic Press, Boston 1990. A. Walcarius, Electroanalysis 2001, I3, 701 . I. M. Thomas, " Multicomponent Glasses from the Sol-gel Process ", in: Sol-gel Technology for Thin Films, Fibers, Preforms, Electronics, and Specialty Shapes, lSt edition, L.C. Klein, Ed., Noyes Publication, New Jersey 1988, p. 2. H. Dislich, "Thin Films from the Sol-gel Process", in: Sol-gel Technology for Thin Films, Fibers, Preforms, Electronics, and Specialty Shapes, 1Sit edition, L.C. Klein, Ed., Noyes Publication, New Jersey 1988, p. 50. M. S. Lee, N. J. Jo, J. Sol-Gel Sci. T echnol. 2002, 24, 175. S. Y. Lee, J. D. Lee, S. M. Yang, J. Mater. Sci. 1999, 34, 1233. 220 [17] [18] [19] [20] [211 [22] [23] [24] [26] [27] [29] 1301 [31] [32] [331 [34] [35] L. Y. L. Wu, E. Chwa, Z. Chen, X. T. Zeng, Thin Solid Films 2008, 516, 1056. E. Dorre, H. Hubner, Alumina: Processing, Properties, and Applications, 1St edition, Springer-Verlag, Berlin 1984, p. 193. X. Liang, S. M. George, A. W. Weimer, N.-H. Li, J. H. Blackson, J. D. Harris, P. Li, Chem. Mater. 2007, I9, 5388. R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4, 835. R. E. Drumright, P. R. Gruber, D. E. Henton, Adv. Mater. 2000, 12, 1841. P. Sangwan, D. Y. Wu, Macromol. Biosci. 2008, 8, 304. B. Braun, J. R. Dorgan, D. M. Knauss, J. Polym. Environ. 2006, 14, 49. M. S. Huda, L. T. Drzal, A. K. Mohanty, M. Misra, Comp. Part B: Eng. 2007, 38, 367 . [25] M. Pluta, M. Murariu, A. Da Silva, A. Galeski, P. Dubois, J. Polym. Sci. Part B: PolymPhys. 2007, 45, 2770. S. Sinha Ray, M. Bousmina, Prog. Mater. Sci. 2005, 50, 962. S. Sinha Ray, M. Okamoto, Prog. Polym. Sci. 2003, 28, 1539. [28] X. Liang, D. M. King, M. D. Groner, J. H. Blackson, J. D. Harris, S. M. George, A. W. Weimer, J. Mem. Sci. 2008, 322, 105. US 4361598 (1982), invs.: B. Yoldas N. Bahlawane, T. Watanabe, J. Am. Ceram. Soc. 2000, 83, 2324. American Society for Testing and Materials. Standard Practice for Conditioning Plastics for Testing. D 618-03. In Annual Book of AST M Standards. AST M: Philadelpia, PA, 2004: 8:01. H.-C. Kao, W.-C. Wei, J. Am. Ceram. Soc. 2000, 83, 362. S. Sinha Ray, K. Yarnada, M. Okamoto, A. Ogami, K. Ueda, Chem. Mater. 2003, I5, 456. R. Auras, B. Harte, S. Selke, J. Appl. Polym. Sci. 2004, 92, 1790. R. Katarnreddy, R. Inman, G. Jursich, A. Soulet, A. Ni‘cholls, C. Takoudis, Thin Solid Films 2007, 515, 6931. 221 [36] [37] [38] [39] [40] [41] [42] [43] 1441 [46] [47] [48] [491 [50] [51] [52] [53] [54] [55] Q. Fu, C.-B. Cao, H.-S. Zhu, Thin Solid Films 1999, 348, 99. S. Sartori, A. Martucci, A. Muffato, M. Guglielmi, J. Euro. Ceram. Soc. 2004, 24, 91 1. ‘ T. Lopez, R. Marmolejo, M. Asomoza, S. Solis, R. Gomez, J. A. Wang, 0. Novaro, J .Navarrete, M. E. Llanos, E. Lopez, Mater. Let. 1997, 32, 325. R. Katamreddy, R. Inman, G. Jursich, A. Soulet, C. Takoudis, J. Electrochem. Soci. 2006, 153, C701. G.-X. Chen, H. Shimizu, Polymer 2008, 49, 943. S. Pilla, S. Gong, E. O'Neill, R. M. Rowell, A. M. Krzysik, Polym. Eng. Sci. 2008, 48, 578. Y. T. Shieh, G. L. Liu, J. Polym. Sci. Part B: Polym. Phys. 2007, 45, 1870. H. Tsuji, Y. Ikada, J. Appl. Polym. Sci. 1997, 63, 855. Y. Wang, S. S. Funari, J. F. Mano, Macromol. Chem. Phys. 2006, 207, 1262. [45] M. Yasuniwa, S. Tsubakihara, Y. Sugimoto, C. Nakafuku, J. Polym. Sci. Part B: Polym. Phys. 2004, 42, 25. M. Yasuniwa, K. Lura, Y. Dan, Polymer 2007, 48, 5398. Y. Wang, J. F. Mano, Eur. Polym. J. 2005, 41, 2335. P. Pan, B. Zhou, W. Kai, T. Dong, Y. Inoue, J. Appl. Polym. Sci. 2008, 107, 54. X. Ling, J. E. Spruiell, J. Polym. Sci. Part B: Polym. Phys. 2006, 44, 3378. X. Ling, J. E. Spruiell, J. Polym. Sci. Part B: Polym. Phys. 2006, 44, 3200. M. L. DiLorenzo, Macromol. Symp. 2006, 234, 176. J. Zhang, K. Tashiro, H. Tsuji, A. J. Domb, Macromolecules 2008, 41, 1352. W. J. Liu, H. L. Yang, Z. Wang, L. S. Dong, J. Appl. Polym. Sci. 2002, 86, 2145. R. C. Streller, R. Thomann, O. Tomo, R. Miilhaupt, Macromol. Mater. Eng. 2008, 293, 218. P. Bhimaraj, H. Yang, R. W. Siege], L. S. Schadler, J. Appl. Polym. Sci. 2007, 222 106, 4233. [56] [57] [58] [60] [61] N. Nakayama, T. Hayashi, Polym. Degrad. Stab. 2007, 92, 1255. D. Wu, L. Wu, L. Wu, M. Zhang, Polym. Degrad. Stab. 2006, 91 , 3149. J. Chang, Y. U. An, G. S. Sur, J. Polym. Sci. Part B: Polym. Phys. 2003, 41, 94. [59] N. Ogata, G. Jimenez, H. Kawai, T. Ogihara, J. Polym. Sci. Part B: Polym. Phys. 1997, 35, 389. M. Pluta, J. Polym. Sci. Part B: Polym. Phys. 2006, 44, 3392. D. Carlson, P. Dubois, R. Narayan, Polym. Eng. Sci. 1998, 38, 311. [62] R. Hiroi, S. S. Ray, M. Okamoto, T. Shiroi, Macromol. Rapid Commun. 2004, 25, 1359. 223 CHAPTER 9 CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK It has been demonstrated that type 4A synthetic zeolites, chabazite natural zeolites, and aluminmn oxide coating layers due to their unique properties have great potential as functional fillers/materials for the bio-based PLA matrix that can be used not only in packaging but also in a plethora of applications. This research provides an understanding about the fabrication method of PLA/zeolite and PLA/aluminum oxide coating layers and their effects on the existing properties of PLA. The feasibility of using synthetic and natural zeolite in bio-based PLA was investigated. It was found that PLA/type 4A synthetic zeolite and PLA/chabazite natural zeolite composites can be successfully fabricated using extrusion followed by injection molding. The morphological studies showed a homogenous dispersion of zeolite type 4A in the PLA matrix. As stress propagated through the composites, zeolite particles remained embedded into the matrix, indicating the existence of good interfacial adhesion between the zeolite particles and the PLA matrix. The fractured surfaces of the PLA/5 wt% type 4A zeolite samples were also more susceptible to crazing and brittle surface deformation than PLA/chabazite composites. Incorporation of both type 4A zeolite and chabazite into PLA had no significant effect on the Tg and Tm of PLA; however, the cold 224 crystallization and percent crystallinity were significantly changed. Both type 4A and chabazite promoted the crystallization of PLA. DMA reveals that PLA/type4A composites have higher storage and loss modulus as compared to PLA/chabazite composites with the same loading level. Thermal degradation studies showed that at temperatures above 300°C, PLA/type 4A synthetic zeolite composites were thermally decomposed more easily than the PLA and PLA/chabazite natural zeolite composites. Inclusion of type 4A markedly reduced the thermal degradation temperature of PLA. The apparent activation energies of thermal degradation of the PLA and PLA/zeolite composites estimated using the Flynn-Wall-Ozawa and Kissinger methods increased in the order of PLA/ type 4A < PLA/chabazite < PLA. As a result, the effects of both type 4A and chabazite zeolites on the thermal degradation behavior of PLA provide a foundation for additional detailed investigation of the PLA/zeolite systems which can facilitate the chemical recycling of PLA. Type 4A zeolite particles drastically accelerate the hydrolytic degradation of the PLA via surface hydrolysis. In addition, ultrathin aluminum oxide coatings were successfully deposited on PLA surfaces using an atomic layer deposition technique at 42°C. The thickness of the coating was found to be 32 run. The AFM results obtained from small scanning areas (10 x 10 m2) revealed that the surface appeared to be as smooth as neat PLA. The newly developed films exhibited high gas barrier properties and good transparency. A stable aluminum oxide colloidal solution was prepared through a sol-gel 225 process using aluminum isopropoxide as a precursor and acetic acid as a peptizer. PLA pellets were coated with the aluminum oxide sol-gel solution using a dip-coating technique. The deposition of aluminum oxide on the PLA pellets was confirmed by FTIR and SEM-EDS. The atomic ratio of Al/O obtained was found to be 0.51:1:0.02. The aluminum oxide coated PLA pellets, particles obtained from the sol-gel solution, and calcinated aluminum oxide (a-A1203) were used to form PLA/Ale composites through extrusion and injection molding processes. SEM studies of PLA/Ale composites showed that Ale particles were dispersed in the PLA matrix uniformly using melt compounding. In conclusion, this research provides knowledge about how PLA properties are affected by synthetic and natural zeolites and aluminum oxide coatings. Clearly, there is considerable room for improvement. Using zeolite or zeolite-like 3 D open frameworks with different pore sizes and surface areas would be beneficial to better understand the relationship between the functional fillers and the PLA matrix. One can firrther improve the properties of functional PLA composites by taking advantage of synergistic effects of using more than one type of functional filler/materials such as a combination of natural and synthetic zeolites or a combination of Ale and SiOx protective barrier layers. 226 1111]11111][11[11111111111111111111